(188ax) An Integrative Approach of Metabolic Network and Bioprocess Modeling in the Strain Design for Succinic Acid Production

An Integrative Approach of Metabolic Network and Bioprocess Modeling in the Strain Design for Succinic Acid Production

Tafur Rangel, A. T^1,2; Gómez Ramírez, J. M¹; González Barrios, A. F¹.

¹ Grupo de Diseño de Productos y Procesos (GDPP), Departamento de Ingeniería Química, Universidad de los Andes, Bogotá, Colombia.

² Grupo de Investigación CINBIOS.Departamentode Microbiología, Universidad Popular del Cesar, Valledupar-Cesar, Colombia.

A large number of chemical products, including succinic acid (SA), polymers, and others, are produced from fossil energy[1], generating environmental concern. The use of the microbial cell as an important environmental friendly factory for renewable feedstocks bioconversion in compounds such as SA has been increasing in the last years. SA is a key chemical platform for the economy of bio-based products, and helpful to reduce environmental impact compared with petroleum-based products. It could be obtained from glycerol as carbon source, one abundant, inexpensive and renewable feedstock formed in great quantities as a by-product of biodiesel production[2][3]. Nowadays, the worldwide production of biodiesel continues increasing, becoming glycerol in a burden for the biofuel industry [4] [5]. This situation required of developing technologies to produce bio-based SA from glycerol with low cost and environmental friendly over petrochemical synthesis.

Bio-based SA production has been targeting the optimization by deletion or overexpression of genes implicated in the bioconversion of glycerol to SA[6]–[9] as well as the recovery and purification process also called downstream processing[10]–[14].

In the last two decades, constraint based modeling of microorganism has been developed as a powerful optimization approach to establish strategies for predicting gene deletion or overexpression in order to improve the yield of SA from several sources and to describe the cellular behavior after metabolic manipulation[15]. It entails the re-engineering of host chassis bacteria, such as Escherichia coli (E. coli) in order to convert low-cost carbon feedstocks, such as glycerol, into SA. In last years, some COBRA (Based on Constraint-Based Reconstruction and Analysis) methods has been developed. However, the ability of metabolic networks modeling to predict metabolic engineering targets depends of the number and confidence of reactions, metabolites and genomic information available for the Genome-Scale Metabolic reconstructions.

The other important element of optimization to increase the SA production is the downstream processing, which represents the highest operational costs. The main reason for high cost in the recovery and purification process is that SA shares physicochemical properties with other carboxylic acids such as lactic acid, citric acid, acetic acid, oxalic acid, and formic acid. Moreover, these carboxylic acids are synthesized in the tricarboxylic acid cycle (TCA) [16], difficulting the recovery and purification process and affecting the purity of SA. Then, the development of a competitive recovery process is a critical step in the production of SA, to obtain a simple process, high purity, and low cost of energy based on physical, chemical and thermodynamic behavior. As the final purification accounts for a significant cost factor, the production of these other by-products (tricarboxylic acids) in the fermentation could be reduced to a minimum by metabolic engineering with the consequential lower processing cost. Nevertheless, both metabolic engineering and downstream processing are usually evaluated separately resulting in no cost-competitive bio-based SA production over petrochemical synthesis.

In this research, a comparative analysis of ECC2 and one reconstruction for E. coli ATCC 8739 to produce SA from glycerol by the prediction of metabolic engineering targets and the effects of the metabolic engineering an Escherichia coli strain in a SA fermentation process on the downstream process in terms of operational cost and energy consumption was carried out. First, potential in silico knockouts were determined using OptKnock in CobraToolbox [44]. Three scenarios were selected for ECC2 and three for E. coli ATCC 8739 model considering the growth rate at least 66%, 20%, and 10% of the Wild-Type and succinate productivity. Secondly, the concentration profiles were predicted departing from the previously mentioned deletions as constraints with Dynamic Flux Balance Analysis based on a Static Optimization Approach. Finally, a SA purification process was simulated, and operational cost and energy consumption were obtained by Aspen Plus.

A maximum of twenty reactions and 20 gene knockouts were considered for each prediction. Results suggest fumarase (FUM) is an important key enzyme to be interrupted to increase the succinate productivity. Although some differences in the reactions predicted were found, the best succinic productivity was reached when 10-12 reactions were blocked, but the growth rate decreased at least 10%. Therefore, it could affect an industrial scale process. These results provide insight of effects after gene knockouts of E. coli to increase the succinate productivity from glycerol, demonstrating core model not necessarily show the best result, but it can even be used as a basis for the identification of metabolic engineering strategies. Lower growth rates in the strain two and three were obtained compared with strain one. The observed total growth time was more significant for the strain 2 and the strain 3 being a potential negative factor for the productivity of the process. However, a significant difference was observed on the SA concentration obtained for the strain 1 compared with the strain 2 and 3. Our results show that the succinic acid productivity constitutes a central parameter when selecting the appropriate gene targets for deletion nonetheless the presence of organic acids in the downstream process and biomass growth rate. The lowest capital cost and period of recovery was obtained for the strain 2 which provided the highest SA production per year clearly demonstrating that when engineering a strain the most important factor to consider is the strain´s productivity regardless of the presence of different by-products such as lactate, ethanol, or acetate that negatively affect the downstream operational cost as the sulfuric acid stream and heater temperature increases.

Here, we proposed a metabolism-downstream coupled model that shows that the bioproduct productivity as the main element to contemplate when bearing in mind the operational cost and the energy consumption in the engineering of strains for industrial-scale production.

Bibliography.

[1] K. H. Zhuang and M. J. Herrgård, “Multi-scale exploration of the technical, economic, and environmental dimensions of bio-based chemical production.,” Metab. Eng., vol. 31, pp. 1–12, Sep. 2015.

[2] M. D. Blankschien, J. M. Clomburg, and R. Gonzalez, “Metabolic engineering of Escherichia coli for the production of succinate from glycerol,” Metab. Eng., vol. 12, no. 5, pp. 409–419, 2010.

[3] G. P. da Silva, M. Mack, and J. Contiero, “Glycerol: A promising and abundant carbon source for industrial microbiology,” Biotechnol. Adv., vol. 27, no. 1, pp. 30–39, 2009.

[4] M. Ayoub and A. Z. Abdullah, “Critical review on the current scenario and significance of crude glycerol resulting from biodiesel industry towards more sustainable renewable energy industry,” Renew. Sustain. Energy Rev., vol. 16, no. 5, pp. 2671–2686, Jun. 2012.

[5] R. Ciriminna, C. Della Pina, M. Rossi, and M. Pagliaro, “Understanding the glycerol market,” Eur. J. Lipid Sci. Technol., vol. 116, no. 10, pp. 1432–1439, 2014.

[6] J. Costanza, G. Carapezza, C. Angione, P. Lió, and G. Nicosia, “Robust design of microbial strains,” Bioinformatics, vol. 28, no. 23, pp. 3097–3104, 2012.

[7] I. Rocha et al., “OptFlux: an open-source software platform for in silico metabolic engineering,” BMC Syst. Biol., vol. 4, no. 1, p. 45, 2010.

[8] P. Pharkya, A. P. Burgard, and C. D. Maranas, “OptStrain: a computational framework for redesign of microbial production systems.,” Genome Res., vol. 14, no. 11, pp. 2367–76, Nov. 2004.

[9] P. Pharkya and C. D. Maranas, “An optimization framework for identifying reaction activation/inhibition or elimination candidates for overproduction in microbial systems,” Metab. Eng., vol. 8, no. 1, pp. 1–13, 2006.

[10] L. Caspeta and J. Nielsen, “Economic and environmental impacts of microbial biodiesel,” Nat. Biotechnol., vol. 31, no. 9, pp. 789–793, Sep. 2013.

[11] J. Kim, “DEVELOPMENT AND APPLICATIONS OF INTEGRATED METABOLIC AND TRANSCRIPTIONAL REGULATORY NETWORK MODELS,” UNIVERSITY OF WISCONSIN–MADISON, 2012.

[12] A. C. Kokossis, M. Tsakalova, and K. Pyrgakis, “Design of integrated biorefineries,” Comput. Chem. Eng., vol. 81, pp. 40–56, 2015.

[13] J. A. Posada, L. E. Rincón, and C. A. Cardona, “Design and analysis of biorefineries based on raw glycerol: Addressing the glycerol problem,” Bioresour. Technol., vol. 111, pp. 282–293, 2012.

[14] M. Kuglarz, M. Alvarado-Morales, D. Karakashev, and I. Angelidaki, “Integrated production of cellulosic bioethanol and succinic acid from industrial hemp in a biorefinery concept,” Bioresour. Technol., vol. 200, pp. 639–647, 2016.

[15] M. Binns, A. Vlysidis, C. Webb, C. Theodoropoulos, P. de Atauri, and M. Cascante, Glycerol metabolic conversion to succinic acid using Actinobacillus succinogenes, vol. 29. Elsevier B.V., 2011.

[16] Y. S. Huh, Y. S. Jun, Y. K. Hong, H. Song, S. Y. Lee, and W. H. Hong, “Effective purification of succinic acid from fermentation broth produced by Mannheimia succiniciproducens,” Process Biochem., vol. 41, pp. 1461–1465, 2006.