(190d) Analysis and Simulation of Multi-Species Dynamic Flux Balance Analysis Models With Obligatory Interactions

Most microorganisms in natural environments do not live in isolation, but exist as part of complex, dynamically changing, microbial consortia. Hence, synthesis of artificial biological process systems based on microbial consortia is a very promising approach. In addition, monocultures optimized for high metabolic production in a sterile laboratory environment are often not economical at production scale. In contrast, naturally occurring multispecies cultures within existing ecologies show high resilience through symbiotic coexistence. Nevertheless, it remains a great challenge to realize such multispecies cultures in industrial applications. To address this challenge, multi-scale models, which integrate the metabolic information available from high-throughput experiments with the ecological scale of the species interaction and the process scale, have to be developed.

 The goal of this study is to introduce a new formulation for dynamic flux balance models that explicitly accounts for obligatory interspecies interaction.

As a structured model for processes involving microorganisms, dynamic flux balance analysis (DFBA) combines the flux balance analysis models (Palsson, 2006) used in metabolic engineering with the dynamic mass balances of the reactor. Multispecies dynamic flux balance models have been previously studied in (Zhuang et al., 2011; Hanly and Henson, 2011). The model description assumes accumulation of the external metabolites determined by the production and uptake rates of the consortia members and does not explicitly consider dependency between the species.

The interaction between microorganisms can be direct communication, using signaling molecules or cell-cell contact, or indirect, through unidirectional or bidirectional exchange of external metabolites. The possible indirect interactions between each pair of organism are mutualism (each organism provides essential nutrients for the other organism), commensalism (only one organism depends on the metabolic product of the other), or neutralism (no direct dependency between the two organisms, but potential competition for the same resource).

In the context of DFBA, obligatory interactions, such as synthropy, can be realized by assuming that the metabolites that are being exchanged do not accumulate in the environment. This enforces additional constraints on the overall process model. The formulation can be interpreted from a game-theoretical perspective in which each player is represented by a microorganism with cost function given by the FBA model and the overall goal is to satisfy all constraints on the external metabolite concentrations given by their interconnection. A similar formulation for steady-state FBA models has been presented by Zomorrodi and Maranas (2012).

Numerical implementation is based on our previous work on simulation of dynamic flux balance models (Hoffner et al., 2013). Enforcing microbial interactions in dynamic models amounts to formulating a differential-algebraic system, in which the interaction constraints on the exchange fluxes are represented as algebraic equations. The linear programs representing the flux balance models of the members of the consortia are interconnected through the algebraic constraints and have to be solved simultaneously.

As a case study, consider a dynamic simulation of a thermophilic phototrophic mat system in Yellowstone National Park (Taffs et al., 2009). The model consist of oxygenic phototrophs, filamentous anoxygenic phototrophs, and sulfate-reducing bacteria represented by three guilds, which each represent a group of species that exploit the same class of environmental resources in a similar manner. There are several synthrophic interconnections between different members of the consortia, which are also changing from day time to night time. A 24h simulation of the mat system is implemented by sequentially simulating the two independent models accounting for diel (day-night)

References

T. J. Hanly and M. A. Henson. Dynamic flux balance modeling of microbial co-cultures for efficient batch fermentation of glucose and xylose mixtures. Biotechnology and Bioengineering, 108(2):376-385, 2011.

K. Hoffner, S. M. Harwood, and P. I. Barton. A reliable simulator for dynamic flux balance analysis. Biotechnology and Bioengineering, 110(3):792-802, 2013.

B. O. Palsson. Systems Biology: Properties of Reconstructed Networks. Cambridge University Press, New York, NY, 2006.

R. Taffs, J. Aston, K. Brileya, Z. Jay, C. Klatt, S. McGlynn, N. Mallette, S. Montross, R. Gerlach, W. Inskeep, D. Ward, and R. Carlson. In silico approaches to study mass and energy flows in microbial consortia: a syntrophic case study. BMC Systems Biology, 3(1):114, 2009.

K. Zhuang, M. Izallalen, P. Mouser, H. Richter, C. Risso, R. Mahadevan, and D.R. Lovley. Genome-scale dynamic modeling of the competition between rhodoferax and geobacter in anoxic subsurface environments. The ISME Journal, 5:305316, 2011.

A.R. Zomorrodi and C.D. Maranas. Optcom: A multi-level optimization framework for the metabolic modeling and analysis of microbial communities. PLoS Computational Biology, 8(2):e1002363, 2012.