An MIQP framework for metabolic pathways optimisation and dynamic flux analysis

Lucas Gerken-Starepravo,Xianfeng Zhu,Bovinille Anye Cho,Fernando Vega-Ramon,Oliver Pennington,Ehecatl Antonio Del Río-Chanona,Keju Jing,Dongda Zhang

doi:10.1016/j.dche.2022.100011

Lucas Gerken-Starepravo, Xianfeng Zhu + Show 6 more

Open Access

https://doi.org/10.1016/j.dche.2022.100011

Copy DOI

Abstract

Dynamic flux analysis methods have been widely used for deciphering complex metabolic fluxes transients. However, many of them require frequent experimental measurements and are ineffective in dealing with under-determined metabolic reaction networks. In this study, we addressed these challenges by (i) integrating a macroscale kinetic model with its dynamic metabolic flux model to enable flux simulation over the entire time course for batch operation, and (ii) constructing a single-level mixed-integer quadratic program (MIQP) to automatically identify the shortest metabolic pathways from substrate inflow to biosynthesis of biomass and desired bioproducts. To demonstrate the advantages of the proposed framework, a X. dendrorhous fermentation process for astaxanthin production was utilised as the case study. It is found that the current framework is able to efficiently identify essential pathways and reduce the size of the original metabolic network by 70% with negligible computational cost. Furthermore, the modelling consistency, robustness, and limitation of this framework were thoroughly investigated. This research provides a new avenue for efficient in-silico design, analysis, and gene knockout of microbial strains for bioproduct synthesis.

Highlights

Over the last few decades, there has been an unprecedented increase in the demand for bio-based pharmaceuticals, cosmetics, foods, aquacultures, and chemically derived products
Many of them require frequent experimental measurements and are ineffective in dealing with underdetermined metabolic reaction networks. We addressed these challenges by (i) integrating a macroscale kinetic model with its dynamic metabolic flux model to enable flux simulation over the entire time course for batch operation, and (ii) constructing a single-level mixed-integer quadratic program (MIQP) to automatically identify the shortest metabolic pathways from substrate inflow to biosynthesis of biomass and desired bioproducts
We have demonstrated a novel metabolic network simulation framework without the need of manual curation and in the absence of costly experimental data measurements by exploiting the merits of carbon constraint FBA (ccFBA) and Metabolic Flux Analysis (MFA) in a dynamic framework

Summary

Introduction

Over the last few decades, there has been an unprecedented increase in the demand for bio-based pharmaceuticals, cosmetics, foods, aquacultures, and chemically derived products. A major limitation to deal with the rising demand of biobased products is due to the fact that the yield of target components in wild-type microbial strains are generally far below the theoretical maximum (Miao et al, 2011). It is essential to develop microbial mutant-type strains with biological enhancement in order to deliver increased yields of substrate-to-product conversion. This is a major area of investigation within the fields of systems biology, industrial biotechnology and metabolic engineering (Leighty and Antoniewicz, 2011; Maia et al, 2016). Flux Balance Analysis (FBA) and Metabolic Flux Analysis (MFA) are two main static approaches used to estimate intracellular metabolic activity and product synthesis when compared with experimental data. Intracellular uptake or production rates of measurable states within each time interval are calculated as the change of their concentration over the time interval divided by average biomass concentration within the time interval

Objectives

Methods

Results

Conclusion