A Constraint-Based Mathematical Model for The Conversion of CO2 to Hyaluronic Acid in Cynobacterium Synechococcus

Abstract

Hyaluronic acid (HA) is a biopolymer of glucuronic acid and N-acetyl-glucosamine connected by β-1,3 and β-1,4 glycosidic bonds. Hyaluronic acid has pharmaceutical, medical and cosmetic applications. Conventionally, HA produced from heterotrophic bacteria, such as Streptococcus zooepidemicus, which requires costly substrates, generates endotoxins, and shows batch-to-batch variation. To target these limitations, explorations of phototrophic pathways using organisms like Synechococcus sp. PCC 7002 utilised CO2 and harvesting photons as a source of energy to generate HA in a carbon-neutral and sustainable way. Here, we explore the production of HA under phototrophic conditions through a reduced-scale metabolic model validated and containing 212 reactions, 141 metabolites, and 238 genes, reconstructed for a microalgal system. The model was thoroughly validated through stoichiometric consistency checks, reaction bound evaluations, biomass and HA optimisation tests, phototrophic boundary condition checks, and leak analyses to provide a solid computational platform for flux-based simulations. Robustness analysis highlighted a exactly linear relationship between nitrate uptake and growth, with growth rates declining from 0.76 h⁻¹ to 0 h⁻¹ (−0.8 to 0 mmol·gDW⁻¹·h⁻¹) under broad sweeps and 0.24 h⁻¹ to 0 h⁻¹ (−0.24 to 0 mmol·gDW⁻¹·h⁻¹) under narrow sweeps, confirming nitrate as the primary growth-limiting nutrient in the model. Conversely, phosphate limitation was thresholdtype and became non-limiting when the uptake flux exceeded some critical value. For the biosynthesis of HA, CO₂ and photon uptake were found to be the primary controlling parameters, and proportional decreases in HA flux were seen with strengthening their lower bounds. Nitrate limitation had little effect on HA flux, establishing a definite sensitivity hierarchy of CO₂ > Photon ≫ NO₃⁻. Flux Variability Analysis at 100 % of optimal HA flux showed minimal variability for photon, CO₂, nitrate, and HA_export reactions, suggesting an exclusive optimal flux distribution. On the other hand, phosphate and water exchanges were more flexible. Single-limiter sweeps also supported the preeminent role of carbon and energy supply, displaying linear decreases in HA flux under photon and CO₂ limitation. Dynamic FBA simulation showed linear HA accumulation up to 0.71 mmol·L⁻¹ within 55 h, driven by steady CO₂ uptake (10 to 0 mmol·L⁻¹), followed by a plateau phase as CO₂ was fully depleted. In comparison, nitrate (3.0 to 2.29 mmol·L⁻¹) and phosphate (~2 mmol·L⁻¹) remained nonlimiting.