BACKGROUND:
There are an increasing number of studies regarding genetic manipulation of cyanobacteria to produce commercially interesting compounds. The majority of these works study the expression and optimization of a selected heterologous pathway, largely ignoring the wholeness and complexity of cellular metabolism. Regulation and response mechanisms are largely unknown, and even the metabolic pathways themselves are not fully elucidated. This poses a clear limitation in exploiting the rich biosynthetic potential of cyanobacteria.
RESULTS:
In this work, we focused on the production of two different compounds, the cyanogenic glucoside dhurrin and the diterpenoid 13R-manoyl oxide in Synechocystis PCC 6803. We ... More
BACKGROUND:
There are an increasing number of studies regarding genetic manipulation of cyanobacteria to produce commercially interesting compounds. The majority of these works study the expression and optimization of a selected heterologous pathway, largely ignoring the wholeness and complexity of cellular metabolism. Regulation and response mechanisms are largely unknown, and even the metabolic pathways themselves are not fully elucidated. This poses a clear limitation in exploiting the rich biosynthetic potential of cyanobacteria.
RESULTS:
In this work, we focused on the production of two different compounds, the cyanogenic glucoside dhurrin and the diterpenoid 13R-manoyl oxide in Synechocystis PCC 6803. We used genome-scale metabolic modelling to study fluxes in individual reactions and pathways, and we determined the concentrations of key metabolites, such as amino acids, carotenoids, and chlorophylls. This allowed us to identify metabolic crosstalk between the native and the introduced metabolic pathways. Most results and simulations highlight the metabolic robustness of cyanobacteria, suggesting that the host organism tends to keep metabolic fluxes and metabolite concentrations steady, counteracting the effects of the heterologous pathway. However, the amino acid concentrations of the dhurrin-producing strain show an unexpected profile, where the perturbation levels were high in seemingly unrelated metabolites.
CONCLUSIONS:
There is a wealth of information that can be derived by combining targeted metabolite identification and computer modelling as a frame of understanding. Here we present an example of how strain engineering approaches can be coupled to 'traditional' metabolic engineering with systems biology, resulting in novel and more efficient manipulation strategies.
