BackgroundNatural aromatic polymers, mainly melanins, have potential and current applications in the cosmetic, pharmaceutical and chemical industries. The biotechnological production of this class of compounds is based on tyrosinase-dependent conversion of L-tyrosine and other aromatic substrates into melanins. The purpose of this work was to apply metabolic engineering for generating Escherichia coli strains with the capacity to synthesize an aromatic polymer from a simple carbon source.ResultsThe strategy was based on the expression in E. coli of the MutmelA gene from Rhizobium etli, encoding an improved mutant tyrosinase. To direct the carbon flow from central metabolism into the common aromatic and the L-tyrosine biosynthetic pathways, feedback inhibition resistant versions of key enzymes were expressed in strains lacking the sugar phosphotransferase system and TyrR repressor. The expressed tyrosinase consumed intracellular L-tyrosine, thus causing growth impairment in the engineered strains. To avoid this issue, a two phase production process was devised, where tyrosinase activity was controlled by the delayed addition of the cofactor Cu. Following this procedure, 3.22 g/L of melanin were produced in 120 h with glucose as carbon source. Analysis of produced melanin by Fourier transform infrared spectroscopy revealed similar characteristics to a pure eumelanin standard.ConclusionsThis is the first report of a process for producing melanin from a simple carbon source at grams level, having the potential for reducing production cost when compared to technologies employing L-tyrosine as raw material.

The expression of the feedback inhibition-insensitive enzyme cyclohexadienyl dehydrogenase (TyrC) from Zymomonas mobilis and the chorismate mutase domain from native chorismate mutase-prephenate dehydratase (PheA CM ) from Escherichia coli was compared to the expression of native feedback inhibition-sensitive chorismate mutase-prephenate dehydrogenase (CM-TyrA p ) with regard to the capacity to produce L-tyrosine in E. coli strains modified to increase the carbon flow to chorismate. Shake flask experiments showed that TyrC increased the yield of L-tyrosine from glucose (Y L-Tyr/Glc ) by 6.8-fold compared to the yield obtained with CM-TyrA p . In bioreactor experiments, a strain expressing both TyrC and PheA CM produced 3 g/liter of L-tyrosine with a Y L-Tyr/Glc of 66 mg/g. These values are 46 and 48% higher than the values for a strain expressing only TyrC. The results show that the feedback inhibition-insensitive enzymes can be employed for strain development as part of a metabolic engineering strategy for L-tyrosine production.

No abstract