Efficient production of indigo in heterologous model systems

Indigo is a naturally occurring pigment that has been used to give blue colouration to clothing, food, paints, and pottery for over 8,000 years. The dyeing of blue denim to make jeans is the most significant use case for the pigment today. Historically, indigo was derived from a number of plants that accumulate glucosylated indigo precursor compounds, however, chemical synthesis of indigo became commercially viable in the early 20th century, displacing plant-derived indigo. In industrial denim production, between three and five litres of blue effluent is produced per kilogram of denim made, adding up to billions of litres of wastewater each year globally. Production of indigo pigment, in planta, in the cotton fibres used to make denim may be a more sustainable alternative.    In this project, a novel two-enzyme indigo production system was developed in the model bacterium Escherichia coli. The E. coli accumulated a maximum of 409.5 µg/L indigo in a 24 h period, when grown in a minimal media environment, with glycerol as a carbon source. Development of this system required the considered selection of a pair of enzymes; an indole synthase (INS) to convert indole-3-glycerol phosphate (I-3GP), produced as part of tryptophan biosynthesis in plants, bacteria and most other organisms, to colourless free indole and an indole oxygenase (INO) to oxidise the free indole to indoxyl, which spontaneously dimerises to form blue indigo. Plant INS enzymes from Arabidopsis thaliana, Zea mays and the indigo plants Isatis tinctoria and Persicaria tinctoria were compared through complementation of three tryptophan auxotrophic E. coli strains. The A. thaliana and Z. mays INS enzymes proved best at complementing all E. coli strains tested, indicating that they produced the most free indole, therefore, they were selected as the candidate INS for the two-enzyme system. P. tinctoria INS failed to complement the E. coli mutant. Five bacterial flavin-dependent monooxygenase-type INO enzymes were expressed in E. coli and the enzymes were compared for their ability to produce indigo when the E. coli were grown on media supplemented with indole. The E. coli expressing Methylophaga aminisulfidivorans and Rhodococcus sp. T104 INO accumulated the most indigo when indole was supplemented at a low level, therefore, these two INO enzymes were chosen as candidates for the two-enzyme system. The selected INS and INO enzyme genes were placed into synthetic operon expression vectors, which contained regulatory elements based on the native proBA and lac operons from E. coli. The E. coli co-expressing Z. mays INS and Rhodococcus sp. T104 INO accumulated a small but visible quantity of indigo when grown in a minimal medium environment, while other INS/INO combinations did not. More indigo was accumulated when additional indole was supplemented into the media or when the native E. coli tryptophanase was converting tryptophan to indole, indicating that production of free indole by INS, due to either the inefficiency of the INS enzyme itself or the limited metabolite coming through the preceding tryptophan biosynthetic pathway, was limiting indigo production. Therefore, a further investigation of INS was undertaken. Six prototype INS enzymes were designed, each with changes made to loops 2 and 6 and protein-protein interaction regions which are hypothesised to be crucial in free indole production by INS. Three of the six Prototype INS enzymes had an enhanced ability to complement tryptophan auxotrophic E. coli, compared to the native INS enzymes from which they were derived, indicating that they may have an enhanced ability to produce free indole. However, expressing the Prototype INS enzymes alongside Rhodococcus sp. T104 INO in E. coli did not result in the E. coli accumulating more indigo pigment than E. coli expressing native Z. mays INS and Rhodococcus sp. T104 INO. Future work should aim to improve the productivity of the two-enzyme indigo production system, before transferring it to the target cotton plant. Several studies have enhanced the tryptophan biosynthetic pathway in E. coli for the purpose of producing the amino acid tryptophan or related products, such as violacein. Most improvements have been realised by overexpression of feedback resistant anthranilate synthase and 3-Deoxy-D-arabino-heptulosonate 7-phosphate (DAHP) synthase or elimination of the tryptophan repressor or tyrosine repressor. Such work has been widely and reliably repeated. Transfer of the two-enzyme system to such an engineered strain is the logical next step, before transferring the system to a model plant, such as a A. thaliana or Nicotiana tabacum and, finally, on to cotton.