Engineering a Carotenoid-Overproducing Strain of Azospirillum brasilense for Heterologous Production of Geraniol and Amorphadiene

Escherichia coil and Saccharomyces cerevisiae have been used extensively for heterologous production of a variety of secondary metabolites. Neither has an endogenous high-flux isoprenoid pathway, required for the production of terpenoids. Azospirillum brasilense, a nonphotosynthetic GRAS (generally recognized as safe) bacterium, produces carotenoids in the presence of light. The carotenoid production increases multifold upon inactivating a gene encoding an anti-sigma factor (ChrR1). We used this A. brasilense mutant (Car-1) as a host for the heterologous production of two high-value phytochemicals, geraniol and amorphadiene. Cloned genes (crtE1 and crtE2) of A. brasilense encoding native geranylgeranyl pyrophosphate synthases (GGPPS), when overexpressed and purified, did not produce geranyl pyrophosphate (GPP) in vitro. Therefore, we cloned codon-optimized copies of the Catharanthus roseus genes encoding GPP synthase (GPPS) and geraniol synthase (GES) to show the endogenous intermediates of the carotenoid biosynthetic pathway in the Car-1 strain were utilized for the heterologous production of geraniol in A. brasilense. Similarly, cloning and expression of a codon-optimized copy of the amorphadiene synthase (ads) gene from Artemisia annua also led to the heterologous production of amorphadiene in Car-1. Geraniol or amorphadiene content was estimated using gas chromatography-mass spectrometry (GC-MS) and GC. These results demonstrate that Car-1 is a promising host for metabolic engineering, as the naturally available endogenous pool of the intermediates of the carotenoid biosynthetic pathway of A. brasilense can be effectively utilized for the heterologous production of high-value phytochemicals. IMPORTANCE To date, the major host organisms used for the heterologous production of terpenoids, i.e., E. coli and S. cerevisiae, do not have high-flux isoprenoid pathways and involve tedious metabolic engineering to increase the precursor pool. Since carotenoid-producing bacteria carry endogenous high-flux isoprenoid pathways, we used a carotenoid-producing mutant of A. brasilense as a host to show its suitability for the heterologous production of geraniol and amorphadiene as a proofof-concept. The advantages of using A. brasilense as a model system include (i) dispensability of carotenoids and (ii) the possibility of overproducing carotenoids through a single mutation to exploit high carbon flux for terpenoid production.