Coronatine is a phytotoxin produced by Pseudomonas syringae spp. that infect a number of commercially important crops. Coronatine is composed of a cyclopropyl amino acid, coronamic acid, and a novel polyketide, coronafacic acid (CFA). This work discloses (1) chemical and (2) biological approaches towards understanding CFA biosynthesis.

1. It was previously known that CFA is assembled from a pyruvate, a butyrate, and three acetate units, but the mode of pyruvate incorporation was obscure. Precursor incorporation experiments with labeled forms of glutamate have now shown that pyruvate is incorporated only after its conversion to one of the starter units, alpha-ketoglutarate, succinic semialdehyde, or glutamate. It was hypothesized that polyketide chain extension of one of these starter units with an acetate unit followed by first ring cyclization would give 2-cyclopenten-1-one 2-carboxylic acid (CPC). This could be extended with a butyrate unit to give 2-[1-oxo-2-cyclopenten-2-ylmethyl]butanoic acid (CPE), a compound previously isolated from P. syringae culture broths. Finally, chain extension with an acetate unit and closure of the second ring would give CFA. This model of CFA biosynthesis (starter unit → CPC → CPE → CFA) has been tested by precursor incorporation experiments involving labeled forms of succinic semialdehyde and CPC. In addition, several synthetic routes to CPE have also been explored.
2. The nucleotide sequence of the remaining unsequenced region of the CFA biosynthetic gene cluster was completed in this work. Analysis revealed the presence of two large ORFs, cfa6 and cfa7, whose gene products show homologies to type I polyketide synthases (PKSs). Previous work had demonstrated the presence of a type II PKS in the CFA cluster, including two adenylating enzymes (Cfl and Cfa5) and a thioesterase (Cfa9). Consideration of all the putative enzymatic activities of the CFA cluster (Cfl and Cfa1--Cfa9) has allowed a refinement of the earlier biosynthetic model, in which a good correlation exists between the enzymatic activities present and the putative biochemical steps of the earlier model (starter unit → CPC → CPE → CFA). Further characterization of the CFA cluster has begun with the soluble expression and partial purification of the Cfl and Cfa5 proteins as maltose binding protein fusions.