Abstract
Heterologous expression of natural product gene clusters in tractable hosts offers great promise for achieving sustainable production of nature-inspired drugs. However, it is common for the roles of some genes in a cluster to be unclear, and this can make it difficult to identify the minimal gene set required to produce the desired molecule. Typically, the function of unknown genes is inferred by time-consuming reductionist techniques, e.g. single and multiple gene knockouts in the native producer, followed by phenotype analysis. Here, we instead present a rapid combinatorial method to assemble individual genes involved in glidobactin biosynthesis into a collection of partial or complete clusters in a heterologous host. Following up with mass spectrometry allowed identification of the minimum genes required for compound production. We applied this synthetic biology approach to characterise the glidobactin gene cluster, for which previous gene knockout studies had yielded conflicting results. In the process, we showed that an added intrinsic advantage of combinatorial assembly is the generation of multiple strains that produce potentially desirable analogues in addition to glidobactin.