Multi-faceted synthetic biology approaches towards aminoglycoside production in Streptomyces spp.

Abstract

The threat of antimicrobial resistance is regarded as a major health concern. While new therapies are in development, the rejuvenation of existing antibiotics will provide further options for treating antibiotic resistant infections. The aminoglycosides are underutilized as antibiotics, due to ototoxic and nephrotoxic side effects. Reducing these side effects could revitalise the group and yield attractive therapeutics for treatment of resistant infections. This PhD project focused on the heterologous expression of a predicted gentamicin biosynthetic gene cluster (Cluster 24) from a novel Micromonospora sp., identified by the industrial partner Demuris. After determining that the ideal Streptomyces coelicolor host strain was sensitive to gentamicin, we generated a series of ‘superhost’ strains by increasing gentamicin resistance. While we were unable to detect gentamicin production in the first trials, we identified an ion of m/z 502.2246 with proposed chemical formula C18H35N3O13, which could represent an aminoglycoside not previously linked to gentamicin biosynthesis. We further engineered Cluster 24 in the non-resistant host strain using targeted knock-outs to produce gentamicin C1a; however, the yield was too low for characterisation of the output. A cluster refactoring approach in the non-resistant host strains also encountered difficulties as it appeared the rebuilt cluster was lost over time, possibly due to product toxicity. Using the resistant strains, a constitutive promoter cassette was knocked into Cluster 24. A yield increase was visualised via antimicrobial bioassay, and we again identified the compound C18H35N3O13 which was found in previous bioactive samples. Lastly, we identified that an incomplete pathway towards the modified nucleoside queuosine co-clustered within certain aminoglycoside clusters. As preQ1 (the final possible product) is the cognate ligand for three riboswitch classes, it was posited that cluster expression could be under riboswitch control. A putative riboswitch was identified in silico and tested alongside two preQ1 riboswitches from Mycobacterium abscessus subsp. abscessus and Lactobacillus rhamnosus in S. coelicolor. While the Cluster 24 region did not respond to preQ1, the riboswitch from M. abscessus subsp. abscessus repressed fluorescence upon preQ1 induction. Deletion of three predicted queuosine biosynthesis genes from Cluster 24 appeared to increase bioactivity; however, the role of these genes in aminoglycoside biosynthesis remains unclear. The work presented in this thesis targets many aspects of heterologous production, providing a platform towards aminoglycoside production in S. coelicolor.

Date of Award	1 Aug 2022
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Sabine Flitsch (Supervisor), Rainer Breitling (Supervisor) & Eriko Takano (Supervisor)