Mémoire

Abstract

Understanding how microorganisms adapt their specialized metabolism in response to environmental signals is a major theme in modern microbiology. This knowledge is crucial to discover new molecules associated with the plethora of crypPc "Biosynthetic Gene Clusters" (BGCs), many of them being transcriptionally silent or low expressed under laboratory conditions. Our strategy to activate the expression of BGCs is "to learn from the biology of the producing microorganism itself ” by identifying the genetic information responsible for the expression of these BGCs and connecting it to environmental signals. For this purpose, our lab is developing the COMMBAT (COndiPons for Microbial Metabolite Biosynthesis Activated TranscripPon) bioinformatic tool which aims at predicting the environmental signals and their associated transcription factors (TF) that control BGC expression. For optimal utilization of this tool, we must in parallel fill the knowledge gap on the signal perceived and transported by specialized metabolite-producing bacteria. The first objective of my Master thesis was to use and test the COMMBAT tool to highlight novel signaling pathways from environmental elicitors to specialized metabolite production. Using the TF/elicitor couple DasR/N-acetylglucosamine (GlcNAc) as test example, we identified a series of BGCs that could possibly have their expression induced by the uptake of GlcNAc. We started to experimentally validate these predictions, by growing the selected strains on media under ON/OFF culture conditions (with and without the inducer: GlcNAc) and visualizing the repercussion on the production level of the metabolites by MALDI Imaging Mass Spectrometry (IMS). The GlcNAc-dependent production of cephamycin C by S. clavuligerus is the most significant new pathway discovered in this work. The second objective of my Master thesis was to contribute to the identification of the carbon source(s) transported by specialized metabolite-producing bacteria. The strategy to identify the carbon source associated with a predicted sugar transporter combined i) in silico analysis of genes of the transcription unit (TU) containing the genes for sugar transport, ii) finding the allosteric effector modulating the DNA-binding activity of the purified TF controlling the TU, and iii) a proteomic analysis of the expression of proteins of the TU in presence of the predicted carbon source. Our approach allowed to identify the genes/proteins required for ribose utilization in Streptomyces species while the identification of genes/proteins for rhamnose utilization still requires further investigation. Overall, the COMMBAT tool seems promising at unveiling novel pathways for the activation of specialized metabolites. The strategy for discovering the signal molecules transported by metabolite-producing bacteria as partially convincing, depending on the level of complexity of the TU investigated.