Ruminococcin-A (RumA) is a peptide antibiotic with post-translational modifications including thioether cross-links formed from non-canonical amino acids, called lanthionines, synthesized by a dedicated lanthionine-generating enzyme RumM. RumA is naturally produced by Ruminococcus gnavus, which is part of the normal bacterial flora in the human gut. High activity of RumA against pathogenic Clostridia has been reported, thus allowing potential exploitation of RumA for clinical applications. However, purifying RumA from R. gnavus is challenging due to low production yields (<1 µg L-1) and difficulties to cultivate the obligately anaerobic organism. We recently reported the reconstruction of the RumA biosynthesis machinery in Escherichia coli where the fully modified and active peptide was expressed as a fusion protein together with GFP. In the current study we developed a scale-up strategy for the biotechnologically relevant heterologous production of RumA, aimed at overproducing the peptide under conditions comparable with those in industrial production settings. To this end, glucose-limited fed-batch cultivation was used. Firstly, parallel cultivations were performed in 24-microwell plates using the enzyme-based automated glucose-delivery cultivation system EnPresso® B to determine optimal conditions for IPTG induction. We combined the bioprocess development with ESI-MS and tandem ESI-MS to monitor modification of the precursor peptide (preRumA) during bioreactor cultivation. Dehydration of threonine and serine residues in the core peptide, catalyzed by RumM, occurs within 1 h after IPTG induction while formation of thioether cross-bridges occur around 2.5 h after induction. Our data also supplies important information on modification kinetics especially with respect to the fluctuations observed in the various dehydrated precursor peptide versions or intermediates produced at different time points during bioreactor cultivation. Overall, protein yields obtained from the bioreactor cultivations were >120 mg L-1 for the chimeric construct and >150 mg L-1 for RumM. The correlation observed between microscale and lab-scale bioreactor cultivations suggests that the process is robust and realistically applicable to industrial-scale conditions.
Ruminococcin A (RumA) is a lanthipeptide with high activity against pathogenic clostridia and is naturally produced by the strict anaerobic bacterium Ruminococcus gnavus E1, isolated from human intestine. Cultivating R. gnavus E1 is challenging, limiting high-quality production, further biotechnological development and therapeutic exploitation of RumA. To supply an alternative production system, the gene encoding RumA-modifying enzyme (RumM) and the gene encoding the unmodified precursor peptide (preRumA) were amplified from the chromosome of R. gnavus E1 and coexpressed in Escherichia coli. Our results show that the ruminococcin-A lanthionine synthetase RumM catalyzed dehydration of threonine and serine residues and subsequently installed thioether bridges into the core structure of a mutant version of preRumA (preRumA∗). These modifications were achieved when the peptide was expressed as a fusion protein together with green fluorescence protein (GFP), demonstrating that a larger attachment to the N-terminus of the leader peptide does not obstruct in vivo processivity of RumM in modifying the core peptide. The leader peptide serves as a docking sequence which the modifying enzyme recognizes and interacts with, enabling its catalytic role. We further investigated RumM catalysis in conjunction with the formation of complexes observed between RumM and the chimeric GFP fusion protein. Results obtained suggested some insights into the catalytic mechanisms of class II lanthipeptide synthetases. Our data further indicated the presence of three thioether bridges, contradicting a previous report whose findings ruled out the possibility of forming a third ring in RumA. Modified preRumA∗ was activated in vitro by removing the leader peptide using trypsin and biological activity was achieved against Bacillus subtilis ATCC 6633. A production yield of 6 mg of pure modified preRumA∗ per liter of E. coli culture was attained and considering the size ratio of the leader-to-core segments of preRumA∗, this amount would generate a final yield of approximately 1-2 mg of active RumA when the leader peptide is removed. The yield of our system exceeds that attainable in the natural producer by several 1000-fold. The system developed herein supplies useful tools for product optimization and for performing in vivo peptide engineering to generate new analogs with superior anti-infective properties.
