Investigating and engineering the post-translational modifications in RiPP natural product biosynthesis (I) Thiopeptides (II) Sactipeptides

RiPPs;natural products;ribosomally-synthesized and post-translationally modified;thiopeptides;sactipeptides;enzymology

Hudson, Graham Arthur

关键词: RiPPs;    natural products;    ribosomally-synthesized and post-translationally modified;    thiopeptides;    sactipeptides;    enzymology;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/104827/HUDSON-DISSERTATION-2019.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Despite remarkable medical advances, including improved sanitation, effective vaccines, and antibiotics, bacterial infections remain a serious threat to human health. Annually, over 17 million people succumb to bacterial infections, with an increasing proportion due to antibiotic resistance. Therefore, there is urgent and continuous need for new antibiotics. Small molecule metabolites from microbes have been a highly productive source of chemical matter that ultimately led to most of today’s clinically used antibiotics. Many of these natural products are derived from polyketide synthase (PKS) or non-ribosomal peptide synthetase (NRPS) families, including well-known antibiotics such as the beta-lactams, tetracyclines, macrolides, and glycopeptides. With the rising resistance to these proven antibiotic classes, alternative sources of antibiotics must be discovered. The ribosomally synthesized and post-translationally modified peptides (RiPPs) have been attracting interest as one such source of untapped potential. Owing to the unique “promiscuous-yet-specific” biosynthetic paradigm that RiPPs offer, this class of natural products have been an area of intense research. Chapter 1 provides an overview of various RiPP classes, their antibiotic activities, the biosynthesis that creates these remarkable molecules, and discusses their engineering potential.Chapters 2, 3, and 4 all elaborate further on the thiopeptide class of RiPP antibiotics. Thiopeptides typically display nanomolar activity against pathogenic Firmicutes, such as Staphylococcus aureus, Enterococcus faecalis, and Clostridium difficile. Despite their potent activity, thiopeptides suffer from poor aqueous solubility which has hampered their clinical development. In order to investigate this problem, we endeavored to investigate the enzymes that give rise to these molecules. Prior to the work described herein, thiopeptide enzymology was a largely undeveloped field, with almost all investigations previously being performed in vivo and introducing the drawback that comes with the cellular “black box.” In order to obviate this issue, we reconstituted the core biosynthetic framework that synthesizes the thiopeptide antibiotic, thiomuracin (Chapter 2). This enabled not only the total in vitro biosynthesis of bioactive thiopeptide, but also the in-depth investigation into the enzymatic steps at each point along the biosynthetic path which was previously impossible due to the inherent obfuscation of in vivo biosynthesis. Armed with this system, we elucidated the details of biosynthetic timing, including order of post-translational modifications, substrate tolerance, and substrate recognition (Chapter 3). These investigations have laid the ground work for further thiopeptide engineering and have already spawned multiple side-projects (see Appendix A).Chapter 4 deviates from the thiomuracin system and focuses instead on bioinformatically surveying the entire thiopeptide class in sequenced organisms. This survey revealed that thiopeptide biosynthetic space is a largely unexplored territory, with many detected gene cluster families having no characterized member. More importantly, we predict that there may exist other macrocycle sizes beyond those already represented in thiopeptides; this metric has previously served as a predictor of bioactivity and thus novel thiopeptides may in turn have novel modes of action. Canvassing the genomic landscape of thiopeptides also enabled an assessment of post-translational modifications. Specifically, we focused on thiopeptide thioamidation. Although in a highly divergent context, thioamidation has been recently shown to be carried out by TfuA/YcaO enzyme pairs in methanogenic archea. Using TfuA/YcaO as a marker, we discovered an additional thiopeptide featuring thioamidation (saalfelduracin), and associated the previously-orphaned two thioamidated thiopeptides with biosynthetic gene clusters (Sch 18640 and thiopeptin). We also demonstrated that thioamidation features “plug and play” activity by incorporating these enzymes into the genome of the thiostrepton producer, S. laurentii, resulting in the production of a thioamidated analog.Lastly, Chapter 5 diverges from thiopeptide work and into the realm of sactipeptides, which feature a radically-installed thioether linkage between a Cys donor residue to the Cα of an acceptor residue. Using bioinformatics, we surveyed the entire family and found a novel sactipeptide featuring a previously-unseen configuration of crosslinks, huazacin. We further delved into this family and found that another thioethercontaining group of RiPPs which had been loosely conflated with sactipeptides, the SCIFFs (six-cysteines in forty-five residues), featured non-alpha linkages. Specifically, we discovered a novel RiPP, freyrasin, which also features radically-installed thioethers, but rather than S–Cα linkages, contains six S–Cβ crosslinks to Asp residues. Lastly, we demonstrated using stable isotope labeling that the only previouslycharacterized SCIFF in the literature, thermocellin, does not contain a S–Cα thioether between Cys and Thr as previously reported, but instead a S-Cɣ. As these findings demonstrate “SCIFFs” as a divergent biosynthetic paradigm from sactipeptides, we proposed to rename this group of natural products to ranthipeptides (radical non-alpha thioether peptides).