Biosynthesis and mode of action of ribosomally synthesized and post-translationally modified antimicrobial peptides

One of the greatest sources of biologically active compounds is natural products. Often these compounds serve as platforms for the design and development of novel drugs and therapeutics. The overwhelming amount of genomic information acquired in recent years has revealed that ribosomally synthesized and post-translationally modified natural products are much more widespread than originally anticipated. Identified in nearly all forms of life, these natural products display incredible structural diversity and possess a wide range of biological functions that include antimicrobial, antiviral, anti-inflammatory, antitumor, and antiallodynic activities. The unique pathways taken to biosynthesize these compounds offer exciting opportunities for the bioengineering of these complex molecules. The studies described herein focus on both the mode of action and biosynthesis of antimicrobial peptides. In Chapter 2, it is demonstrated that haloduracin, a recently discovered two-peptide lantibiotic, possesses nanomolar antimicrobial activity against a panel of bacteria strains. The potency of haloduracin rivals that of nisin, an economically and therapeutically relevant lantibiotic, which can be attributed to a similar dual mode of action. Moreover, it was demonstrated that this lantibiotic of alkaliphile origin has better stability at physiological pH than nisin. The molecular target of haloduracin was identified as the cell wall peptidoglycan precursor lipid II. Through the in vitro biosynthesis of haloduracin, several analogues of Halα were prepared and evaluated for their ability to inhibit peptidoglycan biosynthesis as well as bacterial cell growth. In an effort to overcome the limitations of in vitro biosynthesis strategies, a novel strategy was developed resulting in a constitutively active lantibiotic synthetase enzyme. This methodology, described in Chapter 3, enabled the production of fully-modified lacticin 481 products with proteinogenic and non-proteinogenic amino acid substitutions. A number of lacticin 481 analogues were prepared and their antimicrobial activity and ability to bind lipid II was assessed. Moreover, site-directed mutagenesis of the constitutively active synthetase resulted in a kinase-like enzyme with the ability to phosphorylate a number of peptide substrates. The hunt for a lantibiotic synthetase enzyme responsible for installing the presumed dehydro amino acids and a thioether ring in the natural product sublancin, led to the identification and characterization of a unique post-translational modification. The studies described in Chapter 4, demonstrate that sublancin is not a lantibiotic, but rather an unusual S-linked glycopeptide. Its structure was revised based on extensive chemical, biochemical, and spectroscopic characterization. In addition to structural investigation, bioinformatic analysis of the sublancin gene cluster led to the identification of an S-glycosyltransferase predicted to be responsible for the post-translational modification of the sublancin precursor peptide. The unprecedented glycosyltransferase was reconstituted in vitro and demonstrated remarkable substrate promiscuity for both the NDP-sugar co-substrate as well as the precursor peptide itself. An in vitro method was developed for the production of sublancin and analogues which were subsequently evaluated in bioactivity assays. Finally, a number of putative biosynthetic gene clusters were identified that appear to harbor the necessary genes for production of an S-glycopeptide. An additional S-glycosyltransferase with more favorable intrinsic properties including better expression, stability, and solubility was reconstituted in vitro and demonstrated robust catalytic abilities.

The cellular and mechanistic study of nisin inhibition of bacillus anthracis spore outgrowth, nisin alteration of infection, and native producer immunity

A critical step during Bacillus anthracis infection is the outgrowth of germinated spores into vegetative bacilli that proliferate and disseminate rapidly within the host. An important challenge exists for developing chemotherapeutic agents that act upon and kill B. anthracis immediately after germination initiation when antibiotic resistance is lost, but prior to the outgrowth into vegetative bacilli, which is accompanied by toxin production. Chemical agents must also function in a manner refractive to the development of antimicrobial resistance. In this thesis we have identified the lantibiotics as a class of chemotherapeutics that are predicted to satisfy these two criteria. The objective of this thesis was to evaluate the efficacy of nisin, a prototypical lantibiotic, in prevention of outgrowth of germinated B. anthracis spores. Like all lantibiotics, nisin is a ribosomally translated peptide that undergoes post-translational modification to form (methyl)lanthionine rings that are critical for antimicrobial activity. Our studies indicate that nisin rapidly inhibits the in vitro outgrowth of germinated B. anthracis Sterne 7702 spores. Although germination initiation was shown to be essential for nisin-dependent antimicrobial activity, nisin did not inhibit or promote germination initiation. Nisin irreversibly killed germinated spores by blocking the establishment of a membrane potential and oxidative metabolism, while not affecting the dissolution of the outer spore structures. The membrane permeability of the spore was increased by nisin, but germinated spores did not undergo full lysis. Nisin was demonstrated to localize to lipid II, which is the penultimate precursor for cell wall biogenesis. This localization suggests two possible independent mechanisms of action, membrane pore formation and inhibition of peptidoglycan synthesis. Structure-activity studies with a truncated form of nisin lacking the two C-terminal (methyl)lanthionine rings and with non-pore forming mutants indicated that membrane disruption is essential for nisin-dependent inhibition of spore outgrowth to prevent membrane potential establishment. Finally, utilizing an in vitro infection model, it was shown that nisin reduced the viability of B. anthracis spores within an infection resulting in increased survival of immune cells while reducing infection-mediated cytokine expression. Fluorescence microscopy indicated that nisin localizes with spores within phagosomes of peritioneal macrophages in germinating conditions. These data demonstrate the effectiveness of nisin, as a model lantibiotic, for preventing spore outgrowth. It is speculated that nisin targeting of lipid II, resulting in membrane perturbations, may be effective at inhibiting the outgrowth of spores prepared from bacteria across a number of species.

Statistical modeling of protein lysate array data

The protein lysate array is an emerging technology for quantifying the protein concentration ratios in multiple biological samples. It is gaining popularity, and has the potential to answer questions about post-translational modifications and protein pathway relationships. Statistical inference for a parametric quantification procedure has been inadequately addressed in the literature, mainly due to two challenges: the increasing dimension of the parameter space and the need to account for dependence in the data. Each chapter of this thesis addresses one of these issues. In Chapter 1, an introduction to the protein lysate array quantification is presented, followed by the motivations and goals for this thesis work. In Chapter 2, we develop a multi-step procedure for the Sigmoidal models, ensuring consistent estimation of the concentration level with full asymptotic efficiency. The results obtained in this chapter justify inferential procedures based on large-sample approximations. Simulation studies and real data analysis are used to illustrate the performance of the proposed method in finite-samples. The multi-step procedure is simpler in both theory and computation than the single-step least squares method that has been used in current practice. In Chapter 3, we introduce a new model to account for the dependence structure of the errors by a nonlinear mixed effects model. We consider a method to approximate the maximum likelihood estimator of all the parameters. Using the simulation studies on various error structures, we show that for data with non-i.i.d. errors the proposed method leads to more accurate estimates and better confidence intervals than the existing single-step least squares method.