Structural and mechanistic studies on RNase J paralogs and associated multi-enzyme complexes in Staphylococcus aureus

Abstract

The Staphylococcus aureus RNome, which includes both coding and non-coding RNAs, has a direct bearing on the virulence of this pathogen as it influences the ability to adapt to diverse environments within the host. Multiple factors influence the RNome. These include the rate of gene expression (which is modulated by environmental conditions), the rate of RNA maturation or degradation, and interaction with protein factors or localization mechanisms that retain the mRNA from subsequent translation or recycling mechanisms. The focus of the studies reported in this thesis is on the mechanism(s) that govern RNA maturation and recycling in S. aureus. There are fifteen annotated ribonucleases in S. aureus that have been characterized in different functional contexts, including RNA maturation, degradation, and the CRISPR-Cas mechanism. The first part of the study was designed to understand the basis for the different functional roles ascribed to the two RNase J paralogs, RNase J1 and RNase J2, despite significant sequence and structural similarity. While RNase J1 with 5’-3 exonuclease activity was shown to be a constituent of the RNA degradosome, RNase J2, a catalytically sluggish enzyme, was shown to play a significant role in the CRISPR-Cas mechanism. Extensive structural, biochemical, and mutational analysis allowed us to decipher the catalytic mechanism of RNase J2, which we could later determine as distinct from the one- or two-metal ion mechanisms assumed for RNase J enzymes. The other study reported in this thesis relates to the interaction of RNase J1 with other ribonucleases and the cold shock helicase, CshA. This study involving biochemical, biophysical, and conformational analysis suggested the possibility of multi-enzyme assemblies that could work in tandem to regulate the S. aureus RNome. Other publications that arose from collaborative studies that were performed during the course of my tenure in this laboratory are presented as appendices to the main body of the thesis. Chapter 1 provides a brief introduction to ribonucleases and previously characterized reaction mechanisms that are particularly relevant in the context of RNase J. The emphasis here is on placing the in silico, in vitro (structural, biophysical, and biochemical) data in the context of S. aureus virulence and pathogenesis. Metal ions have been shown to play a crucial role in the catalytic mechanism of different RNases. This information has been summarized as it is particularly relevant in the context of the S. aureus RNase J paralogs. Details of other ribonuclease enzymes are limited to those that interact with the two S. aureus RNase J paralogs. S. aureus, like some other pathogenic bacteria, changes its phenotype in specific growth or environmental contexts. Two distinct phenotypes are the persistent phase (characterized by biofilm) and the virulent phase (characterized by the secretion of pore-forming toxins). While biofilms have been best characterized in the context of implants, the virulent phase has been evaluated in the context of phagocytosis and the toxic shock syndrome. A key molecular determinant of the switch between the two phenotypic states is the accessory gene regulator (agr). The agr mechanism involves the coordinated action of the response regulator triggered by the binding of an auto-inducing peptide to AgrC, a membrane-sensing kinase, leading to the activation of the response regulator, AgrA. AgrA drives the expression of a long non-coding RNA (RNAIII) alongside δ-hemolysin, a potent toxin. The finding that the RNAIII content could alter the phenotype state, alongside the observation that multiple ribonucleases (PNPase, RNase J1) and a helicase (CshA) interact with RNAIII, led to the examination of the effect of these multi-enzyme complexes. The quorum response mechanism and the resultant change in the RNome is described as a prelude to the studies described in Chapter 3. Chapter 2 of this thesis focuses on the catalytic mechanism of two RNase J paralogs- RNase J1 and RNase J2. These paralogs demonstrate varied catalytic efficiencies despite extensive sequence and structural similarity. RNase J1 is substantially more active than RNase J2. Electronic properties of active site residues were evaluated using density functional theory within a quantum mechanics molecular mechanics framework. This analysis suggested that multiple residues at the active site can function as a Lewis base or acid in RNase J2. On the other hand, the bond dissociation energy suggested that the Mn+2 ion in RNase J2, located at a structurally identical location to one of the Mn+2 ions in RNase J1, is crucial for overall structural integrity. Mutation of active site residues revealed that only H80 and E166 were critical for nuclease activity. Structures of mutant enzymes lacking the metal ion were seen to adopt a different orientation between the substrate binding and catalytic domain than wild-type RNase J2. A surprising finding was that the RNase J2 H78A mutant was five-fold more active than the wild-type enzyme. Structural and biochemical experiments performed in the light of this observation revealed that the RNase J2 catalytic mechanism is distinct from both the two-metal ion and one-metal ion reaction mechanisms proposed for RNase J nucleases. Different activity levels in RNase J paralogs can thus be ascribed to the diversity in catalytic mechanisms. The assembly of multi-enzyme complexes involving CshA, RNase J1, and PNPase was evaluated using diverse biochemical, biophysical, and structural biology approaches. This study, described in chapter 3 of this thesis, suggests that these multi-enzyme complexes perhaps operate in tandem to regulate the RNome in S. aureus. A noteworthy finding is that these protein-protein interactions were seen even in the absence of a substrate RNA. Initial inputs on pair-wise protein-protein interactions were obtained using microscale thermophoresis. These interactions were subsequently validated by analytical ultracentrifugation (AUC). The AUC data also provided an insight into the stoichiometry of the interaction. A characteristic feature of the three enzymes (CshA, RNase J1, and PNPase) is the presence of significant intrinsically disordered (or partially structured in the case of RNase J1) segments. Indeed, previous structural descriptions of these enzymes lacked information on large polypeptide segments due to this intrinsic disorder. The multi-enzyme complex identified by interaction and AUC analysis was further analyzed by small-angle X-ray scattering (SAXS). SAXS data provided the shape profiles of the multi-enzyme complexes. We note that the oligomeric state of the enzyme could vary in the presence of other interacting proteins. A fluorescence assay was designed to probe the effect of the individual enzymes vis-à-vis the multi-enzyme complexes on RNA degradation. Enzyme assays revealed an intriguing observation that these multi-enzyme complexes modulate substrate degradation. We hypothesize that this modulation of enzymatic activity could either be due to allostery or occlusion effects, wherein parts of the substrate RNA remain out of bounds for the other enzymes in the mixture. These findings that the association between these degradosome components could result in significant conformational changes suggest that the RNA degradosome is likely to be pleiomorphic. Together, these studies provide a structural basis for directed experiments in the future to understand the rationale for these multi-enzyme complexes and their impact on S. aureus virulence. Studies on the molecular mechanisms that influence RNA degradation in S. aureus revealed a diversity in this cellular process that was hitherto unknown. The diversity in the RNA degradosome across the bacterial kingdom, on the other hand, was predicted from the observation that homologs of E. coli RNase E, the product of an essential gene that has been demonstrated to be the scaffold of the E. coli RNA degradosome, has no homolog in B. subtilis (or as in this case, S. aureus). Furthermore, S. aureus has two RNase J paralogs- whereas E. coli has none. Another variation between bacteria is the observation that the multi-enzyme complexes governing RNA degradation can be either cytosolic or membrane-associated. The cellular and environmental context that governs the localization of the degradosome complex is a topic that needs further investigation. The work reported in this thesis thus places the RNA degradosome- at least in the context of S. aureus- as an assembly of pre-formed multi-enzyme building blocks. The observations on the multi-enzyme complexes and the cellular or environmental triggers for these complexes, if any, need to be evaluated further to understand the impact of these multi-enzyme assemblies on the virulence of this human pathogen.