STRUCTURAL AND METABOLOMICS ANALYSIS OF BACTERIOPHAGE ENDOLYSINS AND THEIR IMPACT ON HOST BACTERIA

Abstract

Bacteriophages are viruses, the natural enemy of bacteria, and are logical agents for controlling bacterial pathogens. Bacteriophages release new phage progeny from their bacterial host cells by expressing phage-encoded enzymes, so called ‘endolysins,’ which effectively rupture the bacterial cell wall from the inside with the help of holins, thus killing the bacterial cell. Due to the rise of antimicrobial resistance (AMR) in pathogens, the application of bacteriophage viruses (phage therapy) to treat bacterial infections is a promising alternative to antibiotics. Phage therapy has its limitations as a biotherapy, due to the high specificity of bacteriophages toward their host bacteria. Additionally, phages are also highly unstable due to their rapidly evolving genome. Hence, phage-encoded endolysins instead of whole phages are being studied to target the pathogens. Phage-based endolysins are being used in various fields, such as food processing and preservation, biofilm eradication, and human medicines. To combat emerging multidrug-resistant bacterial pathogens, several strategies including diverse protein engineering techniques, combinations of endolysins with antimicrobial peptides, and encapsulation of endolysins are currently being employed to improve the stability, catalysis, and specificity of phage-based endolysins. Developing phage lysin formulations that can be administered to combat disease-causing pathogens is challenging due to their limited pharmacokinetic and toxicity studies, as immune responses are expected due to the proteinaceous nature of phage lysins. In this context, in-vivo/preclinical studies, including metabolomics studies of these entities, will help in identifying the metabolites as potential biomarkers during lysin-mediated therapy. Indeed, different families of endolysins may trigger variable extent of immune responses, and henceforth it is essential to identify the erratic set of biomarkers (metabolic fingerprints) and their pathways studied by metabolomic studies to substantiate the role of endolysins as next-generation enzybiotics. Endolysins that infect Gram-positive bacteria are comprised of two distinct domains, namely the enzymatic catalytic domain (ECD) and the cell binding domain (CBD). Conversely, the majority of endolysins that target Gram-negative bacteria possess solely the enzymatic catalytic domain. Among various endolysins, T4 and T7 are two single-domain endolysins known to lyse E. coli cells. T4 endolysin (T4 lysozyme-T4L, 18.66 kDa) belongs to the muramidase family of endolysins, which cleave the β-1,4 glycosidic bond between the N-acetylglucosamine and N-acetylmuramic acid of the peptidoglycan layer. T7 bacteriophage endolysin (T7 lysozyme-T7L), also known as N-acetylmuramoyl-L-alanine amidase or T7 lysozyme, is a 17 kDa protein. T7 lysozyme lyses a range of Gram-negative bacteria by hydrolyzing the amide bond between N-acetyl muramyl residues and the L-alanine of the peptidoglycan layer. Zinc works as a cofactor and is in the catalytic cleft and anchored to the protein by three side-chain ligands, His18, His123, and Cys131, and connected to the hydroxyl group of Tyr47 through a water molecule. Zinc is required for the amidase activity of T7 endolysin. Understanding the structure-stability characteristics of endolysin helps in designing phage endolysins as broadband antimicrobial agents. The present thesis focuses on the global metabolomics profiling of host E. coli upon overexpression of two endolysins (T4L and T7L) and two T7 endolysin variants (H37A and H48K). Further, the catalytic and structural stability features of T7L holoenzyme with Zn as a cofactor compared to apoenzyme. The specific details of the thesis chapters (1-4) are as follows: Chapter 1 provides an overview of antibiotic resistance among pathogenic bacteria and brief information about various alternative strategies applicable to antibiotics. A brief review of catalytic and structural features of zinc metalloproteins. The chapter also covers NMR-based metabolomics and its applications in various fields. Chapter 2 elucidates the impact of two endolysins (T4L, T7L) overexpression on the metabolic fingerprint of E. coli using NMR spectroscopy. The 1H NMR-based metabolomics analysis revealed global metabolite profiles of E. coli in response to endolysins. The acquired data was categorized into two categories; the first category belongs to the metabolites extracted from IPTG treated and RNAP overexpressing cells were compared with that of control (untreated). In the second category, the metabolic profile of T4L and T7L expressing cells were compared with RNAP expressing cells to mark the variations in the metabolites upon endolysin(s) overexpression. The study has identified nearly 75 metabolites, including organic acids, amino acids, sugars, and nucleic acids. The RNAP overexpression has shown minimal perturbation in the metabolic profile of E. coli cells and is thus taken as a control to compare the variations upon overexpression of heterologous proteins (endolysins). The data suggested downregulation of the central carbon metabolic pathway in both endolysins overexpression but to a different extent. Also, the endolysin overexpression has highlighted the enhanced metabolic load and stress generation in the host cells, thus leading to the activation of osmoregulatory pathways. The overall changes in the metabolic fingerprint of E. coli highlight the enhanced perturbations during the overexpression of T4L compared to T7L. These untargeted metabolic studies shed light on the regulation of molecular pathways during the heterologous overexpression of these lytic enzymes that are lethal to the host. Chapter 3 assesses the impact of overexpression of T7L variants (T7L H48K and T7L H37A) on the metabolic profiles of E. coli using NMR spectroscopy. The two variants considered for the study include T7L H37A, which possesses enhanced amidase activity (~40%) than its wildtype counterpart, and T7L H48K which is a dead mutant without any significant activity. In this study, the overexpression of T7L wildtype (T7L-WT) protein and its variants (T7L H48K and T7L H37A), was compared to RNAP overexpression in E. coli cells using 1H NMR-based metabolomics. A total of 75 metabolites including organic acids, amino acids, sugars, and nucleic acids were annotated and analyzed. Based on fold change and multivariate analysis, the two T7L variants showed distinct and variable clustering patterns to that of T7L Wildtype, in which the dead mutant H48K group showed close clustering to that of RNAP. Further pathway impact analysis revealed H48K variant group showed minimal perturbations in energy and amino acids pathways to sustain the osmotic stress. In contrast, the most significant metabolites were observed in T7L WT and H37A groups. Indeed, the differential fold change and metabolic pathway perturbations related to energy/carbohydrate, amino acid, and nucleotide metabolism, were observed between WT and H37A. This comparative analysis of metabolic signatures by different endolysin variants with variable lytic activity in the host (E. coli) helps in optimizing/improving the suitable and robust endolysins as an alternative to antibiotics. Chapter 4 unravels the characterization and influence of pH on structural stability and enzymatic activity of T7L apo and holoenzymes. T7L protein was successfully expressed and purified by transforming E. coli BL 21 (DE3) cells. Purification of T7L was done by using a combination of chromatographic techniques. The mechanism of pH-dependent differential activity profiles and structural features of both apo and holoenzyme were investigated using cell lysis assay and various biophysical techniques, including Circular Dichroism (CD), fluorescence, NMR techniques, and MD simulations. The turbidimetric assay revealed that cofactor Zn is essential for the optimal lytic activity of T7L. The apoenzyme exhibited minimal amidase activity at pH 7 and 8, suggesting that even though it exists in native folded conformation, the optimal lytic activity is only achieved when the Zn cofactor is attached at the catalytic site. Structural studies based on CD and fluorescence experiments revealed slight structural changes between apo and holoenzymes. Urea-dependent unfolding experiments based on far-UV CD demonstrated the differential and more stable structural state of holoenzyme compared to apoenzyme at pH 7. Thermal denaturation studies (20 to 90 °C) on the holoenzyme at pH 7 showed an irreversible melting reaction as the protein started forming insoluble oligomers and/or aggregates. NMR-based 15N−1H HSQC spectra of T7L holoenzyme at different pH conditions revealed stable confirmation with clear chemical shift dispersion up to pH 5 and loss of the chemical shift dispersion at lower pH, suggesting a substantial conformational change. Indeed, once the buffer was exchanged back to pH 7, reversibility was observed with an identical spectrum, confirming the pH-induced reversible structural transition of the T7L holoenzyme. Based on molecular dynamics simulations, T7L holoenzyme (including crystal water) with Zn (including crystal water) as a cofactor showed significantly stable protein conformation compared to apoenzyme in terms of root-mean-square deviation (RMSD), the radius of gyration (Rg), and residue-wise root-mean-square fluctuation (RMSF). The crystal water bound to Tyr47, which is crucial for the hydrolysis of peptidoglycan (PG) moves closer to the Zn and remains intact till the end of the simulation. Altogether, these results depict the role of Zn in the catalytic and stability features of the T7L holoenzyme. In summary, the present thesis elucidated the global metabolomics features of two endolysins, T4L and T7L, and T7L variants: H48K and H37A, in the host E. coli. The molecular interactions, catalytic and stability features of T7L holoenzyme with Zn cofactor elucidated the enhanced stability and activity features of T7L holo form. In summary, unravelling these untargeted metabolic signatures of the host and the biophysical features of endolysins serves as a benchmark for designing phage lysins as next generation enzybiotics.