Proteins modularity enhances the multi-functionality and versatility of proteins by providing such properties as multiple and various ligand-binding sites, increased ligand affinity through the avidity effect, and the juxtaposition of ligand-binding modules near catalytic domains. An NMR-based "dissect-and-build" approach to studying modular protein structure and function has proven very successful, whereby modules are initially characterized individually and then correlated with the overall function of a protein. We have used the dissect-and-build approach and NMR to study two modular protein systems. Chapter 2 details the NMR solution structure of the weak-lysine-binding kringle IV type 8 (KIV8) module from the apolipoprotein(a) (apo(a)) component of lipoprotein(a) was determined and its ligand-binding properties assessed. In vitro studies have demonstrated the importance of the apo(a) KIV7 and KIV8 modules in mediating specific lysine-dependent interactions with the apolipoproteinB-100 (apoB-100) component of LDL in the initial non-covalent step of lipoprotein assembly. Notable differences identified in the lysine binding site (LBS) of the KIV8 were deemed responsible for the differential modes of apoB-100 recognition by KIV7 and KIV8. In addition, the KIV8 structure has brought to light the importance of an RGD sequence at the N-terminus of the apo(a) KIV8 module, which may mediate important apo(a)-integrin interactions. In Chapters 3-6, structure-function studies of the CpGH84C X82 and the CpGH84A dockerin-containing modular pair were conducted to understand how the varying modularity unique to the C-terminal regions of the secreted multi-modular family 84 glycoside hydrolases influences the spreading of Clostridium perfringens. Identification of a CpGH84C cohesin module (X82), and the structural characterization of a dockerin-containing modular pair provides the first evidence for multi-enzyme complex formation mediated by non-cellulosomal cohesin-dockerin interactions. The formation of large hydrolytic enzyme complexes introduces a novel mechanism by which C. perfringens may enhance its role in pathogenesis. / Thesis (Ph.D, Biochemistry) -- Queen's University, 2007-09-27 11:46:38.753

“Neglected Rickettsiaceae” (i.e. those harboured by non-hematophagous eukaryotic hosts) display greater phylogenetic variability and more widespread dispersal than pathogenic ones; yet, the knowledge about their actual host range and host shift mechanism is scarce. The present work reports the characterization following the full-cycle rRNA approach (SSU rRNA sequence, specific in situ hybridization, and ultrastructure) of a novel rickettsial bacterium, herewith proposed as 'Candidatus Megaira polyxenophila' gen. nov., sp. nov. We found it in association with four different free-living ciliates (Diophrys oligothrix, Euplotes octocarinatus, Paramecium caudatum, and Spirostomum sp., all belonging to Alveolata, Ciliophora); furthermore it was recently observed as intracellular occurring in Carteria cerasiformis and Pleodorina japonica (Chlorophyceae, Chlorophyta). Phylogenetic analyses demonstrated the belonging of the candidate new genus to the family Rickettsiaceae (Alphaproteobacteria, Rickettsiales) as a sister group of the genus Rickettsia. In situ observations revealed the ability of the candidate new species to colonize either nuclear or cytoplasmic compartments, depending on the host organism. The presence of the same bacterial species within different, evolutionary distant, hosts indicates that 'Candidatus Megaira polyxenophila' recently underwent several distinct host shifts, thus suggesting the existence of horizontal transmission pathways. We consider these findings as indicative of an unexpected spread of rickettsial infections in aquatic communities, possibly by means of trophic interactions, and hence propose a new interpretation of the origin and phylogenetic diversification of rickettsial bacteria.

Under hostile conditions, bacteria elicit stress response. Such stress response is regulated by a secondary messenger called (p)ppGpp. (p)ppGpp is involved in wide range of functions such as GTP homeostasis, biofilm formation and cell growth. Its regulation and mode of action is not well understood. This work has been initiated with an aim to gain insights into the molecular basis of stress response. (p)ppGpp was discovered on the chromatogram of cell extract from starved E. coli cells. (p)ppGpp is synthesized and hydrolyzed by Rel/SpoT in Gram negative bacteria (such as E. coli), and by bifunctional enzyme called Rel in Gram positive bacteria (such as Mycobacteria). The obvious question that comes in our mind is how bifunctional Rel enzyme decides on synthesis or hydrolysis in Gram positive bacteria such as Mycobacterium? In our laboratory, it has been shown that N-terminal domain of Rel shows unregulated (p)ppGpp synthesis implying regulatory role of C-terminal domain. Also, concurrent increase in anisotropy of Rel C-terminal domain with the increase in concentration of pppGpp has been observed indicating the binding of pppGpp to the C-terminal domain. We performed Isothermal Calorimetry experiment to confirm that pppGpp binds with C-terminal domain of Rel enzyme. For identification of the binding region, small molecule analogue 8-azido-pppGpp has been synthesized. This analogue is UV-crosslinked with C-terminal domain of Rel and specificity of the interaction has been determined by gel based crosslinking experiments. Crosslinked protein has been subjected to the ingel¬trypsin digestion and analyzed by mass spectrometry. We identified two crosslinked peptides in the mass spectra of trypsin digest in case of the crosslinked protein where identity of the parent peptide is confirmed by MS-MS analysis. Site directed mutagenesis has been carried out based on the conservation of residues in the crosslinked peptides. Isothermal Calorimetry analysis has been done where Rel C-terminal domain mutants are titrated with pppGpp in order to detect any defect in binding due to the mutations. Mutations leading to the reduced binding affinity of pppGpp to Rel C-terminal domain have been introduced in the full length Rel protein and activity assays are carried out so as to evaluate the effects of mutations on synthesis and hydrolysis activity. In mutants, synthesis activity is found to be increased with the concomitant reduction in hydrolysis activity. This indicates the feedback loop where pppGpp binds to Rel C-terminal domain to regulate it own synthesis and hydrolysis. In E. coli, pppGpp binds to RNA polymerase and modulates the transcription. The region where it binds is controversial. In addition, whether ppGpp and pppGpp have different binding site on RNA polymerase is not known. The latter question becomes important in the light of evidence where differential regulation of transcription by ppGpp and pppGpp have been indicated. We found that ppGpp and pppGpp have an overlapping binding site on RNA polymerase. The 8-azido-ppGpp has been mapped on β and β’ subunits whereas binding site of 8-azido-pppGpp has been located on the β’ subunit. We observed that the 8-azido¬pppGpp labels RNA polymerase more efficiently than ppGpp. pppGpp can compete out ppGpp as illustrated by DRaCALA assay and gel based crosslinking experiment. However, the RNAP from B. subtilis does not bind to (p)ppGpp. (p)ppGpp is ubiquitous in bacteria but absent in mammals. Thus, blocking (p)ppGpp synthesis would impede the survival of bacteria without having any effect on humans. Recently, Relacin compound has been synthesized by another group in order to inhibit (p)ppGpp synthesis. The limitations of this compound are the requirement of high concentration (5mM) for inhibition and low permeability across the membrane. Taking hints from the latter compound, we acetylated the nd 2’, 3’ and 5’ position of ribose ring and benzoylated the 2position of guanine moiety in guanosine molecule. We observed significant inhibition of in vitro pppGpp synthesis and biofilm formation. More studies will be conducted in near future to test these compounds for their plausible functions. In collaboration with Prof. Jayaraman (Organic Chemistry, IISc), many artificial glycolipids are synthesized and tested for biological function. We observed that synthetic glycolipids exhibit a profound effect as inhibitors of the key mycobacterial functions. These analogs impede biofilm formation and can plausibly affect long term survival. Glycolipid analogs can compete with natural glycolipids, thus may help in understanding their functions. Our past and recent studies have showed that the synthetic glycolipids act as inhibitors of mycobacterial growth, sliding motility and biofilm formation. The major lacuna of these glycolipid inhibitors is the requirement of high concentration. Their inhibitions at nanomolar concentrations remain to be achieved. Issues surrounding the thick, waxy mycobacterial cell wall structures will continue to be the focus in manifold approaches to mitigate detrimental effects of mycobacterial pathogens. In chapter 1, introduction to the research work has been written and role of (p)ppGpp and its functions have been discussed. In chapter 2, novel binding site of pppGpp on Rel C-terminal domain and its regulatory role have been discussed. In chapter 3, differential binding of ppGpp and pppGpp to RNA polymerase has been discussed. In chapter 4, studies on natural and synthetic analogues of pppGpp have been presented. In chapter 5, synthetic glycolipids studies have been described. Chapter 6 summarizes all the chapters.

“Neglected Rickettsiaceae” (i.e. those harboured by non-hematophagous eukaryotic hosts) display greater phylogenetic variability and more widespread dispersal than pathogenic ones; yet, the knowledge about their actual host range and host shift mechanism is scarce. The present work reports the characterization following the full-cycle rRNA approach (SSU rRNA sequence, specific in situ hybridization, and ultrastructure) of a novel rickettsial bacterium, herewith proposed as 'Candidatus Megaira polyxenophila' gen. nov., sp. nov. We found it in association with four different free-living ciliates (Diophrys oligothrix, Euplotes octocarinatus, Paramecium caudatum, and Spirostomum sp., all belonging to Alveolata, Ciliophora); furthermore it was recently observed as intracellular occurring in Carteria cerasiformis and Pleodorina japonica (Chlorophyceae, Chlorophyta). Phylogenetic analyses demonstrated the belonging of the candidate new genus to the family Rickettsiaceae (Alphaproteobacteria, Rickettsiales) as a sister group of the genus Rickettsia. In situ observations revealed the ability of the candidate new species to colonize either nuclear or cytoplasmic compartments, depending on the host organism. The presence of the same bacterial species within different, evolutionary distant, hosts indicates that 'Candidatus Megaira polyxenophila' recently underwent several distinct host shifts, thus suggesting the existence of horizontal transmission pathways. We consider these findings as indicative of an unexpected spread of rickettsial infections in aquatic communities, possibly by means of trophic interactions, and hence propose a new interpretation of the origin and phylogenetic diversification of rickettsial bacteria.

MSc (Microbiology) / Department of Microbiology / See the attached abstract below.

Au laboratoire, une famille de tyrosine kinases propres aux bactéries et ne présentant aucune ressemblance structurale avec les protéine-kinases d’origine eucaryote a été identifiée. Ces enzymes, appelées BY-kinases, sont notamment impliquées dans la biosynthèse des polysaccharides extracellulaires, mais leurs rôles précis ainsi que leurs mécanismes catalytiques sont encore peu compris.Dans la première partie de ce travail, nous avons caractérisé le rôle physiologique de la phosphorylation sur la tyrosine de la protéine Ugd, une UDP-glucose déshydrogénase, par les BY-kinases Wzc et Etk d’E. coli. Nous avons démontré que la phosphorylation d’Ugd sur un site commun à Wzc et Etk augmente son activité. Nous avons également établi que la phosphorylation d’Ugd par Wzc participe à la régulation de la quantité d’acide colanique produit, tandis que la phosphorylation d’Ugd par Etk influence la résistance de la bactérie à la polymyxine.Nous avons également effectué une analyse structure-fonction du domaine cytoplasmique de deux BY-kinases, CapA1/CapB2 de S. aureus et Wzc d’E. coli. Nous avons montré que ces deux protéines s’associent en octamère, grâce au motif EX2RX2R et qu’elle s’autophosphoryle selon un mécanisme intermoléculaire. Nous avons, de plus, identifié le mécanisme d’activation de ces protéines et révélé l’importance d’un domaine particulier dans l’autophosphorylation de Wzc et la biosynthèse de l’acide colanique.La caractérisation structurale et fonctionnelle des BY-kinases représente une approche prometteuse et originale en vue de l’élaboration de molécules inhibant spécifiquement leur activité et pouvant affecter le pouvoir virulent des bactéries pathogènes. / A new class of bacterial enzymes, named BY-kinases, has been shown to catalyze protein-tyrosine phosphorylation. These enzymes share no structural and functional similarities with their eukaryotic counterparts. Evidence of their involvement in extracellular polysaccharide biosynthesis has been provided, but their accurate functions and their catalytic mechanism remain largely unknown.First, we characterized the physiological role of tyrosine phosphorylation of Ugd, a UDP-glucose dehydrogenase, by the BY-kinases Wzc and Etk of E. coli. We demonstrated that Ugd phosphorylation by Wzc or Etk occurs on the same site and increases its activity. We also established that Wzc-mediated phosphorylation of Ugd participates in the regulation of colanic acid production whereas Ugd phosphorylation by Etk influences resistance to polymyxin.In addition, we performed a structure-function analysis of the cytoplasmic domain of two BY-kinases, namely CapA1/CapB2 from S. aureus and Wzc from E. coli. We showed that these two proteins associate in a ring-shaped octamer in which the motif EX2RX2R plays a crucial role. In addition, we showed that BY-kinases autophosphorylate using an intermolecular mechanism. We also identified the activation mechanism of BY-kinases and we revealed the role of a particular domain, found specifically in BY-kinases from proteobacteria, in Wzc autophosphorylation and colanic acid biosynthesis.Structural and functional characterization of BY-kinases represents an original and promising approach in order to develop new molecules inhibiting specifically these enzymes and to affect the virulence of bacterial pathogens.

Les écosystèmes aquatiques soumis à des pressions anthropiques sont des lieux d'évolution rapide des communautés microbiennes. Cet environnement participe certainement à l'émergence d'agents infectieux résistants aux antibiotiques. La ville de Montpellier est située dans un petit bassin versant qui subit d’une part des épisodes de pluies brutales et d’autre part de fortes pressions démographiques. Le principal hôpital est situé dans une zone de ruissellement comprenant deux petites rivières urbaines provenant d'eaux souterraines karstiques à quelques kilomètres en amont. L’objectif de cette étude est d'explorer les communautés bactériennes dans les rivières urbaines qui coulent près du centre hospitalier afin d'évaluer l'influence des ruissellements sur la résistance aux antibiotiques dans les communautés bactériennes. Les communautés bactériennes sont également décrites dans les aquifères karstiques en amont. Une section introductive présente les méthodes disponibles pour l'étude de la résistance aux antimicrobiens dans l'environnement et une revue de la littérature expose les données actuelles sur la résistance aux antibiotiques dans l’eau de ruissellement urbain. Cette partie justifie les stratégies expérimentales. La méthode développée ici, appelée détermination de la concentration inhibitrice à l’échelle de la communauté bactérienne (c-IC, pour community inhibitory concentration), est combinée à une description de la richesse taxonomique pour donner une description instantanée des communautés bactériennes résistantes dans les environnements aquatiques. Une stratégie dérivée de l'approche c-IC permet d'explorer la résistance bactérienne dans le système hydrologique urbain près de l'hôpital et dans les aquifères karstiques. Les données microbiologiques recueillies sont complétées par des données hydrologiques, hydrogéologiques, climatiques et physico-chimiques.L'impact de très faibles concentrations d'antibiotiques sur la structure de la communauté bactérienne dans divers environnements hydriques a été démontré et apparaît comme un indicateur de la vulnérabilité des écosystèmes face aux pressions antimicrobiennes. Le répertoire taxonomique des communautés fluviales urbaines a été décrit et sa dynamique a été confrontée aux conditions environnementales. Le voisinage des hôpitaux augmente significativement la prévalence des bactéries résistantes par rapport à une zone urbaine similaire éloignée de l'hôpital. Des bactéries d’intérêt médical résistantes aux céphalosporines et aux carbapénèmes ont été isolées. De façon surprenante, un Escherichia coli producteur de NDM-5, pathogène émergeant hautement résistant, a été signalé pour la première fois dans une rivière française. Le clone a été détecté dans deux échantillons indépendants montrant sa persistance. Le gène blaNDM-5 et son environnement génétique ont été décrits sur un plasmide IncX3 transférable, indiquant un transfert génétique horizontal possible. La résistance aux antimicrobiens dans l'eau souterraine karstique a varié dans le temps et dans l'espace et est difficilement comparable à celle décrite dans les rivières associées. En milieu urbain, la qualité de l'eau et le risque infectieux sont généralement évalués sur les eaux usées et les effluents des stations d'épuration. Cette étude montre que les eaux de ruissellement dans des zones urbanisées contribuent à l'émergence et à la dissémination de la résistance aux antimicrobiens. Compte tenu de l'épidémiologie inquiétante des maladies infectieuses, cette étude invite à explorer les potentiels réservoirs environnementaux de bactéries résistantes afin de compléter les connaissances sur le cycle épidémiologique de la résistance aux antimicrobiens et essayer de l’interrompre ou de le ralentir. / Aquatic ecosystems subjected to anthropic pressures are places of rapid evolution of microbial communities. They are likely hotspots for emergence of infectious disease agents resistant to antibiotics. The city of Montpellier is located in a small watershed that undergoes brutal rainfall episodes and strong demographic pressures. A hospital is located in a runoff area including two small urban rivers originating from karstic groundwater few kilometers upstream. The aim of the study is to explore bacterial communities in urban rivers flowing near hospital settings in order to evaluate the influence of runoffs on antibiotics resistance in the bacterial communities. Bacterial communities are also described in upstream karstic aquifers.An introductive section presents the methods available for studying antimicrobial resistance in environment and then reviews comprehensively bibliography on antibiotics resistance in urban runoffs. This part supports the experimental strategies. The method developed herein, called community Inhibitory Concentration (c-IC) determination is combined to taxonomic richness description to provide a tool that gives a rapid snapshot of resistant bacterial communities in aquatic environments. A strategy derived from c-IC approach allows the exploration of bacterial resistance in the urban hydrologic system near the hospital and in karstic aquifers. The collected microbiological data has been completed by hydrological, hydrogeological, climatic and physico-chemical data.The impact of very low concentration of antibiotics on the bacterial community structure in various water bodies was demonstrated and appeared as an indicator of the vulnerability of ecosystems to antimicrobial pressures. The taxonomic repertory of the urban river communities was described and its dynamics was compared to environmental conditions. Hospital vicinity significantly increase the prevalence of resistant bacteria compared to a similar urban area remote from hospital. Diverse clinically relevant cephalosporins and carbapenems resistant bacteria have been isolated. Surprisingly, a NDM-producing Escherichia coli, which is a highly resistant and emerging pathogen was reported for the first time in a French River. The clone was detected in two independent sampling showing its persistence. The blaNDM-5 gene and its surrounding sequences were described on a transferable IncX3 plasmid, indicating possible genetic transfer to other bacteria. The antimicrobial resistance in karst groundwater varied in time and space and was hardly compared with that described in related rivers.In urban settings, water quality and infectious risk is generally assessed on sewers and wastewater treatment plants effluents. This study shows that runoff waters in urbanized area contribute to the emergence and dissemination of antimicrobial resistance. Considering the worrisome epidemiology of infectious diseases, it urges to decipher all environmental reservoirs for resistant bacteria in order to complete knowledges about the epidemiological cycle of antimicrobial resistance and try to break or slow down it.

The three-dimensional structures of biological macromolecules and molecular assemblies are becoming increasingly important with the changing methodologies of drug discovery. The structures aid in understanding of protein function at the molecular level: be it a macromolecular assembly, a cytosolic enzyme or an intermembrane receptor molecule. X-ray crystallography is the most powerful technique to obtain the three-dimensional structures of such molecules at or near atomic resolution. With such a wide-spread importance, crystallography is an integral part of structural biology and also of the current drug discovery programs. The present thesis mainly deals with application of the crystallographic techniques for understanding the structure and function of adenylosuccinate lyase (ASL) from bacterial pathogens Salmonella typhimurium and Mycobacterium tuberculosis as well as its non-pathogenic counterpart Mycobacterium smegmatis. Studies were also carried out to understand the structure-function relationship of the protease in the plant virus Sesbania Mosaic Virus (SeMV). The thesis has been divided into six chapters. The first chapter contains an introduction to nucleotide synthesis and ASL superfamily of enzymes known as the aspartase/fumarase superfamily based on the published literature. Chapter 2 provides the details of the techniques used for the investigations presented in this thesis. Chapters 3-5 deal with the structural and functional studies carried out on ASL from the three bacterial organisms. Chapter 6 deals with the simulation studies carried out on SeMV protease. Mechanism and importance of nucleotide synthesis is introduced in Chapter 1, with special emphasis on purine de novo and salvage pathways. ASL is introduced as an important enzyme for purine synthesis. Its superfamily, the aspartase/fumarase superfamily of enzymes is described in detail with respect to its structure, function and pathophysiology. Objectives of the present study are outlined towards the end of the chapter. The experimental and computational techniques utilized during the course of my research are described in Chapter 2. These techniques include gene cloning, protein expression and purification, kinetic and biophysical characterization of proteins, crystallization, X-ray diffraction, data collection and processing, structure solution, refinement, model building, validation and structural analysis, phylogenetic studies, molecular docking and molecular dynamic simulation studies. Adenylosuccinate lyase is an important enzyme participating in purine biosynthesis. With the emergence of drug resistant variants of various pathogens, ASL has been recognized as a drug target against microbial infections. Chapter 3 deals with the structural and functional characterization of ASL from Salmonella typhimurium. Two constructs of the StASL gene were cloned and expressed leading to the purification of truncated (residues 1-366) and full-length (residues 1-456) polypeptides. Crystallization of the two polypeptides resulted in three independent structures. The full-length structure was very similar to the E. coli ASL structure consistent with 95% amino acid sequence identity between the two polypeptides. However, the truncated structures showed large distortions, especially of the active site residues, accounting for the catalytic inactivity of the truncated polypeptide in spite of retaining all residues considered important for function. The full-length ASL was catalytically active. A unique feature observed in StASL, not reported in other ASLs, was its allosteric regulation by the substrate. Kinetic studies also revealed hysteretic behavior of the enzyme. The electron density map of the full-length structure showed two novel densities on the molecular 2-fold axis into each of which a molecule of cadavarine could be fitted. Docking studies revealed a ligand-binding site at the inter-subunit interface between the two observed densities which might represent a potential allosteric site. Combining the structural and kinetic results, a possible morpheein model of allosteric regulation of StASL was hypothesized. Chapter 4 deals with the crystallographic and kinetic investigations on ASL from Mycobacterium smegmatis and Mycobacterium tuberculosis. MsASL and MtbASL were cloned, purified and crystallized. The X-ray crystal structure of MsASL was determined at 2.16 Å resolution. It is the first report of an apo-ASL structure with a partially ordered active site C3 loop. Diffracting crystals of MtbASL could not be obtained and a model for its structure was derived using MsASL as a template. Most of the active site residues were found to be conserved with the exception of Ser 148 and Gly 319 of MsASL. Ser 148 is structurally equivalent to a threonine in most other ASLs. Gly 319 is replaced by an arginine residue in most ASLs. The two enzymes were catalytically much less active when compared to ASLs from other organisms. Arg319Gly substitution and reduced flexibility of the C3 loop might account for the low catalytic activity of mycobacterial ASLs. The low activity is consistent with the slow growth rate of Mycobacteria, their high GC containing genomes as well as with their dependence on other salvage pathways for the supply of purine nucleotides. Chapter 5 deals with the identification of the catalytic residues important for ASL catalysis in view of the earlier conflicting reports on the identity of these residues. pH-dependent kinetic studies were performed on full-length StASL. The theory behind these studies is also described in this chapter. Two residues with pKa values of 6.6 and 7.7 were identified as essential for the enzymatic activity. These results were interpreted along with structural comparison of MsASL and other superfamily enzymes with ordered C3 loops. They suggest that His 149 and either Lys 285 or Ser 279 of MsASL are the residues most likely to function as the catalytic acid and base, respectively. The final Chapter 6 of the thesis deals with the structural and dynamic studies carried out on Sesbania mosaic virus (SeMV) protease. The chapter begins with a general introduction to viruses, followed by a brief summary of SeMV. The goal of this study is to understand the interactions between the protease and VPg at a structural level using the information available from biochemical studies. Crystallographic studies initiated for the mutant H275APro and Y315APro were unsuccessful due to the insolubility of the proteins. Co-crystallization or soaking experiments of wild type protease with cognate peptides were unsuccessful due to the inability of the enzyme to bind to its substrates in the absence of VPg. Higher resolution structure of wild type protease did not yield any new insights when compared to the earlier reported structure determined at a lower resolution. In the absence of structural insights, molecular dynamic simulations were carried out on wild type protease structure and in silico generated mutants using GROMACS package. The studies showed the importance of flipping of residue Phe 301 and opening-closing of the loop region corresponding to residues 301-308 for the catalytic mechanism. The thesis concludes with Future perspectives of the various studies carried out on ASL and SeMV protease. The atomic coordinates determined from the work presented in this thesis have been deposited in the PDB and the assigned PDB codes are reported in the respective chapters. Publications cited in the thesis are listed in the Bibliography section.

H. pylori has a panmictic population structure due to high genetic diversity. The homoplasy index for H. pylori is 0.85 (where 0 represents a completely clonal organism and 1.0 indicates a freely recombining organism) which is much higher than homoplasy index for E. coli (0.26) or naturally competent Neisseria meningitides (0.34). It undergoes both inter as well as intra strain transformation. Intergenomic recombination is subject to strain specific restriction in H. pylori. Hence, a high homoplasy index means that competence predominates over restriction in H. pylori. Annotation of the genomes of H. pylori strains 26695 and J99 show the presence of nearly two dozen R-M systems out of which 16 were postulated to be Type II for J99. H. pylori has been described to be an ideal model system for understanding the equilibrium between competing tension of genomic integrity and diversity (42). R-M systems allow some degree of sexual isolation in a population of competent cells by acting as a barrier to transformation. The mixed colonizing population of H. pylori has a polyploidy nature where each H. pylori strain adds to ‘ploidy’ of the colonizing population. Maintenance of polyploidy nature of mixed colonizing population in a selective niche of stomach needs a barrier to free gene flow. Restriction barrier maintains a polyploidy nature of H. pylori population which is considered as yet another form of genetic diversity helping in persistence of infection. Thus, according to the model proposed by Kang and Blaser, where H. pylori are considered as perfect gases like bacterial population, transformation and restriction both add to genetic diversity of the organism. Again, restriction barriers are not completely effective, which could be due to cellular regulation of restriction system. Thus, a perfect balance between restriction and transformation in turn regulates the gene flow to equilibrate competition and cooperation between various H. pylori strains in a mixed population. RecA, DprA and DprB have been shown to be involved in the presynaptic pathway for recombination substrates brought in through the Com system. Biochemical characterization of HpDprA, during this study revealed its ability to bind to ssDNA and dsDNA. Binding of HpDprA to both ssDNA as well as dsDNA results in large nucleoprotein complex that does not enter the native PAGE. However, DNA trapped in the wells could be released by the addition of excess of competitor DNA, illustrating that the complex are formed reversibly and do not represent dead-end reaction products. Transmission electron microscopy for SpDprA interaction with ssDNA established that a large nucleoprotein complex consisting of a network of several DNA molecules bridged by DprA is formed which is retained in the well. A large DNA-protein complex that sits in the well has also been observed with other DNA binding proteins like RecA. It has been observed for ssDNA binding protein (SSB) that they bind non-specifically to dsDNA under low salt condition (20 mM NaCl) in the absence of Mg2+. The non specific binding of SSB to dsDNA was prevented under high salt conditions (200 mM NaCl) or in the presence of Mg2+. HpDprA interaction with both ssDNA and dsDNA was stable under high salt condition (200 mM NaCl) and in the presence of Mg2+ indicating that these interactions are specific. The interaction of HpDprA with dsDNA is significant since dsDNA plays an important role in natural transformation of H. pylori. The pathway of transformation by dsDNA is highly facilitated (nearly 1000 fold) as compared to ssDNA. However, dsDNA is a preferred substrate for REases which are a barrier to horizontal gene transfer. This implies that the decision of ‘restriction’ or ‘facilitation for recombination’ of incoming DNA might be taken before the conversion of dsDNA into ssDNA. The incoming DNA has been shown to be in the double-stranded form in periplasm and in single-stranded form in cytoplasm. Hence, the temporal and spatial events surrounding endonuclease cleavage remain to be understood. Taken together, these results suggest a very important role of dsDNA in natural transformation in H. pylori. Hence, binding and protection of dsDNA by HpDprA is possibly of crucial importance in the success of natural transformation process of the organism. DprA is characterized by presence of a conserved DNA binding domain. The DNA binding domain adopts a Rossman fold like topology spanning most region of the protein. Rossman fold consists of alternating alpha helix and beta strands in the topological order of β-α-β-α-β. It generally binds to a dinucleotide in a pair as a single Rossman fold can bind to a mononucleotide only. All homologous DprA proteins characterized till date show that in addition of the prominent Rossman fold domain they consist one or more smaller domains. RpDprA consists two more domains other than the Rossman fold domain i.e., N- terminal SAM (sterile alpha motif) domain and a C-terminal DML-1 like domain. SpDprA consist of an N-terminal SAM domain other than Rossman fold domain. While the main function of Rossman fold is to bind DNA, the supplementary domains are highly variable in sequences and functions. For example, the SAM domain in S. pneumoniae plays a key role in shut-off of competence by directly interacting with ComE~P. HpDprA consist of an N-terminal Rossman fold domain and a C-terminal DML-1 like domain. Both these domains are found to be prominently α-helical in nature. Amino acid sequence analysis of the protein suggests that NTD is basic and CTD is acidic in nature. NTD is sufficient for binding with ssDNA and dsDNA, while CTD plays an important role in formation of higher order polymeric complex with DNA. For HpDprA and SpDprA, dimerization site was mapped in Rossman fold domain. Gel filtration data revealed an important observation that HpDprA can exist as a monomer (dominant species at lower concentration) as well as a dimer (dominant species at higher concentration) in solution. However, the exchange between these two forms is very fast resulting in a single peak of elution. Since, HpDprA binds to DNA in dimeric form, the dimer species will be favoured in presence of DNA. Hence, even at lower concentrations HpDprA will be mainly a dimer in presence of DNA. Interestingly, both domains of HpDprA i.e., NTD and CTD were able to form dimers but no higher oligomeric form. On the other hand, HpDprA was seen to form oligomeric forms higher than dimer in gluteraldehyde cross linking assay. The strength of CTD dimer was much lower that NTD dimer, therefore it could be proposed that there are two sites of interaction present in HpDprA - a primary interaction site (N-N interaction) and a secondary interaction site (C-C interaction). The N-N interaction is responsible for dimer formation but further oligomerization of HpDprA necessitates the interaction of two dimers using C-C interaction site. It was shown that NTD binds to ssDNA but forms lower molecular weight complex. SPR analysis of DprA and NTD – DNA interaction pointed out that deletion of CTD leads to faster dissociation of the protein from DNA. Concomitantly, reduction in binding affinity was observed for both ss and ds DNA upon deletion of CTD from full length protein. These results suggest that CTD does play an important role in interaction of full length HpDprA with DNA. Two possible roles of CTD were proposed by Wang et al (2014) group to explain their observation of formation of lower molecular weight complex in absence of CTD. (i) CTD possesses a second DNA binding site but much weaker than site present in NTD. (ii) CTD is not involved in DNA binding but mediates nucleoprotein complex formation through protein – protein interaction. EMSA and SPR analysis with purified CTD protein confirmed that there is no secondary DNA binding site present in CTD. As discussed above, it was observed that CTD can mediate interaction between two HpDprA through C-C interaction. Since the interaction is weaker it is lesser likely to be responsible for dimer formation but in trimer or higher oligomeric form of HpDprA, the presence of N-N interaction will facilitate and stabilize C-C interaction. These observations together bring forward an interesting model for HpDprA – DNA interaction. HpDprA forms dimer through N-N interaction (favourably in presence of DNA) and many HpDprA dimers bind to DNA owing to their high affinity and sequence independent nature of binding. These dimers interact with each other through C-C interaction resulting in higher molecular weight nucleoprotein complex. HpDprA - DNA complex formation is slower than NTD – DNA complex but the former one is more stable (Fig. 2). According to the above proposed model there are two binding events (DNA – protein and protein – protein) in case of HpDprA – DNA complex formation and hence it would take longer time than NTD-DNA complex formation which involves only one binding event. But the resulting higher order complex with HpDprA – DNA would be much more stable. NTD is able to offer equally efficient protection from nuclease to ssDNA and dsDNA (Fig. 7). This shows that NTD alone is sufficient to completely coat single molecule DNA. AFM images confirm the difference in binding pattern of HpDprA full length protein and NTD. As can be seen in Fig. 8F, NTD binds a DNA molecule by entirely occupying all the available space but forms nucleoprotein filaments isolated from each other. In contrast to full length HpDprA, which forms tightly packed, condensed, extensively cross linked polynucleoprotein complexes, NTD forms much thinner complexes with DNA. In the electron micrographs of SpDprA – DNA complex, extensive cross filament interaction was observed resulting in a dense molecular aggregate. Similar kinds of complexes with DNA were also observed for Bacillus subtilis DprA in atomic force microscope images. Thus, it could be proposed that HpDprA binds to a single DNA molecule (single strand or double strand) mainly as a dimer formed through N-N interaction. Such multiple individual nucleoprotein filaments come together and interact with each other through C- C interaction resulting in dense and intricate poly – nucleoprotein complex. HpDprA is proposed to undergo conformational changes from closed state to open state in presence of ssDNA. In agreement with this, structural transition (resulting in reduction of α-helicity of the protein) was observed in presence of ssDNA. Similar structural transitions were observed for dsDNA indicating possibly a common mode of interaction for both forms of DNA. Further, mutation of the residues shown to be involved in binding ssDNA from crystallographic data, resulted in decrease of binding affinity with dsDNA as well. The fold reduction in binding affinity of dsDNA was lower than that for ssDNA despite that it is obvious that the same positively charged pocket which is primarily involved in ssDNA interaction is also responsible (atleast partially) for binding with dsDNA. However, the residues crucial for interaction with these two forms of DNA may be different. Both DprA and R-M systems have been shown to have presynaptic role in natural transformation process. While DprA has a protective role, R-M systems have an inhibitory role for incoming DNA suggesting a functional interaction between them. Results of this study show that HpDprA interacts with dsDNA, inhibits Type II restriction enzymes from acting on it and at the same time stimulates the activity of MTases resulting in increased methylation of bound DNA. This observation is of significance from the view of genetic diversity as the only way a bacterial cell discriminates between self and nonself DNA is through the pattern of methylation. Binding of HpDprA to incoming DNA inhibits its access to restriction endonucleases but not to methyltransferases. As a result DNA will be methylated with the same pattern as that of the host cell. Hence, it no longer remains a substrate for restriction enzymes. HpDprA thus, effectively alleviates the restriction barrier. However, it remains to be understood as to how DNA in complex with HpDprA, while not accessible to REases or other cellular nucleases, is accessible to a MTase? A possible explanation could be that HpDprA interacts with MTase and recruits it on DNA. It has been shown that there is a overlap between DprA dimerization and RecA interaction interfaces and in presence of RecA, DprA-DprA homodimer is replaced with DprA-RecA heterodimer allowing RecA nucleation and polymerization on DNA followed by homology search and synapsis with the chromosome. A similar scenario can be thought for interaction of HpDprA with the MTase. R-M systems play an important role in protection of genomic DNA from bacteriophage DNA. Hence, downregulation of restriction barrier by HpDprA may not be desirable by host during the entire life cycle. Therefore, the expression of HpDprA, which is ComK dependent and that which takes place only when competence is achieved is noteworthy. In H. pylori, DNA damage induces genetic exchange via natural competence. Direct DNA damage leads to significant increase in intergenomic recombination. Taken together it can be proposed that when genetic competence is induced, R-M systems are down regulated to allow increased genetic exchange and thus, increasing adaptive capacity in a selective environment of stomach. There is an evolutionary arms race between bacterial genomes and invading DNA molecules. R-M systems and anti-restriction systems have co-evolved to maintain an evolutionary balance between prey and predator. Phages and plasmids employ anti-restriction strategies to avoid restriction barrier by a) DNA sequence alteration, b) transient occlusion of restriction sites and c) subversion of restriction-modification activities. DNA binding proteins have been shown to bind and occlude restriction sites. On the other hand, λ Ral protein alleviates restriction by stimulating the activity of Type IA MTases. The observations of MTase stimulation and site occlusion of restriction sites by HpDprA appears to be analogous to anti restriction strategies, otherwise employed by bacteriophages. Thus, DprA could be a unique bacterial anti-restriction protein used by H. pylori for downregulating its own R-M systems to maintain the balance between fidelity and diversity. In conclusion, HpDprA has unique ability to bind to dsDNA in addition ssDNA but displays higher affinity towards ssDNA. Binding of HpDprA to DNA results in a compact complex that is inert to the activity of nucleases. A novel site of oligomerization for HpDprA was observed which suggests the role of C-C interaction in inter-nucleoprotein filament interaction. It would be interesting to further study the effects of CTD deletion on the transformation efficiency of H. pylori, to understand these mechanisms better. It has been well demonstrated that R-M systems offer a barrier to incoming DNA, but our understanding of the regulation of R-M systems has been poor. While other factors like regulation of cellular concentration of restriction enzymes and conversion of dsDNA into ssDNA might play crucial roles in striking the perfect balance between genome diversity and integrity, one of the factors that regulate R-M systems could be DprA.

Bovine mastitis adversely affects the dairy industry around the world. This disease is caused by a diverse range of bacteria, broadly categorised as minor and major pathogens. In-line tools that help identify these bacterial groupings in the early stages of the disease are advantageous as timely decisions could be made before the cow develops any clinical symptoms. The first objective of this research was to identify the most informative milk parameters for the detection of minor and major bacterial pathogens. The second objective of this research was to evaluate the potential of supervised and unsupervised neural network learning paradigms for the detection of minor infected and major infected quarters in the early stages of the disease. The third objective was to evaluate the effects of different proportions of infected to non-infected cases in the training data set on the correct classification rate of the supervised neural network models as there are proportionately more non-infected cases in a herd than infected cases. A database developed at Lincoln University was used to achieve the research objectives. Starting at calving, quarter milk samples were collected weekly from 112 cows for a period of fourteen weeks, resulting in 4852 samples with complete records for somatic cell count (SCC), electrical resistance, protein percentage, fat percentage, and bacteriological status. To account for the effects of the stage of lactation on milk parameters with respect to days in milking, data was divided into three days in milk ranges. In addition, cow variation was accounted for by the sire family from which the cow originated and the lactation number of each cow. Data was pre-processed before the application of advanced analytical techniques. Somatic cell score (SCS) and electrical resistance index were derived from somatic cell count and electrical resistance, respectively. After pre-processing, the data was divided into training and validation sets for the unsupervised neural network modelling experiment and, for the supervised neural network modelling experiments, the data was divided into training, calibration and validation sets. Prior to any modelling experiments, the data was analysed using statistical and multivariate visualisation techniques. Correlations (p<0.05) were found between the infection status of a quarter and its somatic cell score (SCS, 0.86), electrical resistance index (ERI, -0.59) and protein percentage (PP, 0.33). The multivariate parallel visualisation analysis validated the correlation analysis. Due to significant multicolinearity [Correlations: SCS and ERI (-0.65: p<0.05); SCS and PP (0.32: p<0.05); ERI and PP (-0.35: p<0.05)], the original variables were decorrelated using principle component analysis. SCS and ERI were found to be the most informative variables for discriminating between non-infected, minor infected and major infected cases. Unsupervised neural network (USNN) model was trained using the training data set which was extracted from the database, containing approximately equal number of randomly selected records for each bacteriological status [not infected (NI), infected with a major pathogen (MJI) and infected with a minor pathogen (MNI)]. The USNN model was validated with the remaining data using the four principle components, days in milk (DIM), lactation number (LN), sire number, and bacteriological status (BS). The specificity of the USNN model in correctly identifying non infected cases was 97%. Sensitivities for correctly detecting minor and major infections were 89% and 80%, respectively. The supervised neural network (SNN) models were trained, calibrated and validated with several sets of training, calibration and validation data, which were randomly extracted from the database in such a way that each set has a different proportion of infected to non-infected cases ranging from 1:1 to 1:10. The overall accuracy of these models based on validation data sets gradually increased with increase in the number of non-infected cases in the data sets (80% for the 1:1, 84% for 1:2, 86% for 1:4 and 93% for 1:10). Specificities of the best models for correctly recognising non-infected cases for the four data sets were 82% for 1:1, 91% for 1:2, 94% for 1:4 and 98% for 1:10. Sensitivities for correctly recognising minor infected cases for the four data sets were 86% for 1:1, 76% for 1:2, 71% for 1:4 and 44% for 1:10. Sensitivities for correctly recognising major infected cases for the four data sets were 20% for 1:1, 20% for 1:2, 30% for 1:4 and 40% for 1:10. Overall, sensitivity for the minor infected cases decreased while that of major infected cases increased with increase in the number non-infected cases in the training data set. Due to the very low prevalence of MJI category in this particular herd, results for this category may be inconclusive. This research suggests that somatic cell score and electrical resistance index of milk were the most effective variables for detecting the infection status of a quarter followed by milk protein and fat percentage. The neural network models were able to differentiate milk containing minor and major bacterial pathogens based on milk parameters associated with mastitis. It is concluded that the neural network models can be developed and incorporated into milking machines to provide an efficient and effective method for the diagnosis of mastitis.