Restriction Modification (R-M) systems prevent the invasion of foreign genetic material into bacterial cells and are therefore important in maintaining the integrity of the host genome. The spread of antibiotic resistance, which is proposed to occur via the transfer of foreign genes to the bacterial genome, makes the subject of R-M systems extremely relevant. R-M systems are currently classified into four types (I to IV) on the basis of differences in composition, target recognition, cofactors and the manner in which they cleave DNA. Kennaway et al (2012) proposed that there is an evolutionary link between Types I and II. Comparing the structures of examples from two of the subfamilies of Type II systems (IIB and IIG) to those of Type I structures, similarities can be observed. Due to the fact that Type II R-M systems cut DNA at fixed positions, they can be used to obtain genetic material selectively. They have therefore proven to be invaluable in molecular biology. One aspect of this project aims to create a novel R-M system, a pseudo-Type II system, by removing the molecular motors from the restriction subunit of a Type I system and fusing the remaining nuclease domain to a known Type I methyltransferase (MTase). This will not only provide evidence to support the theory that evolution has produced a pared down form of the Type I systems in the Type II systems, but it may also become a useful biological tool. This thesis describes the several attempts at doing this and how the subsequent constructs were expressed, purified and assayed to varying degrees of success. An important characteristic of the Type I systems is their ability to methylate DNA, and it is the mechanism via which host DNA is protected from restriction. This is another subject investigated in this project. As with the nuclease activity of the Type I systems, the site at which DNA is methylated is dictated by the HsdS subunit. It is described here how this subunit can be altered to change the sequence of DNA that is recognised by the system. Again, using Type II system subtypes as a reference, various mutations were made to the HsdS subunit of an MTase from Staphylococcus aureus. This is in an effort to bring about a new mode of action, but also to provide further evidence for an evolutionary link between the two system types. The HsdM and HsdS subunits are expressed from two separate genes at the same locus. There is a frameshift between the genes where the start of the hsdS gene occurs a few base pairs upstream from the stop codon of the hsdM gene. This work shows that removing this frameshift creates an MS fusion product, and in vivo studies show that this product has methylase activity and can form an active restriction complex when the HsdR subunit is added. The product can also be over-expressed and purified, and shows in vitro restriction activity on addition of the HsdR subunit protein. The HsdS subunit is composed of two target recognition domains (TRDs), each dictating one part of the bipartite recognition sequence. These TRDs can be altered, bringing about a change in the sequence of DNA recognised by the enzyme. In this thesis, it is shown that the C-terminal TRD can be removed and that the subsequent “Half S” enzyme possesses both methylase and restriction activity in vivo and that its recognition sequence is different from that of the wild-type enzyme. After the successful creation of both “MS fusion” and “Half S” recombinant proteins of the Sau1, Type I system from a CC398 strain of Staphylococcus aureus, a further construct was produced. This possesses both in vivo and in vitro activity. The novel “M Half S Fusion” enzyme not only links the two aspects of this project but also creates a structure similar to some seen in the Type II systems. This shows that the Type I systems can be manipulated to change their mode of action but also supports the idea that Types I and II are evolutionarily linked. By making the alterations in a step-wise fashion identifies that these structural changes can create viable enzymes, and that they could have occurred through the process of evolution.

Bacterial type I restriction-modification systems are composed of three different subunits: one HsdS subunit is required for identification of target sequence and anchoring the enzyme complex on DNA; two HsdM subunits in the methyl-transferase complex serve for host genome modification accomplishing a protective function against self-degradation; two HsdR (or motor) subunits house ATP-dependent translocation and consequent cleavage of double stranded DNA activities. The crystal structure of the 120 kDa HsdR subunit of the Type I restriction-modification system EcoR124I in complex with ATP was recently reported. HsdR is organized into four approximately globular structural domains in a nearly square-planar arrangement: the N-terminal endonuclease domain, the RecA-like helicase domains 1 and 2 and the C-terminal helical domain. The near-planar arrangement of globular domains creates prominent grooves between each domain pair. The two helicase-like domains form a canonical helicase cleft in which double-stranded B-form DNA can be accommodated without steric clash. The helical domain, probably involved in complex assembly, exhibits only a few specific interactions with helicase 2 domain. Molecular mechanism of dsDNA translocation, cleavage and ATP hydrolysis has not been yet structurally investigated. Here we propose a translocation cycle of the restriction-modification system EcoR124I based on analysis of available crystal structures of superfamily 2 helicases, strutural modeling and complementary biochemical characterization of mutations introduced in sites potentially inportant for translocation in the HsdR motor subunit. Also a role of the extended region of the helicase motif III in ATPase activity of EcoR124I was probed.

Restriction modification (R-M) systems are important components of the prokaryotic arsenal against invading genomes. R-M systems directly target the foreign DNA and are often considered as primitive immune systems in bacteria. The defense system comprises of two contrasting enzymatic activities – a restriction endonuclease (REase) and a methyltransferase (MTase). Functionally, REases cleave a specific DNA sequence endonucleolytically at the phosphodiester bonds generating 5' or 3' overhangs or blunt ends. MTases catalyze the transfer of a methyl group from S-adenosyl-Lmethionine to adenine or cytosine. Four types of R–M systems are found in bacteria, viz., Types I, II, III and IV. Type II R-M systems, comprising of a separate REase and MTase, are the most abundant and well-studied enzymes. Type II REases recognize and cleave DNA within or near their recognition sequences. Surprisingly, these enzymes share little or no sequence homology amongst them. All the enzymes identified so far can be grouped into conventional PD-(D/E)XK, ββα-Me, GIY-YIG, phospholipase-derived and half-pipe endonucleases according to their folds and active site structures. Owing to their high specificity and defined cleavage pattern, they have become indispensable tools in molecular biology and have been widely exploited for studying protein–DNA interactions. The work presented in this thesis deals with R.KpnI, which belongs to the HNH superfamily of nucleases and is characterized by the presence of a ββα-Me finger motif. The REase isolated from Klebsiella pneumoniae recognizes the palindromic DNA sequence GGTAC/C and cleaves DNA as indicated. The enzyme is unique in exhibiting promiscuous DNA cleavage in the presence of Mg2+, a natural co-factor for a vast majority of REases. Surprisingly, Ca2+ and Zn2+ completely suppress the Mg2+ mediated promiscuous activity and induce high fidelity cleavage. These unusual features of R.KpnI led to the functional characterization of the ββα-Me finger active site motif. In addition, the studies were aimed at understanding the mechanism and the biological significance of substrate and co-factor promiscuity exhibited by the enzyme. The salient aspects of the thesis are summarized below. A general introduction and overview of the literature on structure-function studies, mechanism of recognition and catalysis by REases with special emphasis on Type II enzymes is presented in the Chapter 1. An account of co-factor specificity in REases, role of metal ions in DNA binding as well as in phosphodiester bond hydrolysis is provided. The various aspects of R-M systems that target the invading DNA elements and counter strategies employed by the foreign genomes to evade the restriction are also covered. The new developments that provide insights in understanding the diversity of R-M systems and additional biological roles that could increase the fitness of the host organism harboring them are described. The features of substrate and metal ion specificity in REases and the efforts undertaken to alter the specificity have been dealt at the end of the chapter. From the structures of the several ββα-Me finger nucleases, the α-helix has been implicated in providing a structural scaffold for the correct juxtapositioning of the catalytic residues. However, no mutagenesis data exists to delineate its role. Homology modeling studies of R.KpnI suggested a crossover structure for the α-helix of the ββαMe finger active site motif, which could possibly form dimeric interface and/or structural scaffold for the active site. Chapter 2 describes the computational modeling and mutational analysis performed to understand the role of the residues present in this α-helix in intersubunit interactions and/or stabilization of the active site. Mutation of the residues present in the α-helix lead to the loss of the enzyme activity, but not dimerization ability. Subsequent biophysical experiments showed that the α-helix of the ββα-Me finger of R.KpnI plays an important role for the stability of the protein–DNA complex needed for its function. In Chapter 3, unusual co-factor flexibility for R.KpnI is shown by using a battery of divalent metal co-factors differing in ionic radii and coordination geometries. A number of alkaline earth and transition group metal ions function as co-factors for DNA cleavage. The metal ions replaced each other readily from the enzyme’s active site revealing the active site plasticity. Mutation of the invariant His residue of the HNH motif caused abolition of the enzyme activity with all the co-factors indicating that the enzyme follows single metal ion mechanism for DNA cleavage. The indispensability of the invariant His in nucleophile activation together with the broad co-factor tolerance of the enzyme indicated the role of metal ions in electrostatic stabilization during catalysis. At higher concentrations, Mg2+, Mn2+ or Co2+ stimulate promiscuous cleavage while Cd2+, Ni2+ or Zn2+ inhibit phosphodiester bond hydrolysis. The underlying molecular mechanisms for the modulation of the enzyme activity by the metal ion binding to the second site are presented. Regulation of the endonuclease activity and fidelity by a second metal ion binding is a unique feature of R.KpnI among REases and HNH nucleases. The identification of additional metal ion binding residues would help in engineering REase variants with enhanced activity and/or specificity. Chapter 4 describes the generation of an R.KpnI variant with altered co-factor specificity by exploiting the active site plasticity of the enzyme. The mutant enzyme is a Mn2+ -dependent endonuclease defective in DNA cleavage with Mg2+ and other divalent metal ions. In the engineered mutant, only Mn2+ is selectively bound at the active site, imparting in vitro activity while being dormant in vivo. In addition to the Mn2+ selectivity, the mutant is impaired in concerted double-stranded DNA cleavage leading to the accumulation of nicked intermediates. The nicking activity of the mutant enzyme is further enhanced by altering the reaction conditions. Thus, a single point mutation in the active site of R.KpnI generates a Mn2+ -dependent REase and a sequence specific nicking endonuclease. The potential applications of such enzymes engineered for selective metal ion dependent activities have been discussed. R.KpnI is peculiar in retaining robust promiscuous cleavage despite being a typical Type II REase in all other characteristics. Chapter 5 presents results of the growth properties and phage titer analysis carried out with R.KpnI and its high fidelity variant to understand the biological significance of promiscuous activity. The enzyme isolated from the K. pneumoniae exhibited biochemical properties similar to that of R.KpnI overexpressed in E.coli. It was observed that the wild type but not the high fidelity variant could effectively restrict bacteriophages methylated at GGTACC. These results show that the REase exhibits promiscuous activity in vivo, which would be advantageous for the organism to better target the incoming foreign DNA. The promiscuous behavior of the R.KpnI could be one of the counter strategies employed by the bacteria against the constantly evolving phages in the co-evolutionary arms race. In conclusion, the work described in this thesis provides new insights about structure, function and biology of REases in general and R.KpnI in particular. The co-factor and substrate promiscuity of R.KpnI may indicate its evolutionarily intermediate form that is yet to attain a high degree of specificity. Alternatively, it is possible that this unique feature is retained during the evolution of the HNH REases serving some unknown function(s) in the cell, in addition to having an edge in countering the phage infections.

Enzyme DNA methylation is an important biochemical process that imprints DNA with additional information. DNA methylation is catalyzed by S-adenosyl-L-methionine (AdoMet)-dependent methyltraferases (MTases). Prokaryotic DNA MTases are usually components of restriction-modification(R-M) systems that enable cells to resist propagation of foreign genomes that would otherwise kill them. Based on the position methyl group transfer on the bases in DNA, MTases are classified into two groups-exocyclic or amino MTases and endocyclic or ring MTases. The amino MTases methylate exocyclic amino nitrogen to form either N6-methyladenine or n4-methycytosine. N6-methyaladenine is mostly found in the genomes of bacteria, archaea protists and fungi. Helicobacter pylori is a gram-negative, flagellated, fastidious bacterium that colonizes the highly acidic environment of the gastric mucosa. Frequently and persistence of H.paylori infection in humans make it attractive model for studying the host- pathogen interaction mechanisms. Analysis of the genome sequence of H.pylori strains 26695, J99.HPAGI, and G27 revealed an abundance of restriction-modification (R-M) systems. Most of the R-M system genes are either conserved among the strains or specific to each strain. Strain specific genes are responsible for different phenotypes in several host adapted pathogens such as H.pylori. Many of the R-M gene homologues exhibit different usages of condon bias and lower G+C content from the average genes suggesting horizontal transfer of the R-M system genes in H. Pylori. Genome analysis of strain 26695 showed the presence of three putative type III R-M systems and hp0592-hp0593 constitutes one such type III R-M system. Based on the conserved motif arrangements, HP0593 MTases belongs to the subgroups of MTases. The amino acid sequence of HP0593 MTases has 38% sequence identity to Ecop11 MTases and EcoP151 MTase, both of which belongs to type IIIR-M systems therefore, it was important to study in detail previously unexplored role of this putative type III DNA MTase (HP0593) in H. Pylori. Investigation of methyltransferease activity and sequence specifically of putative DNA adenine MTase (HP0593) HP0593 (N6-adenine) - DNA MTase is a member of a type III R-M system in H. pylori strain 26695. HP0593 MTase has been cloned, over expressed and purified heterologously in Escherichia coli. Matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) was carried out with purified HP0593 and profile showed a single peak with expected molecular mass of 70.6kDa. The protein was determined as-5.8. HP0593 MTase exits predominantly as monomer and a small fraction as dimer in solution as determined by size exclusion chromatography and glutaraldehyde cross-linking studies. The recognition sequence of the purified MTase was determined as 5’GCAG-3’ and the target base of methylation is adenine. Dot-blot assay using antibodies that reacted specifically with DNA containing m6A modification confirmed that HP0593 MTase is an adenine specific MTase. Exocyclic MTase have a conserved catalytic motif (D/N/S/SPPY/F/W). Most interestingly, the amino acid sequence analysis of HP0593 MTase revealed the presence of a PCQ-like motif, which is the catalytic motif for C5-cytosine MTase in addition to DPPY motif. In order to check the role of both these MTase by glycine. HP0593 –Y107G and C54G mutant proteins were purified to near homogeneity. It was found that the Y107G mutant protein was catalytically inactive as compared to wild-type HP0593 MTase. On the other hand the C54G mutant protein was found to be as active as the wild-type HP0593 MTase indicating that HP0593 MTase is an adenine MTase and not a C5- cytosine MTase. Kinetic and catalytic properties of HP0593 DNA adenine methyltransferase DNA binding studies were carried out by electrophoretic mobility shift assay using DNA having cognate site and either in absence or presence of AdoHcy or sinefungin. In all the three cases two different DNA-protein complexes were observed-a fast running complex I and a slow running complex 2. It can be surmised that the fast running complex could be HP0593 monomer-DNA and the slow running complex could be a HP0593 dimer-DNA complex. With non specific DNA (lacking 5’-GCAG-3’ sequences) no complexes were formed even in the presence of cofactors. Based on the above observations it is suggested that a specific interactions of HP0593 MTase with DNA occurs on cognate recognition site. The activity of HP0593 MTase is optional at pH 5.5. This is a unique property in context of natural adaptation of H. pylori in its acidic niche. When initial velocities were plotted against varying concentrations of duplex DNA having a single 5’GCAG-3’ site a rectangular hyperbola was obtained confirming that HP0593 MTase obeys michaelis menten kinetics. From non-linear regression analysis of the plot of initial velocity versus DNA concentration Km (DNA) and kcat were calculated. Analysis of initial velocity with AdoMet as a substrate showed that two molecules of AdoMet bind to HP0593 MTase. The nonlinear dependence of methylation activity on enzyme concentration indicated that more than one molecule of methylation activity on enzyme concentration indicated that more than one molecule of enzyme is required for its activity. Metal ion cofactors such as CO 2, Mn2+ and Mg2+ stimulated the HP09593 MTase activity. As Mn2+ showed maximum stimulation of methyaltion activity compared to other metal ions, surface plasmon resonance spectroscopy was used to determine the kinetics of DNA binding by HP0593 MTase in the absence and presence of Mn2+. In the presence of Mn2+, HP0593 MTase showed~1000-fold increase in affinity to duplex DNA. DNA MTase bind substrates in random or sequential order. Preincubation study demonstrated that the preformed enzyme-DNA complex is competent than the preformed enzyme-AdoMet complex. This suggests that MTase binds to DNA first followed by AdoMet. Isotope partitioning analysis indicated that HP0593 MTase shows a distributive mechanism of methylation DNA having more than one recognition site. Effects of inactivation of HP0593 DNA MTase in Helicobacter pylori 26695 strain and its functional role. DNA dot-blot assay using hp0593 gene specific primer showed that this gene is present in 25.15% of the clinical strains checked suggesting that hp0593 is strain-specific gene. Strain-specific genes in many host-adapted pathogene impart strain specific phenotype. Wild-type 26695 strain grew slightly faster at the initial phase of growth in PH 4.5 compared to pH 7.4. A~5-fold enhanced level of hp0593 mRNA expression was growth under acidic condition HP0593 MTase could play an important role in H. pylori physiology through methylation. To elucidate the possible role(s) played by the MTase in H.pylori physiology, an hp0593 knock-out in 26695 strain was generated by chloramphenecol cassette mediated insertional gene inactivation. Growth kinetic study was carried out with both wild-type and hp0593 knock-out strain at pH7.4, the growth of the hp0593 strain. At pH 4.5 no major differences were observed in the growth compared to the wild-type hp0593 knock-out strain. To further investigate the effect of the knock-out, cell-morphology study was carried out after growing the strains at pH 7.4 till mid-exponential phase. Transmission electron microscopy studies reveled changes in cell shape, presence of sheathed structure and production of outer membrane vesicles (OMVs) in the hp0593 knock out strain. OMVs contain effectors molecules during infection helps in pathogenicity caused by H.pylori.This is the first report where inactivation of DNA MTase causes shedding of vesicles. OMVs are also known to modulate the production of IL-8 by gastic epitheial cells. To check weather H.pylori strains could produce IL-8, both wild-type and hp0593 knock-out strains were co-cultured with AGS cell infected with the hp0593 knock out strain. This was further confirmed by semi-quantitative RT-PCR analysis. To analyze the different phenotypes observed in the hp0593 knock-out strain, transcriptome profile were compared by microarray and RT-PCR analysis. In thehp0593 knock-out strain peptidologlycan and murein synthesis genes like pbp2, murC and neu4 showed upregulation which could be responsible for the changes in cell shape presence of sheathed structure and OMVs production. The RT-PCR data showed ~9-fold down-regulation of dank chaperone which might play a key role in slow growth phenotype in the hp0593 knock-out strain. Considering the occurrence of GCAG sequence in the potential promoter regions of physiologically important genes such as dank, neuA, murC, fliH, filP and cag5, the results presented in this study provide impetus for exploring the role of HP0593 DNA MTase in the cellular processes of H.pylori. However, R-M systems are not absolutely essential, but different methylation patterns may contribute to strain-specific epigenetic gene regulation and may contribute to variability among the strains.