DNA repair processes are crucial for mutation avoidance and the maintenance of genetic integrity in all organisms. Organisms rely on repair processes to combat genotoxic stress imposed by hostile host environment, and sometimes by therapeutic agents. Most pathogens rapidly generate genetic variability to acquire increased virulence and evade host immune response. Therefore, there needs to exist a fine balance between mutation avoidance and fixation, which is perhaps regulated by repair processes. Haemophilus influenzae and Helicobacter pylori contribute significantly to morbidity and mortality caused by bacteria worldwide. H. influenzae is an obligate commensal of upper respiratory tract with the potential to cause a variety of diseases in humans like meningitis and respiratory infections. H. pylori, which inhabits the human stomach, is associated with gastric and duodenal ulcers and cancerous gastric lesions. One of the striking differences between these two genetically diverse bacterial species is the absence of recognized DNA mismatch repair (MMR) pathway homologs in H. pylori. MMR is a highly conserved post-replicative process, which corrects base pairing mismatches and small loops arising during DNA replication and recombination due to misincorporated nucleotides, insertions, and deletions. Defective MMR results in increased mutation frequency that can alter the pathogenic potential and antibiotic resistance of pathogens. MMR has been extensively studied in Escherichia coli, and requires an orchestrated function of different proteins like MutS, MutL, MutH, UvrD, SSB, RecJ, ExoVII, ExoI, ExoX, beta-clamp, DNA polymerase III and DNA ligase. A growing body of evidence suggests that bacteria other than the well-characterized E. coli paradigm differ in basic DNA repair machinery. MMR proteins involved in mismatch recognition and strand discrimination like MutS, MutL and MutH from H. influenzae have been characterized, but other downstream repair genes like UvrD helicase and exonucleases like RecJ have not been studied functionally in detail. H. pylori harbors a UvrD homolog, which shares limited homology with other UvrD proteins (29% identity with E. coli UvrD and 31 % with H. influenzae UvrD) and its cellular functions are not clear. Moreover, it is not well-understood how the activities of UvrD and RecJ proteins are regulated within these pathogens. It was, therefore, envisaged that biochemical characterization of UvrD and RecJ would lead to a better understanding of the mechanistic aspects of repair processes within these pathogens. The following sections summarize the results presented in this investigation. Functional characterization of UvrD from H. influenzae UvrD or DNA helicase II is a member of superfamily I of DNA helicases with well-documented roles in nucleotide excision repair (NER) and MMR, in addition to roles in replication and recombination. The 727-amino acid H. influenzae Rd KW20 UvrD (HiUvrD) protein was purified as an N-terminal (His)6-tagged protein to near homogeneity, and its authenticity was confirmed by peptide mass fingerprint analysis. HiUvrD displayed robust binding with single-stranded (ss) DNA as compared to double-stranded (ds) DNA. HiUvrD was found exhibit ~ 1000-fold higher affinity for ssDNA as compared to dsDNA as determined by surface plasmon resonance (SPR). In addition, to gain insights into the role of HiUvrD in replication, repair, recombination and transcription, the ability of HiUvrD to bind different DNA structures resembling intermediates of these processes was investigated using electrophoretic mobility shift assays. HiUvrD exhibited relatively high affinities for a number of branched DNA substrates and the order of affinity observed was; splayed-duplex ≥3’-flap ≥ ssDNA > 3’-overhang > four-way junction > three-way junction > nicked duplex > looped duplex ≥ duplex. Concurrent with its high affinity for ssDNA, HiUvrD exhibited a robust ssDNA-specific and Mg2+ - dependent ATPase activity. HiUvrD was able to unwind different DNA structures with varying efficiencies (3’ flap ≥ 3’-overhang > three-way junction > splayed-duplex > four-way junction > nicked > loop = duplex >>> 5’-overhang) and with a 3’-5’ polarity, which underpins its role in replication fork reversal, recombination and different DNA repair pathways. Multiple sequence alignment of HiUvrD with other helicases showed the presence highly conserved helicase motifs of which motif I and II are essential for ATP binding and hydrolysis. Mutation of an invariant glutamate residue (E226Q) in motif II of HiUvrD resulted in a dominant negative growth phenotype since, it was not possible to recover transformants when wild-type E. coli expression strains BL21(DE3)plysS or BL21(DE3)plysE were transformed with expression vector carrying hiuvrDE226Q. Mutation of a conserved arginine residue to alanine (R288A) in motif IV resulted in approximately 80 % reduction in ATP hydrolysis, and abrogation of helicase activity as compared to the wild-type protein. This can be attributed to ~ 70 % reduced ATP binding by HiUvrDR288A as determined by UV-crosslinking of radioactive ATP without change in affinity for ssDNA. HiUvrD was found to exist predominantly as a monomer with small amounts (~ 2-3 %) of higher oligomers like dimers and tetramers in solution. Deletion of 48 amino acid residues from distal C-terminus of HiUvrD resulted in abrogation of the oligomeric species implicating C-terminus to be involved in protein oligomerization. Interplay of UvrD with MutL and MutS in H. influenzae, and its modulation by ATP To investigate the effects of H. influenzae MutS (HiMutS) and MutL (HiMutL) on the helicase activity of HiUvrD, two different nicked DNA substrates were generated- a homoduplex and a heteroduplex DNA with a GT mismatch. HiMutL and HiMutS did not exhibit any helicase activity on either homoduplex or heteroduplex DNA, and unwinding of these substrates was observed only in presence of HiUvrD. In the presence of HiMutL the helicase activity of HiUvrD was stimulated on both homoduplex and heteroduplex nicked substrates whereas no significant modulation of HiUvrD ATPase activity in presence of HiMutL was observed. A much higher stimulation of unwinding of heteroduplex DNA was obtained, in presence of increasing concentrations of HiMutS. With increasing concentrations of HiMutL a progressive increase in HiUvrD mediated unwinding of the radiolabeled DNA strand was observed, which was ~ 15-fold higher than unwinding by HiUvrD alone. To investigate the effect of ATP in the stimulation of HiUvrD by HiMutL, two mutants of HiMutL–E29A (E29 is involved in ATP hydrolysis in E. coli UvrD), and D58A (D58 is essential for ATP binding in E. coli UvrD) were generated. HiMutLE29A retained only ~ 30 % of the wild-type ATPase activity, which was completely abolished in HiMutLD58A. Similar to wild-type protein, HiMutLE29A was able to stimulate HiUvrD helicase activity whereas HiMutLD58A failed to stimulate this activity. This indicated that ATP-bound form of MutL was essential for stimulation and perhaps interaction with UvrD. SPR analysis was carried out to validate and quantitate the direct protein-protein interaction between HiUvrD and HiMutL in absence or in presence of ATP, AMPPNP, and ADP. In the presence of ATP as well as AMPPNP, almost ~ 10,000-fold increase in the affinity between HiMutL and HiUvrD was observed but the same was not the case in presence of ADP. This clearly suggested that ATP binding rather than its hydrolysis promotes the interaction of MutL with UvrD. The effect of HiMutS on MutL-stimulated DNA unwinding by HiUvrD was determined using a heteroduplex nicked DNA with a GT mismatch. Interestingly, in the presence of HiMutS ~ 20-fold activation of DNA unwinding was observed, which is higher than the stimulation by HiMutL alone. The role of ATP-hydrolysis by MutS in regulation of UvrD helicase was studied by replacing wild-type protein with HiMutSE696A in the helicase assays. HiMutSE696A failed to hydrolyze ATP but was able to bind ATP with the same affinity as the wild-type protein and interacted with heteroduplex DNA with ~ 8-fold reduced affinity as compared to wild-type MutS. Intriguingly, increasing concentrations of HiMutSE696A failed to stimulate HiUvrD helicase activity in presence of HiMutL indicating that ATP hydrolysis by HiMutS is essential for stimulation of HiUvrD helicase activity post MutH-nicking during MMR. SSB, an essential component of all DNA metabolism pathways, possibly functions to stabilize the ssDNA tract generated by UvrD and exonucleases during MMR. ATPase and helicase activities of HiUvrD were inhibited by the cognate SSB protein. This inhibition could be overcome by increasing the concentration of HiUvrD helicases thus, pointing out the fact that SSB and UvrD perhaps compete with each other for ssDNA substrate. Noticeably, MutL and MutS proteins could alleviate the inhibition of HiUvrD by HiSSB. Functional characterization of UvrD from H. pylori In H. pylori, UvrD has been reported to limit homologous recombination and DNA-damage induced genomic recombinations but the protein has not been functionally studied. UvrD from H. pylori strain 26695 (HpUvrD) was over-expressed and purified as an N-terminal (His)6-tagged protein, and its authenticity was confirmed by peptide mass fingerprint analysis. HpUvrD exhibited high affinity for ssDNA as compared to dsDNA as determined by electrophoretic mobility shift assays and SPR. In addition, HpUvrD was able to bind a number of branched DNA structures (splayed duplex > ssDNA > 3’-flap > 3’overhang > three-way junction = four-way junction > loop >>> nicked ≥ duplex) suggesting its role in different DNA processing pathways. HpUvrD exhibited a Mg2+ - dependent ssDNA-specific ATPase activity, and a 3’-5’ helicase activity. HpUvrD was able to unwind different branched DNA structures with 3’-ssDNA regions like splayed duplex, 3’-overhang and 3’-flap. Blunt-ended duplex, duplexes with nick and loop as well as three-way and four-way junctions were unwound with less efficiency. Interestingly, the helicase activity of HpUvrD was supported by GTP and dGTP to almost the same level as ATP and dATP, which is in stark contrast to other characterized UvrD proteins. Moreover, HpUvrD was able to hydrolyze GTP albeit with ~ 1.5-fold reduced rate as compared to ATP. However, motifs associated with GTP binding and hydrolysis were not found in HpUvrD and it is possible that GTP binds in the same site as ATP. To investigate this possibility, helicase assay was done in the presence of ATP together with different concentrations of GMP-PNP, which is a non-hydrolysable analog of GTP, and did not support HpUvrD helicase activity. With increasing concentrations of GMP-PNP, a progressive inhibition of DNA unwinding by HpUvrD was observed suggesting that GMP-PNP could compete with ATP for a common binding site within HpUvrD. Replacement of a highly conserved glutamate residue with gluatamine (E206Q) in Walker B motif of HpUvrD resulted in ~17-fold reduced ATPase activity, and abrogation of helicase activity as compared to the wild-type protein. HpUvrDE206Q was able to bind ssDNA and ATP with comparable affinities as the wild-type protein suggesting the role of E206 in ATP hydrolysis. Like HiUvrD, HpUvrD was found to exist predominantly as a monomer in solution together with the presence of small amounts of higher oligomeric species. However, unlike HiUvrD, deletion of distal C-terminal 63 amino acids in HpUvD did not abrogate the oligomeric species suggesting that additional regions of the protein may be involved in protein oligomerization. The ATPase and helicase activities of HpUvrD were inhibited by the cognate SSB protein, and this inhibition could be overcome by increasing HpUvrD concentrations again suggesting that both UvrD and SSB proteins compete for ssDNA substrate. To investigate the role of UvrD in the physiology of H. pylori, a knock-out of hpuvrD was constructed in H. pylori strain 26695 by insertion of chloramphenicol cassette in its open reading frame. The mutant H. pylori strain 26695 obtained after disruption of hpuvrD was extremely slow growing under the normal microaerophilic conditions compared to the wild-type strain. Growth defect of H. pylori strain 26695ΔhpuvrD highlights the importance of UvrD in H. pylori cellular processes and in vitro fitness. Characterization of H. influenzae RecJ and its interaction with SSB Among the four exonucleases involved in MMR pathway, RecJ is the only known nuclease that degrades single-stranded DNA with 5’ to 3’ polarity. RecJ exonuclease plays additional important roles in base-excision repair, repair of stalled replication forks, and recombination. RecJ exonuclease from H. influenzae (HiRecJ) is a 575 amino acid protein, which harbors the characteristic motifs conserved among RecJ homologs. Due to limited solubility of HiRecJ, the protein was purified as a fusion protein with maltose binding protein (MBP). The purified protein exhibited a Mg2+ or Mn2+- dependent, and a highly processive 5’ to 3’ exonuclease activity, which is specific for ssDNA. MBP did not affect the exonuclease activity of HiRecJ. The processivity of HiRecJ was determined as ~ 700 nucleotides per binding event, using a ssDNA substrate labelled internally with 3H and at its 5’-terminus with 32P. Cd2+ inhibited the Mg2+ - dependent exonuclease activity of RecJ, which could not be overcome by increasing Mg2+ concentration. Site-directed mutagenesis of highly conserved residues in HiRecJ- D77A, D156A and H157A abolished the enzymatic activity. Interestingly, HiRecJD77A was found to interact with ssDNA with a 10-fold higher affinity than wild-type protein suggesting that this conserved aspartate residue may function to coordinate the binding of metal ion or DNA to hydrolysis of DNA. E. coli HU protein inhibited the HiRecJ exonuclease activity in a concentration-dependent manner possibly due to sequestration of ssDNA, thus making it unavailable for HiRecJ. During MMR, ssDNA tracts generated by UvrD helicase activity are most probably stabilized by SSB and hence, the in vivo substrate for RecJ would be SSB-ssDNA complex. The exonuclease activity of HiRecJ was stimulated approximately 3-fold by H. influenzae SSB (HiSSB) protein. HiSSB was able to stimulate HiRecJ exonuclease activity on a ssDNA substrate, which formed either a very strong secondary structure or on a homopolymeric ssDNA substrate, which did not form any secondary structure, suggesting that HiRecJ exonuclease was stimulated independent of the ability to HiSSB to melt secondary structures and stabilize ssDNA. Significantly, steady-state-kinetic analysis clearly showed that HiSSB increases the affinity of HiRecJ for ssDNA. H. influenzae SSBΔC and T4 gene 32 protein, a SSB homolog from bacteriophage T4, failed to enhance the HiRecJ exonuclease activity suggesting a specific functional interaction between HiSSB and HiRecJ mediated by C-terminus tail of HiSSB. More importantly, HiRecJ was found to directly associate with its cognate SSB. The C-terminus of HiSSB protein was found to be essential for this interaction. To delineate the regions of HiRecJ that interact with HiSSB, different truncated forms of HiRecJ were generated in which regions external to conserved motifs required for exonuclease activity were deleted. Different deletion mutants of HiRecJ- RecJ∆N34, RecJ∆C76 and the core catalytic domain (which contains amino acid residues 35-498) were purified as fusion proteins with MBP. HiSSB was found to interact with all the truncated forms of HiRecJ suggesting that its core-catalytic domain harbors a site for interaction with SSB. Taken together, the results presented in this study lead to a better understanding of the structure-function relationships of the UvrD helicase and RecJ exonuclease. Importantly, they provide insights into the interplay between various proteins in DNA MMR pathway. Characterization of repair proteins that are involved in multiple genome fidelity pathways is of fundamental importance to understand repair processes, more so in pathogenic bacteria wherein they regulate mutation rates, which can alter the fitness and virulence of the pathogens. Publication Sharma R., and Rao, D.N. (2009). Orchestration of Haemophilus influenzae RecJ exonuclease by interaction with single-stranded DNA-binding protein. J. Mol. Biol., 385, 1375-1396.

The work described in this dissertation focuses on developing simple yet effective assays integrating nucleic acid (NA) aptamers with rolling circle amplification (RCA) for the detection of non-NA biomarkers. The first project, a comprehensive literature review, highlights the current state of the art in functional NA-based RCA applications, and identifies shortcomings in the detection of non-NA targets with RCA. Biosensor design is critically evaluated from four key perspectives: regulation, efficiency, and detection of RCA, and the integration of all three components for point of care (POC) applications. The second project investigates how target-binding to a linear aptamer can be utilized to regulate RCA in a simple and inexpensive format. Phi29 DNA polymerase (DP) exhibits difficulty processing DNA strands that are bound to non-NA materials such as proteins. The work uses this restriction of phi29 DP as a feature by utilizing protein-binding aptamers as primer strands (aptaprimers) for RCA. The simplicity is showcased by adapting the method to a cellulose paper-based device for the real-time detection and quantification of PDGF or thrombin within minutes. As the second project is a turn-off sensor, the third project exploits the inherent 3’-exonuclease activity of phi29 DP to generate a simple turn-on assay instead. As target-bound aptamers were shown to be resistant to exonuclease activity, the phi29 DP preferentially digests target-free aptaprimers instead of target-bound aptaprimers. The target-bound aptaprimer could be liberated by a circular template (CT) by incorporating toehold-mediated strand displacement (TMSD), and used for RCA. Sensitivity was improved relative to project two, though the dynamic range was narrow owing to difficulty liberating target-bound aptaprimer at high target concentrations. Project four instead used RecJ, which has 5’-exonuclease activity, to modulate aptaprimer availability. Similarly to project three, target-binding conferred protection on the aptaprimer from 5’-exonuclease digestion by RecJ. By including a free 3’ terminus on the aptaprimer, inhibition of RCA due to target binding was avoided and CT-mediated TMSD was not needed, simplifying the assay. As well, this approach was generalizable as it was demonstrated using both a protein (thrombin) and a small molecule (ochratoxin A) target. This turn-on method further improved the assay compared to project three with a 100-fold enhancement in sensitivity and a restoration of the dynamic range. In sum, this work contributed multiple simple and sensitive approaches for the real-time fluorescent detection of proteins and small molecules with the RCA of linear aptamers. / Thesis / Doctor of Science (PhD)