<p>Interstrand cross-linking (ICL) damage to DNA is cytotoxic as it blocks replication and transcription. This cytotoxicity is exploited in anti-cancer therapies, but increased ICL repair limits the efficacy of these chemotherapies. SNM1A (sensitive to nitrogen mustard 1A), of the beta-CASP family of nucleases, has been shown to participate in the initiation of one of the ICL repair processes. Biochemical studies of SNM1A have been limited due to insolubility and instability of SNM1A in bacteria and insect cell lines and toxicity in human cell lines. Work reported in this thesis describes a novel and efficient method of generating active protein from inclusion body expression of the beta-CASP domain of SNM1A. This refolded beta-CASP domain shows 5’ exonuclease activity on single stranded and double stranded DNA in vitro. Nevertheless, this domain alone is unable to complement <em>pso2</em> null ICL repair defects in<em> S. cerevisiae</em> after exposure to ICL agents. These functional studies of the beta-CASP domain of SNM1A will be helpful in directing future research on its role in ICL repair. Additionally, this will aid future structural and inhibitor studies of this essential interstrand cross-link repair protein, SNM1A.</p> / Master of Science (MSc)

<p>DNA interstrand cross-links (ICLs) constitute one of the most severe types of DNA damage. ICLs covalently tether both strands of duplex DNA, preventing unwinding and polymerase access during replication and transcription. This obstruction is exploited in cancer chemotherapy since it leads to replication fork collapse, double strand breaks (DSBs), and cell death. Mechanistic understanding of how eukaryotic cells repair these specific lesions, however, is still in its infancy. It is understood that ICL repair consists of a multitude of intersecting and connecting repair pathways that rely on interplay between critical protein factors. Interestingly, Pso2 has been identified as an integral component of the ICL repair pathway in <em>Saccharomyces cerevisiae</em>. Pso2 is a yeast nuclease from the β-CASP family of proteins that function predominantly in the repair of ICLs. It has been recognized as the only protein that does not serve a redundant function in any other DNA repair pathway. It remains unclear how the ICL repair pathway generates DNA intermediates suitable for high fidelity repair dependent on Pso2 nuclease activity. Here we show that Pso2 possesses structure-specific endonuclease activity that may be essential to its role in ICL repair. Direct <em>in vitro</em> activity assessment of the protein on a site-specific ICL proved to be inconclusive due to the heat-labile nature of the cross-linking agent employed. <em>In vitro </em>activity testing was also performed on various substrates resembling intermediates generated during ICL repair. Biochemical analysis demonstrated that Pso2 cleaves hairpins, stem loops, heterologous loops, and symmetrical bubbles. Although the precise cleavage sites vary between substrates, Pso2 demonstrates preference for the single- to double-stranded junction in the DNA backbone, with similar activity to that previously demonstrated for its human homologue, Artemis. This specific endonuclease activity is stimulated by increased concentrations of phosphate. Through two-dimensional gel electrophoresis, the presence of unique DNA intermediates generated in response to ICL damage <em>in </em><em>vivo </em>was also monitored. Results suggest the generation of hairpin-like intermediates that resemble those tested <em>in vitro</em>. These intermediates persist in the absence of Pso2 but are resolved by exogenous addition of control endonucleases. Our findings expand on previous data that established hairpin-opening activity for this protein and suggest that the structure-specific endonuclease activity demonstrated by Pso2 is important for ICL repair. We anticipate that Pso2 acts on a hairpin-containing DNA substrate in the ICL repair pathway and the resolution of this intermediate is uniquely dependent on Pso2 for the effective repair of ICL damage in yeast. Taking into consideration the current models of ICL repair, both in yeast and humans, possible roles for Pso2 have been described. Achieving a complete mechanistic perspective of this pathway is critical for the therapeutic exploitation of the human homologue, SNM1A. Implications include the potential inhibitory target for increased efficacy of chemotherapy with cross-linking agents.</p> / Master of Science (MSc)

No description available.

Abasic sites (Ap site, from apurinic/apyrimidinic) are one of the most common lesions generated in DNA by spontaneous base loss or DNA repair processes. There are two equilibrating forms of an Ap site - ring-open aldehyde and cyclic hemiacetal. Ring- opened aldehydes are reactive electrophilic groups capable of formation covalent adduct with nucleophilic sites in DNA. DNA interstrand cross-link (ICL) resulting from the Ap sites is formed spontaneously as a covalent bond between ring-open aldehyde and amin group of adenin residue in the opposite strand of double stranded DNA. ICLs block DNA replication and transcription. The formation of Ap site derived ICL is relatively long process taking several hours. We assume that the ring-opening of an abasic site is the rate-limiting step in the formation of the thermodynamic ICL. However, formation, stability and DNA repair of Ap-ICL are still poorly understood processes. Here, I have set up mechanistic in vitro experiments to reveal and calculate the probability of Ap-ICl formation in vivo. In more detail, I study the rates of formation of Ap-ICLs in the sequence context of neighbouring nucleotides of freshly formed covalent bond of ICL. I focus on sequence preference, the influence of AT/ GC rich regions and the length of oligonucleotides. I have...

Interstrand cross-links (ICLs) are a type of DNA damage that prevents strand separation required for basic cellular processes. ICL-based anti-cancer therapies exploit the cytotoxic consequences of replication and transcription inhibition, however, they are limited by the ability of the cell to repair DNA crosslinks. The challenge of ICL repair involves coordinating multiple DNA repair pathways to remove damage occurring on both strands of DNA. Participation of factors that are both exclusive and essential to crosslink repair suggests a pathway requirement to process unique structures and/or intermediates arising only in ICL repair. SNM1A is a nuclease required for survival of human cells in response to ICL exposure, but the specific function and role of SNM1A remain unclear. Here we show that, in addition to known 5’-3’exonuclease activity, SNM1A possesses single-strand specific endonuclease activity. Furthermore, SNM1A exhibits translesion nuclease activity on crosslinks which deform the helical backbone, but not non-distorting stable ICLs. We report the identification and characterization of nine small molecules inhibitors of SNM1A, isolated from an in vitro high-throughput screen of nearly 4,000 bioactive compounds. Finally, we demonstrate that inhibitors of SNM1A potentiate the cytotoxicity of ICL-inducing agent cisplatin in HeLa cells. The work in this thesis expands the possible roles of SNM1A in ICL repair and lays the groundwork for SNM1A inhibition in ICL sensitization efforts. / Thesis / Doctor of Philosophy (PhD)

Platinum-based chemotherapeutics are used for the treatment of a wide range of cancers and a number of attempts have been made toward developing compounds with better cellular stability and similar or enhanced cytotoxicity as compared to their predecessors. The first part of the work reported here focuses on the cellular effects of the metabolically stable dinuclear platinum compound, BBR3610-DACH. Comet assay showed this compound to form interstrand crosslinks, a highly toxic DNA lesion in HCT116 cells, at equimolar concentrations to its parental compound, BBR3610. Cell cycle studies showed that BBR3610-DACH causes G1/S and G2/M cell cycle arrest with S phase depletion, which was p21 dependent and partially p53 dependent in contrast to BBR3610 which showed initial S phase accumulation followed by a classical G2/M arrest. BBR3610-DACH-induced G1/S and G2/M cell cycle arrest interestingly was found to be independent of the DNA damage response mediated via the activation of ATM and ATR kinases. Also, the cell cycle arrest culminated in apoptosis, although apparently through a non-canonical pathway. The second project explores the cellular effects of trans-4-NBD which is a fluorescent derivative of transplatin. Like cisplatin, trans-4-NBD induced interstrand crosslinks in HCT116 cells as detected by the comet assay. Treatment with trans-4-NBD showed a G2/M arrest in HCT116 cells and a transient S phase accumulation in A2780 cells, with a marked increase in p53 and p21 protein levels. A robust apoptotic response is also seen via caspase activation and PARP cleavage in both the cell lines. Finally, the focus is shifted toward the nucleolar targeting platinum complex, TriplatinNC. Confocal studies in TriplatinNC-treated HCT116 and A2780 cells showed disruption of rRNA transcription as an early event followed by a robust G1 cell cycle arrest. Apoptotic induction was observed with the onset of cellular morphological changes and apparent caspase activation which was independent of the p53 status of the cells. Overall, these studies explore novel platinum based compounds that show promising anti-cancer activities by affecting various facets of cellular signaling.

[Formulae and special characters can only be approximated here. Please see the pdf version of the Abstract for an accurate reproduction.] Several inert platinum complexes were synthesised: [(en)Pt([special character]-dpzm)2Pt(en)]4+, [{Pt(dien)}2[special character]-dpzm]4+, [{Pt(dien)}2[special character]-H2N-(CH2)6-NH2]4+, cis-[(NH3)2Pt([special character]--dpzm)2Pt(NH3)2]4+, trans-[Pt(NH3)2([special character]-dpzm)2]2+. Three active complexes, all with chloro ligands, were also synthesised: trans-[{Pt(NH3)Cl2}2[special character]-dpzm)], trans-[{Pt(NH3)2Cl}2[special character]-dpzm]2+ (di-Pt) and trans-[trans-{Pt(NH3)2Cl}2{trans-[Pt(NH3)2([special character]-dpzm)2]}]4+ (tri-Pt). 1H NMR established that multi-nuclear platinum complexes will preferentially associate in the DNA minor groove with a preference for A/T sequences, and with a binding constant [special character]-105 M-1, regardless of the charge, linking ligand, length or shape. Using [(en)Pt([special character]-dpzm)2Pt(en)]4+ and the oligonucleotide d(GC)5 it was determined that the metal complex binds G/C rich sequences also in the minor groove, but with a much reduced binding constant, 103 M-1. CD studies showed [(en)Pt([special character]-dpzm)2Pt(en)]4+ was able to induce a DNA conformation change from B-type to what appeared to be a partial Z-type. Transcription assays showed that even though the metal complex does not bind DNA covalently, it is still able to inhibit DNA transcription at particular sites. The complexes di-Pt, tri-Pt, [{Pt(dien)}2[special character]-dpzm]4+ and trans-[Pt(NH3)2([special character]-dpzm)2]2+ were tested for anti-cancer activity in the L1210 murine leukaemia cell line, and gave values of 3.8, 2.5, [special character]200 and 64 [special character]M respectively. In the cisplatin resistant line (L1210/DDP), trans-[Pt(NH3)2([special character]-dpzm)2]2+ showed an increase in activity with a drop to 32 [special character]M, while both di-Pt and tri-Pt showed decreases in activity to values of 8.8 and 3.6 [special character]M. In the human ovarian carcinoma 2008 cell line and its cisplatin resistant derivative C13[special character]5, both complexes showed good activity with values of 2.5 and 20.9 [special character]M respectively, but again both showed decreases in activity in the resistant line with values of 17.8 and 37.7 [special character]M respectively. To help explain the difference between activity of these complexes and the complexes BBR3464 and BBR3005, cell uptake and DNA interstrand cross-linking experiments were performed. The cell uptake studies showed that both di-Pt and tri-Pt are taken up by cells at very high levels, when administered at 100 [special character]M, thus indicating that the difference is unlikely to be due to large differences in cell uptake. The DNA interstrand cross-linking studies showed both complexes readily form interstrand adducts (50% interstrand cross-linking at 12 nM and 22 nM respectively, c.f cisplatin 3 [special character]M). These results suggest that the rigid nature of the dpzm linker may be affecting the DNA adducts formed, with more interstrand links being formed than BBR3464. Possibly, it is this that causes the large differences in cytotoxicity. The DNA binding of di-Pt and tri-Pt was examined with the nucleosides adenosine and guanosine and the dinucleotide d(GpG). Both complexes bound at the N7 of guanosine, but 2-fold slower than cisplatin. In addition, di-Pt bound at the N7 and either the N1 or N3 of adenosine, 7-fold slower than guanosine. Di-Pt forms a large variety of cross-links between two d(GpG) molecules, however it could not be established whether the 1,2-intrastrand adduct could be formed. Di-Pt, however, forms a 1,2-GG interstrand adduct with the oligonucleotide d(ATGCAT)2 resulting in a conformation change away from B-type DNA. The sugar pucker of the G3 nucleoside changes from 2[special character]-endo towards 3[special character]-endo, and the position of the nucleotide relative to the sugar changes from anti to syn. The ability of multi-nuclear platinum complexes to form covalent adducts in the DNA minor groove remains unclear. It appears that di-Pt can form up to 33% minor groove adducts with the oligonucleotide d(AT)5, but when added to the oligonucleotide d(GCCAAATTTCCG)2 no definite minor groove adducts are seen and the major adduct appears to be a 1,2-interstrand cross-link between the two A6's or between the G1 and G11. Finally, a study of the encapsulation of platinum complexes within cucurbit[7]uril (Q7) as a means of reducing drug toxicity was made. For complex A and di-Pt, encapsulation of the linker ligand occurred. The effect of Q7 on the rate of hydrolysis of di-Pt was at least a 3-fold reduction as compared to free di-Pt with guanosine. Studies with [{Pt(dien)}2[special character]-dpzm]4+/Q7 and the oligonucleotide d(CGCGAATTCGCG)2 showed that the metal complex could dissociate from the Q7 and associate with the oligonucleotide, where an equilibrium is achieved with 15 % of the metal complex bound to the oligonucleotide and 75 % encapsulated in Q7. Tests in the L1210 and L1210/DDP cancer cell lines showed that di-Pt/Q7 has almost the same activity compared to free di-Pt.