Search

Cancer initiation and progression is an evolutionary process where somatic alterations convey fitness advantages to cells that exist in a complex environment acting to shape cell populations. Homeostasis, being the normal state of a tissue prior to disease, prioritizes and protects long lived stem cells that are capable of incurring these somatic alterations and contributes to the early evolutionary dynamics from which cancer initiation occurs. Research over the past two decades has made substantial inroads into our understanding of somatic mutations through characterization of healthy tissue. In parallel, agent based mechanistic models have emerged as an important tool for understanding somatic mutation and stem cell dynamics in both cancer and homeostatic tissues. In this thesis we combine first principles mathematical modeling of homeostasis, complete with high-resolution genotypic information, with healthy patient sequence data in various tissue types to answer critical biological questions about somatic evolution, stem cell dynamics, and cancer initiation. We first develop a three-dimensional mechanistic model of the homeostatic epidermis and methodology necessary to incorporate base pair resolution mutation data. We use this model to examine neutrally expanding subclonal populations and compare simulations to healthy patient data. We then introduce functionally heterogenous oncogenic drivers in our in-silico model to examine non-neutral dynamics in the epidermis. Next, we develop a novel lineage tracing method using fluctuating methylation sites through analysis of patient data combined with a spatial model of a homeostatic intestinal crypt to determine stem cell dynamics in glandular tissue. Our spatial model reveals the necessary dynamics of fluctuating methylation sites to be used as fluctuating methylation clocks that allows us to determine stem cell behavior over short timescales in vivo. In the last chapters of this thesis, we focus on using our novel fluctuating methylation clock and somatic evolution as an important biomarker to understand clonal hematopoiesis. As the hematopoietic stem cells become homogenized through disease, a distinctive distribution emerges within patient data that can be exploited to diagnose clonal hematopoiesis without the need to understand somatic drivers. Combined, the work in this thesis reveals the power of somatic evolution on various timescales as a marker that can provide tissue level insights. We conclude by outlining a future direction where combining fluctuating methylation markers with DNA alteration could be a powerful tool in diagnostics and prognosis within the clinic.

Cervical malignancy is the fourth most common cause of cancer-related mortality in women worldwide; infection with high-risk human papillomavirus (HRHPV) is responsible for over 500,000 cases of cervical carcinoma each year, approximately 90% of which are squamous cell carcinomas (SCCs). Over half of all HPV-positive cervical SCCs are caused by the deregulated expression of HPV16 oncogenes E6 and E7 in proliferating basal cells of the cervical squamous epithelium. The major risk factor associated with cervical neoplastic progression is integration of HRHPV into the host genome, which is detected in $~$85% of HPV16-positive cervical carcinomas. The work presented in this doctoral thesis sought to provide insights into our understanding of the process of HPV16 integration as well as to elucidate mechanisms that deregulate both virus and host gene expression following integration. The W12 cell model system used in this project is a polyclonal cervical keratinocyte line generated by explant culture of a low-grade cervical squamous intraepithelial lesion (LSIL) that arose following natural infection with HPV16. The W12 clones were isolated in the absence of selective pressure, and as such represent the range of integration events that occur in a pre-malignant lesion at the early stages of carcinogenesis, prior to integrant selection. Despite identical genetic backgrounds, expression levels of oncogenes E6 and E7 varied up to 16-fold between the W12 clones. Expression of HPV oncogenes is ultimately determined by transcription factor binding to the non-coding long control region (LCR) of the viral genome. The initial result of this study found that genomic mutations affecting transcription factor binding at the LCR of the W12 clones was not a cause of differential viral expression, concluding that epigenetic control may be at play. Higher levels of virus expression per template were associated with increased levels of histone post-translational modification (PTM) hallmarks of transcriptionally active chromatin and reduced levels of repressive hallmarks. There was greater abundance of the active/elongating form of the RNA polymerase-II enzyme (RNAPII-Ser2P), together with CDK9, the component of positive transcription elongation factor-b (P-TEFb) responsible for the Ser2 phosphorylation. The changes observed were functionally significant, as cells with higher HPV16 expression per template showed greater sensitivity to depletion and/or inhibition of histone acetyl transferases and CDK9, as well as reduced sensitivity to histone deacetylase inhibition. Employing next generation sequencing data available for five representative W12 clones, the sites of HPV16 host integration were identified. The three-dimensional (3D) structure of the nucleus and physical interactions between stretches of the genome over long distances (i.e. enhancer and promoters) are known to exert an additional level of gene regulation. Identification of 3D virus-host interactions in the W12 clones employing the newly developed and unique 'Sequence Capture of Regions Interacting with Bait Loci Hi-C' (SCRiBL-Hi-C) protocol showed that both short- ($~$50 kb), and long-range ($~$1 Mb) interactions occur during the early stages of carcinogenesis. Together, the data in this thesis indicate that transcription and subsequent expression of the HPV16 genome is controlled by multiple layers of epigenetic regulation.

Folates are water-soluble B vitamins, which maintain DNA stability by regulating nucleotide synthesis and DNA methylation. Folates influence CRC risk and their ability to prevent or promote carcinogenesis may be dependent on several variables here investigated. No in vitro study has yet modelled the physiological folate status that human colon cells are exposed to in vivo. This study evaluates the ability of different forms and concentrations of folate to maintain DNA stability and normal cell function in folate-sufficient and stable human colonocytes and to modify DNA instability and the acquisition of abnormal characteristics in folate-deficient and genomically-unstable colonocytes. Non-malignant human NCM460 colonocytes cultured at physiologically-relevant concentrations of 5-methyl-THF or FA, representing the average deficient (2.5 ng/mL), sufficient (10 ng/mL) or highest post-supplementation (100 ng/mL) folate levels found in human plasma were used in this study as a model of colon-folate interaction. This work established that FA is taken up and/or retained to a lesser extent than 5-methylTHF and is less efficient at maintaining DNA stability and normal cellular characteristics in folate-sufficient and genomically-stable colonocytes at baseline, particularly at deficient and sufficient concentrations in the medium to longer term (14-21 days). During repletion of folate-deficient and genomically-unstable cells, sufficient concentrations of FA do not increase intracellular folate status and worsen the unstable phenotype, by perpetuating DNA instability and enabling the acquisition of a more pro-malignant protein expression. On the contrary, employing 5-methyl-THF sufficiency for repletion positively modifies the abnormal protein profile and morphological features of folate-deficient cells, mitigating potential progression to malignant transformation. When high post-supplementation concentrations are employed, both folate forms increase intracellular folate status, but drive a more promalignant and stress-induced proteome profile and, in the case of 5-methyl-THF, promote abnormal cell morphologies. In conclusion, the folate type, concentration employed, baseline folate status and timing of exposure to folate supplementation are important variables that should be taken into account by future studies evaluating the potential impact of mandatory FA fortification on CRC.

The underlying etiology of numerous disease states results from perturbations in the maintenance of cellular proteostasis. Carcinogenesis relies on these perturbations to foster uncontrolled cell growth and eventual metastases, while neurodegenerative diseases are a consequence of such perturbations. Control of these processes occurs at numerous molecular levels, commonly starting with transcription. A key transcriptional complex that is involved is the CDK8 Kinase Module (CKM). The CKM is conserved from yeast to man, forming a tetrameric complex consisting of MED12, MED13, CDK8, and CCNC. The CKM has not only been implicated in a variety of cancers but also in a spectrum of developmental disorders. Using RNA sequencing on murine and human cell lines, I showed that the CKM regulates the transcription of genes mediating conserved proteostasis pathways including translation, protein degradation, and differentiation. In addition, I showed that these processes are druggable targets, as chemotherapeutics aimed at translation and protein degradation, sensitized CCNC-/- cells to cell death. The use of CKM inhibitors phenocopied this sensitivity when cyclin C was present. As customized drug treatments become a reality, these data represent an approach that may be used to target cancers. These data further present a model of how mutations in the CKM can contribute to a diverse spectrum of disease states.

Young children are particularly sensitive to environmental pollutants. They can directly ingest soil by putting dirty hands and objects in their mouths. The reliance on animal derived models for human health risk and exposure assessment has several limitations. In this investigation, a tool-kit was developed and optimised to facilitate more accurate, reliable and representative predictions of soil contaminants that might pose a significant hazard to young children. The tool-kit was developed and optimised using an in vitro human digestion bioassay. This procedure was followed by the optimisation of several mutagenicity bioassays to link to the bioaccessible fraction which quantified by the in vitro bioassay. The application of novel and sensitive environmental-based biosensors requires them to work in parallel with effective and proven extraction techniques. In this study, chemical analysis was used to quantify the bioaccessible (human assimilated portion) of pollutants in soils. Acute toxicity was measured using constitutively marked bioluminescent bacterial biosensors and these were indicative of the total contaminant burden. A range of mutagenic assays were applied and optimised. In the Ames assay, any compound exhibiting a greater than two-fold increase in the number of revertants colonies over the number of spontaneous revertants was considered as a mutagen. Mutagenic-responsive SOS-lux based microbial biosensors were compared to the Ames assay. Mutagenicity assessment of a broad range of environmental pollutants (i.e. B[a]P, DiB(a,h)A, B[a]A, Ni and Cu), was performed using four SOS-lux microbial biosensors; E. coli DPD1718, E. coli K12C600, S. aureus pAmiUmuC and S. aureus pAmiRecA. The results substantiated that the four biosensors were unable to be induced by these pollutants. Nevertheless, E. coli DPD1718 and E. coli K12C600 were successfully induced by Mitomycin C (MMC) in a dose response manner. The Ames assay was performed for the above pollutants in the absence and the presence of the metabolic activation S9 mix. The standard plate incorporation assay and a modification protocol for the Ames assay were applied. Results reported from the Ames assay confirmed mutagenicity responses of the tested pollutants except Cu and Ni. MMC was selected and introduced into soil samples as a case study to assess the performance of the developed tool-kit. Soils amended with MMC were extracted by the in vitro human digestion bioassay, and the mutagenicity of the bioaccessible fraction was measured using the Ames assay and the biosensors. A comparison was made between the permissible concentrations of MMC obtained from the developed tool-kit and RISC4 derived concentrations. The four microbial biosensors applied in this study were incapable to detect the mutagenicity of the tested pollutants. On the other hand, the Ames assay was more robust and sensitive to a broad range of environmental pollutants. The in vitro human digestion bioassay enabled the quantification of the human bioaccessible fraction of the tested pollutants. This fraction posed a concern due to its estimation of the doses that would reach the blood circulation and cause harm to human. While the permissible concentration of MMC measured by the developed tool-kit was less than 10 μg MMC/g, the RISC4 model calculated that it should be 40 μg MMC/g. This revealed that, in this situation, risk assessment model was less conservative than empirical study for human health risk assessment. This study enabled the assessment of the permissible concentrations of environmental pollutants that could remain in a soil and pose permissible harm to humans. This approach also enabled a comparison of modelled and empirical data to allow a measure of sensitivity to be judged. There is a need to develop bioassay techniques more able to assess the potency of hydrophobic compounds both in isolation and combination.

Cellular senescence is an irreversible growth-arrested state seen in primary human cells after a finite number of cell divisions both in culture and in vivo. The escape from senescence to immortalisation is often thought to be a prerequisite for carcinogenesis. Many changes in senescent cells are consistent with changes in tissue and organ function with age. I used serial analysis of gene expression (SAGE) to analyse global gene expression profiles in senescent and early passage human foetal fibroblasts (hff). A total of over 20,000 SAGE tags were sequenced and characterised, corresponding to 2,675 unique transcripts. Relative to the early passage hff, transcripts which were found to be markedly increased in senescent hff included those encoding p21WAF1, Cyclin D1, ferritin heavy chain, transforming growth factor-beta induced gene (BIGH3), skin collagenase and amyloid. Amongst them, genes such as skin collagenase and amyloid are known to be up-regulated in human ageing. Together with these known genes, a number of "unknown" genes were up-regulated. Seven differentially expressed genes, up-regulated in senescence (BIGH3, prion, dickkopf (Xenopus laevis) homolog 1 (dkk1), Cbp/p300-interacting transactivator, with Glu/Asp-rich carboxy-terminal domain, 2 (CITED2), Rap1a and Cysteine knot superfamily 1, bone morphogenetic protein antagonist 1 and retinoic acid receptor, alpha (RARA)) were selected for validation with reverse transcriptase-polymerase chain reaction (RT-PCR). The roles of these genes are discussed in relation to cancer and neurodegenerative diseases. My study demonstrates that replicative senescence is a complex phenomenon involving genes from the wingless-type (Wnt), mitogen-activated protein (Map) kinase kinases extracellular signal-regulated (Erk) kinase, transforming growth factor beta signalling and epigenetic pathways. The challenges remaining now are to identify senescence-related tumour suppressor gene(s) and to elucidate further the biochemical pathways of senescence.

The ability to engineer mice which carry exogenous genes targeted for expression in specific tissues, or have an endogenous gene functionally switched off has allowed us to examine the role of particular proteins in an in vivo model. This technology, in combination with a well-characterised model of chemically-induced skin carcinogenesis in the mouse, has allowed us to start dissecting the changes that occur during a particular pathway of tumorigenesis. TGFbeta1 has been shown to act as an inhibitor of both normal cellular proliferation and tumour formation. However, it can also accelerate malignant conversion at later stages of carcinogenesis. Previously in this laboratory, mice were generated that overexpressed Tgfbetal in suprabasal layers of skin epidermis under the control of a K10 promoter. These transgenic mice showed an increased proliferation rate in the basal compartment of their epidermis compared to wild-type controls. In this study, it is shown that the change in proliferative index is not dependent on p53, a molecule central to growth control and apoptosis. The K10-Tgfbeta1 mice were crossed on to a Tgfbeta1-null background and the proliferative phenotype of the transgene was examined. An increased basal cell proliferative index was observed whether the background was Tgfbeta1-null or wild-type. TGFbeta1 has been shown to induce apoptosis and a line of transgenic mice overexpressing BCL-2 in the epidermis became available, providing an opportunity to investigate the role of an apoptotic pathway in the skin. These K10-BCL-2 mice were shown to express the transgene at both mRNA and protein levels in suprabasal cells of the epidermis. Normal skin homeostasis, however, was not disrupted by the exogenous BCL-2 and expression of other apoptotic markers was not altered in these mice. Purthermore, K10-5CL-2 mice responded to chemical insult by a skin tumour promoter in the same way as wild-type littermates. K10-Tgfbeta1 mice had demonstrated resistance to papilloma formation during chemical carcinogenesis and tumours on the transgenic mice had also shown a higher rate of malignant conversion in the same experiment. In this study, the effects of epidermally-targeted Tgfbeta1 during tumorigenesis were investigated further by crossing the K10-Tgfbeta1 mice to a tumour-developing K5-RAS transgenic line. Mice carrying both the K5-RAS and K10-Tgfbeta1 transgenes showed no difference in tumour formation compared to those harbouring just the K5-RAS transgene.

Human papillomavirus (HPV) infection has been widely implicated in cervical carcinogenesis, but it appears to be an early event, with other genetic abnormalities required for biological transformation. The initial aim of this thesis was to examine the relationship between HPV presence and type, and numerical chromosome abnormalities in 60 low and 30 high grade squamous intraepithelial lesions (SIL) and 25 invasive squamous carcinomas. This involved the technical development of; i) a sensitive HPV in situ hybridisation technique; ii) a PCR system for accurate determination of HPV type; and iii) interphase cytogenetic methodology using pericentromeric probes to chromosomes 1, 3, 11, 17, 18 and X. Eleven of 33 low grade lesions infected with high risk HPVs (HPV 16, 18, 30, 31, 33, 35, 39, 45, 51, 52, 56, 58 and 66) but none of 24 lesions infected with low risk HPVs (HPV 6, 11, 42, 43 and 44) showed basal cell tetrasomy of all six chromosomes in the HPVinfected areas. Of the high grade lesions, all of which were infected with high risk HPV, 8 showed basal cell tetrasomy and 6 were aneusomic. Most of the invasive carcinomas contained high risk HPV and were also aneusomic. Two distinct patterns were observed; i) mainly disomic with additional loss/gain of individual chromosomes; and ii) mainly tetrasomic with additional loss/gain of chromosomes. This suggested that two different pathways/mechanisms are involved in neoplastic transformation and that tetrasomy may in some lesions, be an intermediate state. Cell cycle progression and genome maintenance are regulated by cyclins and cyclin-dependent kinases. These not only provide controlled passage through the cell cycle but also provide 'checkpoints' or restraints to prevent replication of damaged DNA. In vitro studies have shown that E6 and E7 oncoproteins can bind cell cycle regulatory proteins and disrupt cell cycle control. This not only accommodates viral replication but may also contribute to the accumulation of chromosome abnormalities that lead to immortalisation and transformation. In the second part of this thesis, the relationship between the expression of cyclins D1, E, A and B1 and both the HPV type and the presence of numerical chromosome abnormalities was examined. The expression of cyclins E, A and B was elevated in lesions infected with both low and high risk viruses. Cyclin D1 expression was absent in the majority of lesions infected with high risk HPVs, in contrast with overexpression observed in the low risk infected lesions. Taken together with the chromosome abnormalities, the data provided evidence that the G;S and G2M cell cycle checkpoints are abnormal in lesions infected with high risk HPVs but not in those infected with low risk HPVs. Finally, raft culture systems of human keratinocytes that were; i) transfected with whole HPV 18 viral genomes; and ii) acutely infected with recombinant retroviruses expressing E6, E7 or E6/E7 genes of HPV 18, were analysed for karyotype and cyclin expression to investigate which HPV genes induced tetrasomy, and under what cell cycle conditions. The data indicated that tetrasomy occurs in association with E7 gene expression.

In-vitro mutagenicity assays have traditionally been used for first line identification of potential genotoxic hazard, purporting to chemical carcinogenesis and heritable genetic damage. The recent advances m combinatorial chemistry and high throughput screening technologies have led to a massive explosion in numbers of possible therapeutic candidates being produced at the early stages of drug discovery. This rapid increase in the number of chemicals to be classified results in a greater need for to acquire alternative methods for the prediction of toxicity. Quantitative StructureActivity Relationships (QSAR) can till this need for early hazard identifications by elucidating the physicochemical basis of biological activity. The assumption with predictive QSARs for toxicity is that "biological activity may be described as a function of chemical constitution". This thesis focuses on the Ames mutagenicity assay data for two compound sets; one of 90 compounds, with limited structural flexibility, comprising a range of chemical classes (non-congeneric series), the second, a set of 30 flavonoid compounds. Three physicochemical descriptor sets were generated: EV A, a theoretical molecular descriptor based on the normal co-ordinate modes of vibration; WHIM, derived from weighting functions applied to the 3D-structural molecular co-ordinates; and TSAR, a series of hydrophobic, electronic and steric parameters traditionally associated with the production of biological QSARs. Various "unsupervised" data pre-treatment methods were adopted, to reduce the level of degeneracy within the individual descriptor sets, prior to the calculation of stepwise linear discriminant classification functions. Good predictive models for Ames mutagenicity, as determined by leave-one-out (jackknife) cross-validation, were obtained with each of the three physicochemical descriptor sets. An increase in the predictive ability was observed following the combination of variables from the individual descriptor sets, inferring that some unique information associated with mutagenic activity is contained within each descriptor set. The predictive stability of the models produced was assessed via independent compound predictions, with a poor overall success rate determined. This failure in external prediction was investigated and fundamental differences in physicochemical data space occupancy revealed. Conclusions on training set composition and general model applicability are made with consideration to individual model physicochemical data space coverage.

The transforming genes of bovine papillomavirus type 4 (BPV-4) are encoded by the E7 and E8 open reading frames (ORFs). E7 is the transforming gene of BPV-4, in that, in co-operation with activated ras, it induces morphological transformation of primary bovine fibroblasts (PalF) in the absence of other viral genes, but the acquisition of anchorage independent growth requires the additional presence of E8. The BPV-4 E8 is a small transforming protein localized to cellular membrane. It consists of two domains: a very hydro'phobic region encompassing the first 30 amino acids of the protein, and a second region of mainly hydrophilic amino acids comprising the C terminal 12 residues. The results in this thesis demonstrated that in addition to the ability of E8 to grow independently of anchorage, PalF cells expressing E8 lose gap junction intercellular communication (GJIC), can grow in low serum, and are not contact inhibited. E8 also transactivates the cyclin A promoter in PalF cells. Mutant forms of E8 were generated to establish if the transforming functions of the protein could be segregated and therefore to define its functional domains. Mutations were introduced both in the hydrophobic domain and in the hydrophilic C-terminal tail, and chimeras with BPV-1 E5 were constructed. Cells expressing either E8 wild type or its mutants were analysed for their ability to: morphological transformation, anchorage independent growth, focus formation, cell population growth in low serum, tumorigenicity in nude mice, trans-activation of the cyclin A promoter, binding to ductin and down regulation of GJIC. The analysis of E8 mutants and chimeras constructed with BPV-1 E5 show that the multiple transforming function of E8 can be segregated and that both the hydrophobic domain and the hydrophilic C-terminal tail of E8 are critical for its functions and for the transactivation of the cyclin A promoter. These results support the hypothesis that the different aspects of cellular transformation produced by E8 might be due to interaction with different cellular targets. The observation from the analysis of the transformation parameters of E8 and BPV-1 E5 expressing cells suggest that E8 acts differently from E5. This study also demonstrates that the separate domains of E5 and E8 are not functionally interchangeable. The short term co-transfection analysis of E8 mutants suggest that substitution of alanine with proline, which is expected to alter the conformation of the hydrophobic domain, may have an effect on cell transformation. Also the short-term co-transfection experiments of E8 mutants in the putative casein kinase II site support the possibility that BPV-4 E8 might be phosphorylated by CKII and that this phosphorylation could have an effect on the biological activities of this protein.

Bovine papillomavirus type 4 (BPV-4) induces papillomas in the upper alimentary canal of cattle which can progress to carcinomas in animals feeding on bracken fern. Viral DNA is rarely detected in either naturally occurring or experimentally induced cancers. Similarly, BPV-4 DNA is seldom found in in vitro transformed established cells. These results suggest that presence of BPV-4 is not required for progression to or maintenance of the transformed state. The frequency with which BPV-4 DNA is lost both in vivo and in vitro also suggests that there may be active selection against whole or part of the BPV-4 genome. Several BPV-4-transfected established lines were analysed to examine whether this proposed negative selection was observable at the DNA level. In common with previous in vitro studies, BPV-4 DNA was found to be progressively lost on continued sub-culture. Analysis of cell lines containing BPV-4 DNA showed that there was no overt mutation or rearrangement of viral DNA sequences. The apparent wildtype organisation of viral genes included the E8 open reading frame (ORF). It was originally hypothesised that this viral ORF, which is the second major transforming gene of BPV-4, was an attractive target for negative selection due to the detrimental effect observed on transfecting primary cells with BPV-4 E8 DNA. Previous work also reported amplification and rearrangement of specific host sequences in BPV-4-transformed lines. It was therefore proposed that induction of cellular DNA amplification may be an important aspect of BPV-4 transformation activity. This was investigated in virally-transformed established cells. No BPV-4-mediated DNA amplification was apparent in any of the lines, although involvement of cellular regions of repetitive sequences in manifestation of such amplification may be indicated. The co-factors involved in BPV-4-associated carcinogenesis have been identified as including the mutagens, carcinogens and immunosuppressants present in bracken fern. One of major mutagens present is the flavonoid quercetin. Quercetin has discernible effect on BPV-4 transformation in vitro as it synergises with the virus to fully transform primary bovine fibroblasts (PalF cells). The work described in this thesis confirmed these initial results and extended these findings. Quercetin-treated cells showed a more aggressive transformed morphology than untreated transfectants, whether they had been transfected with whole genome BPV-4 or sub-genomic fragments. However, whereas in non-treated cells the E8 gene was required for anchorage independence, quercetin-treated cells containing the BPV-4 E7 gene alone were found to be capable of anchorage-independent growth. Conversely, and contrary to expectation, quercetin- treated cells transfected with the E8/E7 genes grew very poorly or not at all in semi-solid media. The reasons for the antagonistic action of E8 and quercetin are not yet understood. Independent of quercetin action, results also provided circumstantial evidence that the E8 oncoprotein is responsible for downregulation of gap junctional intercellular communication in BPV-4-transformed cells. BPV-4 does not have an E6 gene and immortalisation of BPV-4-transformed cells is achieved only in the presence of an exogenous E6 gene. Quercetin treatment however conferred immortality on cells transformed by whole genome BPV-4 or by the E7 gene alone. The observed synergism between BPV-4 and quercetin was also found to be dependent on time of treatment. Treatment of cells with quercetin either immediately before or after DNA transfection resulted in an increased degree of cellular transformation compared to that seen on lengthening the interval between treatment with the chemical and transfection with BPV-4 genes. PalF cells treated with quercetin immediately after transfection with the E7 gene were tumourigenic. The tumours grew far more aggressively than when cells were treated with quercetin before transfection with whole genome BPV-4. This indicated that the timing of exposure of viral products and quercetin is crucial and that in certain circumstances only the E7 gene is required for full malignant transformation.

The cytochromes P450 (CYPs), epoxide hydrolases (EHs) and glutathione S-transferases (GSTs) are three of the major families of enzymes involved in the metabolism of xenobiotics in the human body. Immunohistochemical analysis revealed a high frequency of expression of xenobiotic metabolising enzymes in all tumour types studied, in contrast to corresponding normal tissue which displayed only low levels of expression. Further examination of the CYP1 family was carried out by immunoblot analysis. All breast tumours studied were found to express CYP1B1, and not CYP1A1 or CYP1A2. Moreover, CYP1B1 was identified in a number of kidney tumours but not in corresponding normal kidney, indicating that CYP1B1 may be a tumour-specific form of CYP, RT-PCR, in combination with restriction digestion and DNA sequencing, was used to identify CYP mRNA species present in several tumour types. Although CYP1A1 mRNA was identified in breast carcinomas, CYP1B1 was found to be the most frequently expressed form of the CYP1 family in this tissue. CYP3A mRNA was also displayed by several breast tumours, and demonstrated by sequencing to be CYP3A5. A similar situation to breast tumours was observed in tumours of the gastro-intestinal and urinary tracts, with CYP1B1 being the most frequently expressed form of the CYP1 family, and only a small number of samples displaying evidence of CYP1A mRNA. The effects of the expression of xenobiotic metabolising enzymes in tumours may be complex, and depend upon the relative amounts of active protein present, but it is likely that they will exert an influence on both the development of carcinogenesis and the anti-cancer drug resistance of tumours.

Human papillomavirus is the most common sexually transmitted pathogen in the US and drives benign and malignant diseases, namely high-risk (e.g. HPV16) and low-risk (e.g. HPV6; HPV11) strains. High-risk HPV strains are responsible for 5% of cancers worldwide while low risk strains cause benign tumors including those in recurrent respiratory papillomatosis (RRP). HPV specifically infects the stratified squamous epithelium, where the virus can either remain latent, thus avoiding immune detection, or actively replicate and express viral genes in synchrony with epithelial differentiation. Despite decades of research, no antivirals exist to clear HPV infection and it remains unknown which specific epithelial cell subpopulation(s) support disease formation and/or persistence. Single-cell RNA sequencing technologies enable a granular understanding of distinct cell types and subpopulations within complex tissues. Thus, we applied scRNAseq to epithelial tissues to (1) identify the transcriptomes of keratinocyte subpopulations and (2) discover HPV reprogramming and viral-driven disease phenotypes in infected epithelium. In Part One, we utilized 3D organotypic epithelial rafts, a model of stratified squamous epithelium, to compare isogenic HPV16+ versus HPV16-naïve tissue. We identified twelve transcriptomically distinct keratinocyte clusters and spatially mapped select subpopulations, including actively replicating cells and those that comprise basal and differentiated layers. HPV16-driven reprogramming was evidenced by changes in gene signatures of defined subpopulations (e.g. basal cells), in the localization of replicating cells, and in the overall composition of the epithelium wherein we discovered an HPV-induced differentiated subpopulation. This subpopulation, termed HPV-induced differentiation dissonant epithelial nonconventional (HIDDEN) cells, was further determined to be driven by the transcription factor ELF3 and to be enriched in stages of HPV+ carcinogenesis, including in dysplastic cervical patient tissue and in published scRNAseq data on HPV+ vs HPV- head neck squamous cell carcinomas (HNSCCs). Additional analysis of the epithelial subpopulations in HNSCCs identified subpopulations enriched in HPV+ and HPV- tumors, which can now be related to HPV-infected and uninfected rafts to elucidate the respective cells of origin. Altogether, this work established an atlas of HPV16-infected and uninfected epithelium, demonstrated its utility for defining viral reprogramming, and informed functional studies to discover HPV and cellular drivers of infection and carcinogenesis with potential broad implications for HPV-associated disease. In Part Two, we (1) analyzed a newly established patient cohort of juvenile-onset RRP, (2) established a translational pipeline of papilloma and non-diseased matched patient tissues and organotypic raft models, and (3) applied sequencing technologies to this pipeline – including bulk RNAseq and the scRNAseq workflow developed in Part One. Importantly, we identified HPV11 (vs HPV6) as a risk factor for more severe disease and progression to malignancy, described a dissociation-type EMT phenotype in papilloma vs non-diseased cells, rafts and tissues, and defined two keratinocyte subpopulations that are greatly enriched in papillomatous tissue. Notably, one of these closely matches the HIDDEN cell signature in HPV16-infected tissues while the other harbors pathways related to epithelial carcinogenesis. This initial work supports further analysis of disease drivers, mechanistic studies targeting these drivers, and drug screens using pharmacogenomics approaches.

Fundamental to cellular differentiation, development, and physiology is the precise regulation of gene expression. At the center of this control lies chromatin, and disruptions in chromatin biology are key contributors to complex human diseases. By convention, gene expression analysis primarily focuses on transcript abundance, however the splicing of transcripts into distinct isoforms is also of biological relevance. One new form of alternative splicing, backsplicing, is responsible for generating circular RNAs (circRNAs), which are emerging as essential components to gene regulatory networks. However, unlike other forms of alternative splicing, the role that chromatin plays in backsplicing and circRNA formation remains poorly understood. This study examines the impact of chromatin on circRNA biogenesis and function. Using a Drosophila model, we investigate how the PARP1-chromatin complex affects backsplicing. Through native elongating transcript sequencing (NETseq), we uncover the role of PARP1 in regulating RNA polymerase II (RNAPII) pausing across circRNA host genes. Our findings reveal that PARP1-chromatin influences circRNAs biogenesis by altering RNAPII pausing within gene bodies, with outcomes contingent upon the architectural context of host genes. We find PARP1 regulates backsplicing by controlling RNAPII pausing, and the outcomes of this regulation are determined by the context of host gene architecture. Furthermore, PARP1's influence extends to RNAPII pausing within introns and exons, influencing the direction of transcriptional output to favor either circRNA or mRNA. We next explore the pathological implications of disrupted chromatin dynamics, particularly in the context of Rett Syndrome, arising from mutations in the chromatin-associated protein MECP2 a PARP1 nucleosome competitor. Employing a single-cell multi-omics approach within patient-derived cortical spheroids, we illuminate cell-type-specific alterations in chromatin accessibility and gene expression mediated by mutant MECP2. These changes correlate with disturbances in the maturation of MEIS2+ neurons and identify potential circRNA candidates associated with these cells in the context of Rett syndrome. The functional presence of these circRNAs in RTT-associated cell types is being tested. Future mechanistic studies will determine how these dysregulated circRNAs function to drive Rett pathology. Finally, we asked how circRNA might function in an epigenetically driven disease—arsenic (iAs)-induced lung cancer. We show that the expression of SATB2, a 3D-chromatin organizer linking chromatin to the nuclear matrix, is upregulated by iAs exposure in bronchial epithelial cells along with its circRNA, circ3915. We show both SATB2 and circ3915 collaborate to promote cancer cell proliferation, migration, and carcinogenic gene expression pathways characterized by acquisition of oncogenic NFE2L2 and KRAS-like gene expression signature. We also show that circ3915 is translated in iAs-transformed cells, where it’s peptide associates with full-length SATB2 to modify chromatin associations in the nucleus. This discovery suggests that iAs triggers SATB2 expression, leading to alterations in chromatin states that govern gene expression to promote oncogenesis. This work advances our understanding of how genome architecture can provide context to fundamental biological processes such as splicing. Our identification of a circRNA that can be translated is exciting and demonstrates how circRNA translation adds to protein biodiversity and function. These findings not only broaden our comprehension of the intricate connections between genome architecture and fundamental biological processes like splicing but also highlight the translational potential of circRNA and its significance in precision medicine as a source of biomarkers and therapeutic targets.

Carcinogenesis is a complex process involving somatic mutations in a number of key biological pathways. Studying cancer evolution is an important task which contributes to better understanding of cancer biology and facilitates identification of new therapeutic targets. We focus on two important questions in cancer evolution. The first question is to delineating the temporal order of pathway mutations during tumorigenesis. And the other question is to cluster patients into biologically meaningful cancer subtypes. We present new statistical methods to 1)leverage functional annotations of mutations to enhance estimation of the order of pathway mutations during carcinogenesis, 2) incorporate intra-tumoral heterogeneity information in pathway mutation order inference and 3) identify cancer subtypes and classify patients into cancer subtypes based on the order of pathway mutation.Our methods use a probabilistic approach to characterize the likelihood of mutational events from different pathways occurring in a certain order, wherein it giving greater weight to the orders that have more evidence to be consistent with the phylogenetic structures of the tumor samples. The functional impact of each mutation is incorporated to weigh more on a mutation that is more integral to tumor development. A maximum likelihood method is used to estimate parameters and infer the probability of one pathway being mutated prior to another. Furthermore, we propose a PMBC(penalized model-based clustering)-like method to leverage the temporal order of pathway mutations for cancer subtype classification, which leads to mechanistically interpretable cancer subtypes. Our method uses a penalized likelihood approach and an EM algorithm to simultaneously cluster patients into subtypes and select important temporal order features that contribute to discriminating different subtypes. Simulation studies and real data analyses demonstrate the ability of our methods to accurately infer the temporal order of pathway mutations and cluster patients into subtypes.

Abstract: Field Cancerization is a hypothesis for the formation of cancer in certain types of tissues. It proposes the idea that a tumour can form in a “field” of cells that are predetermined for the development of cancer. Further, it is hypothesized that these fields are mainly caused by the onslaught of carcinogens on the tissue. Lastly, field cancerization proposes that tumour recurrence is related to the tumour being excised without fully removing the surrounding field. The model we propose is a hybrid cellular automaton (CA) used to verify the previously stated propositions and determine how long cancer development will take. The CA is considered to be hybrid due to its’ rule depending on the results of partial differential equations (PDEs) and a multi-layer perceptron (MLP). The PDEs represent the spread of carcinogens in the tissue, while the MLP computes the effects of the carcinogens on the gene expression of the genes related to cancer development in the tissue under consideration. We apply the model to field cancerization of the tongue. Most of the parameters of the model were chosen and are not based upon real data, as the necessary data was not available. This includes the choice of substituting nicotine, which is a mutagen but not currently listed as a carcinogen, to represent the carcinogen impacts of smoking tobacco. According to Health Canada tobacco contains over 4,000 chemicals, of which more than 70 are carcinogens. Many researchers are investigating how nicotine contributes to the development of cancer due to its use in non-tobacco products such as e-cigarettes and nicotine patches. One such study by Sanner & Grimsrud suggests that nicotine has several cancer-causing effects including speeding up cell growth, it is poisonous to cells, it kick-starts a process that is an important step in the path toward malignant cell growth, and it decreases the tumour suppressor CHK2. Therefore, we considered the readily available data with regards to nicotine as an appropriate choice to substitute for the over 70 carcinogens in tobacco. The other carcinogen considered in this thesis is ethanol to represent alcohol consumption. It was found that nicotine was a more potent carcinogen than ethanol. The combined impact of both ethanol and nicotine resulted in more aggressive cancer growth. It was verified that removing the field results in recurrence taking longer to occur than if the field is not removed. We also tracked cell lineages and found that as the field develops, the number of distinct cell lines decreases. Finally, we found that most tumour masses formed via polyclonal origin instead of monoclonal origin, though both occur within the simulations.

Lung cancer is the deadliest form of cancer and resulted in 1.8 million deaths worldwide in 2020. While cigarette smoking is the most familiar cause of lung cancer, up to 25% of cases occur in non-smokers, thus other environmental agents are also causative. Hexavalent chromium [Cr(VI)] is a known lung carcinogen and poses occupational and environmental exposure risks relevant to humans, wildlife, and ecosystems. This dissertation considers the carcinogenic mechanisms of a highly potent, particulate, hexavalent chromium compound, zinc chromate. The molecular mechanism of carcinogenesis induced by Cr(VI) is not fully understood, but it is known that chromosome instability is a key effect. Chromosome instability refers to structural instability characterized by breaks and translocations, and numerical instability characterized by changing numbers of chromosomes. This dissertation focuses on how hexavalent chromium causes numerical chromosome instability in human lung cells and uses the One Environmental Health approach to gain insights into the associated mechanism using whale cells as a comparative model. The hypothesis of this project is: Prolonged hexavalent chromium exposure targets securin in human lung cells, leading to centrosome amplification and numerical chromosome instability, while the ability of whale cells to retain normal securin levels confers resistance to these effects. A main driver of numerical chromosome instability is centrosome amplification, defined in this study as a single cell with more than two centrosomes. We previously found Cr(VI) induces centrosome amplification, which increased with duration and concentration of exposure and correlated with Cr(VI)-induced aneuploidy. In Aim 1 we focused on a novel potential target of Cr(VI), securin. Securin is an important centrosome regulator because it is the canonical inhibitor of separase. Separase is the enzyme that causes centriole disengagement and permits centrosome duplication. Prolonged Cr(VI) exposure decreased securin protein levels in a dose-dependent manner. Securin protein loss was not due to changes in protein degradation, but rather a loss of securin mRNA. Three measures of securin function were analyzed to determine if decreased securin levels were sufficient to control separase activity. Separase cleaves itself, kendrin, and cohesin. Prolonged Cr(VI) exposure caused increased separase autocleavage, increased kendrin cleavage, and increased separation at centromeres caused by cohesin cleavage. Securin knockdown increased levels of aneuploidy after acute Cr(VI) exposure, in contrast to untransfected cells which retained normal background levels. Together these data showed Cr(VI) disrupts securin, a key protein in the maintenance of numerical chromosome stability. Aim 2 sought to uncover the mechanism of Cr(VI)-induced securin loss described in Aim 1. Transcription factors bind to gene promoter regions to enhance or inhibit gene transcription and have been shown to be disrupted by Cr(VI). Thus, we measured three securin-promoting transcription factors, Sp1, NF-YA, and E2F1, and two repressing transcription factors, KLF6 and p53. Cr(VI) elevated levels of securin-promoting factors in nuclear fractions, indicating Cr(VI) is not repressing their levels to decrease securin expression. E2F1 was previously shown to decrease with prolonged Cr(VI) exposure, but experimental knockdown revealed E2F1 is not a driving factor for Cr(VI)-induced securin loss. Securin expression loss was also not explained by p53 activation, as protein levels and phosphorylation did not correlate with securin loss. KLF6 nuclear protein levels were increased at prolonged timepoints, which may begin to explain reduction in securin levels. MicroRNA (miRNA) regulation is altered by Cr(VI) and is a possible mechanism for securin loss. Several miRNAs were found to be significantly up- and down-regulated at all tested concentrations and timepoints. Securin was not specifically targeted by any of the altered miRNAs. However, several centrosome-associated proteins are putative targets of Cr(VI)-induced miRNA alteration. The One Environmental Health approach acknowledges humans, wildlife, and ecosystems are connected by a shared environment and common toxicants such as Cr(VI). Aim 3 characterizes the effects of prolonged Cr(VI) exposure on whale cells. We measured securin levels and function in bowhead whale cells and found securin levels remained normal under all exposure conditions and neither premature centromere separation nor centrosome amplification were increased. We found Cr(VI) did not cause aneuploidy in sperm whale or bowhead whale cells. Interestingly, chromosome damage still occurred, indicating Cr(VI) is active inside whale cells. These data indicate maintenance of securin under Cr(VI) exposure may protect whale cells from developing centrosome amplification and numerical chromosome instability. Overall, this dissertation shows securin is targeted by prolonged Cr(VI) exposure in human lung cells and centrosome regulation pathways are central in the mechanism of Cr(VI) carcinogenesis. We showed whales are resistant to securin loss and the proposed downstream phenotypes, centrosome amplification and numerical chromosome instability. Together these data support the hypothesis Cr(VI)-induced securin loss leads to numerical chromosome instability.

Hexavalent Chromium [Cr(VI)] is a known human lung carcinogen and general health hazard. The mechanism of carcinogenesis remains poorly understood, but chromosome instability (CIN) is the major theory in its carcinogenic mechanism. Homologous recombination (HR) repair is a DNA repair pathway that prevents CIN by repairing DNA double-strand breaks. RAD51, a key mediator protein of HR repair, along with the RAD51 paralogs (RAD51B, C, D, XRCC2, and 3) are required for HR repair. During HR, RAD51 loads and forms a helical nucleoprotein filament structure to promote DNA strand exchange and stimulate pairing activity of DNA. Cr(VI) exposures have been shown to target RAD51 and prevent its loading in lung fibroblasts. The mechanism by which Cr(VI) impacts RAD51 paralogs to cause RAD51 dysfunction remains unknown. In this study, we investigate the effects of Cr(VI) on these paralogs and their complexes in human lung cells. This study found both acute and prolonged Cr(VI) exposure inhibits RAD51D repair response evidenced by decreased RAD51D foci formation, protein levels and gene expression. In contrast, Cr(VI) had minimal effect on XRCC3 repair function, suggesting RAD51D as a part of the BCDX2 complex may be a key initial target in Cr(VI)-induced loss of RAD51 function and HR repair.

Particulate hexavalent chromium [Cr(VI)] is a well-established human lung carcinogen. RAD51, a key protein in homologous recombination repair pathway, is inhibited after prolonged exposure to Cr(VI), leading to an increase in chromosome instability after prolonged exposures in human lung fibroblasts. chromosome instability is the proposed driver of Cr(VI) carcinogenesis. Since tumors from chromate workers develop from epithelial cells, we sought to translate these findings from human bronchial fibroblasts to human bronchial epithelial cells. We hypothesized Cr(VI) inhibits RAD51 after prolonged exposure leading to an increase in chromosome instability in human bronchial epithelial cells (BEP2D). We characterized the cytotoxicity and measured intracellular Cr ion levels, chromosome instability and RAD51 response. Altogether, the data show, in BEP2D cells, Cr(VI) induces DNA double strand breaks and targets RAD51 leading to an increase in chromosome instability, successfully translating the outcomes seen in human bronchial fibroblasts to human bronchial epithelial cells.

El desarrollo del trasplante hepático es el fruto de múltiples avances médicos. Los más importantes son la mejora en la técnica quirúrgica y el desarrollo de la inmunosupresión. A pesar de esto persiste una disminución de la supervivencia de los trasplantados hepáticos a largo plazo. Las principales complicaciones tardías del trasplante hepático son fundamentalmente las complicaciones renales, el desarrollo de neoplasias de novo y el aumento del riesgo cardiovascular. La inmunosupresión de mantenimiento de los receptores de trasplante hepático se basa en los inhibidores de calcineurina. Esto fármacos son parcialmente responsables de los efectos secundarios que conducen a las principales causas de mortalidad tardía de estos pacientes. Un planteamiento razonable es la disminución de la inmunosupresión, teniendo en cuenta el riesgo potencial de rechazo. Los objetivos de nuestro estudio fueron conocer la frecuencia con que se utiliza la minimización de la inmunosupresión en la práctica clínica en España y las indicaciones para la misma, el efecto de la minimización de la inmunosupresión sobre la función renal a largo plazo en los receptores de trasplante hepático y conocer el efecto de la inmunosupresión de mantenimiento en monoterapia con micofenolato mofetil sobre el desarrollo de neoplasias. Se realizó un estudio multicéntrico, transversal, acerca de la minimización de la inmunosupresión en receptores de trasplante hepático en diferentes centros de trasplante hepático de España y un estudio observacional retrospectivo, unicéntrico en el que se evaluó la posible influencia de la minimización de la inmunosupresión en forma de monoterapia con micofenolato mofetil sobre el desarrollo de neoplasias. El estudio multicéntrico es el primer estudio que analiza la minimización de los inhibidores de calcineurina en la práctica habitual, en este se objetiva que aproximadamente el 50% de los pacientes recibe inmunosupresión minimizada a largo plazo y que la minimización de la inmunosupresión varía entre los distintos centros hospitalarios. La principal indicación de minimización de la inmunosupresión es la insuficiencia renal y la minimización de la inmunosupresión puede ayudar a preservar o a mejorar la función renal. Finalmente, la inmunosupresión en monoterapia con micofenolato mofetil (interrumpiendo los inhibidores de calcineurina) se asocia a una reducción del riesgo de desarrollo de neoplasias tras el trasplante hepático. Este efecto se produce tanto sobre el desarrollo de neoplasias cutáneas como sobre el desarrollo de neoplasias no cutáneas.

Cadmium (Cd) is a ubiquitous pollutant in the environment and a known carcinogen for lung cancer. Cd has been shown to act as a weak mutagen, which suggests that it may exert tumorigenic effect through non-genotoxic ways, such as epigenetic mechanisms. The goal of this project is to investigate the mechanisms of Cd carcinogenesis focusing on the role of lncRNA dysregulations. The Cd-exposed cells formed significantly more colonies in soft agar, displayed cancer stem cell (CSC)-like property and formed tumors in nude mice. Mechanistically, the lncRNA microarray analysis revealed that chronic Cd exposure dysregulates lncRNA expressions. Q-PCR analysis confirmed the significant upregulation of the oncogenic lncRNA DUXAP10 level in Cd-transformed cells. Knockdown of DUXAP10 in Cd-transformed cells significantly reduced their CSC-like property. Further mechanistic studies showed that DUXAP10 activates the Hedgehog pathway to promote Cd-induced CSC-like property. Furthermore, it was determined that chronic Cd exposure upregulates DUXAP10 expression by inducing Pax6 expression. In addition to oncogenic lncRNA upregulation, Cd exposure was also found to downregulate the expression of a tumor suppressive lncRNA MEG3 in Cd-transformed cells. Meanwhile, the levels of DNMTs in Cd-transformed cells were found significantly elevated. Bisulfite-sequencing study revealed that the differentially methylated region (DMR) upstream of MEG3 is hypermethylated in Cd-transformed cells, indicating that the promoted DNMTs activity contributed to downregulation of MEG3. Stably expressing MEG3 in Cd-transformed cells decreased cell proliferation and induced CSC-like property. Further studies showed that MEG3 inhibits cell transformation by limiting cell proliferation and inducing apoptosis. Mechanistic studies revealed that MEG3 reduced cell proliferation by regulating the levels of cell cycle proteins and induced apoptosis by inhibiting the level of Bcl-xL. These findings suggest that dysregulations of lncRNAs play important roles in Cd carcinogenesis. As well as epigenetic dysregulations, we also determined the effect of genetic factor contributing to lung cancer. Enhanced EGFR signaling contributes to 60% of NSCLC cases. However, there is an unmet need to solve acquired resistance to tyrosine kinase inhibitors and low response rate for immunotherapy in lung cancer patients. This study was performed to investigate the role of SOCS3 in EGFR mutation-driven lung cancer and to explore the potential of its regulatory axis as therapeutic target for the development of novel approach. In our transgenic mouse model, overexpression of SOCS3 significantly inhibited tumor formation with mutated EGFR. Further investigation for the underlying mechanism revealed that SOCS3 downregulates YAP protein, which further suppressed Bcl-2 family proteins. External YAP inhibitor was shown to efficiently inhibit the growth of tumor organoids. In vivo studies demonstrated that SOCS3 downregulating YAP leads to less immunosuppressive tumor microenvironment. Lastly, SOCS3 was often found to be silenced in cancers. To mimic this circumstance, the therapeutic efficacy of utilizing external YAP inhibitor combined with anti-PD-L1 was assessed and showed promising outcomes. These results suggest the critical role of SOCS3 as a biomarker for the oncolytic immune environment and provide a novel insight for improving lung cancer immunotherapy.

Lung cancer is the leading cause of cancer death. Lung cancer is commonly associated with smoking, however, 1 in 5 women and 1 in 12 men who develop lung cancer are never-smokers. Environmental exposures, therefore, account for a significant portion of lung cancer cases. Hexavalent chromium [Cr(VI)] is a global environmental contaminant and known human lung carcinogen. Cr(VI) and other carcinogenic metals induce chromosome instability, an early event in lung cancer. Structural chromosome instability arises in part due to failed DNA repair. Particulate Cr(VI), the most potent form of Cr(VI), induces DNA double strand breaks and inhibits the high-fidelity DNA repair mechanism, homologous recombination. Specifically, the effector step of homologous recombination is affected shown by RAD51 failure. RAD51 failure is due to inhibited expression, inhibited localization to double strand breaks, or a combination of these two mechanisms. Little is known about the mechanisms of Cr(VI)-inhibited expression. However, Cr(VI) exposure results in downregulation of global expression, and it has been suggested epigenetic changes affect expression profiles after Cr(VI) exposure. Studies show changes in acetylation of the RAD51 promoter affect E2F1-mediated RAD51 transcription by altering the “histone code” as potential epigenetic mechanisms of inhibited expression. Studies also show changes in microRNAs are an additional epigenetic mechanism of Cr(VI)-altered expression, and this may provide an additional mechanism of inhibited RAD51 expression. The mechanisms of particulate Cr(VI)-induced RAD51 failure were investigated in a human lung cells, and key events were confirmed in a wildlife model, leatherback sea turtle (Dermochelys coriacea) cells. The leatherback model was included as a part of the One Environmental Health Approach to investigate particulate Cr(VI) carcinogenesis across species. This type of analysis is used to identify how two species with different environmental adaptations may have alternative responses to chemical exposures. Therefore, the hypothesis of this dissertation is: Prolonged exposure to particulate Cr(VI) inhibits RAD51 expression through E2F1-inhibited transcription and alteration of microRNA expression profiles, and these effects are paralleled in a leatherback sea turtle model. We found particulate Cr(VI) inhibits RAD51 and E2F1 nuclear and whole cell protein and mRNA levels in human lung cells. Therefore, we aimed to show E2F1 modulates the RAD51 response to particulate Cr(VI). We found E2F1 overexpression did not rescue particulate Cr(VI)-induced RAD51-failure after prolonged (120 h) exposure. However, when we knocked down E2F1 we found E2F1 knockdown does not inhibit RAD51 mRNA or protein expression but does reduce nuclear foci formation after acute (24 h) particulate Cr(VI) exposure when RAD51 is normally functional. These results suggest E2F1 may affect RAD51 localization to double strand breaks, but not expression after Cr(VI) exposure. As an alternative mechanism of inhibited RAD51 expression we next performed RNA sequencing (RNAseq) analysis to asses Cr(VI)-altered microRNA (miRNA) expression. This study showed Cr(VI) significantly affected global miRNA expression, a subset of which target homologous recombination genes and RAD51 expression directly. These data advance our understanding of how Cr interferes with a critical cellular pathway that contributes to carcinogenesis. We previously reported particulate Cr(VI) induces structural chromosome instability in leatherback lung cells similarly to data in human lung cells with some differences. In this dissertation we confirmed particulate Cr(VI) induces DNA double strand breaks in leatherback lung cells. In analysis of DNA repair we found lower levels RAD51 foci after prolonged particulate Cr(VI) exposure compared to acute exposure in leatherback lung cells. However, the sister chromatid exchange assay showed homologous recombination is functional after prolonged particulate Cr(VI) exposure. These results are dissimilar to results in human lung cells indicating there are significant differences in the mechanistic response to particulate Cr(VI) exposure between human and leatherback lung cells.

Upon entry into a host, carcinogens are subjected to a variety of Phase I and Phase II metabolic pathways that result in bioactivation or detoxification. The bioactivation pathways are of particular importance because they often generate DNA-damaging compounds. It would stand to reason that fully understanding these activation pathways, their outcomes, and their differences amongst individuals would aid in combating cancer. This dissertation focuses on the metabolic activation of two tobacco carcinogens: 4- (methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and N'-nitrosonornicotine (NNN). Concepts pertinent to this work such as tobacco carcinogenesis, chemistry of N-nitroso compounds, cytochrome P450 metabolism, and tobacco-specific nitrosamines are reviewed. The first study of this dissertation evaluates a hypothesized metabolic pathway for N-nitrosamines: processive P450 oxidation of NNK and NNN to N-nitrosamides. In this study, the three corresponding N-nitrosamides were synthesized, tested for stability, and monitored for formation in vitro. This study shows for the first time that N-nitrosamides are direct products of N-nitrosamine metabolism by cytochrome P450s. While these compounds were minor metabolites, their relative stability and DNA-damaging properties could impart biological relevance. Determining the generality of this metabolic pathway requires future work. The second study sought after the structures and abundance of stable 2'- deoxyadenosine (dAdo) damage (DNA adducts) induced by NNK bioactivation. This was accomplished by synthesizing hypothetical dAdo-adduct structures based on known reactivity and applying them to in vitro and in vivo assays. In vitro data indicates that N6- and N1-adducts are formed, however, in vivo data only shows N6-adduct formation, indicating extensive repair of N1-adducts. The relative abundance of these adducts were determined in rat liver and lung for three different treatment groups. The biological activity of these adducts requires future study. The last study measured direct biomarkers for human NNN metabolic activation for the first time by using [pyridine-D4]NNN-enriched tobacco. The deuterium-labelling allows NNN metabolites to be selectively measured by mass spectrometry and removes all interference by competing nicotine metabolites. This study is ongoing but current data suggests metabolic activation of NNN varies among individuals and is at least partially due to the activity of P450 2A6, the dominant enzyme for NNN bioactivation.

Genome instability is related to disease development and carcinogenesis. DNA lesions are caused by genotoxic compounds but also by the dysregulation of fundamental processes like transcription, DNA replication and mitosis. Recent evidence indicates that impaired expression of RNA-binding proteins results in mitotic aberrations and the formation of transcription-associated RNA–DNA hybrids (R-loops), events strongly associated with DNA injury. We identify the splicing regulator SLU7 as a key mediator of genome stability. SLU7 knockdown results in R-loops formation, DNA damage, cell-cycle arrest and severe mitotic derangements with loss of sister chromatid cohesion (SCC). We define a molecular pathway through which SLU7 keeps in check the generation of truncated forms of the splicing factor SRSF3 (SRp20) (SRSF3-TR). Behaving as dominant negative, or by gain-of-function, SRSF3-TR impair the correct splicing and expression of the splicing regulator SRSF1 (ASF/SF2) and the crucial SCC protein sororin. This unique function of SLU7 was found in cancer cells of different tissue origin and also in the normal mouse liver, demonstrating a conserved and fundamental role of SLU7 in the preservation of genome integrity. Therefore, the dowregulation of SLU7 and the alterations of this pathway that we observe in the cirrhotic liver could be involved in the process of hepatocarcinogenesis.

Nanoparticles, which measure 100 nm in at least one dimension, have surged in development, production, and use for a wide range of applications. However, the rapid pace of development for these emerging materials with unclear/unknown toxicity profiles makes it difficult to adequately assess health risk associated with exposure. One critical obstacle which limits scientific research to fill these critical knowledge gaps is the lack of accurate and predictive models for nanotoxicology studies, particularly those which involve occupationally relevant exposure scenarios (pulmonary exposure to low dose of particles in the circulating air). Typically, animal models are used to assess potential systemic toxicity. However, the time, cost, and resource heavy methods are insufficient due to the sheer number of new nanoparticles being produced and used each day. Cell culture-based systems have been suggested as a more rapid alternative which could be used to predict and rank potential toxicity of such emerging materials. However, the translation of an existing in vivo study to an in vitro model requires additional parameters to be considered for experimental design, and a lack to do so may contribute to the discrepancies and lack of clarity in the existing nanoparticle toxicity literature. The key objective for my dissertation work was to illustrate how the better integration of in vitro and in vivo may be used to evaluate potential toxicity of two nano-metal oxides with unclear toxicity profiles: iron oxide nanoparticles (IONP) and cerium oxide nanoparticles (CONP), as well as how alterations in specific physicochemical properties may contribute to severity of the pulmonary adverse outcomes they may induce. Some studies report IONP to be biologically benign, whereas others have linked IONP exposure to cancer-related adverse outcomes, including genotoxicity, neoplastic-like cell transformation, and tumor promotion in vivo. To assess this potential toxicity, a specific IONP (nFe2O3) was evaluated in a physiologically relevant low dose/long term in vitro exposure model to identify its general toxicity and potential carcinogenic capacity, as well as how alterations in particle surface chemistry (with the addition of an amorphous silica coating) may impact overall toxicity. Results were compared to previously published in vivo data which showed nFe2O3 would promote tumor formation in mice. CONP has a similarly unclear toxicity profile, with some studies suggesting CONP may reduce oxidative stress, whereas others show this particle to induce robust oxidative stress and inflammatory response. The chemical composition and valence state of CONP is known to be a critical component to its toxicity, and the inability to control for this in previous studies likely obscured any other physicochemical properties which may also play a role. Therefore, we evaluated size dependent toxicity of CONP using chemically identical particles, including their ability to induce inflammation, pro-fibrotic, and potential systemic toxicity within an occupationally relevant in vivo exposure model. Corresponding dose was used in an in vitro model system for a better understanding of potential mechanism underlying these outcomes, as well as to illustrate the potential use of an in vitro/in vivo integrated model system. Overall, my dissertation study results showed that an occupationally relevant low dose/long term exposure to IONP would induce neoplastic-like cell transformation in human bronchial epithelial cells, which was dramatically reduced with the alteration of its surface chemistry with an amorphous silica coating; Alteration of CONP size affected its inflammatory and pro-fibrotic response in male C57BL/6J mice, with smaller sizes inducing a robust and sustained inflammatory response, while larger sizes induced a milder response but with increased fibrotic potential. These results clearly show the impact of specific physicochemical properties on overall toxicity profile, as well as the importance of careful physiologically relevant model design for better in vitro to in vivo translation, with the ultimate goal of protecting those at risk of exposure to these new materials with unclear/unknown toxicity profiles.

The increasing frequency and severity of harmful algal blooms (HABs) has quickly become an environmental health concern worldwide. As rates of anthropogenic nutrient use rise with growing food demand, and as extreme precipitation events are expected to increase with the changing climate, rates of nutrient influx into watersheds are expected to increase. With increasing nutrient loads and temperatures in watersheds, many of the world’s water bodies and reservoirs are becoming eutrophic, establishing optimal conditions for HAB formation. HABs may change the ecology of water bodies; produce hypoxic zones resulting in fish deaths; and produce cyanotoxin compounds toxic to the liver, nervous system, and reproductive system of most eukaryotic organisms. Human and animal exposure to the most commonly occurring cyanotoxin, microcystin (MC), has been linked with hepatotoxicity, nausea, vomiting, and death in extreme circumstances. Within the MC family of toxins, microcystin-LR (MC-LR) is the most common, possesses the highest toxicity, and is cited as a suspected carcinogen. While HAB mitigation and MC exposure prevention efforts have often focused on large lakes and bodies of water, small lakes and ponds (SLaPs) remain understudied and unmonitored. SLaPs are the most numerous lentic bodies of water worldwide, providing vital ecosystem services and biodiversity support. Due to their location, low volume, and seasonal water level changes, SLaPs are at an increased risk for eutrophication and HAB formation. SLaPs have many uses including recreation, aquaculture, irrigation, and even drinking water sources in economically stressed areas of the world. The utility of SLaPs presents a potential exposure pathway to HABs and MC compounds. While our knowledge base of MC-related health outcomes continues to grow, several gaps exist, including: low dose, chronic exposure outcomes; differences in toxicity between pure toxic compounds and crude cyanobacterial extracts; and the cancer promoting role of MCs. Finally, current satellite-based remote sensing methods demonstrate several limitations in the study and monitoring of HAB formation in SLaPs, including: land adjacency effect contamination, prohibitive costs, and inadequate spatial and temporal resolutions. The overall goal of this work is to address the existing knowledge gaps pertaining to HAB appearance in SLaPs, MC toxicity from chronic and acute exposures, and to improve the current monitoring methods.

With the worldwide prevalence of cancer, this disease has constituted a major public health problem for many years without great success in the development of treatment. This is due in part to the large genetic heterogeneity that exists, even between cancers of the same class. The current standard of care focuses on targeting shared elements such as rapid cell division, yet treatments like chemotherapy have tremendous cytotoxic and off-target adverse effects. Recently, a different link has been discovered that is shared by cancers of diverse types: the disruption in the regulation and expression of microRNAs (miRNAs). miRNAs, produced endogenously from discrete genes that exist for them, have been shown to be master regulators of diverse processes, including growth and division. They serve to downregulate or silence the expression of certain genes at the post-transcriptional level, with each miRNA regulating multiple target messenger RNAs. This provides an important therapeutic potential because the correction in the expression of any one miRNA, especially if it has tumor-suppressive functions, has the ability to inhibit cancer formation, growth, and metastasis by multiple related pathways. The mechanisms by which the expression of miRNA is altered in cancer are reciprocal; that is, the processes that lead to oncogenesis alter the production of miRNA and altered production of miRNA can lead to increased tumorigenicity. In particular, it has been found that miRNA is globally downregulated in most cancers. The downregulation of miRNA can arise incidentally because of chromosomal abnormalities that occur with cancer, or miRNA can be directly repressed through transcriptional control, epigenetic modification, or disruption in the biogenic production process. When miRNAs are downregulated, the condition can contribute to the main hallmarks of cancer: uncontrolled growth, angiogenesis, metastasis, and evasion of anti-growth and apoptotic signals. A therapeutic strategy that has been investigated in numerous preclinical studies is miRNA replacement therapy, and this treatment has been shown to be extremely efficacious with a wide range of candidates and targets in diverse cancer systems. However, these advances have only led to two clinical trials, partly because of the lack of efficacious delivery methods. Nevertheless, with the rapid advances being made in gene therapy to address these shortcomings, a translation of these benefits from the bench-to-bedside is almost certain.

Cancer is the second leading cause of death in developed countries. Despite advances in both therapies and our understanding of the carcinogenesis process itself, consistently effective treatments remain elusive for a large number of cancer subtypes. For this reason, the minimization of cancer risk is a high priority subject. At the moment, cancer risk management is generally characterized by abstention rather than active engagement. The importance of avoiding activities associated with carcinogenesis-inducing genotoxicity, such as smoking, is well ingrained in the public consciousness as an anticancer strategy. The development of treatments that positively reduce genotoxicity poses a novel method of protection against general cancer risk, as well as the mitigation of risk from scenarios such as inflammation-inducing infections or genotoxic radiation exposure. The first part of this thesis describes the development of a genotoxicity mitigating compound, Yel002. In yeast and mouse models, Yel002 reduced both genotoxicity and lethality resulting from radiation exposure. Administration of this compound up to 24 hours after an otherwise lethal dose of ionizing radiation reduced radiation-induced hematopoietic depletion and systemic DNA damage markers. Although issues with compound stability made it difficult to use consistently, its development establishes a pipeline for the testing and use of similar compounds.The second part of this thesis work describes the characterization of a Lactobacillus strain, L. johnsonii 456, associated with anti-inflammatory and anti-genotoxic outcomes in mice. Although originally derived from a murine host, the strain survives in both in vivo and in vitro models of the human gut, suggesting that these protective effects could be applicable to humans as well. Importantly, the strain’s ability to inhibit pathogen growth and binding to human gut monolayer cells was demonstrated experimentally. As pathogenic bacteria are a major source of inflammation in the gastrointestinal tract, future probiotic intervention with this strain could result in active intervention strategies to manage local and systemic genotoxicity. The third part of this work is a reprint of the previously published review article, Mouse models for radiation-induced cancers, included with the gracious approval of Mutagenesis. The mouse models described represent important animal models for the testing of active genotoxicity management interventions. Finally, two appendices are included to describe subtle factors that influence the carcinogenesis pathway. Appendix I is a version of the article in submission Glyphosate Based Herbicides and Cancer Risk: A Post IARC Decision Review of Potential Mechanisms, Policy, and Avenues of Research, describing the heavily debated carcinogenic potential of the pesticide glyphosate and how it might exert effects through the microbiome. Appendix II is a reprint of the article BALR-6 regulates cell growth and cell survival in B-lymphoblastic leukemia, and elucidates a novel lncRNA-based mechanism of gene dysregulation in leukemic progression, to which I contributed bone marrow data analysis. As Yel002 suppresses leukemic progression in DBA/2 mice by an unknown mechanism, lncRNAs should be considered as a potential mechanism by which radiation mitigators affect hematopoietic stem cells. The active management of systemic genotoxicity by both probiotic and pharmaceutical methods is a relatively untapped area in cancer risk reduction. This work describes the characterization process of two potential mitigators of genotoxicity-induced cancer risk, and establishes baseline criteria by which future interventions of these types can be identified and tested.