Tesis sobre el tema "Cancer epigenetics"

Breast cancer it is the most common type of cancer and leading cause of cancer mortality among women worldwide. Women who inherit mutations in the breast cancer 1 susceptibility gene (BRCA1) are five times more likely to develop breast cancer than women who do not. However, only ~5-10% of breast cancer cases are due to germline mutations in tumor suppressor genes. There are currently no targeted therapies available triple negative breast cancers (TNBC), which often lack BRCA1 expression. BRCA1 is epigenetically silenced by the activated aryl-hydrocarbon receptor (AhR), suggesting that dietary antagonists of the AhR may inhibit BRCA1 silencing. Genistein is an isoflavone abundant in soy foods and its high consumption levels is thought to underlie the lower prevalence of breast cancer in Asian countries compared to Western countries. The hypothesis of this work is that genistein antagonizes AhR-dependent epigenetic silencing of BRCA1. To test this hypothesis we first determined the capacity of genistein to prevent AhR-dependent silencing of BRCA1 in estrogen receptor-alpha (ERα) expressing cells, with wild-type BRCA1 and inducible AhR (MCF-7). We also determined the effectiveness of genistein in reversing silencing of BRCA1 in ERα-negative cells with hypermethylated BRCA1 and constitutively active AhR (UACC-3199). The effect of genistein on BRCA1 promoter methylation and markers of cell proliferation was also determined in both cell lines.

Cancer is the 2nd most common cause of death in our society and is associated with high morbidity and costs. The word ‘cancer’ amalgamates the complex interplay between cells which have acquired genetic alterations leading to uncontrolled proliferation, i.e. the malignant cells, and genetically ‘normal’ host cells, i.e. stromal cells, vascular cells and inflammatory cells which all acquire modified biological phenotypes in the presence of malignant cells. This community of cells and their secreted proteins defines the tumour microenvironment. Stromal cells in the tumour microenvironment display characteristic biological changes which promote cancer growth. Little is known on the underlying regulatory mechanisms defining this phenotype. Epigenetics describes inheritable changes not encoded by the nucleic acid sequence. Epigenetic regulation has been described to occur in stromal cells in the tumour microenvironment, but little is known about its role on myofibroblasts. In this work I describe how oesophageal cancer derived stromal cells, i.e. cancer associated myofibroblasts (CAMs) accelerate tumour growth in vivo. I observed that CAMs not only affect the local tumour microenvironment but might also accelerate tumour growth at a distant site. I also show how myofibroblasts play an important role in early tumour niche formation in xenograft models and describe their disappearance and replacement by murine stromal cells during tumour progression. Oesophageal CAMs were shown to be epigenetically distinct from matched adjacent tissue myofibroblasts (ATMs). They exhibited a global DNA hypo- methylation compared to ATMs. We identified distinct DNA methylation signatures between oesophageal cancer CAMs and ATMs with the use of the Illumina 450k bead chip methylation array. The methylation array data showed altered methylation signatures of genes implicated Wnt/β-catenin signalling pathway. The transcription factor paired like homeodomain (PITX) 2 and the regulatory protein secreted frizzled like protein (SFRP) 2 both showed altered methylation signatures and expression patterns between oesophageal cancer CAMs and ATMs. I found that upregulation of SFRP2 in myofibroblasts induces angiogenesis and I hypothesise that epigenetic modification regulates myofibroblasts-derived SFRP2 expression which may play an important role in tumour neovascularisation. Based on these findings I conclude that ATMs and CAMs are epigenetically distinct and altered protein expression is at least partially regulated by altered DNA methylation. This work also presents a model for epigenetic modification of tumour stroma cells: exposure of myofibroblasts to the DNA methyl transferase inhibitor 5’Aza-3’deoxycytosine (DAC) lead to a mild decrease of global DNA methylation and induced persistent biological changes in myofibroblasts. These epigenetically modified myofibroblasts induced an accelerated xenograft growth when injected together with oesophageal cancer cells. Based on these experiments I conclude that DAC epigenetically modifies myofibroblasts which induces an activation of normally silenced genes leading to a biologically more active cell.

The risk of breast cancer transiently increases immediately following pregnancy. Hispanic women have one of the highest rates of postpartum breast cancers of all racial/ethnic minority groups in the US. The biology that underlies this risk window and the effect on the natural history of the disease is unknown. MicroRNAs (miRNAs) are small non-coding RNAs that have been shown to be dysregulated in breast cancer. In this study, we measured the miRNA expression of 56 tumors from a case series of multiparous Hispanic women and assessed the pattern of expression by time since last full-term pregnancy. A data-driven splitting analysis on the pattern of 355 miRNAs separated the case series into two groups: a) an early group representing women diagnosed with breast cancer ≤ 5.2 years postpartum (n=12), and b) a late group representing women diagnosed with breast cancer ≥ 5.3 years postpartum (n=44). We identified 15 miRNAs that are differentially expressed between the early and late postpartum groups; 60% of these miRNAs are encoded on the X chromosome. Ten miRNAs had a two-fold or higher difference in expression; miR-138, miR-660, miR-31, miR-135b, miR-17, miR-454, and miR-934 were overexpressed in the early versus the late group; while miR-892a, miR-199a-5p, and miR-542-5p were under expressed in the early versus the late postpartum group. The DNA methylation of three out of five tested miRNAs (miR-31, miR-135b, and miR-138) was lower in the early versus late postpartum group, and negatively correlated with miRNA expression. Taken together, the results of this study show that miRNAs are differentially expressed and differentially methylated between tumors of the early versus late postpartum, suggesting that potential differences in epigenetic dysfunction may be operative in postpartum breast cancers.

Changes in the epigenetic landscape are widespread in neoplasia, with de novo methylation and histone repressive marks commonly occurring in association with gene silencing. However, understanding the dynamics of epigenetic changes is often hindered due to the absence of adequate in vitro model systems that accurately reflect events occurring in vivo. Human mammary epithelial cells (HMECs) grown under standard culture conditions enter a growth arrest termed selection, but a subpopulation is able to escape from arrest and continue to proliferate. These cells, called post-selection cells, have many of the hallmarks seen in the earliest lesions of breast cancer, including transcriptional silencing and hypermethylation of the p16INK4A tumour suppressor gene. The overall aim of my thesis was to use post-selection HMECs as model system to identify and dissect the mechanism involved in early epigenetic aberrations. Firstly, using a microarray approach, I found that multiple members of the TGF-β signalling pathway were concordantly suppressed in post-selection cells, and this was associated with functional disruption of the TGF-β pathway. Interestingly, concordant gene suppression was not associated with aberrant DNA methylation, but with repressive chromatin remodelling. Secondly, to further understand the mechanism underpinning epigenetic silencing, I demonstrated using laser capture technology, that p16INK4A silencing is a precursor to DNA methylation and histone remodelling. Thirdly, I found that individual post-selection HMEC strains during the early passages shared a common 'wave' pattern of regional-specific methylation within the p16INK4A CpG island. Interestingly, the 'wave' pattern of early de novo methylation correlated with the apparent footprint of nucleosomes within the p16INK4A CpG island. Lastly, to further characterise the properties of the HMEC culture system, I demonstrated that post-selection cells do not possess a natural tumour-inducing property when transplanted into the mammary fat pad of immunocompromised mice. However, post-selection HMECs were associated with high expression of a variety of stem/progenitor markers, as well as stem/progenitor associated polycomb genes, thus demonstrating that these cells share some common features of stem/progenitor cells. The research presented in this thesis demonstrate that epigenetic changes occur early in the growth of post-selection HMECs and many of these changes are common in breast cancer.

Mitjançant tècniques d’anàlisi de tot el genoma hem sigut capaços d'identificar regions diferencialment metilades entre teixit de colon normal i tumoral provinents del mateix pacient. Un dels gens més importants és AKR1B1, l'illa CpG associada al promotor d’aquest gen esdevé hipermetilada en un 90% dels càncers colorectals estudiats (més de 200), aquesta troballa ha estat confirmada en dos sets de mostres independents. Sorprenentment, la hipermetilació no està acompanyada per una clara baixada de l’expressió d’AKR1B1, de fet vam observar que aquesta hipermetilació estava associada amb el silenciament dels gens AKR1B10 i AKR1B15 (tots ells membres de la mateixa família d’aldo-keto reductases), localitzats a 60 Kbs de l'illa CpG. Utilitzant tècniques per esbrinar l'estructura de la cromatina dins del nucli vam observar que l'illa CpG es troba en contacte amb el promotor de AKR1B10, el que indica una possible funció d’enhancer per a l'illa CpG. Aquesta funció enhancer va ser corroborada mitjançant assajos de luciferasa i demostrada in situ mitjançant l’alteració de la regió enhancer, utilitzant la tècnica d’edició genòmica CRISPR. A més a més, l’acció enhancer va ser recuperada en models cel·lulars gràcies a l’acció de diverses drogues epigenètiques, que produïen l’activació de l’enhancer (observable per la reaparició de la marca H3K27ac) i la reactivació de l’expressió d’AKR1B10 i AKR1B15. En aquest treball també s’avalua l’impacte de la hipermetilació en la via de l’àcid retinoic, a la qual pertanyen els enzims de la família AKR1B. Observem que la via està fortament hipermetilada i, en conseqüència, silenciada. El fet més remarcable a nivell clínic és que aquestes alteracions poden servir com a marcador de pronòstic de la malaltia i com a marcador de diagnòstic no invasiu.
By applying a genome-wide approach to detect differential methylation between colorectal tumors and their paired normal tissue we identified a number of genes that suffer hypermethylation in colorectal cancer. One of the most interesting genes was AKR1B1, which promoter CpG island became hypermethylated in about 90% of all the colorectal cancers samples analyzed. Unexpectedly, this hypermethylation was not accompanied by a clear downregulation of the AKR1B1 transcript. After extending the analysis to the neighboring genes, we realized that this hypermethylation was actually associated with silencing of AKR1B10 and AKR1B15 genes (all of them members of the same aldo-keto reductase gene family), located 60 Kb upstream of the methylated CpG island. Using techniques to elucidate the chromatin structure within the nucleus, we realized that the CpG island was in close contact with the AKR1B10 promoter, what indicates a putative enhancer role for the CpG island. This enhancer function was checked using luciferase assays and was demonstrated in situ through the alteration of the enhancer region, using the genome editing technique CRISPR. Moreover, the enhancer activity was recovered using cellular models through the treatments with different epigenetic drugs, which produced the activation of the enhancer (with the characteristic H3K27ac mark) and the reexpression of AKR1B10 and AKR1B15 genes. In this work we also evaluated the impact of the hypermethylation in the retinoic acid pathway, where AKR1B enzymes belong to. We observed a strong hypermethylation of the retinoic acid pathway and, as a consequence, silencing. The most remarkable finding is that those alterations can be used to predict the outcome of the disease and also as a non-invasive diagnostic marker.

Epigenetics plays an important role in carcinogenesis with heritable changes in DNA methylation and histone modifications intricately linked to the initiation, promotion, and progression of cancer. Evidence shows that a number of chemical and physical agents can induce epigenetic changes during carcinogenesis. Two such agents, estrogen and ionizing radiation, are generally recognized as being carcinogenic. Yet the epigenetic repercussions of these carcinogens remain relatively unknown. More importantly, the combined effect of these carcinogens has never been addressed in vivo from an epigenetic standpoint. Therefore, we focused on the effect of estrogen and ionizing radiation applied separately or in conjunction. We have found that the exposure to estrogen, either alone or in combination with radiation, induced pronounced morphological alterations, which was paralleled by modifications to the epigenomic landscape in the mammary gland. The results obtained from these rodent models can potentially be extrapolated to humans.
xiv, 190 leaves : ill. (chiefly col.) ; 29 cm

The aims of the work presented in this thesis were: to investigate the role of methylation of 14-3-3σ (a key regulator of p53-mediated G2/M arrest and of translational control during mitosis) in colorectal cancer using colorectal cancer cell lines and fresh colorectal tumours; to investigate any relationship between 14-3-3σ methylation status and gene expression; to determine whether aberrant methylation is associated with cell cycle defects and other factors known to contribute to colorectal carcinogenesis. PCR bisulphite sequencing showed that 78% (7/9) of colorectal cancer cell lines were unmethylated in the 14-3-3σ upstream promoter region (UPR). The unmethylated cell lines expressed high levels of 14-3-3σ, while methylated cell lines expressed negligible levels of 14-3-3σ protein or mRNA. Methylated colorectal cancer cell lines were treated with 5-aza-2’-deoxycytidine and demethylation was confirmed by MSP analysis. However, demethylation did not induce 14-3-3σ re-expression in the methylated cell lines, suggesting that CpG methylation may not be the only mechanism of transcriptional control. In contrast to colorectal cancer cell lines, 90% (89/99) of fresh colorectal tumours were methylated at CpG dinucleotides within the 14-3-3σ UPR. Bisulphite sequencing analysis of individual clones from 14-3-3σ methylated tumours (n =3) demonstrated that the clones displayed methylated CpG levels of approximately 41%. In agreement with previous PCR bisulphite sequencing analysis, there were a low percentage of methylated CpG dinucleotides (~ 15%) in clones from the 14-3-3σ unmethylated tumours (n =3). Unmethylated tumours expressed significantly higher levels of 14-3-3σ in comparison to methylated tumours (p =0.03), indicating that 14-3-3σ methylation may be associated with expression. PCR bisulphite sequencing analysis of matched normal mucosa tissues indicated that the 14-3-3σ UPR was methylated in all samples. Preliminary studies therefore suggest that there is tumour-specific loss of 14-3-3σ methylation in colorectal tumours within the 14-3-3σ UPR and CpG island. There were no apparent clinico-pathological correlations with 14-3-3σ methylation status. Whilst 14-3-3σ methylation was associated with expression in fresh colorectal tumours, there was no significant difference in expression levels between unmethylated colorectal tumours and matched methylated normal tissue. Bisulphite sequencing analysis of individual clones from normal tissues (from patients free of cancer) revealed that the 14-3-3σ UPR and CpG island was methylated at the majority of CpG sites analysed in colonic tissue (422/495, ~85.2%) and approximately half (795/1557, 51.1%) of CpG sites in skin samples (n =3). Furthermore, higher levels of 14-3-3σ protein were observed in skin tissue samples compared to normal colonic tissue, suggesting that 14-3-3σ CpG island methylation may be associated with tissue-specific expression. Experiments to assess the relationship between 14-3-3σ methylation and general methylation defects, suggest that methylation differences in 14-3-3σ were not simply a consequence of more general methylation phenomena well described in colorectal cancer. Nearest Neighbor analysis showed no evidence of generalised hypomethylation. Furthermore, MethyLight analysis of the CpG Island Methylator Phenotype (CIMP) showed no relationship between 14-3-3σ methylation status and CIMP; since, 1/5 (20%) tumours methylated at 14-3-3σ UPR and 1/5 (20%) tumours unmethylated at 14-3-3σ UPR were CIMP positive. In vitro functional assays showed that overexpression of 14-3-3σ in SW480 cells (14-3-3σ methylated) delayed the apoptotic response to UV-C, compared to control SW480 cells. This suggests that 14-3-3σ may protect colorectal cancer cells from apoptosis. MTT assays showed that overexpression of 14-3-3σ in SW480 cells resulted in a trend of increasing proliferation with a significant increase on day 4, compared to controls SW480 cells (p <0.01). Furthermore, FACS-sorted SW480 cells overexpressing 14-3-3σ, showed a significant shift to S-phase from G1 compared to control SW480 cells (p <0.01). Western blot analysis and immunohistochemistry revealed no relationship between p53 status and methylated 14-3-3σ in fresh tumours, while there was no relationship between published p53 status for colorectal cancer cell lines and 14-3-3σ methylation status defined experimentally. I have presented data which shows that methylation status of 14-3-3σ varies between colorectal cancer tissue, colorectal cancer cell lines and normal colonic tissue. Overexpression of 14-3-3σ appears to contribute to colorectal cancer carcinogenesis, raising the hypothesis that 14-3-3σ expression and function may at least in part be dependent on CpG methylation.

Colorectal cancer (CRC) arises from a multi-step process leading to the progressive accumulation of genetic and epigenetic mutations, thus causing deregulation in homeostasis followed by neoplastic transformation.Epigenetic and genetic alterations are able to induce a constitutive activation of the WNT signaling pathway, whose aberrant activity converges into deregulation of proliferation, differentiation and cell death pathways. The most common causes of WNT pathway hyper-activation are APC loss of function, or b-catenin constitutive activation mutations. Despite this knowledge of aberrant WNT activity, upstream interference with this signaling pathway induces adverse effects due to high cross-talk with other pathways, highlighting a need to find alternative ways to indirectly target the effectors of this pathway. Deregulated pathways in CRC provoke aberrant signaling that converges into the nucleus where transcription and chromatin-remodeling factors cooperate to maintain or modify the identity of a cell. In recent years, several studies have been focused on epigenetic players, which act by depositing specific and reversible post-translational modifications. For this reason they are being recognized as promising new targets for the development of cancer therapeutic strategies. In this context, my project takes advantage of murine 3D intestinal organoid cultures, carrying oncogenic deregulations of the WNT pathway, as a platform for pooled and arrayed RNA interference screens, to identify novel regulators that control the nuclear/transcriptional aspect of this oncogenic pathway. I also implemented the validation of selected targets in human metastatic colorectal cancer organoids to highlight their clinical relevance. Moreover, since several chromatin modifier inhibitors have been already developed, the findings of this project should prompt the development of new molecules for CRC treatment to target the novel sensitivities I identified. Finally, this project generated important technical knowledge through this pioneering approach that will open up the possibility of performing functional screens in other tissues from which organoid cultures have already been established.

Epigenetic deregulation contributes significantly to the development of multiple human diseases, including cancer. While great effort has been made to elucidate the underlying mechanism, our knowledge on epigenetic regulation is still fragmentary, an important gap being how the diverse epigenetic events coordinate to control gene transcription. In the first part of our study, we demonstrated an important link between Snail-mediated transcriptional control and epigenetic regulation during cancer development. Specifically, we found that the highly conserved SNAG domain of Snail sequentially and structurally mimics the N-terminal tail of histone H3, thereby functions as a molecular “hook”, or pseudo substrate, for recruiting histone lysine specific demethylase 1 (LSD1) repressor complex to the E-cadherin promoter. Furthermore, we showed that Snail and LSD1 are both required for E-cadherin repression and EMT induction, and their expression is highly correlated with each other in multiple human tumor tissues. Our findings have important clinical ramifications in that compounds mimicking the SNAG domain may disrupt Snail-LSD1 interaction and inhibit EMT and metastasis. In the second part of our study, we designed a batch of compounds based on the structure of the SNAG domain and are currently screening for candidates capable of competing with SNAG peptide for LSD1 binding. In addition, we applied a peptide pulldown/mass spectrometry-coupled analysis to identify SNAG-interacting proteins, among which are many chromatin enzymes and modulators. Functional characterization of these proteins will help to elucidate the Snail-mediated epigenetic regulation process. In the third part of our study, we found that Snail interacts with poly(ADP-ribose) polymerase 1 (PARP1) through a potential pADPr-binding motif and is subject to poly(ADP-ribosyl)ation, which can stabilize the Snail-LSD1 complex for enhanced PTEN suppression under DNA damage condition. Our findings added another layer to the delicate Snail transcriptional machinery, and indicated that PARP inhibitors may be applied in combination with conventional chemotherapies to target cancers with high expression of Snail and LSD1. In summary, we demonstrated that Snail cooperates with multiple epigenetic machineries to induce EMT as well as survival of tumor cells. Our findings contribute to a better appreciation of Snail-mediated epigenetic network as well as diversification of therapeutic strategies against cancer.

Integrin α6β4 is upregulated in pancreatic carcinoma, where signaling promotes metastatic properties, in part by altering the transcriptome. Such alterations can be accomplished through DNA demethylation of specific promoters, as seen with the pro-metastatic gene S100A4. I found that signaling from integrin α6β4 dramatically upregulates expression of amphiregulin (AREG) and epiregulin (EREG), ligands for the epidermal growth factor receptor (EGFR), and that these ligands promote pancreatic carcinoma invasion. To determine if AREG and EREG are regulated by DNA methylation, pancreatic cancer cells with low AREG and EREG expression were treated with the DNA methyltransferase inhibitor 5-aza-2’-deoxycytidine (5-Aza-CdR), resulting in stable overexpression of AREG and EREG, and this induction required signaling from integrin α6β4. Similarly, treatment of cells with high integrin α6β4 with the methyl donor S-adenosylmethionine inhibited gene expression of AREG and EREG. Whole genome bisulfite sequencing on pancreatic cancer cells reveled hypomethylation of the promoter regions of AREG and EREG when integrin α6β4 is high, and these regions correspond to H3K27Ac, indicative of enhancer location. Interestingly, I also observed genome-wide DNA demethylation, and a large proportion of altered CpGs correspond to potential enhancers. It is currently accepted that active DNA demethylation occurs via DNA repair. I tested this hypothesis by treating cells with Gemcitabine, which inhibits multiple components of DNA repair, including DNA demethylation mediated by GADD45A. Gemcitabine treatment resulted in marked reduction in AREG and EREG expression. To further test the involvement of GADD45A, I used RNAi-mediated knockdown or cDNA overexpression to alter GADD45A levels. In both instances, AREG and EREG expression positively correlated with GADD45A, particularly when integrin α6β4 is high, indicating that GADD45A is a rate-limiting step in AREG and EREG overexpression. Similarly, using stable shRNA, I show that Thymine DNA Glycosylase (TDG), and TET1 known modulators of DNA demethylation, are required for AREG and EREG expression in integrin α6β4 high cells, and nuclear localization of TDG is much higher in cells with high integrin α6β4. Using a specific inhibitor I found that AREG and EREG expression is dependent on Parp-1. Finally, I determined that integrin α6β4 signaling enhances cells ability to respond to and survive in the presence of DNA damage, and that active DNA repair is required for integrin α6β4 mediated DNA demethylation. Taken together, these data indicate that DNA repair is required to maintain overexpression of AREG and EREG in response to signaling from integrin α6β4 and that integrin α6β4 promotes this overexpression by enhancing DNA repair.

I have investigated epigenetic mechanisms of acquired endocrine-resistance in breast cancer using an in vitro model system based on estrogen-dependent MCF7 cells and their derivatives, LCC1 and LCC9. LCC1 cells, derived from MCF7 after passage in ovariectomised mice and routinely cultured in vitro in the absence of estrogen, exhibit estrogen-independent growth. They retain sensitivity to tamoxifen and fulvestrant. LCC9 cells, derived from LCC1 cells by growing them in increasing concentrations of fulvestrant, are completely estrogen-independent and are resistant to fulvestrant and cross-resistant to tamoxifen. When compared to MCF7 cells, LCC1 cells have marked up-regulation of the estrogen receptor α (ERα) protein that is not concomitant with increased estrogen receptor 1 (ESR1) transcription, suggesting a role for estrogen in controlling the proteasomal degradation of ERα. However, despite being grown in the same estrogen-deprived conditions, LCC9 cells do not have up-regulated ERα levels. As LCC1 cells retain sensitivity to tamoxifen and fulvestrant, these data suggest that LCC1 have developed estrogen-independence through ERα uncoupled from its ligand. However, LCC9 cells appear to have developed an alternative mechanism which is not dependent on ERα, presumably explaining their resistance to fulvestrant. I have studied global gene expression changes in the presence and absence of estrogen in these cell lines, using oligonucleotide microarrays, and correlated these data with global DNA methylation data derived from methylation arrays, which interrogate the methylation status of approximately 27,000 CpG dinucleotides in the genome. The analysis led to the discovery of more than 5,000 genes that were potentially either up-regulated or down-regulated by estrogen in MCF7 cells, either directly or indirectly. The transcriptional response to estrogen was generally muted in LCC1 and LCC9 compared with MCF7, but was not completely absent. I used various methods based on differential gene expression to parse the data, including gene ontology analysis, aiming to select genes for further mechanistic study. However, none of these methods led to the conclusive identification of a specific gene (or set of genes) that might have accounted for the physiological differences between the cell lines. In one strategy, I reasoned that, as the endocrine-resistant cells had lost their estrogen-dependence, genes involved might be regulated in an estrogen-dependent manner in MCF7 cells, without exhibiting misregulation in LCC9. This led to the identification of DUSP1 as a candidate gene, which was taken forward for mechanistic study because of its potential role in regulating ERα expression. However, when over-expressing DUSP1 in LCC9 cells, I could not demonstrate any effect on ERα levels. The final approach taken was to identify genes that might have been epigenetically deregulated, being both estrogen-regulated and deregulated in association with aberrant DNA methylation in the estrogen-independent cell lines. Surprisingly, given the phenotypic differences between the cell lines, only a very few genes were significantly methylated between cell lines. Of those that were differentially methylated between MCF7 cells and LCC1/9, only three exhibited the expected inverse correlation between methylation and expression. Of these, the gene CYBA was selected for further investigation. CYBA is a critical component of the NAPDH oxidase complex which is involved in generating oxygen free-radicals. My work suggests CYBA expression is estrogen-dependent, and that chronic estrogen deprivation leads to the epigenetic inactivation of CYBA in breast cancer cells. I speculate that the epigenetic suppression of CYBA may protect cells from the oxidant damage that results from estrogen deprivation and may be part of the mechanism that leads to acquired endocrine-resistance in previously sensitive cells.

Ovarian cancer is the eleventh most frequently diagnosed cancer in women and the fifth most common cause of cancer-related deaths. Epithelial ovarian cancer accounts for ~90% of cases. Serous epithelial ovarian cancer (SEOC) is the most common and aggressive histological subtype. Clinically, SEOC can be divided into type I and II tumours. Type II tumours (high grade serous ovarian carcinoma (HGSOC)), are generally diagnosed at later stages, grow aggressively and possess a mutation in the TP53 gene, which codes for the tumour suppressor protein p53. Mortality rates of HGSOC have remained largely unchanged for decades due to a lack of sensitive biomarkers and poor response to primary treatment. Finding new strategies to treat patients and identify early disease is critical to improving patient outcomes. Post-translational histone modifications describe the chemical and protein alteration of histone tail residues which are capable of altering gene expression. Monoubiquitination of lysine 120 on histone H2B (H2Bub1) disrupts chromatin strands and has important physiological functions including transcription, DNA repair, differentiation and crosstalk. Recent research has demonstrated that H2Bub1 is lost in cancer suggesting it may play an important role in malignancy. Interestingly, several studies have reported a link between the tumour suppressor p53 and H2Bub1, including a direct interaction of p53 with the H2Bub1 E3 ligase RNF20, and H2Bub1 enrichment at the p53 target gene CDKN1A. Consequently, this link may prove important in HGSOC where TP53 mutations occur almost ubiquitously and result in the loss of wildtype p53 function. 2 At the commencement of this work no study had comprehensively investigated the relationship between TP53 mutation status and p53 immunohistochemistry (IHC) using massively parallel sequencing (MPS) in HGSOC. The aim of this work was to determine the accuracy of p53 IHC to determine TP53 mutation status. We used MPS and IHC to characterise HGSOCs for TP53 mutation and p53 expression. Mutations in TP53 were identified in 94% (68/72) of HGSOCs, 62% of which were missense. Missense mutations demonstrated high p53 staining by IHC, as did 35% (9/26) of non-missense mutations. Low p53 staining was seen by IHC in 62% of HGSOC associated with non-missense mutations. Most wildtype TP53 tumours (75%, 6/8) displayed intermediate p53 expression levels. The overall sensitivity of detecting a TP53 mutation based on classification as ‘Low’, ‘Intermediate’ or ‘High’ for p53 IHC was 99%, with a specificity of 75%. We suggest p53 IHC can be used as a surrogate marker of TP53 mutation in HGSOC; however, this will result in misclassification of a small proportion of TP53 wildtype and mutant tumours. H2B monoubiquitination can be catalysed by five E3 ligases, and removed by up to eleven deubiquitinases. Loss of H2Bub1 has been reported in a range of malignancies, where it associates with higher tumour stage/grade, metastasis and poor patient survival. The aim of this section of work was to investigate H2Bub1 levels in HGSOC and correlate them with p53 mutation status/IHC, the deubiquitinase USP7 and the polycomb repressive complex subunit enhancer of zeste homolog 2 (EZH2), which is capable of negatively regulating H2Bub1 levels by competing for the same substrate. H2Bub1 was lost in 68% of HGSOC samples, but did not correlate with overall patient survival. P53 intermediate IHC and wildtype TP53 status correlated with higher H2Bub1 IHC staining, while USP7 expression was inversely correlated with H2Bub1 IHC. EZH2 expression did not correlate with H2Bub1 IHC or with patient overall survival. In vitro experiments using wildtype and mutant p53 3 constructs transfected into p53 functionally null (herein referred to as null) cell lines appeared to contradict the p53 and H2Bub1 IHC data, warranting further investigation. Along with other studies, our data suggested a link between p53 and H2Bub1. Consequently, this work aimed to investigate the effects of p53 on H2Bubl levels. We treated p53 wildtype cell lines with cisplatin, which activated p53 and resulted in global loss of H2Bub1 levels. In contrast, treatment of p53 null cell lines did not significantly alter H2Bub1 levels. Transfection of wildtype p53 into the same null cell lines resulted in a significant decrease in H2Bub1 levels compared to the vector only control. Downregulation of the main H2Bub1 E3 ligase, RNF20, and treatment with cisplatin showed a decrease of p53 levels in RNF20 sRNA treated cells compared to control siRNA. Chromatin immunoprecipitation coupled with next generation sequencing (ChIP-seq) of a p53 wildtype cell line treated with cisplatin, demonstrated H2Bub1 enrichment at p53 target genes. RNA-seq demonstrated an increased expression of these H2Bub1 enriched p53 target genes. Downregulation of RNF20 using siRNA, sensitised MCF7 and A2780 cells to cisplatin treatment. Together, this work suggested a functional link between H2Bub1 and p53. Chemoresistance is a hallmark of HGSOC and has been linked to the presence of cancer stem cells (CSCs). Ovarian CSCs make up a small subpopulation of cells within the primary tumour mass and express the markers CD133 and ALDH. Recent reports have suggested H2Bub1 plays an important role in stem cell differentiation and maintenance. The aim of this section was to investigate the role of H2Bub1 in ovarian CSCs. H2Bub1 was shown to be depleted in CD133+ cells and spheroids enriched for ALDH, consistent with a more stem-like phenotype. Decreasing H2Bub1 levels through RNF20 downregulation resulted in a profound effect on the CSC populations, where ALDH+ populations increased and the CD133+ 4 populations decreased. Spheroid culture of the RNF20 downregulated cells resulted in larger and more numerous spheroids than the shRNA control suggesting RNF20 and hence H2Bub1, can influence spheroid formation. Together, this data suggests H2Bub1 may play an important role in CSC differentiation.

Biomedical Sciences
Ph.D.
Epigenetic silencing is often altered in cancer and is a target for drug discovery. Unbiased screens in live cells are performed to identify potential novel targets of epigenetic therapy, and these screens have identified drugs that were not previously recognized to be involved in epigenetic reactivation of gene silencing such as cardiac glycosides and a CDK9 inhibitor. Recently, our lab performed a whole genome siRNA screen in combination with DNMT inhibition. One of the top targets revealed in this screen was the splicing factor SF3B1. SF3B1 is a well-known crucial splicing factor and is mutated in several cancers. However, its role in epigenetic regulation has not been well studied. I propose SF3B1 is a novel target for epigenetic therapy in cancer. In the YB5 colon cancer cell line where GFP is under the control of a methylated CMV promoter, I validated the screen results and found 0%, 1.0% and 5.3% GFP+ cells after treatment with siControl, siSF3B1 or the DNA methyltransferase inhibitor decitabine (DAC), respectively. DAC and siSF3B1 were synergistic, inducing 17.2% GFP+ cells. This synergy was also seen in an additional live cell assay and with other SF3B and SF3A family proteins. RNA-Seq analyses showed 423 genes upregulated by siSF3B1, 430 genes induced by DAC, and 1190 induced by the combination. siSF3B1 resulted in aberrant splicing of 695 genes, but there were only 27 genes overlapping between splicing alterations and gene expression changes, suggesting different mechanisms. Genes regulated upon siSF3B1 treatment were enriched for the TATA motif in their promoters, and the TATA-Box binding protein (TBP) was among the genes differentially spliced after siSF3B1. DNA methylation analyses showed demethylation synergy between siSF3B1 and DAC. Finally, the effects of siSF3B1 were phenocopied by treatment with the pan-SF3B inhibitor Pladienolide B (PB). GFP was reactivated in two separate colon cancer cell lines upon treatment with PB with synergistic activation when combined with DAC in YB5 cells. Thousands of genes were regulated and alternatively spliced with PB treatment alone, and among the differentially spliced genes was TBP. Furthermore, PB treatment with DAC induced demethylation significantly more than with DAC treatment alone. Genes regulated upon SF3B1 loss and inhibition were enriched for p53 target genes. Indeed, there was reduced cell proliferation and cell cycle arrest when SF3B1 was inhibited. This study demonstrates that the splicing factor SF3B1 has unexpected effects on gene transcription and targeting SF3B1 is synergistic with DNA methylation inhibition suggesting clinical potential for the combination.
Temple University--Theses

Biomedical Sciences
Ph.D.
DNA hypermethylation is known to contribute to the formation of cancer and this process has been widely studied. However, DNA hypomethylation has received far less attention and the factors controlling the balance between hypo and hypermethylation and its impact on tumorigenesis remains unclear. TET1 is a DNA demethylase that regulates DNA methylation, hydroxymethylation and gene expression. Full length TET1 (TET1FL) has a CXXC domain that binds to un-methylated CG islands (CGIs). This CXXC domain allows TET1 to protect CGIs from aberrant methylation but it also limits its ability to regulate genes outside of CGIs. This dissertation reports a novel isoform of TET1 (TET1ALT) that has a unique transcription start site from an alternate promoter in intron 2, yielding a protein with a unique translation start site. Importantly, TET1ALT lacks the CXXC domain but retains the catalytic domain. TET1ALT is repressed in ESCs but becomes activated in embryonic and adult tissues while TET1FL is ex
Temple University--Theses

Transfer RNAs (tRNAs) are essential molecules that allow the translation of the genetic code into amino acids. Extensive research during the last 50 years have revealed that, despite their apparently simple structure and function, tRNAs are more than simple adaptors in protein synthesis –they are of high importance in normal cell functions. Reinforcing this, tRNA levels are tightly regulated to match the codon usage patterns of a given cell type or cellular status to meet the cellular specific needs and adapt to stress. Moreover, tRNA nucleoside modifications are critical for their function at multiple levels, such as translation efficiency and fidelity, wobbling and fragmentation. The relevance of tRNA regulation in cell physiology is emphasized by the recent discovery that these molecules and their derived fragments are deregulated in cancer. Not only tRNA biology imbalance is associated to malignant transformation, but it also actively participates in it. These alterations occur at multiple levels of tRNA biology, such as expression, nucleoside modification and fragmentation, but many open questions remain unanswered. Cancer- specific tRNA deregulation is a very new and still unexplored discipline, and further studies are required to fully understand the molecular mechanisms that account for these alterations and their relevance in tumor biology. Because alterations in DNA methylation constitute a frequent mechanism by which transformed cells acquire their malignant characteristics, the cornerstone of this thesis is the description of epigenetic lesions that support the cancer-associated tRNA deregulation. To this end, we have designed and performed two independent studies to unveil the epigenetic regulation of tRNA biology in cancer. In the first study, we highlighted the tumor-specific epigenetic silencing of TYW2 as a mechanism to induce tRNAPhe hypomodification at position 37, a phenomenon that was observed for the first time more than forty years ago but whose cause and consequences have remained obscure. Our results established the connection between this epigenetic defect and a phenotype that enhances -1 ribosome frameshifting events to ultimately confer increased migratory capacities and mesenchymal features to the transformed colon cells. In the second study, we established a founded connection between cancer-associated DNA methylation defects with alterations in the expression of specific tRNAs. Our analyses also revealed that the oncogenic tRNA-Arg-TCT-4-1 overexpression in endometrial cancer was guided by DNA hypomethylation. Most importantly from the clinical perspective, the epigenetic alterations identified in both studies can anticipate the patients’ outcome, for which they may serve as biomarkers to allow the identification of high-risk patients that may benefit from a more comprehensive surveillance or complementary therapeutic strategies.
Els ARN de transferència (tRNAs) són d’una importància clau en la regulació de la síntesi proteica i l’expressió gènica. La seva rellevància en la fisiologia cel·lular es veu reforçada pel descobriment que aquestes molècules i els seus derivats estan alterats en patologies com el càncer, on contribueixen activament. Les alteracions dels tRNAs en càncer suposen una nova disciplina d’estudi on encara moltes preguntes romanen obertes per tal d’arribar a comprendre quines són les causes d’aquestes defectes i quin impacte tenen sobre la malaltia. Aquesta tesi té com objectiu identificar i caracteritzar alteracions en la metilació de l’ADN subjacents als desequilibris en la biologia dels tRNAs de les cèl·lules tumorals. En el primer estudi, hem descobert el silenciament epigenètic de l’enzim TYW2 en càncer colorectal com a causa de la hipomodificació del tRNAPhe, un fenomen que va ser descrit per primer cop fa més de quaranta anys però les causes i conseqüències del qual no van ser mai estudiades. Els nostres resultats estableixen una clara connexió entre aquest defecte epigenètic i un fenotip que és propens a potencial el frameshift dels ribosomes, cosa que augmenta la capacitat migratòria de les cèl·lules de càncer de colon. El segon estudi ha servit per caracteritzar la relació entre els canvis en la metilació de l’ADN i les alteracions en l’expressió dels tRNAs en càncer. Els nostres resultats han revelat que l’expressió de tRNA-Arg-TCT-4-1 augmenta en càncer d’endometri arrel de la hipometilació del seu gen. Més enllà d’aquests dos mecanismes epigenètics per modular la biologia dels tRNAs, els nostres estudis estableixen una connexió entre aquestes lesions epigenètiques i la prognosi dels pacients amb certs tipus de tumor, per la qual cosa podrien proposar-se com biomarcadors per identificar pacients de risc.

Les cellules placentaires portent un génome différent du génome maternel, puisque 50% de leurs gènes proviennent du génome paternel. Cependant, comme les cellules cancéreuses après la transformation néoplasique, elles réussissent à envahir les tissus de leur hôte, échapper à son système immunitaire et induire une angiogenèse afin d’établir la grossesse. Les cellules cancéreuses et placentaires arborent aussi une différence majeure : alors que de tels mécanismes typiques des cancers sont incontrôlés dans les cellules cancéreuses, ils sont spatialement et temporairement contrôlés dans les cellules placentaires saines. Ainsi, le recherche sur le « concept cancer/placenta » – l’utilisation du placenta pour mieux comprendre le cancer – peut aboutir à l’identification de biomarqueurs et d’approches thérapeutiques innovantes en oncologie, tout comme en gynécologie-obstétrique. Par exemple, les efforts de recherche portant sur l’expression des gènes CGB, codant pour la sous-unité ß de l’hormone chorionique gonadotrope humaine, dans les cellules cancéreuses et placentaires a mené au développement d’un biomarqueur largement utilisé pour la prise en charge de multiples cancers. Il est aussi intéressant de noter que ce même biomarqueur est aussi utilisé pour le dépistage d’aneuploïdies fœtales. De même, le clonage d’INSL4, codant pour le précurseur du peptide placentaire précoce ressemblant à l’insuline (pro-EPIL), dans des cellulaires placentaires précoces, a mené au développement d’un biomarqueur faisant actuellement l’objet d’études cliniques. Avec l’émergence de l’épigénétique, des études de la méthylation de l’ADN, la caractéristique épigénétique la mieux comprise, ont montré que les loci de gènes CGB et INSL4 sont hypométhylés dans les cellules cancéreuses et placentaires ; ce qui pourrait refléter l’hypométhylation globale caractéristique de ces deux types cellulaires. Par conséquent, le projet doctoral présenté dans cette thèse a exploré les modifications des paysages épigénétiques des cellules placentaires au cours de la grossesse et des cellules cancéreuses au cours de la transformation néoplasique. Ce projet a contribué initialement au développement d’un test d’immunoanalyse qui détecte l’hCGß de type II, spécialement codée par un sous-groupe de gènes CGB et détectée dans le sérum de patients atteints de cancers non-placentaires et de trisomie 21 fœtale. Ce test d’immunoanalyse, avec un test similaire développé pour la détection de pro-EPIL, a aussi été utilisé pour des études de preuve de concept précoces quant à l’effet de la méthylation de l’ADN sur l’expression de l’hCGß de type II et de pro-EPIL dans des surnageants de culture cellulaire. En fin de compte, ce projet a mené à la première comparaison directe et pan-génomique de la méthylation de l’ADN dans des cellules cancéreuses au cours de la transformation néoplasique et dans des cellulaires placentaires au cours de la grossesse. Cette étude a porté sur des données, disponibles publiquement, générées à partir de biopsies de 13 types de tumeurs, de villosités choriales (tissus placentaires) et d’autres tissus sains. Elle a également porté sur des données originales générées par nos soins à partir d’échantillons placentaires uniques : des cellules cytotrophoblastiques isolées de villosités choriales ex vivo. Toutes les données inclus dans cette étude ont été générées sur une plateforme de puces à ADN pour la mesure de la méthylation au niveau de 485 512 sites CpG pour chaque échantillon. En combinant, des logiciels innovants reposant sur la puissance d’algorithmes de lissage statistique et sur un solide rationnel biologique, cette étude a ainsi contribué à l’identification de motifs d’hypométhylation à l’échelle du mégabase distinguant les cellules placentaires du début de la grossesse de celles de la fin de la grossesse tout comme ils distinguent les cellules cancéreuses des cellules normales. (...)
Placental cells carry a genome different from the maternal genome, as 50% of it originate from the paternal genome. However, like cancer cells after neoplastic transformation, they successfully invade their host tissues, escape its immune system and induce angiogenesis in order to establish the pregnancy. Cancer and placental cells also display a major discrepancy: while such hallmarks of cancer mechanisms are uncontrolled in cancer cells, they are spatially and temporally controlled in healthy placental cells. Thus, research on the “cancer/placenta concept” – the use of the placenta to better understand cancer – can lead to innovative biomarkers and therapeutic approaches in oncology as well as in gynecology and obstetrics. For example, research efforts on the expression of the CGB genes, encoding for the human chorionic gonadotropin beta subunit (hCGß), in cancer and placental cells have led to the development of a biomarker widely used for the management of various cancers. Interestingly, this same biomarker is also used for the screening of fetal aneuploidies. Likewise, the cloning of INSL4, encoding for the precursor of the early placenta insulin-like peptide (pro-EPIL) in early pregnancy placental cells, has led to the development of a biomarker currently investigated in the clinical setting. Following the rise of epigenetic, studies on DNA methylation, the most well understood epigenetic mark, showed that the loci of CGB genes and INSL4 are hypomethylated in cancer and placental cells, which may reflect a global hypomethylation also characteristic of these cells. Therefore, the doctoral project presented in this dissertation had explored modifications in the epigenetic landscape of placental cells throughout pregnancy and cancer cells throughout neoplastic transformation. This project initially contributed to the development of an immunoassay detecting type II hCGß, specifically encoded by a subset of CGB genes and detected in the serum of patients with non-placental cancers and fetal Down Syndrome. This immunoassay, along with another one directed to pro-EPIL, was also used for an early proof of concept study regarding the effect of DNA methylation on the expression of type II hCGß and pro-EPIL in cell culture supernatants. Ultimately, this project led to the first direct genome-wide comparison of DNA methylation in cancer cells throughout neoplastic transformation and in placental cells throughout pregnancy. It included publically available data generated from biopsies of 13 types of tumors, chorionic villi (placental tissues) and other normal tissues. It also included original data generated from unique placental samples: villous cytotrophoblastic cells isolated ex vivo from chorionic villi. All datasets were generated on a microarray platform measuring DNA methylation at 485,512 CpG sites in each sample. Combining innovative software that leverages the power of statistical smoothing algorithms and a strong biological rationale, this study thus contributed to the identification of megabase-scale patterns of hypomethylation distinguishing early pregnancy from late pregnancy placenta cells as they distinguish normal from cancers cells. Strikingly, the affected genomic regions encompassed genes related to hallmarks of cancer mechanisms such as epithelial-mesenchymal transition (EMT), innate and acquired immune response, and hypoxia. Taken together, these results suggest the hypothesis that patterns of DNA methylation might contribute to “cancer/placenta epigenetic switches” allowing placental implantation and neoplastic transformation when turned “on”, while preventing the placenta to degenerate into an aggressive tumor when turned “off”

Au-delà de l'information génétique contenue dans la séquence de l'ADN des cellules, il existe une mémoire cellulaire, dite épigénétique comprenant l'ensemble des circuits génétiques avec rétroactions positives permettant d'amplifier ou de maintenir une réponse cellulaire dans le temps. Nous nous sommes intéressés, à travers deux exemples, aux boucles de rétrocontrôle positif comme élément de réponse à un signal, permettant de fixer, de manière à la fois dynamique et robuste, le comportement cellulaire. Dans un premier temps, nous avons identifié une boucle d'auto-amplification dans la production de vitellogénine chez la truite et permettant d'expliquer l' « effet mémoire de la vitellogénèse » (une seconde stimulation à l'œstradiol induit une plus forte production de vitellogénine et plus rapidement que lors de la première stimulation, alors même que le niveau de vitellogénine retombe à zéro entre les deux stimulations). Le modèle que nous proposons implique un récepteur tronqué à l'œstradiol possédant une activité basale même en l'absence de son ligand, permettant de maintenir la cellule dans un état d'aptitude à répondre sans pour autant produire de vitellogénine. Dans un deuxième temps, nous nous sommes intéressés à une des causes possibles provoquant la transition épithélio-mésenchymateuse (EMT), responsable des métastases dans les cancers. L'EMT témoigne d'un état plus agressif des cellules tumorales et s'accompagne notamment d'un changement du métabolisme des cellules cancéreuses, diminuant la part de la phosphorylation oxydative au profit de la glycolyse (effet Warburg). Cela entraîne une baisse d'efficacité de la production d'ATP, obligeant les cellules à prélever davantage de nutriments dans leur milieu. Cette observation a suscité le développement de thérapies basées sur la privation de glucose et qui, a priori, devraient nuire principalement aux cellules cancéreuses. Nous avons étudié les effets d'un faible contenu cellulaire en ATP sur la transformation cellulaire. Nous avons observé qu'un traitement par un analogue non métabolisable du glucose diminue drastiquement le contenu en ATP des cellules ayant passé l'EMT et induit des changements morphologiques et génétiques orientés vers le phénotype mésenchymateux. La protéine MKL1, cofacteur de transcription dont l'activité est régulée par la polymérisation de l'actine, pourrait être un relais génétique entre l'état métabolique cellulaire et le maintien de l'EMT. Ces résultats suggèrent de fortes connections entre l'EMT et le niveau énergétique des cellules, faisant d'une privation d'énergie une cause possible de l'aggravation du phénotype mésenchymateux et remettant en cause les bienfaits sur le long terme de thérapies visant à « affamer » les cellules tumorales
Beyond the genetic information contained in the DNA sequence of cells, there is a cellular memory called epigenetic, including genetic circuits with positive feedback loops amplifying or maintaining cellular states in time. We studied through two examples, the positive feedback loops as part of response to a signal, able to set cell behavior, in a dynamic and robust way. As a first step, we identified a self-amplification loop in the production of trout vitellogenin explaining the "vitellogenesis memory effect" (a second estradiol stimulation induces higher and faster vitellogenin production than during the first stimulation, even though the vitellogenin level falls to zero between the two stimuli). The model we propose involves a truncated estradiol receptor, with a basal activity even in the absence of its ligand, which is able to maintain the cell in an estrogen-responsive state without producing vitellogenin. In a second step, we studied one of the possible causes leading to the epithelial-mesenchymal transition (EMT), involved in cancer metastasis. The EMT reflects a more aggressive state of tumor cells and is associated with a particular change in the metabolism of cancer cells, reducing the part of oxidative phosphorylation in favor of glycolysis (Warburg effect). This leads to a reduction in the efficiency of ATP production, forcing the cells to take more nutrients from their environment. This observation led to the development of treatments based on glucose deprivation which should mainly affect cancer cells. We studied the effects of a low cellular ATP content on cell transformation. We observed that a treatment with a non-metabolizable glucose analogue drastically reduces the ATP content of cells that had undergone EMT and induces morphological and genetic changes enforcing the mesenchymal phenotype. We identified the transcriptional coactivator MKL1, whose activity is regulated by actin polymerization, as a possible genetic link between the cellular metabolic state and maintenance of EMT. These results suggest strong connections between the EMT and the energy level of the cells, and raise serious questions about the benefits of the long-term therapy "starving" tumor cells, considering that energy deprivation could aggravate the mesenchymal cell phenotype

Cytotoxic chemotherapy is extremely important in adjuvant treatment of breast cancer. Yet, tumours frequently acquire chemoresistance that correlates with increased aggressiveness and poor prognosis. Three theories exist describing how the resistance develops: genetic, epigenetic and karyotypic theory. The epigenetic theory is the least explored. Here we analyzed the role of the epigenetic phenomena in the acquisition of drug resistance. To do so, we employed genome wide screens of microRNA and gene expression, DNA methylation and complete genome hybridization. We identified three novel microRNA interactions involved in the chemoresistant phenotype. These three microRNAs displayed depressed expression in the resistant cell lines and we were able to re-establish some level of drug sensitivity through ectopic expression of these under expressed microRNAs. In addition, we described the role of DNA methylation in impacting expression of a wide range of genes, thus, contributing to the phenotype of chemoresistance. Furthermore, we revealed a distorted global DNA methylation pattern that coincides with massive instability of the resistant genome. Finally, our results present a striking similarity between gene expression, epigenetic profiles and chromosomal aberrations in two different drug resistant cell lines. Taken together, this project suggests that the acquisition of chemoresistant phenotype is epigenetic in nature and may arise with a predictable pattern. Elucidating the specifics of this pattern may in the future prove useful in developing treatment and prognostic chemoresistance biomarkers.
xiii, 116 leaves : ill. (some col.) ; 29 cm

Breast cancer is the most frequently diagnosed form of cancer in women and the second leading cause of cancer-related deaths. Breast cancer is a heterogeneous disease consisting of many types of tissue neoplasia, and there appears to be no model of how a particular lesion develops into an aggressive, malignant, invasive carcinoma. Genetic mutation and aberrant epigenetic regulation are among the most common events that lead to neoplasia. In breast cancer, p53 mutation is the most common genetic defect related to a single gene. Therefore, this dissertation focuses on the mechanisms and consequences of p53 mutation during breast tumorigenesis. Genome-wide analysis of gene expression and epigenetic modifications in a panel of breast cancer cell lines suggested that p53 mutation and aberrant epigenetic silencing were cooperating mechanisms in the silencing of wild-type p53 target genes during cancer progression. Therefore, models of p53 inactivation were created in non-malignant human mammary epithelial cells to determine the role of p53 mutation on the epigenetic status of its target genes and the acquisition of malignant phenotypes. Comparisons of each model demonstrated that differing modes of p53 inactivation produced different functional consequences. Loss of wild-type p53 function alone ablated the normal cellular response to external stress stimuli, but had no affect on the expression of genes or epigenetic status in untreated cells. Introduction of missense mutant p53 protein caused very few changes when the protein was expressed at low levels. However, accumulation of mutant p53 caused a variety of gene expression changes and interfered with endogenous wild-type p53. The accumulation of mutant p53 also caused an increase in migration and invasion of the cells that expressed it. Interestingly, epigenetic aberrations were not detected in response to any of the p53 manipulations. These data suggest that accumulation of missense mutation is particularly dangerous to normal cells. They also suggest that p53 mutation and epigenetic aberration are two distinct mechanisms, which overlap and cooperate during tumorigenesis. These data suggest that treatment strategies for human breast cancer should include modalities to target both defects for increased efficacy.

Improved survival for women receiving chemotherapy for breast cancer (BC) has been accompanied by the development/persistence of psychoneurological symptoms (PNS) that compromise their quality of life. The biological basis for these PNS is unknown, but could reflect the acquisition of soma-wide chromosomal/epigenetic alterations. An important first step in testing this hypothesis is to determine if somatic genetic/epigenetic changes arise and persist following treatment. To answer this question we longitudinally studied 71 women (ages 23-71) with early-stage BC and collected measures before chemotherapy (baseline), and 4 weeks (mid-chemo); six months (during radiation therapy for a subset of women); and one year following the initiation of chemotherapy. Acquired lymphocyte chromosomal instability (scored by micronuclei frequencies [MNF]) showed a significant increase in post-treatment compared to baseline time-points (p<0.0001), with these increases persisting for at least one year following chemotherapy. Significant predictive associations were observed between MNF and tumor characteristics [luminal B (lower MNF; p=0.0182); triple negative (higher MNF; p=0.0446)], radiotherapy (higher MNF; p=0.0004), the type of chemotherapy received (p=0.0463), race (Caucasians > African Americans; p=0.0037), perceived stress levels (positive-association; p=0.0123), and cognitive flexibility domain measures (positive-association; p=0.0238). Genome-wide acquired methylation changes were also measured in peripheral blood cells, with 1265 sites showing significant differential methylation following chemotherapy. These sites were localized to open sea, shores, shelves, and CpG island sequences and included sites within genes involved in cell cycle, DNA repair, transcription regulation, signal transduction pathways, neuronal regeneration, and immunity. To determine if the genetic/epigenetic alterations acquired in peripheral blood cells correlated with those in tumor cells, BC tumors from 10 participants were analyzed using a genome-wide copy number/targeted mutations (CN/M) microarray. While no clear blood-tumor cell correlations were detected, genome-wide CN/M evaluations showed promise for stratifying tumors. Lastly, in an unrelated project studying a rare case of fetuses in fetu, methylation changes acquired in embryogenesis were shown to be influenced by both environmental and genetic cues. In summary, acquired chromosomal/epigenetic alterations do arise following chemotherapy (and in embryogenesis). Further delineation of these acquired changes could increase our understanding of the biological basis for cancer-related side-effects and help to identify “at risk” individuals.

L’organisation nucléaire de la chromatine n’est pas aléatoire. Sa structure est parfaitement contrôlée, suivant un modèle hiérarchique avec différents niveaux d’organisation et de compaction. A large échelle, chaque chromosome occupe son propre espace au sein du noyau. A plus fine résolution, un chromosome est subdivisé en compartiments actifs ou répressifs, caractérisés par un état de la chromatine plus ou moins compact. A l’échelle du méga-base, cette organisation hiérarchique peut encore être divisée en domaines topologiques (ou TADs), jusqu’à la caractérisation de boucle d’ADN facilitant les interactions entre promoteurs et régions régulatrices. Très brièvement, et bien que les méchanismes exactes restent à déterminer, il a récemment été démontré que l’organisation spatiale de la chromatine dans une cellule normale joue un rôle primordial dans la régulation et l’expression des gènes. L’organisation en domaines topologiques implique la présence de complexes protéiques insulateurs tel que CTCF/cohésine. Ces facteurs jouent un rôle de barrière en restreignant et favorisant les interactions entre éléments régulateurs et gènes à l’intérieur d’un domaine, tout en limitant les interactions entre domaines. De cette façon, deux régions appartenant au même domaine topologique pourront fréquemment interagir, alors que deux régions appartenant à des domaines distincts auront une très faible probabilité d’interaction. Dans la cellule cancéreuse, l’implication de l’épigénome et de l’organisation spatiale de la chromatine dans la progression tumorale reste à ce jour largement inexplorée. Certaines études récentes ont toutefois démontré qu’une altération de la conformation de l’ADN pouvait être associée à l’activation de certains oncogènes. Même si les mécanismes exacts ne sont pas encore connus, cela démontre que l’organisation de la chromatine est un facteur important de la tumorigenèse, permettant, dans certains cas, d’expliquer les méchanismes moléculaires à l’origine de la dérégulation de certains gènes. Parmi les cas rapportés, une alération des régions insulatrices (ou frontières) entre domaines topologiques permettrait à des régions normalement éloignées spatialement de se retrouver en contact, favorisant ainsi l’activation de certains gènes. Une caractérisation systématique de la conformation spatiale des génomes cancéreux pourrait donc permettre d’améliorer nos connaissances de la biologie des cancers. Les techniques haut-débit d’analyse de la conformation de la chromatine sont actuellement largement utilisées pour caractériser les interactions physiques entre régions du génome. Brièvement, ces techniques consistent à fixer, digérer, puis liguer ensemble deux régions du génome spatialement proches. Les fragments d’ADN chimériques ainsi générés peuvent alors être séquencés par leurs extrémités, afin de quantifier le nombre de fois où ces régions ont été trouvées en contact. Parmi les différentes variantes de ces techniques, le Hi-C associé à un séquençage profond permet l’exploration systématique de ces interactions à l’échelle du génome, offrant ainsi une vue détaillée de l’organisation tri-dimensionnelle de la chromatine d’une population cellulaire
The chromatin is not randomly arranged into the nucleus. Instead, the nuclear organization is tightly controlled following different organization levels. Recent studies have explored how the genome is organized to ensure proper gene regulation within a constrained nuclear space. However, the impact of the epigenome, and in particular the three-dimensional topology of chromatin and its implication in cancer progression remain largely unexplored. As an example, recent studies have started to demonstrate that defects in the folding of the genome can be associated with oncogenes activation. Although the exact mechanisms are not yet fully understood, it demonstrates that the chromatin organization is an important factor of tumorigenesis, and that a systematic exploration of the three-dimensional cancer genomes could improve our knowledge of cancer biology in a near future. High-throughput chromosome conformation capture methods are now widely used to map chromatin interaction within regions of interest or across the genome. The Hi-C technique empowered by next generation sequencing was designed to explore intra and inter-chromosomal contacts at the whole genome scale and therefore offers detailed insights into the spatial arrangement of complete genomes. The aim of this project was to develop computational methods and tools, that can extract relevant information from Hi-C data, and in particular, in a cancer specific context. The presented work is divided in three parts. First, as many sequencing applications, the Hi-C technique generates a huge amount of data. Managing these data requires optimized bioinformatics workflows able to process them in reasonable time and space. To answer this need, we developped HiC-Pro, an optimized and flexible pipeline to process Hi-C data from raw sequencing reads to normalized contact maps. HiC-Pro maps reads, detects valid ligation products, generates and normalizes intra- and inter-chromosomal contact maps. In addition, HiC-Pro is compatible with all current Hi-C-based protocols

L’organisation nucléaire de la chromatine n’est pas aléatoire. Sa structure est parfaitement contrôlée, suivant un modèle hiérarchique avec différents niveaux d’organisation et de compaction. A large échelle, chaque chromosome occupe son propre espace au sein du noyau. A plus fine résolution, un chromosome est subdivisé en compartiments actifs ou répressifs, caractérisés par un état de la chromatine plus ou moins compact. A l’échelle du méga-base, cette organisation hiérarchique peut encore être divisée en domaines topologiques (ou TADs), jusqu’à la caractérisation de boucle d’ADN facilitant les interactions entre promoteurs et régions régulatrices. Très brièvement, et bien que les méchanismes exactes restent à déterminer, il a récemment été démontré que l’organisation spatiale de la chromatine dans une cellule normale joue un rôle primordial dans la régulation et l’expression des gènes. L’organisation en domaines topologiques implique la présence de complexes protéiques insulateurs tel que CTCF/cohésine. Ces facteurs jouent un rôle de barrière en restreignant et favorisant les interactions entre éléments régulateurs et gènes à l’intérieur d’un domaine, tout en limitant les interactions entre domaines. De cette façon, deux régions appartenant au même domaine topologique pourront fréquemment interagir, alors que deux régions appartenant à des domaines distincts auront une très faible probabilité d’interaction. Dans la cellule cancéreuse, l’implication de l’épigénome et de l’organisation spatiale de la chromatine dans la progression tumorale reste à ce jour largement inexplorée. Certaines études récentes ont toutefois démontré qu’une altération de la conformation de l’ADN pouvait être associée à l’activation de certains oncogènes. Même si les mécanismes exacts ne sont pas encore connus, cela démontre que l’organisation de la chromatine est un facteur important de la tumorigenèse, permettant, dans certains cas, d’expliquer les méchanismes moléculaires à l’origine de la dérégulation de certains gènes. Parmi les cas rapportés, une alération des régions insulatrices (ou frontières) entre domaines topologiques permettrait à des régions normalement éloignées spatialement de se retrouver en contact, favorisant ainsi l’activation de certains gènes. Une caractérisation systématique de la conformation spatiale des génomes cancéreux pourrait donc permettre d’améliorer nos connaissances de la biologie des cancers. Les techniques haut-débit d’analyse de la conformation de la chromatine sont actuellement largement utilisées pour caractériser les interactions physiques entre régions du génome. Brièvement, ces techniques consistent à fixer, digérer, puis liguer ensemble deux régions du génome spatialement proches. Les fragments d’ADN chimériques ainsi générés peuvent alors être séquencés par leurs extrémités, afin de quantifier le nombre de fois où ces régions ont été trouvées en contact. Parmi les différentes variantes de ces techniques, le Hi-C associé à un séquençage profond permet l’exploration systématique de ces interactions à l’échelle du génome, offrant ainsi une vue détaillée de l’organisation tri-dimensionnelle de la chromatine d’une population cellulaire
The chromatin is not randomly arranged into the nucleus. Instead, the nuclear organization is tightly controlled following different organization levels. Recent studies have explored how the genome is organized to ensure proper gene regulation within a constrained nuclear space. However, the impact of the epigenome, and in particular the three-dimensional topology of chromatin and its implication in cancer progression remain largely unexplored. As an example, recent studies have started to demonstrate that defects in the folding of the genome can be associated with oncogenes activation. Although the exact mechanisms are not yet fully understood, it demonstrates that the chromatin organization is an important factor of tumorigenesis, and that a systematic exploration of the three-dimensional cancer genomes could improve our knowledge of cancer biology in a near future. High-throughput chromosome conformation capture methods are now widely used to map chromatin interaction within regions of interest or across the genome. The Hi-C technique empowered by next generation sequencing was designed to explore intra and inter-chromosomal contacts at the whole genome scale and therefore offers detailed insights into the spatial arrangement of complete genomes. The aim of this project was to develop computational methods and tools, that can extract relevant information from Hi-C data, and in particular, in a cancer specific context. The presented work is divided in three parts. First, as many sequencing applications, the Hi-C technique generates a huge amount of data. Managing these data requires optimized bioinformatics workflows able to process them in reasonable time and space. To answer this need, we developped HiC-Pro, an optimized and flexible pipeline to process Hi-C data from raw sequencing reads to normalized contact maps. HiC-Pro maps reads, detects valid ligation products, generates and normalizes intra- and inter-chromosomal contact maps. In addition, HiC-Pro is compatible with all current Hi-C-based protocols

Il mesotelioma pleurico maligno (MPM) è un tumore molto aggressivo e rapidamente progressivo che si sviluppa a livello del mesotelio che compone la pleura; questa neoplasia può assumere diversi sottotipi istologici (epitelioide, bifasico e sarcomatoide), i quali sono strettamente correlati alla prognosi. Le modificazioni epigenetiche che avvengono nelle fasi di iniziazione e progressione del MPM possono svolgere un ruolo fondamentale nel regolare negativamente il crosstalk tra tumore e sistema immunitario, contribuendo a mantenere un microambiente tumorale immunosoppressivo. Conoscere più dettagliatamente il panorama epigenetico del MPM può contribuire a definire il razionale per nuove terapie antitumorali e porre le basi per studi di combinazione che prevedano l’utilizzo di farmaci epigenetici con farmaci immunoterapeutici. Con il presente studio abbiamo voluto valutare, in un primo momento, le modificazioni nel profilo di espressione genica di 10 linee di MPM, di diverso istotipo, trattate con la guadecitabina, un agente demetilante il DNA di seconda generazione, tramite la piattaforma nCounter di Nanostring. I risultati ottenuti tramite Ingenuity Pathway Analysis (IPA) hanno mostrato che la guadecitabina era in grado di indurre l’attivazione dei geni coinvolti nel crosstalk tra cellule dendritiche e natural killer nel 50% delle linee cellulari di MPM indagate, accompagnata dall’attivazione di altre componenti coinvolte nella risposta immunitaria a infezioni e infiammazioni. I fattori trascrizionali “upstream” più frequentemente attivati appartenevano al pathway di segnalazione dell’interferon (IFN)-γ. Inoltre, è stata riscontrata l’up-regolazione (fold change medio (mFC) ≥ 1.5) di molecole immuno-relate, come NY-ESO-1 (mFC=13.16), MAGE-B2 (mFC=13.09), CD70 (mFC=5.27) e CTLA-4 (mFC=4.81). Abbiamo inoltre effettuato analisi istotipo-specifiche per esplorare le modificazioni molecolari indotte dalla guadecitabina nei 3 sottotipi di MPM. La guadecitabina ha indotto l’up-regolazione dell’espressione di marcatori del fenotipo epiteliale (es. CDH1, EPCAM e PECAM1), osservata ad alti livelli nelle linee cellulari sarcomatoidi; ciò è stato associato alla down-regolazione di molecole di origine mesenchimale (es. CDH2 e NCAM) e induttori della cascata metastatica (es. CDH11). Successivamente abbiamo comparato gli effetti immunomodulatori della guadecitabina con quelli di altri farmaci epigenetici (gli inibitori delle iston acetiltransferasi (HDAC) VPA e SAHA o l’inibitore di EZH2 EPZ-6438) da soli o in combinazione con la guadecitabina in 5 linee cellulari di MPM (2 sarcomatoidi, 1 bifasica e 2 epitelioidi). Analisi citofluorimetriche e molecolari hanno rivelato che la guadecitabina up-regolava l’espressione delle molecole immuno-relate, quali HLA di classe I (mFC=1.59), ICAM-1 (mFC=3.27), PD-L1 (mFC=2.13), e NKG2DL (MICA mFC=1.88, MICB mFC=2.42, ULBP2 mFC=3.16), inducendo/up-regolando l’espressione dei Cancer Testit Antigens (CTA) NY-ESO-1, MAGE-A1 e MAGE-A3; il VPA up-regolava l’espressione degli antigeni di HLA di classe I (mFC=1.50), PD-L1 (mFC=2.76), NKG2DL (MICA mFC=1.69, MICB mFC=2.67, ULBP2 mFC=3.26) e quella dei CTA MAGE-A1 e MAGE-A3, rispettivamente in 2/5 e 3/5 linee cellulari di MPM; il SAHA up-regolava l’espressione di MICA (mFC=1.57), MICB (mFC=4.05) e MAGE-A1 e MAGE-A3, rispettivamente in 2/5 e 4/5 linee cellulari; per contro, l’EPZ-6438 ha mostrato minime capacità immunomodulanti, inducendo solamente NY-ESO-1 e up-regolando l’espressione di PD-L1, MICB e ULBP2 in 1 linea cellulare ciascuno. Contrariamente ai risultati eterogenei ottenuti dai singoli farmaci, l’associazione di VPA, SAHA o EPZ-6438 alla guadecitabine ha rafforzato le capacità immunomodulanti di quest’ultima, influenzando l’espressione di tutte le molecole indagate. Specificatamente, le combinazioni di guadecitabine con VPA, SAHA o EPZ-6438 up-regolavano l’espressione degli antigeni HLA di classe I (mFC=2.21, 2.03, o 2.29 rispettivamente), di ICAM-1 (mFC=4.09, 4.63, o 5.33), di PD-L1 (mFC=6.95, 2.42, o 2.50), di MIC-A (mFC=3.48, 2.00, o 2.23), di MIC-B (mFC=6.80, 2.48, o 2.81) e di ULBP2 (mFC=13.45, 3.40, o 4.11). Infine, livelli di up- regolazione/induzione maggiori sono stati osservati per i CTA a seguito di tutti e 3 i trattamenti combinati rispetto alla guadecitabina in singolo. La modulazione delle caderine è stata influenzata dal sottotipo istologico di MPM: l’espressione di CDH1 è stata indotta dalla guadecitabina in singolo e dalla sua combinazione con VPA, SAHA e EPZ-6438 nelle 2 linee cellulari sarcomatoidi, costitutivamente negative per l’espressione del gene; l’espressione di CDH2 è stata up-regolata dal VPA e dal SAHA singoli in 1/5 linee cellulari e dalle combinazioni di guadecitabina con VPA o SAHA, rispettivamente in 3/5 o 1/5 linee cellulari di MPM; ciononostante, non è stata osservata alcuna up-regolazione del gene nelle 2 linee cellulari epiteliodi, costitutivamente negative per l’espressione di CDH2. In conclusione, dalle analisi approfondite del pannello di espressione genica abbiamo confermato che la guadecitabina è in grado di up-regolare/indurre l’espressione di molecole immunitarie e immuno- relate cruciali per il crosstalk tra il tumore e il sistema immunitario; inoltre, abbiamo dimostrato che essa induce l’attivazione di geni correlati all’IFN, soprattutto nel fenotipo sarcomatoide, supportando l’ipotesi che i demetilanti possano aumentare la risposta immunitaria contro il MPM, potenzialmente anche del tipo istologico più aggressivo; la modulazione delle molecole di adesione tendente verso il fenotipo epitelioide suggerisce la possibilità di revertire la transizione epitelio-mesenchima, cruciale nel processo di metastatizzazione. Infine, combinando la guadecitabina con farmaci inibitori delle HDAC/EZH2 ha rafforzato la sua attività immunomodulante, fornendo il razionale per studi di associazione di farmaci epigenetici e agenti immunoterapici in modo da aumentare l’efficacia di questi ultimi nel trattamento del mesotelioma.
Malignant pleural mesothelioma (MPM) is a highly aggressive and rapidly progressive tumor that affect the mesothelium componing the pleura; it can acquire different histological subtypes (mainly epithelioid, biphasic, and sarcomatoid MPM), which are of prognostic significance. Epigenetic modifications occurring during MPM initiation and progression may play a relevant role in negatively regulating the crosstalk between the tumor and the immune system, as well as contributing to the highly immunosuppressive microenvironment. A better understanding of MPM epigenetics will contribute to refine antitumor strategies, laying the ground for epigenetic-based immunotherapy. The present study evaluated, in the first instance, changes in the gene expression fingerprint of 10 MPM cell lines of different phenotype treated with the second-generation DNA hypomethylating agent (DHA) guadecitabine, through the Nanostring Oncology panel with nCounter readout. Ingenuity pathway analysis results revealed that guadecitabine induced the activation of natural killer and dendritic cells signaling pathways in 50% of MPM cell lines, followed by the activation of other components involved in the immune system response to infections and inflammation. Besides, the most frequently activated upstream regulators belonging to the interferon (IFN)-γ signaling pathway. Also, the up- regulation (mean fold change (mFC) ≥ 1.5) of key immune-related molecules, such as the NY-ESO-1 (mFC=13.16), MAGE-B2 (mFC=13.09), CD70 (mFC=5.27), and CTLA-4 (mFC=4.81) was reported. We also performed histological type-specific investigations to explore molecular changes induced by guadecitabine among the 3 histotypes. Guadecitabine induced the up-regulation of the expression of epithelial markers (e.g., CDH1, EPCAM, PECAM1), observed at higher levels in sarcomatoid cell lines; this was accompanied by the down-regulation of mesenchymal origin molecules (e.g., CDH2, NCAM), and inductor of metastatic signals (e.g., CDH11). Secondly, the immunomodulatory effects of guadecitabine were compared to those of different epigenetic drugs (the histone deacetylase (HDAC) inhibitors VPA and SAHA, or the EZH2 EPZ- 6438), alone or in combination with guadecitabine, in 5 MPM cell lines (two sarcomatoid, one biphasic, and two epithelioid). We performed cytofluorimetric and molecular qRT-PCR analyses and, in this regard, results showed that guadecitabine up-regulated the expression of immune-related molecules, such as HLA class I antigens (mFC=1.59), ICAM-1 (mFC=3.27), PD-L1 (mFC=2.13), and NKG2DLs (MIC-A mFC=1.88, MIC-B mFC=2.42, and ULBP2 mFC=3.16), and up-regulated/induced Cancer Testis Antigens (CTA: NY-ESO-1, MAGE-A1, and MAGE-A3) expression; VPA up-regulated the expression of HLA class I antigens (mFC=1.50), PD-L1 (mFC=2.76), NKG2DLs (MIC-A mFC=1.69, MIC-B mFC=2.67, and ULBP2 mFC=3.26), and the expression of CTA MAGE-A1 and MAGE-A3 in 2/5 and 3/5 MPM cell lines, respectively; SAHA up- regulated the expression of MICA (mFC=1.57), MICB (mFC=4.05), MAGE-A1 and MAGE-A3 in 2/5and 4/5 MPM cell lines, respectively; conversely, EPZ-6438 induced minimal immunomodulatory effects, inducing only NY-ESO-1 and up-regulating PD-L1, MIC-B, and ULBP2 expression in 1 MPM cell line each. Despite the heterogeneous activities of single epigenetic drugs, the addition of both VPA, SAHA, and EPZ-6438 to guadecitabine strengthened the immunomodulatory effects of the latter, by affecting the expression of all investigated molecules. Specifically, guadecitabine plus VPA, SAHA, or EPZ-6438 upregulated the expression of HLA class I antigens mFC=2.21, 2.03, or 2.29; ICAM-1 mFC=4.09, 4.63, or 5.33; PD-L1 mFC=6.95, 2.42, or 2.50; MIC-A mFC=3.48, 2.00, or 2.23; MIC-B mFC=6.80, 2.48, or 2.81; ULBP2 mFC=13.45, 3.40, or 4.11, respectively. Lastly, higher levels of upregulated/induced CTA expression were observed after all 3 combination treatments versus guadecitabine alone. Cadherins modulation was MPM histotype-related: CDH1 expression was induced in the 2 constitutive-negative sarcomatoid MPM cell lines by guadecitabine alone or combined with VPA, SAHA, or EPZ-6438; CDH2 expression was upregulated by VPA or SAHA in 1/5 cell lines, and by guadecitabine plus VPA or SAHA in 3/5 or in 1/5 MPM cell lines, respectively; however, no induction of CDH2 have been reported in the constitutive negative epithelioid cell lines. Overall, from comprehensive gene expression panel analyses, we confirmed that guadecitabine induced/up-regulated the expression of immune and immune-related molecules, pivotal in the tumor- immune system crosstalk; also, we highlighted that guadecitabine-induced activation of IFN-related genes, especially in the sarcomatoid phenotype, supporting the hypothesis that DHA could increase the immune response against MPM, potentially also with sarcomatoid features; moreover, the modulation of adhesion molecules towards the epithelial type suggests the possibility to revert the epithelial-to- mesenchymal transition (EMT) event, crucial in the invasion-metastasis cascade. Also, combining guadecitabine with HDACi/EZH2i strengthened its immunomodulatory capabilities, laying the rationale for epigenetic drugs-based immunotherapies, to enhance efficacy of these strategy in the MPM clinic.

L'UHRF1 est une cible de médicaments pour traiter le cancer. Il est très présent dans de nombreux cancers, ce qui peut causer des problèmes de méthylation des gènes. Notre travail vise à étudier le potentiel anticancéreux des inhibiteurs de l'UHRF1-SRA (AMSA2, MPB7 et UM63), et à comprendre comment ils agissent et comment ils sont sélectifs envers les cellules cancéreuses. On a utilisé des techniques de biologie cellulaire et moléculaire. Cela a montré que ces composés ont un effet anticancéreux. Ils empêchent que le gène UHRF1 et le gène DNMT1 se retrouvent au même endroit. Ils contrôlent aussi leur niveau de protéines. Cela fait baisser la méthylation de l'ADN. On a aussi vu que le développement et le cycle des cellules cancéreuses ralentissaient, et que les protéines qui induisent l'apoptose augmentaient. L'analyse du méthylome a montré que ces inhibiteurs diminuaient l'hyperméthylation des TSG, ce qui réactivait leur rôle de protection contre le cancer. Ces composés n'ont eu qu'un faible effet sur les cellules non cancéreuses, ce qui a été confirmé dans des conditions de culture cellulaire en 2D et en 3D
UHRF1 has been identified as a druggable epigenetic target for cancer therapy as it is overexpressed in many cancers promoting hypermethylation/silencing of tumor suppressor genes (TSGs), which lead to uncontrolled cell proliferation. This thesis aimed to investigate the anticancer potential of UHRF1-SRA inhibitors (AMSA2, MPB7 and UM63), and to explore their mechanism of action as well as their selectivity towards cancer cells. Using multiple cell and molecular biology techniques, we revealed that these compounds exert anticancer activity. They prevent co-localization of UHRF1/DNMT1 tandem and also downregulate their protein levels which lead to a decrease in global DNA methylation. Furthermore, a significant arrest in cancer cell proliferation and cell cycle was observed, followed by an upregulation of pro-apoptotic proteins resulting in apoptosis. Methylome analysis revealed that these inhibitors decreased the hypermethylation at TSGs, reactivating their onco-protective role. Interestingly, these compounds exerted minimal impact on non-cancerous cells, validated in both 2D and 3D cell culture conditions

Colorectal tumors are hierarchically organized and governed by populations of self-renewing cancer stem cells, representing one of the deadliest types of cancers worldwide. Emergence of a cancer stem-like phenotype depends on epigenetic reprogramming, associated with profound transcriptional changes. As described for pluripotent reprogramming, epigenetic modifiers play a key role in developing and maintaining cancer stem cells by establishing embryonic stem-like transcriptional programs, thus altering the balance between self-renewal and differentiation. Through my work, I have identified overexpression of histone methyltransferase G9a as a risk factor for colorectal cancer, associated with shorter relapse-free survival. Moreover, using human transformed pluripotent cells as a surrogate model for cancer stem cells, I demonstrate that G9a activity is essential for the maintenance of an embryonic stem-like transcriptional signature that is required to promote self-renewal, tumorigenicity and an undifferentiated state. Such a role was also applicable to colorectal cancer, where inhibitors of G9a histone methyltransferase function induced intestinal differentiation while restricting tumor-initiating activity in patient-derived colorectal tumor samples. By integrating transcriptome profiling with G9a/H3K9me2 loci co-occupancy, the canonical Wnt pathway, epithelial-to-mesenchyme transition and extracellular matrix organization were identified as potential targets of such a chromatin regulation mechanism in colorectal cancer stem cells. Considering such novel insights on the role of G9a as a driver of the cancer stem cell phenotype, as well as a promoter of self-renewal, tumorigenicity and an undifferentiated state, I established and executed a multi-step drug screening pipeline to identify new repurposed drugs that selectively alter G9a functions in human CSCs. This pipeline revealed 3 new drug candidates that inhibit H3K9me2 deposition and impair human CSCs in culture. Future in-depth characterization of those candidates will represent an important step toward the development of novel CSC-targeting therapeutics.

The development of several organ systems through modeling and shaping of the tissue structure occurs from signaling through axon guidance molecules. The Slit family of ligands has been shown to regulate branching morphogenesis in mammary gland duct development and loss of Slit gene expression during this time leads to the formation of hyperplastic, disorganized lesions suggesting a potential role for Slits in cancer formation. Characterization of human pancreatic ductal adenocarcinoma cell lines showed a loss of SLIT2 expression in cells that contain activated Kras. Loss of SLIT2 expression was associated with DNA methylation of CpG sites within the SLIT2 core promoter and chromatin enrichment of repressive histone modifications at the SLIT2 transcriptional start site. Additionally, treatment of pancreatic ductal adenocarcinoma cell lines with demethylating agent 5-aza-2'-deoxycytidine led to SLIT2 re-expression while treatment with histone deacetylase inhibitor Trichostatin A did not. Mir-218-1 is an intronic microRNA encoded within intron 15 of the SLIT2 gene. Expression of mir-218-1 does not correlate with SLIT2 mRNA expression suggesting that it is transcribed from a promoter independent of the SLIT2 gene promoter. Pancreatic ductal adenocarcinoma cell lines showed a peak of H3K4me3 chromatin enrichment localized to a 1kb region within intron 4 of the SLIT2 gene denoting a candidate alternative promoter for mir-218-1. A concordant peak of H4ac chromatin enrichment overlapped the peak of H3K4me3 enrichment and transcriptional activity was measured from the 1kb region in all pancreatic ductal adenocarcinoma cell lines. A NF-κB binding site was also predicted to exist within the 1kb region. Transfection with two independent siRNAs to NF-κB led to an increase in both pre-mir-218-1 and mature mir-218-1 while treatment with an inhibitor to IκB kinase led to an increase in pre-mir-218-1 expression. Additionally, the p65 subunit of NF-κB was found to bind to the candidate mir-218-1 alternative promoter in pancreatic ductal adenocarcinoma cell lines that do not contain DNA CpG methylation at the predicted NF-κB binding site. It was discovered that miR-218 is a modulator of ARF6 expression suggesting a role in the inhibition of pancreatic ductal adenocarcinoma cell invasion through modulation of the actin cytoskeleton. Overexpression with a miR-218 precursor showed that miR-218 is an inhibitor of pancreatic ductal adenocarcinoma cell invasion in two dimensions. Additionally, it was found that while miR-218 does not have an affect on the ability of pancreatic ductal adenocarcinoma cells to form functional invadopodia, miR-218 is an inhibitor of the extracellular matrix degradation properties of mature invadopodia. Interestingly, the effect of miR-218 on pancreatic ductal adenocarcinoma cell invasion or extracellular matrix degradation is not reliant on the cell's dependency on Kras signaling for growth and survival. Collectively, these observations indicate that understanding the transcriptional regulation of SLIT2 and mir-218-1 expression as well as their signaling properties may provide a step toward the development of diagnostic tests and therapeutic treatments for patients with invasive or metastatic pancreatic ductal adenocarcinoma.

In this thesis I present an investigation of a novel steroid hormone receptor cochaperone, FKBPL, and its role in male infertility. I present evidence of proposed functional mutations in the FKBPL coding sequence within a subset of the infet1ile male population, namely, those presenting with non-obstructive azoospermia, that were not present within fertile control populations. FUl1her, I show that FKBPL exhibits a cell specific pattern of expression in human testis consistent with a role in androgen receptor signalling. I then demonstrate that increased expression of FKBPL can potentiate androgen responsive gene expression in vitro. The androgen receptor and its co-factors are important also in the development and progression of prostate cancer. I present an examination of the histone modifying enzyme and androgen receptor co-activator, LSD 1, and report that it is of limited clinical utility as a prostate cancer biomarker and is not a critical factor in the maintenance of aberrant DNA methylation at the GSTP 1 locus in prostate cancer cell lines. I present further examination of the epigenetic landscape of prostate cancer using a panel of prostate cell lines representing progression from normal to late stage androgen-insensitive prostate cancer. I have demonstrated that a loss of epigenetic control at a specific set of gene loci correlates with malignant transformation while epigenetic control at other gene classes is maintained. Transcriptional analysis of candidate genes, coupled with analysis of associated promoter CpG density, reveals some genes that are repot1ed to be hypermethylated and silenced in prostate cancer to possess low density CpG promoters whose methylation is likely to be inconsequential to transcriptional regulation. This suggests that a reevaluation of purported candidate biomarkers in light of recent developments in our understanding of CpG promoter regulation might be necessary. FUl1her, I find that coordinate changes in the expression of in silica identified targets of methylation that are also androgen responsive is not characteristic of disease progression in prostate cancer, though some genes identified in this screen were able to discriminate normal from disease cells based on their transcriptional profile. I also show that epigenetic aberrations may be pharmacologically reversed, but that normal control of a subset of genes is lost as a consequence. This work collectively highlights the potential diagnostic and therapeutic translational applications of our growing understanding of the human epigenome in prostate cancer, and highlights some important considerations in the interpretation of epigenetic data in the prostate cancer field.

En France comme dans la majorité des pays développés, le cancer de la prostate est le plus fréquent chez l’homme. Il est clairement établi que les altérations génétiques et épigénétiques sont des événements communs dans les cancers de la prostate, se traduisant par l’expression aberrante de gènes critiques. La méthylation des histones participe à la régulation de l’expression des gènes dans la cellule. La marque épigénétique H3K27me3 est associée à la répression génique et se trouve dérégulée dans les cancers de la prostate. Ses niveaux sont déterminés par l’équilibre entre les activités de la méthyltransférase d’histone EZH2 et de la déméthylase d’histone JMJD3. Afin de comprendre le mécanisme de dépôt de H3K27me3 dans la tumorigenèse prostatique, le travail de cette thèse s’est orienté sur l’évaluation simultanée de l’impact de JMJD3 et de EZH2. Dans un premier temps, les niveaux d’expression de JMJD3 et de EZH2 ont été montrés augmentés simultanément dans le cancer de la prostate. Cette augmentation est corrélée à un enrichissement de ces deux protéines sur le promoteur des gènes RARβ2, ERα, RGMA, AR et PGR. Dans un deuxième temps, une analyse transcriptomique a permis d’identifier une signature génique corrélée avec le niveau d’agressivité de la tumeur. L’utilisation des « épidrogues » GSK-J4 et DZNeP ciblant JMJD3 et EZH2 permettent de moduler l’expression de ces gènes. L’ensemble de ces résultats caractérise JMJD3 et EZH2 comme des facteurs clés dans le processus de tumorigenèse prostatique. Le panel de gènes identifié devrait permettre de développer de potentiels marqueurs de diagnostic mais également de pronostic dans le cancer de la prostate et sa modulation par les « épidrogues » permettra de développer de nouvelles stratégies thérapeutiques
In France like in majority of developed countries, prostate cancer is the most common cancer in men. It has been clearly established that genetic and epigenetic alterations are common events in prostate cancer resulting in aberrant gene expression. Histone methylation are involved in gene expression of cells. The H3K27me3 epigenetic mark is a repressive mark and it is deregulated in prostate cancer. H3K27me3 levels are determined by the balance between histone methyltransferase EZH2 and histone demethylase JMJD3 activities. In order to understand the mechanism of H3K27me3 deposition in prostatic tumorigenesis, this thesis focused on the simultaneous assessment of the impact of JMJD3 and EZH2.Firstly, expression levels of JMJD3 and EZH2 were shown to be simultaneously increased in prostate cancer. The increase is correlated to both protein enrichments on RARβ2, ERα, RGMA, AR and PGR gene promotors. Secondly, transcriptomic analysis identified gene signature correlated with tumor aggressiveness. The utilization of GSK-J4 and DZNeP epidrugs targeting JMJD3 and EZH2 allowed us to modulate gene expressionOur results characterized JMJD3 and EZH2 as key factors in prostatic tumorigenesis process. The identified gene panel would be able to develop potential diagnostic and prognostic markers in prostate cancer and their modulation by epidrugs would make new therapeutic strategies

Epigenetic states constitute heritable features of the chromatin to regulate when, where and how genes are expressed in the developing conceptus. A special case of epigenetic regulation, genomic imprinting, is defined as parent of origin-dependent monoallelic expression. The Igf2-H19 locus is considered as paradigm of genomic imprinting with a growth-promoting gene, Igf2, expressed paternally and a growth antagonist, H19 encoding a non-coding transcript, expressed only from the maternal allele. The monoallelic expression patterns are regulated by the epigenetic status at an imprinting control region (ICR) in the 5´-flank of the H19 gene. The chromatin insulator protein CTCF interacts with only the maternal H19 ICR allele to prevent downstream enhancers to communicate with the Igf2 promoters. Mutations of these CTCF binding sites lead to biallelic Igf2 expression, increased size of the conceptus and predisposition for cancer. Reasoning that these effects cannot be explained by the regulation of Igf2 expression alone, a technique was invented to examine long-range chromatin interactions without prior knowledge of the interacting partners. Applying the circular chromosomal conformation capture (4C) technique to mouse neonatal liver cells, it was observed that 114 unique sequences interacted with the H19 ICR. A majority of these interactors was in complex with only the maternal H19 ICR allele and depended on the presence of functional CTCF binding sites. The functional consequence of chromosomal networks was demonstrated by the observation that the maternal H19 ICR allele regulated the transcription of two genes on another chromosome. As the chromosomal networks underwent reprogramming during the maturation of embryonic stem cells, attention was turned to human cancer cells, displaying features common with mouse embryonic stem cells. Subsequently, chromatin folding at the human H19 ICR suggested that stable chromatin loops were organized by synergistic interactions within and between baits and interactors. The presence of these interactions was linked to DNA methylation patterns involving repeat elements. A "flower" model of chromatin networks was formulated to explain these observations. This thesis has unravealed a novel feature of the epigenome and its functions to regulate gene expression in trans. The identified roles for CTCF as an architectural factor in the organization of higher order chromatin conformations may be of importance in understanding development and disease ontogeny from novel perspectives.

Cancer is considered nowadays a genetic and epigenetic disease. Aberrancies in epigenetic marks in DNA and histone tails together with alterations in epigenetic regulators responsible of catalyzing these marks have been shown to be crucial in tumorigenesis. These epigenetic regulators are commonly known as writers, readers and erasers. The plasticity of the epigenetic landscape compared to the unchangeable nature of genetic alterations has led to an increasing interest in the last years in finding specific drugs able to modulate and correct the epigenetic aberrancies present in tumors. At present, there are six epigenetic drugs already used in the clinics for the treatment of hematological cancers, two DNA methyltransferase inhibitors and four Histone deacetylases (HDACs) inhibitors. There is also a vast number of clinical trials ongoing with several drugs targeting the epigenetic modulators of the epigenome. The fact that some HDAC inhibitors are already in clinics makes particularly interesting the study of histone acetylation and its enzymatic regulators. Thus, lysine acetylation is regulated by: 1) Histone acetyltransferases (HATs) responsible of adding the acetylation mark in histone tails, being the ‘writers’; 2) Histone Deacetylases (HDACs) that remove the acetyl group acting as ‘erasers’ and 3) the bromodomains are the ‘readers’, that bind to acetyl groups and doing so recruit to specific sites in chromatin other molecular machinery involved in DNA-related processes. Moreover, it has also been described that HDACs and HATs also regulate acetylation in proteins different from histones but also very important in cancer such as the well-known p53 or Myc. The present Doctoral Thesis has been devoted to study epigenetic regulators involved in acetylation of histones and non-histones substrates, as also its targeting with small-inhibitors in cancer. The project was divided in three lines of study. Study I: We investigated the role of the HAT KAT6B in Small Cell Lung Cancer (SCLC). We reported that KAT6B undergoes homozygous deletion in SCLC and that it has tumor suppressor-like properties in vitro and in vivo in this type of cancer. KAT6B catalyzes the acetylation of lysine 23 of histone H3, being the first acetylation site described for this protein. Moreover, KAT6B impairment predicts an increased sensitivity to Irinotecan in SCLC models. Study II: Our objective was to unveil the molecular implications of bromodomain inhibitor JQ1 treatment in breast cancer. We found that JQ1 decreases cell viability in human luminal breast cancer cell lines and downregulates PDZK1 and BCAS1, two important genes in breast cancer tumorigenesis. In addition, JQ1 used as curative treatment in a luminal breast cancer mice model leads to the appearance of smaller tumors. As a preventive treatment in the same mice model JQ1 treatment increases overall survival and delays the offset of the tumors. Study III: We studied a new HDAC6 inhibitor (QTX125) in cancer. We found that this new drug is highly specific for HDAC6 over the other HDACs and increases acetylation levels of α-tubulin, a well-known target of HDAC6, in a dose-dependent manner. It has antitumoral effect in 48 human cancer cell lines and Mantle Cell Lymphoma (MCL) cell lines are highly sensitive to QTX125. This drug induces apoptosis by cleavage of Caspase 3, 8 and 9 and PARP in vitro in MCL and it also exerts antitumoral effect by decreasing tumor growth in MCL xenografts. Interestingly, we observed that MCL primary cells from patients are more sensitive to QTX125 than PBMCs, CD3+ and CD19+ cell from healthy donors
El cáncer se define actualmente como una enfermedad genética y epigenética. En los tumores son frecuentes las alteraciones en marcas epigenéticas en el ADN y en colas de histonas, así como en las enzimas reguladoras de estas marcas, denominadas escritoras, lectoras o borradoras. Concretamente, la marca de acetilación está regulada por: 1) Histonas acetiltransferasas (HATs); 2) Histonas deacetilasas (HDACs) y 3) bromodominios. Además, estas enzimas también pueden regular la acetilación de otras proteínas importantes en cáncer diferentes a las histonas. Una de las ventajas de la epigenética frente a la genética es la naturaleza reversible de las marcas epigenéticas. Por ello, es interesante el estudio de fármacos que permitan modular y corregir las aberraciones epigenéticas presentes en los tumores. Esta tesis doctoral tiene como objetivo el estudio de los reguladores epigenéticos de la acetilación de histonas y otras proteínas en cáncer, así como su tratamiento con inhibidores específicos. El proyecto fue dividido en tres líneas de estudio. Estudio I: Estudiamos el papel de la HAT KAT6B en cáncer de pulmón de células pequeñas (SCLC). Describimos la deleción homocigota de KAT6B y su papel como supresor tumoral en SCLC y encontramos que KAT6B acetila la lisina 23 de la histona H3. Además, esta deleción predice una elevada sensibilidad al fármaco irinotecán en SCLC. Estudio II: Centrado en estudiar las implicaciones moleculares del fármaco inhibidor de bromodominios JQ1 en cáncer de mama de tipo luminal. JQ1 reduce el crecimiento tumoral in vitro disminuyendo la expresión de PDZK1 y BCAS1, dos genes importantes en cáncer de mama. JQ1 también disminuye el desarrollo tumoral, aumenta la supervivencia y retrasa la aparición de tumores en un modelo murino de cáncer de mama luminal. Estudio III: Consiste en el estudio de un nuevo fármaco (QTX125) inhibidor de HDAC6 en cáncer. Reportamos su efecto antitumoral en cáncer, y una elevada sensibilidad en linfoma de células del manto (MCL). QTX125 es altamente específico para HDAC6 e inhibe el crecimiento tumoral in vitro e in vivo en este tipo de linfoma. Además, células primarias de pacientes de MCL son más sensibles a QTX125 que PBMCs procedentes de donantes sanos.

Nicht-kleinzelliger Lungenkrebs (NSCLC) hat bis heute einen hohen medizinischen Bedarf an effektiveren Therapien. Inhibitoren der Bromodomain and extra-terminal domain (BET) Familie wie JQ1 wirken in verschiedenen Krebsarten, einschließlich Lungenkrebs. Während ihre Aktivität auf die Expression von Onkogenen wie c-Myc in vielen Studien untersucht wurde, bleibt der Effekt von BET-Inhibition auf den Apoptose Signalweg weitgehend unbekannt. In dieser Arbeit wurde die Aktivität von BET-Inhibitoren auf den Zellzyklus und auf Komponenten der Apoptose-Antwort der Zelle untersucht. Genomweite Transkriptionsanalysen haben zusammen mit Chromatin Immunpräzipitation und anschließender Sequenzierung geholfen das MYC Gen und dessen assoziierte Super-enhancer als primäres Ziel des BET-Inhibitors JQ1 zu identifizieren. Mittels einer Gruppe von NSCLC Modellen belegt diese Arbeit, dass Zelllinien die auf die BET-Inhibitoren reagieren in Apoptose gehen und eine Reduktion der S-Phasen Population zusammen mit gleichzeitiger de-regulation der c-Myc Expression aufwiesen. Andererseits konnte die ektopische Überexpression von c-Myc der anti-proliferativen Wirkung entgegenwirken. Die Auswirkung von BET-Inhibition auf die Expression von 370 Genen, die in der Apoptose Regulation involviert sind, wurde in sensitiven und resistenten Zellen verglichen und dabei wurde die starke BET-Abhängigkeit der Expression von zwei Schlüsselgenen der Apoptose FLIP und XIAP festgestellt. Die Kombination von JQ1 mit dem tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) oder dem Chemotherapeutikums Cisplatin die verstärke die Induktion von Apoptose in sowohl BET-Inhibitor sensitiven als auch in resistenten Zellen. Des Weiteren zeigte die Kombination einen verbesserten Antitumor-Effekte in A549 tumortragenden Mäusen. Insgesamt zeigen diese Ergebnisse, dass die Identifizierung von BET-abhängigen Genen unterstützend für die Wahl von therapeutischen Kombinationspartnern in der Krebsbehandlung sein kann.
Non-small cell lung cancer (NSCLC) has a high medical need for more effective therapies. Small molecule inhibitors of the bromodomain and extra terminal domain (BET) family such as JQ1 are active in different cancer types, including lung cancer. While their activity on oncogene expression such as c-Myc has been addressed by many studies, the effects of BET inhibition on the apoptotic pathway remain largely unknown. This work evaluates the activity of BET bromodomain inhibitors on cell cycle distribution and on components of the apoptotic response. Genome-wide transcriptional analyses together with chromatin immunoprecipitation followed by sequencing helped to identify the MYC gene and associated super-enhancers as a primary target of JQ1. Using a panel KRAS-mutated NSCLC models, it was found that cell lines responsive to BET inhibitors underwent apoptosis and reduced their S-phase population, concomitant with down-regulation of c-Myc expression. Conversely, ectopic c-Myc overexpression rescued the anti-proliferative effect of JQ1. The effects of BET inhibition on the expression of 370 genes involved in apoptosis were compared in sensitive and resistant cells and the expression of the two key apoptosis regulators FLIP and XIAP was found to be highly BET-dependent. Consistent with this, combination treatment of JQ1 with the tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) or the pro-apoptotic chemotherapeutic agent cisplatin enhanced induction of apoptosis in both BET inhibitor sensitive and resistant cells. Furthermore the combination of JQ1 with cisplatin led to significantly improved anti-tumor efficacy in A549 tumor-bearing mice. Altogether these results show that the identification of BET-dependent genes provides guidance for the choice of drug combinations in cancer treatment.