Repository logo

Theses - MRC Cancer Unit


Recent Submissions

Now showing 1 - 20 of 34
  • ItemOpen Access
    Investigating the regulation of APOBEC mutagenesis in cancer
    Ahmed, Karim
    The sequencing of cancer genomes has revealed that cancers harbour recurrent patterns of mutation, known as mutational signatures. One common mutational signature, known as the APOBEC signature, is found in over 70% of cancer types. The APOBEC signature is thought to be mediated by the activity of the APOBEC enzyme family. However, the underlying cause of APOBEC activity in cancer is not fully understood. This thesis investigates the regulation of APOBEC mutagenesis in cancer. There is a focus on APOBEC3A and APOBEC3B as likely mediators of APOBEC signature mutations, and a focus on the hypothesis that APOBEC activity might be driven by the activity of LINE-1 retrotransposons. Firstly, the thesis details experiments to investigate APOBEC activity in cultured cancer cells. The experiments conducted suggest that p53 inactivation leads to the upregulation of LINE-1 and APOBEC3B, and that their expression may be downregulated when p53 activity is promoted. An enzymatic assay for cancer-associated APOBEC activity is established. Reverse transcriptase inhibitors, which inhibit LINE-1 activity, appear to modulate APOBEC3B expression and associated enzymatic activity. This appears to occur when cells are p53-deficient, but not when p53 is intact. Secondly, the thesis reports exploratory bioinformatic analyses conducted using large genomic datasets. These indicate that APOBEC3A expression is associated with interferon signalling in cancer while APOBEC3B expression is associated with cell cycle signalling in cancer. A deletion of a consensus interferon response factor binding site in the APOBEC3B promoter is identified. Analyses of regulatory data suggest that APOBEC3B might be transcriptionally insulated from syntenic APOBEC3 genes by CTCF. In addition, p53 deficiency in cancer appears to be associated with the upregulation of APOBEC3A and APOBEC3B. Thirdly, the thesis reports bioinformatic analyses of an RNA sequencing dataset from patients with the rare genetic disease Aicardi-Goutières syndrome. It is thought that the disease process of the genotypes studied might driven by LINE-1 activity, as suggested in part by successful patient trials of reverse transcriptase inhibitors. The analyses conduced appear to identify changes to the transcriptome in Aicardi-Goutières syndrome that might mirror those associated with APOBEC activity in cancer. In sum, these experiments provide evidence for possible regulators of APOBEC mutagenesis in cancer, including evidence that broadly supports the hypothesis that it may be driven by LINE-1 activity. The experiments also identify a class of drugs that might enable the pharmacological modulation of cancer-associated APOBEC activity. APOBEC mutagenesis is thought to mediate cancer initiation, progression, intratumour heterogeneity and responses to therapy, including immunotherapy. The findings detailed might therefore contribute to the ability to understand and control the natural history of cancer across multiple cancer types.
  • ItemEmbargo
    Investigating the chromatin dynamics that underlie CRISPRi-validated enhancer-promoter interactions
    Pirvan, Liviu
    Discovered in the 1980s, enhancers have grown to become important cis-regulatory elements responsible for both temporal and tissue-specific gene regulation coordinating cell differentiation processes, pathological responses and disease progression, and cellular functions in normal physiological states. However, due to their complex nature and function, research into enhancer activity remains highly active, producing novel insights based on multiple types of data and experimental assays. Firstly, advances in Next Generation Sequencing technologies in the 2000s allowed extensive enhancer characterisations through mass detection of genome wide epigenetic marks, chromatin conformation, chromatin accessibility, and expression data. Nevertheless, the process of validating enhancer targets remained costly until recent developments of CRISPRi methods enabled large scale perturbation experiments for identifying the target genes of enhancers. However, CRISPRi methods produced results that were not always consistent with the previous understanding of enhancer-promoter interactions (EPIs), raising questions regarding the mechanism by which EPIs are produced. Then, this thesis aims to investigate the ways by which chromatin dynamics enable the function of CRISRPi-validated EPIs, in order to explain the discrepancies observed. Firstly, using experimental data from mouse cells, I consider EPIs based on chromatin folding data while highlighting the limitations of current frameworks. Then, I consider more complex representations of biological features to propose predictive models for improving EPI detection. Next, to investigate the chromatin dynamics that underlie CRISPRi-validated EPIs, I use unsupervised machine learning to identify distinct biological classes of EPIs that show mechanistic differences. With respect to these classes, I propose that they are underlied by different chromatin structures, and their functions may be related to enabling cancer progression. Thus, the former insight motivates the development of a graph-based deep learning model that can model complex chromatin structures and transcription factor binding data in order to predict EPIs and interpret their structures. Similarly, the second insight motivates the design of a CRISPRi experiment that can be used to validate the classification system in a second independent cell type. Then, I conclude that the relationship between the functions of CRISPRi-validated EPIs and chromatin dynamics reveals novel complexities that can be modelled computationally to describe important processes in gene regulation.
  • ItemControlled Access
    The host response to cancer and cancer-associated cachexia
    Ferrer Gonzalez, Miriam
    Cachexia, a severe wasting syndrome, affects up to 90% of patients with cancer and causes significant morbidity and mortality worldwide. Patients with cachexia suffer from weight loss, anorexia, sarcopenia, and behavioural changes that eventually lead to death. Given its incidence and prevalence, its negative impact on prognosis and quality of life, and the consequent poor tolerance/response to treatment, cachexia is a major global public health burden. In the first chapter of my thesis, I analyse the usage of ketogenic diets (KD) containing high levels of fats as adjuvant therapies in end-stages of cancer that are associated with cachexia. Metabolism of fats through non-enzymatic lipid peroxidation is a recognized source of highly reactive and mutagenic molecules: lipid peroxidation products (LPPs). The biosynthetic pathway of corticosterone, the main regulator of metabolic stress, and the pathway for detoxification of LPPs require NADPH as cofactor but their biochemical interdependency has not been explored. Using two murine models of cancer-associated cachexia, I demonstrate that KD slows down tumour growth but accelerates onset of cachexia in tumour-bearing mice. Reduced tumour size results from accumulation of LPPs, saturation of the GSH detoxifying pathway and ferroptotic death of cancer cells. Moreover, systemic redox state imbalance in KD-fed tumour-bearing mice causes NADPH depletion and primary hypoadrenalism. Dexamethasone treatment delays the onset of cachexia and extends the survival of tumour-bearing mice fed with KD compared to untreated tumour-bearing mice on either KD or standard feeding by improving metabolic homeostasis and utilization of nutritional substrates. Thus, lack of appropriate corticosterone synthesis during cachexia leads to metabolic maladaptation and an inability to use energy sources in mice fed KD. Dexamethasone administration to KD fed mice improves tissue preservation, energy expenditure and survival while preserving reduced tumour growth. In the second chapter I describe neutrophilia as an early event during cancer progression. Transcriptomic and metabolic assessment reveals that neutrophils in tumour-bearing animals utilize aerobic glycolysis, similar to cancer cells. Although pharmacological inhibition of aerobic glycolysis slows down tumour growth in tumour-bearing mice, it precipitates cachexia. This negative effect may be explained by the observation that acute depletion of neutrophils in pre-cachectic mice impairs systemic glucose homeostasis secondary to altered hepatic lipid processing. Thus, changes in neutrophil number, distribution, and metabolism play an adaptive role in host metabolic homeostasis during cancer progression. These findings provide insight into early events during cancer progression to cachexia, with implications for therapy. Lastly, I discuss preliminary data on other potential research avenues in the context of cancer cachexia. These includes understanding the underlying mechanisms behind the behavioural changes observed in cancer that are mediated by the nervous system, the investigation of sexual dimorphism and genetic background with regards to susceptibility to cachexia, disentangling the specific role of interleukin-6 in the metabolic reprogramming associated with cachexia, and the host’s metabolic response to caloric restriction. My results demonstrate that the host response is an important determinant of cancer outcome and argue for the absolute necessity of a combined analysis regarding the effects of candidate cancer treatments on both the tumour and the host.
  • ItemOpen Access
    Investigating the Role of Oncogenic KRAS G12 Mutations in Cell Signalling
    Patel, Khushali
    Cancer is a multistep process reflecting genetic alterations that drive progressive transformation of normal cells into highly proliferative malignant cells. Deregulation of cellular signalling is one of the key traits in cancer, allowing cells to breach anticancer defence mechanisms. The most frequently mutated oncogene in cancer is KRAS, encoding a small GTPase protein involved in controlling the activity of critical signalling pathways that regulate normal cellular proliferation, such as the PI3K and ERK pathways. The prevalence of different codon substitutions in the KRAS gene varies in different tissues. There is emerging evidence supporting the notion that different codon substitutions may trigger different biological effects. We hypothesise that different codon substitutions in KRAS can trigger different feedbacks and signalling dynamics that may result in varying fitness advantages in different tissues. I use quantitative Western blotting and Modular Response Analysis with KRAS isogenic cell lines to characterise how KRAS substitutions at codon G12 perturb the topology and dynamics of the ERK signalling network as a first step to test this hypothesis. My work has identified two mutant-specific interactions in the ERK pathway: a MEK to RAF inhibition seen strongest in G12A, G12C and G12D cells, and a RAF to ERK activation (or loss of inhibition) seen in all mutants compared to WT. Antibody array data suggests the potential role of JNK and TYK2 in mediating these interactions, respectively, and thus my work has provided preliminary, albeit testable, hypotheses for elucidating the possible mechanisms responsible for this rewiring of the ERK pathway. My work has also identified the presence of a second BRAF form present only in the mutants with the strongest MEK to RAF inhibition. It appears this KRAS-mutant specific BRAF form is most likely a splice variant that has enhanced dimerisation capabilities with CRAF. RAF dimerisation is one mechanism via which RAF inhibitors fail to be effective in treating KRAS mutant cancers. This exemplifies the importance of characterising the identity and role of this BRAF form, as it may have implications on development of mutant-specific therapies for KRAS-driven cancers. ERK pathway activation can also regulate the transcription of many downstream targets. I have attempted to characterise KRAS-dependent gene expression changes with RNA sequencing data, with the aim of understanding key differences between mutant and WT KRAS cells and to also understand how the gene expression profiles change over time during the very early steps in oncogenesis. Preliminary data suggests mutant-specific differential gene transcription that may be linked to the RAF inhibition via MEK and/or the second BRAF form. My work currently does not explain the specific mechanisms behind the interactions identified with MRA, however, it does lay down the foundation and provides hypotheses that can be tested in the lab. Gaining mechanistic insight into the pathogenicity of cancer driver mutations and their differential role in different tissues is of fundamental importance to understand how different mutations shape the evolution of cancer clones during carcinogenesis and to design optimal targeted therapeutic strategies.
  • ItemOpen Access
    Development of a hydrogel-based microfluidic model of the lymph node
    Mazzaglia, Corrado
    Lymph nodes (LNs) are immunological hubs where antigen and naïve lymphocytes meet to mount antigen specific immune responses. In cancers, the tumour-draining lymph node (TDLN) is the site of initial anti-tumour immune response. The TDLN undergoes dramatic reprogramming in response to tumour drainage that is still partly unclear. In vivo models do not easily permit us to follow events longitudinally or to manipulate the environment, while currently there is a lack of in vitro platforms that recapitulate the LN complexity to study its role in response to antigens and in disease. First, I have tried to adapt an existing PDMS-based device to model the LN. Having encountered some limitations, I have then developed a novel hydrogel-based LN-on-chip device. I have optimized a fluidic system, cell culture conditions and hydrogel composition to obtain long term culture of lymphoid primary cultures from spleen with a stromal cell line of fibroblastic reticular cells (FRCs). Different hydrogel formulations were tested and characterized with multiple techniques focusing on maintaining physiological cell morphology in the new 3D environment. In addition, a LN-like architecture was achieved where B cells are segregated from T cells, into B cell follicles. After optimization of the 3D cellular system with built-in fluidics, I applied the model to different context. Initially, PMA/Ionomycin, a strong cell activator, was perfused to obtain a general activation response. Then, LPS was used to trigger a reaction to a bacteria-derived antigen. Finally, using the OT-I model, I have used the device to trigger an antigen specific immune response to an OVA peptide. In the final part of the thesis, I modelled a TDLN in vitro. The LN device was incubated with tumour conditioned media (TCM) and IL-7 downregulation was observed, similarly to in vivo murine melanoma models. Then, to increase the complexity of the source of the tumour cues, a new tumour model was developed through 3D bioprinting, which also includes a shell of cancer associated fibroblasts (CAFs) and immune cells to form a complex engineered tumour microenvironment (TME). Incubation of the LN device with conditioned medium from the engineered TME also induced downregulation of IL-7. v Unlike current in vivo models that require node dissection at each discrete time point, this system allowed the monitoring of tissue remodelling in real-time via live imaging and measure molecular changes at a surface markers level as well as gene expression. Overall, this platform can represent a valid tool to study in vitro the LN complexity in a more tractable way.
  • ItemControlled Access
    Resolving the evolution of Oesophageal Adenocarcinoma using spatial and temporal sampling
    Ococks, Emma Louise
    Introduction Oesophageal adenocarcinoma (OAC) remains a poor prognosis cancer type. The pre-malignant lesion for this type of cancer is called Barrett’s Oesophagus (BO), which has two subtypes, Gastric Metaplasia (GM) and Intestinal Metaplasia with GM (IM). Both BO and OAC are genetically heterogeneous lesions and little is known about the genetic relationship between OAC and pre-cancerous BO, as well as the clinical significance of GM which is not considered to be BO in the US guidelines. Furthermore, the considerable intra-tumour heterogeneity (ITH) may underlie the observed chemo-resistance. There have not been any large-scale studies to assess ITH in OAC and associated BO. Aside from multi-region sequencing, another way of overcoming sampling bias of the heterogeneous tissue is through circulating tumour DNA (ctDNA). Liquid biopsy sampling also has been proven to be highly prognostic in multiple cancer types and recent advances in ctDNA detection have now provided a potential avenue for this sampling approach to utilised in OAC. Hypothesis Deciphering the tissue heterogeneity of OAC and associated BO will help determine the evolutionary relationship between these tissue types with clinical relevance. Blood biopsy may provide a method to overcome tissue heterogeneity and inform clinical management in OAC. Aims • Characterise the intra tumour heterogeneity in OAC • Better understand the evolution of OAC, by using phylogenetic analysis of BO and OAC • Compare the genetic architecture of Barrett’s IM and GM • Evaluate the clinical potential of ctDNA detection for minimal residual disease using two different strategies, namely cancer panel versus personalised assay. Materials and Methods Multi-region whole exome sequencing was performed on chemotherapy-naïve 398 samples from 79 patients with OAC. The samples were manually micro-dissected and then sequenced to a mean depth of 150x. Clustering was performed using PyClone and tree building by CloneEvol. Mutational signatures were identified using deconstructSigs. The gene panel ctDNA analysis included 245 double-spun plasma samples and 78 peripheral blood samples from 97 patients. The samples were prepared using the Roche expanded Avenio ctDNA panel. The personalised panel included 53 double-spun plasma samples collected from 20 patients and processed using the Signatera assay. Results Multi-region sequencing revealed considerable ITH in OAC, with the majority of mutations being subclonal. Moreover, SMAD4 which has been previously shown to be a biomarker for poor prognosis, was more often subclonal. In addition, the majority of the BO mutations are also subclonal, indicating substantial ITH in this premalignant lesion. As expected, BO had less driver mutations than OAC, but the driver mutations in BO were more likely to be subclonal. Comparison between IM and GM, demonstrated they had distinct mutational signature profiles and GM was significantly less mutated than IM. Interestingly, the variant allele frequency (VAF) of the tumour and IM mutations was similar, however the VAF of the GM mutations was significantly lower than the tumour and IM mutations, which was surprising given that GM samples generally had a higher cellularity than IM samples. Also, driver gene mutations in BO were not always shared with the tumour, and in some cases there were no shared clones between the BO and tumour samples. Both the gene panel and personalised panel demonstrated ctDNA to be prognostic for disease free and cancer-specific survival in the post-surgery setting despite the ITH. Using the Roche gene panel assay that is comprised of 77 cancer genes, 14 of which are OAC driver genes, TP53 was the most frequently mutated gene (15%), followed by APC (8%), then KRAS (6%). The sequencing of peripheral blood cell samples (available for 80%) revealed that 23% of cases had at least one variant derived from Clonal Haematopoiesis of Indeterminate Potential (CHIP) that could confound the analysis. Once CHIP was removed, median cancer-specific survival in the post-surgical setting for ctDNA positive patients was 10 months compared to 30 months for ctDNA negative patients (hazard ratio 5.55, 95% CI: 2.42-12.71, P = 0.0003). In order to overcome the diverse genetic landscape and the low ctDNA shedding in OAC, a bespoke panel based on the WGS of the primary may be advantageous. The pilot evaluation (n=20) of a personalised panel was weakly prognostic for disease free survival (p= 0.042) and could detect minimal-residual disease down to 0.01%. Moreover, this method gave an average lead time of nearly one year. Discussion Multi-region analysis revealed that OAC is a highly heterogenous cancer type, which could explain why OAC is frequently resistant to chemotherapy. In addition, the results from this study suggest that BO is also heterogeneous. Furthermore, GM and IM are quite distinct in terms of their evolution, with differences in their mutational signatures, mutation burden, and mutation VAF providing biological rationale for the lower cancer conversion rate that has been observed for GM. This finding may substantiate the clinical decision not to include GM cases in Barrett’s diagnosis and surveillance programmes. Based on these results, an alternative approach to take could be to design surveillance programmes that are further stratified to distinguish between patients with GM and IM when the BO segment exceeds 3 cm. The results from the plasma analysis demonstrates that ctDNA is prognostic post-operatively in OAC. Moreover, the technology available is highly sensitive, indicating the potential clinical application of liquid biopsy sampling. However, if a gene panel approach is used, a peripheral blood cell sample should also be sequenced to eliminate variants derived from clonal haematopoiesis and this is suboptimal given the heterogeneous nature of the driver gene landscape. A personalised approach has promise for detection of minimal residual disease that should be further evaluated in larger prospective trials.
  • ItemOpen Access
    An Investigation of Mutational Signatures in the Evolution of Oesophageal adenocarcinoma
    Abbas, Sujath; Abbas, Sujath [0000-0002-2541-4969]
    An Investigation of Mutational Signatures in the Evolution of Oesophageal adenocarcinoma Sujath Abbas Oesophageal adenocarcinoma (OAC) remains a public health challenge with dismal survival rates and increasing incidence. This PhD study aimed to investigate how mutational processes act across different stages of OAC development and in metastasis for better understanding of the influence of mutational forces during tumour formation. To identify these signatures in clinical samples this study also aimed to develop a cost-effective DNA sequencing method in clinical formalin fixed OAC samples. A large study cohort was assembled comprising of 161 Barrett’s, 777 OAC primary tumours and 59 metastatic samples. Mutational signature analysis revealed 14 distinct single base substitution (SBS) mutational signatures in these genomes, SBS17b/a were most prevalent and presented early in Barrett’s. Traces of BER (SBS30), MMR(SBS44) and colibactin associated signature (SBS41) were uncovered for the first time, as well as a platinum signature (SBS35). Mostly signatures increased in their proportions from Barrett’s to invasive tumours and further in metastasis. SBS17 showed strong bias towards untranscribed and lagging strands. Nucleosome periodicity patterns were similar across the stages and SNVs were enriched in the inward facing minor groove suggesting a common mutational process throughout the disease evolution. Evaluation of evolutionary bottlenecks uncover a distinct SBS17b shift, with a decrease sub-clonally in Barrett’s, OACs and metastasis and this was by far the most dominant signal during OAC evolution. Clinical risk factors including alcohol, smoking and NSAIDs were positively correlated with signature proportions. APOBEC and colibactin processes were informative for Barrett’s and OAC classification, suggesting a role in transformation, and the BER signature (SBS30) was most prognostic in our cohort. Given that signatures have the potential to be clinically informative, a novel cost-effective DNA sequencing method to extract mutational signatures from archival FFPE tissues was developed successfully. Computational simulations on pan-cancer WGS and an experimental confirmation of the method showed very good concordance and mirrored the WGS-derived signatures (cosine similarity >0.9%). It is hoped that this work will pave the way for further studies to understand how mutations are laid down and determine their clinical application.
  • ItemOpen Access
    Selection and competition of somatic mutations in normal epithelia
    Hall, Michael
    Tumourigenesis occurs when a series of genome alterations occur in the same group of cells and cause uncontrolled cell proliferation. Therefore, to understand the journey from healthy to cancerous tissue, it is important to study the accumulation and spread of mutations in pre- cancerous normal tissues. Recent studies have shown that apparently normal epithelium contains a high burden of mutations in cancer-associated genes. This thesis explores the behaviour of mutant clones in normal epithelium and how this affects cancer development. The nature of mutant clonal growth and competition in normal epidermis has been a subject of debate. A study found that mutant clone sizes inferred from DNA sequencing of normal human eyelid skin were consistent with a mathematical model of neutral cell dynamics, appearing to contradict a genetic analysis of the same dataset that found several genes under positive selection. I investigate this debate using computational modelling that takes into account the tissue structure and experimental tissue-sampling methods. The results show that mutant clone sizes in skin and oesophagus are consistent with non-neutral clonal competition. Further evidence for non-neutral selection in normal epithelium is found in patterns of mutations detected by DNA sequencing. By adapting a statistical method used for driver gene discovery, I look for enrichment or depletion of structural categories of missense mutations and find biologically meaningful patterns of selection in several proteins. The method can associate changes to protein structure or function with cell fitness, even in the absence of hotspot mutations and in the presence of passenger mutations. I demonstrate how the method is flexible and could be widely applicable, but can also produce misleading results if confounding sources of selection are not accounted for. Clonal competition in normal oesophageal epithelium is dominated by Notch1 loss-of- function mutations. I fit stochastic models of clonal dynamics to lineage tracing data to show that haploinsufficiency greatly accelerates Notch1 mutant expansion and that the loss of the second Notch1 allele provides a further strong selective advantage, consistent with the high frequency of NOTCH1 loss-of-heterozygosity events observed in human oesophagus. Finally, I examine a consequence of the spread of these highly fit mutant clones in the normal tissue. I use a mathematical model to analyse the results of a series of experiments in mutagen-treated mouse oesophagus, finding that microscopic tumours can be eliminated by highly fit clones in the surrounding normal tissue.
  • ItemOpen Access
    SMARCB1 Maintains Lineage Fidelity in Clear Cell Renal Cell Carcinoma
    (2022-06-15) Wesolowski, Ludovic
    Lineage-specific transcription factors have emerged as a promising class of essential genes in cancer. The best examples of leveraging this phenomenon in the clinic is targeting the androgen receptor and the oestrogen receptor in prostate and breast cancer respectively. Despite the success of these therapies, the mechanisms that maintain lineage fidelity in advanced cancer clones, and whether lineage factor pathways could be exploited in other cancer types remain poorly understood. Using clear cell renal cell carcinoma (ccRCC) as a model, I characterise mechanisms that underlie lineage factor dependence in cancer. Using CRISPR/Cas9 loss-of-function screening coupled with in vitro and in vivo validation I show that the loss of SMARCB1, a member of the SWI/SNF chromatin remodelling complex, confers an advantage to ccRCC cells upon inhibition of the essential renal lineage factor PAX8. SMARCB1 knockout (KO) leads to large-scale loss of a kidney-specific enhancer program, conversion to a cellular state resembling that of rhabdoid tumours, and the re-activation of proliferative pathways. Using a second CRISPR/Cas9 screen, I show that these proliferative pathways are underpinned by the acquisition of new transcriptional dependencies. These dependencies represent rare essential genes across different lineage-specific and oncogenic pathways, a principle validated in a large-scale CRISPR/Cas9 screening data set comprising hundreds of cancer cell lines. In summary, dependence on tissue-specific lineage factors in cancer can be modulated via epigenetic remodelling.
  • ItemOpen Access
    Interrogating the functional adaptions of Tumour-draining lymph nodes (TDLNs) and the subsequent impact of checkpoint inhibition
    Cridge, Jake
    Lymph nodes (LNs) function as immunological hubs whereby antigen and naïve lymphocytes meet to facilitate an immune response. In cancers, the tumour-draining lymph node (TDLN) is where the initial anti-tumour immune response occurs. Previous studies have demonstrated that TDLNs undergo dramatic reprogramming in response to tumour drainage. However, the precise nature of these alterations and their contribution to tumour-induced immune escape requires further exploration. Utilising the B16-F10 melanoma and E0771 breast cancer (BC) model I identified significant changes in the T cell landscape of TDLNs. While the CD8-mediated immune response is effectively suppressed in TDLNs, the CD4 compartment undergoes dramatic alterations. Most notably, regulatory T cells (Tregs) expand and become more active, upregulating suppressive markers, including TGF-B, CTLA-4 and FasL. Concurrently, CD4 and CD8 T cells show diminished secretion of IL-2 and IFNy, common markers of T cell activation. Localisation of Tregs also differed in TDLNs, displaying closer contacts with stromal fibroblastic reticular cells (FRCs) and lymphatic endothelial cells (LECs). Supporting these observations, in vitro studies demonstrated that tumour-conditioned FRCs display an enhanced capacity to recruit and adhere T cells. These findings support a model whereby tumour drainage reprograms the stromal and immune populations in TDLNs to create an immunosuppressive environment, thereby hindering the anti-tumour immune response. Immune checkpoint inhibitors (ICIs) have emerged as a novel strategy to reinvigorate the immune system promoting tumour clearance; however, therapeutic success is often followed by resistance and tumour resurgence. To date most research has focused on the impact of ICIs at the tumour site, and the TDLN remains largely ignored. In this regard, I also examined the T cell response of ND- and TDLNs following a-PD-1 or a-CTLA-4 monotherapy. Immunotherapy bolstered the anti-tumour immune response in TDLNs, driving the differentiation of both CD4 and CD8+ T cells into effector and CM subsets, which circulated to mediate tumour cell clearance. Interestingly, checkpoint blockade also induced Treg expansion in TDLNs, with cells adopting a more immunosuppressive phenotype, as confirmed by in vitro and ex vivo suppression assays. Correspondingly, tumour cell death was reduced in LN isolates from therapy versus control treated LNs, demonstrating that the inhibitory niche is sufficient to overcome the immunostimulatory effects afforded by checkpoint blockade. Consequently, we have devised the “Homeostatic Hypothesis” to explain this phenomena. Immunotherapy effective bolsters the anti-tumour immune response within the TDLN, enhancing cancer clearance. Tregs subsequently expand in TDLNs to quell the immune response and prevent overt activation, resulting from checkpoint blockade. Complete Treg depletion within TDLNs therefore holds promise to maximize therapeutic effect, following checkpoint blockade, and mitigate resistance.
  • ItemOpen Access
    Oncogene-induced remodelling of cellular networks
    Oriol Valls, Pablo
    Biochemical networks maintain cellular homeostasis by processing information from the intra- and extra-cellular environments and coordinating appropriate responses to stimuli and stressors. During carcinogenesis, cells accrue genetic alterations that rewire signalling pathways, eventually leading to the partial loss of homeostatic control and cancer- related phenotypes. Cancer is a very heterogeneous disease with genetic alterations that exhibit vast variability within a tumour, between tumours of different tissues, and patients. Despite the enormous advances the community has achieved in understanding the molecular causes of cancer, the mechanisms of pathogenicity of specific mutations are often unclear, severely limiting the efficacy of the therapeutic intervention and disease management. Moreover, we are recognizing that driver mutations once considered sufficient to induce initiation and promotion of cancer are more frequent than originally expected also in healthy tissues. In addition to genetic variability, cell-to-cell variability of non-genetic origin is increasingly recognized – yet poorly understood – as a fundamental force fuelling carcinogenesis and resistance to therapies. We therefore develop biochemical imaging techniques to investigate aspects of cancer heterogeneity that are currently challenging – if not impossible – to study. KRAS is a driver gene significantly involved in lung, colorectal and pancreatic carcinogenesis. Various types of KRAS mutations respond differently to therapy, and some have been linked to the emergence of resistance. However, the mutation-specific mechanisms leading to these phenotypes is not well understood. We used KRAS as a clinically relevant model to study the effects of mutations at codon G12 on signalling dynamics and phenotypic changes, including cell-to-cell variability. In this thesis, first I introduce the concepts of evolution-driven oncogenesis, genetic and non-genetic heterogeneity and signal transduction pathways. In Chapter 2, I describe the characterisation of KRAS mutation-specific MAPK signalling. To do this, I generated an isogenic panel of cells stably expressing a FRET sensor and developed an experimental pipeline including automated biochemical imaging, microfluidic-based stimulation, and custom image analysis. Using these tools, I have identified differ- ential signalling dynamics and heterogeneity in response to epidermal growth factor (EGF). Taken together with other results, we hypothesise the presence of a weakened negative feedback loop in cells with a G12D mutation in KRAS. In Chapter 3, I report on methodological developments aimed at drastically improving our capability to characterise biochemical networks in single living cells. I have optimized novel pairs of fluorescent proteins dedicated to biochemical multiplexing and capable of simultaneously monitoring three biochemical reactions. Moreover, I developed an ex- pandable software pipeline that yields single-cell ERK signalling data and several types of analyses from microscopy FRET imaging. Finally, I further develop optogenetic systems designed to dynamically and optically activate KRAS in single cells. In conclusion, I show the effects of KRAS mutations on signalling dynamics and cell- to-cell variability, a first step towards a deeper understanding of how genetic and non-genetic heterogeneity may cooperate to make of cancer the terrible disease we know. At the same time, I have improved methodologies designed for the novel study of biochemical networks, laying down the foundations of system-level investigation in single living cells.
  • ItemOpen Access
    Identification of tumour microenvironment-derived signals that modulate the development and functionality of MDSCs
    Zimarino, Carlo; Zimarino, Carlo [0000-0001-8650-5912]
    Myeloid Derived Suppressor Cells (MDSCs) are a heterogeneous immune population found within the tumour microenvironment (TME). We sought to explore the potential role of microenvironment components on the suppressive behaviour, development and maintenance of MDSCs, focusing mainly on the role of the SIRPα-CD47 signalling axis. In an orthotopic melanoma model, we observed an increase in Ly6C-expressing myeloid cells (indicating monocytic-myeloid derived suppressor cells (M-MDSCs) and monocytic dendritic cells (moDC)) as tumours developed. These cells were suppressive, able to block T CD8+ cell proliferation. To model the effect of tumour-derived factors on M-MDSCs and moDCs development and function, we developed a culture system using hematopoietic stem cells cultured with GM-CSF and melanoma tumour condition media (TCM). Similar to the in vivo setting, exposure to TCM skewed the myeloid compartment towards an M-MDSC and moDC phenotype (based on Ly6C expression) that potently suppressed CD8+ T cell proliferation to a greater extent than GM-CSF induced MDSCs. Further characterisation of the TME by single-cell RNA sequencing and flow cytometry revealed specific expression of the signal-regulatory protein alpha (SIRPα) in M-MDSC and moDC cells fraction and elevated expression of its cognate ligand, CD47 by other immune cells. Thus, we investigated the impact of CD47-SIRPα interaction on MDSC function. Engagement of SIRPα on moDCs and M-MDSCs by recombinant CD47 in vitro induced intracellular signalling via SHP2, and inhibited the phagocytic capability of these cells. Moreover, persistent activation of this programme translated to an increase in their suppressive phenotype quantified by elevated expression of immune checkpoint molecules, inhibitory factors and reactive oxygen species. Knowing this axis promoted a pro-tumour, suppressive phenotype, we then investigated the consequence of its disruption on tumour growth in vivo. Neutralization of SIRPα on moDCs and M-MDSCs in established tumours resulted in a significant decrease in growth, which was driven by a reprogramming of moDCs and M-MDSCs. Disruption of the CD47-SIRPα axis was sufficient to rescue their phagocytic capability, which in turn enhanced their ability to process and present antigen to tumour infiltrating T cells. These functional changes were accompanied by metabolic adaptations. In summary, we report the CD47-SIRPα axis functions as a mechanism used to support moDC and M-MDSC suppressive function in the TME, and its disruption in early tumour-infiltrating monocyte progenitors shows potential to restore anti-tumour features of myeloid cells and in turn promote the T cell mediated anti-tumour immune response.
  • ItemOpen Access
    The generation of a human cell line to elucidate the role of Fumarate Hydratase loss in cancer
    Schmidt, Christina
    The reprogramming of cellular metabolism is an established hallmark of cancer, which enables cancer cells to survive, proliferate, and metastasize even under harsh environmental conditions. These cancer-associated metabolic changes can affect several pathways one of which is mitochondrial metabolism. The suppression of mitochondrial metabolism has been associated with poor clinical outcomes and mitochondrial dysfunction has been associated with some hereditary and sporadic forms of cancer that arise from mutations in mitochondrial genes. Understanding the mechanisms responsible for cellular transformation and subsequent tumour formation in these hereditary, metabolically-impaired tumours could link dysregulated mitochondrial function and tumour formation. Hereditary mutations and subsequent loss of the mitochondrial TCA cycle enzyme fumarate hydratase (FH) leads to Hereditary Leiomyomatosis and Renal Cell Cancer (HLRCC), an aggressive form of renal cancer associated with poor clinical outcome. The loss of FH triggers the accumulation of fumarate, which induces a multi-layer cellular reprogramming that contributes to tumorigenesis. Yet, it is unclear how FH loss influences the whole gene expression landscape and if the gene expression is regulated on the level of DNA-methylation, transcription or translation. In this thesis, I generated the first FH-deficient human renal epithelial cell lines using CRISPR/Cas9-based genome editing, and applied proteomics, metabolomics, and transcriptomics approaches to investigate how the loss of FH alters these cellular layers. First, I confirmed that this model faithfully recapitulates the biochemical and phenotypic markers of FH-deficiency as previously reported. Next, I developed a novel multi-omics tool, SiRCle (Signature Regulatory Clusters) to disentangle this interconnected network of signalling cascades. Using SiRCle, I extracted clusters of increased/decreased gene expression that are regulated at the level of DNA methylation, transcription, and/or translation, and identified which clusters drive which phenotype of FH loss. By mapping the transcription factors that drive the genes of each cluster, I identified unique drivers that could be responsible for the cellular rewiring after FH loss. It is now clear that the tumour microenvironment affects the phenotype of cancer cells, and hence that metabolic rewiring becomes essential for tumour cells to strive even under harsh environmental conditions. Yet, its effect on FH-deficient cells’ behaviour is currently unknown. In this part of the thesis, I used a Tumour Roll for Analysis of Cellular Environment and Response (TRACER), a 3D scaffold that develops oxygen and nutrient gradients similar to those observed in tumours. Using TRACER, I show that the main metabolic signature of FH loss, which is driven by the high levels of fumarate, is not influenced by the nutrient and oxygen gradients generated in this 3D model. Consequently, FH loss is a stronger driver of the metabolic signature than environmental cues. Moreover, by applying linear modelling to the metabolic profile of the cells over the different layers, I identify specific layer-dependent metabolic signatures in FH-deficient cells that are not observed in 2D culture. These results imply that in vivo FH loss could undergo previously unacknowledged compensatory metabolic changes, which underlines the important role of the microenvironment in dictating the phenotype of cancer cells.
  • ItemOpen Access
    Modelling timing in blood cancers
    Talarmain, Laure
    Dysregulation of biological processes in normal cells can lead to the abnormal growth of tumours. Oncogenesis requires the acquisition of advantageous mutations to expand in a fluctuating environment. Cancer cells gain these genetic and epigenetic alterations at different timing in their development, resulting in the formation of heterogeneous cell populations which interact and compete with each others inside tumours. At later stages, by escaping the immune system and acquiring malignant properties, some cancer cells manage to evade the primary tumour and spread in different organs to form metastases. Hence, tumour development in healthy tissues endure several biological changes whilst progressing and the order between these molecular and cellular events may modify prognosis. This thesis addresses the influence of biological event timing on blood cancer progression and clinical outcomes. It first investigates the therapeutic efficacy of p53 restoration in a lymphoma mouse model. While several therapy schedules are tested, all fail due to resistance emergence. Computational modelling establishes the cell dynamics in these tumours and how to use it to propose alternative treatment strategies. Data availability leads this work to explore the impact of molecular evolution in myeloid malignancies. Notably, one study has found that Myeloproliferative Neoplasms patients with both JAK2 and TET2 mutations have different disease characteristics with distinct mutation order. My analyses identify HOXA9 as a potential prognosis marker and biological switch responsible for patient stratification in these patients and in Acute Myeloid Leukemia. Additionally, a molecular network identifies the hematopoietic regulators involved in the branching evolution of Myeloproliferative Neoplasms. Further investigations of the Acute Myeloid Leukemia data show the possible involvement of APP, a gene associated to Alzheimer disease, in early cell fate commitment in hematopoiesis and in poor survival prognosis in undifferentiated leukemia when lowly expressed. Finally, this thesis examines the regulatory dynamics behind three clusters of Acute Myeloid Leukemia patients with distinct levels of HOXA9 and APP expression. By building a program inferring molecular motifs from biological observations, genes which may interact with HOXA9 and APP are identified.
  • ItemOpen Access
    Functional Characterisation of a RECQL4 Mutation in Rothmund-Thomson Syndrome
    Wu, Tianyi; Wu, Tianyi [0000-0002-5386-2355]
    Germline mutations affecting the RECQL4 DNA helicase cause Type II Rothmund-Thomson syndrome (RTS), a human disease characterised by defects in skeletal development and predisposition to specific types of cancer, including osteosarcoma (OS). RECQL4 has been implicated in multiple cellular functions that mediate accurate DNA replication and repair. How germline RECQL4 mutations associated with Type II RTS affect these functions to cause disease remains unclear, in part due to the paucity of appropriate cellular models. In this work, CRISPR/Cas9 gene editing was used to generate cell lines containing a prevalent RTS patient RECQL4 mutation, the “Mut-2” c.2269C>T. The resulting Mut-2 clones exhibited greatly reduced RECQL4 protein levels, similar to decreases observed in RTS patient cells. Unexpectedly, the major effect of this predicted nonsense mutation was the upregulation of the use of an alternative splice site in exon 14 which skipped the premature stop codon and resulted in the deletion of 66 amino acids in the RECQL4 ATPase domain. Despite the lower overall RECQL4 expression, single cell clones bearing the Mut-2 mutation showed mostly normal cell cycle distribution with a slight increase in population doubling times. When challenged with various DNA damaging agents, these Mut-2 clones exhibited increased sensitivities to DNA alkylators and topoisomerase inhibitors, and mild sensitivities to DNA crosslinkers and PARP inhibitors, a sensitivity profile suggestive of defects in DNA double-strand break (DSB) repair. When further assayed using flow cytometric GFP reporters, the Mut-2 clones showed decreased DNA DSB repair capacities in the homologous recombination (HR) and microhomology mediated end joining (MMEJ) pathways, providing evidence that RECQL4 disruption impacted replication-specific DNA DSB repair in particular. Additional RECQL4 reconstitution studies confirmed that the decreased HR repair was a result of structural changes to RECQL4 due to the Mut-2 mutation. Finally, the formation of RAD51 foci—a commonly used marker of HR function—in the Mut-2 clones post-DNA DSB induction was investigated. Surprisingly, upon DNA DSB challenge, all Mut-2 clones were as proficient at forming RAD51 foci as parental HEK293. This suggested that the RECQL4 Mut-2 mutation disrupted its function further downstream in the HR pathway than had been previously reported. The work presented in this dissertation is a novel approach to studying the effects of clinical RTS RECQL4 mutations. These studies have illuminated mechanisms of RECQL4 disruption in Type II RTS as well as the roles of the RECQL4 helicase in cellular DNA damage repair. Because about 30% of Type II RTS patients are diagnosed with osteosarcoma, a common and deadly primary malignancy of the bone, the results presented here could shed new light on potential mechanisms underlying osteosarcoma tumour development and ultimately suggest new avenues and strategies for targeted clinical intervention.
  • ItemOpen Access
    Defining the pro-tumour impact of the evolving stromal microenvironment
    (2020-10-26) Davidson, Sarah
    Much like normal tissues, tumours require a supporting microenvironment for growth and survival, known as the tumour stroma. However, tumours represent a dynamic and turbulent environment, in which factors such as hypoxia, fibrosis, nutrient deprivation and the local cytokine milieu continually fluctuate as the tumour grows and develops. These factors influence stromal phenotypes and create heterogeneity, which can confound our understanding of their role within the microenvironment. In particular, cancer associated fibroblasts (CAFs) represent a diverse population of cells, which cannot be identified by one universal marker. CAFs promote tumour growth and dissemination by secreting growth factors, stimulating angiogenesis, aiding the development of tumour-promoting inflammation and remodelling the extra-cellular matrix (ECM). However, recent investigations have shown these functions belong to defined populations that differ between tumour types. This project aimed to investigate stromal heterogeneity across melanoma development, with a specific focus on the CAF compartment. Whilst conventional techniques such as IF and flow cytometry showed varied expression of fibroblast markers, they lacked the resolution to discern functional subsets. Thus, we employed single cell RNA sequencing (scRNAseq) to profile CAF populations at different stages of tumour development. To avoid bias, CAFs were isolated from the B16-F10 melanoma model using a ‘negative selection’ approach. Our data revealed the presence of 3 functionally distinct fibroblast populations, termed ‘immune’, ‘desmoplastic’ and ‘contractile’, which expressed genes involved in immune cross talk, matrix remodelling and stress fibre contraction respectively. Furthermore, these populations are dynamic, changing in prevalence as the tumour grows. While ‘immune’ and ‘desmoplastic’ populations were present from early stages, ‘contractile’ CAFs were more abundant at later time points. Owing to their unique marker profiles, we were able to identify these populations within the tumour stoma and validate their temporal nature. Subsequent investigation into the contribution of these subsets to tumour growth, revealed that ‘immune’ CAFs promoted accumulation of suppressive macrophages by production of C3 and its cleavage product C3a. Significantly, inhibition of the C3a/C3aR axis reduced the number of macrophages and decreased tumour volume. This reduction in tumour growth was accompanied by increased CD8 T-cell infiltration, implying that ‘immune’ CAFs may inhibit adaptive anti-tumour immunity through controlling the myeloid compartment. The interaction between C3 producing CAFs and C3aR expressing macrophages was conserved in different murine tumours and human cancer. Thus, ‘immune’ CAFs and C3a signalling may represent therapeutic targets in multiple cancer types. Overall, our data highlights the complexity of stromal phenotypes and microenvironment interactions, which likely reflects the convoluted climate of the developing tumour.
  • ItemOpen Access
    The role of space in homeostasis and preneoplasia in stratified squamous epithelia
    Kostiou, Vasiliki
    A major subject of study in biological research is the dynamics of stem cells in squamous epithelia. Given that most common human cancers develop from epithelia, understanding the rules of cell fate decision in these systems is key to explaining not only healthy tissue growth and maintenance but also the processes of mutagenesis and cancer. The aim of my project was to investigate the dynamics in squamous epithelial tissues both in homeostasis and preneoplasia, using cellular automata (CA) models. Stem cell dynamics has been shown to be accurately described by a simple mathematical model, the single progenitor (SP) model. Reliable parameterisation of this model would give access to valuable quantitative information on epithelial tissue maintenance and enable investigating how mutations affect tissue dynamics. I initially identified the most appropriate method for accurately parameterising the homeostatic system. I then sought to account for the spatial patterning of cells by implementing the SP model in two-dimensional space. The spatial model was able to reproduce the key signatures of homeostatic dynamics, thus showing that restrictions imposed by tissue organization do not alter the neutral dynamics. Furthermore, I studied non-homeostatic dynamics in stratified squamous epithelial tissues by spatially modelling the growth and competition of non-neutral mutations as well as the effects of wounding in the tissue. The studied dynamics of Notch and p53 mutant clones in mouse epithelia has been found to be highly distinct, with the former fully colonizing the tissue whereas the latter only partially. I demonstrated that the two mutants’ tissue takeover dynamics can be recapitulated by two distinct spatial feedback rules, on the basis of response to crowding, providing a mechanistic explanation of the observed distinct growth modes. Finally, mutant competition was explored. A striking effect resulting from the spatial interaction of the two mutations in a wild-type background is that the p53 mutant cell population was always outcompeted by the Notch mutant population and appeared to shrink. Considering this consistent emergent behaviour in the competition simulations and given the paucity of Notch mutations in human cancer datasets, it is tempting to speculate that the aggressive fitness of Notch may offer a tumour-protective effect.
  • ItemOpen Access
    Mechanisms of human RAD51 regulation by RAD52 and BRCA2
    Constantinou, Stephanie
    The RAD51 recombinase assembles as helical nucleoprotein filaments on single-stranded (ss)DNA substrates to mediate homologous DNA recombination (HR) and replication fork protection, processes vital in human cells for the maintenance of genome stability. RAD51 assembly is controlled by two key mediator proteins in eukaryotic organisms – the tumour suppressor, BRCA2, and RAD52. Recent evidence suggests that human RAD52 becomes essential for viability in cancer cells lacking BRCA2, making its activity an attractive target for potential therapeutic strategies. However, the mechanisms by which RAD52 and BRCA2 coordinate RAD51 regulation during HR or replication fork protection remain unclear. Therefore, I sought to elucidate these mechanisms and determine the functional redundancy, if any, between the two proteins. Here, I show that human RAD52 co-localises with the ssDNA-binding protein RPA and RAD51 following ionising radiation (IR)-induced damage. Moreover, RAD52 controls the chromatin recruitment and DNA assembly of RAD51, as well as subsequent HR-mediated DNA repair in BRCA2-deficient cells, but is dispensable for these processes in cells that are heterozygous or wild-type for BRCA2. In contrast, RAD52 protects nascent DNA at reversed replication forks from excessive degradation by the MRE11 endonuclease, not only in BRCA2-deficient cells, but also in cells that are heterozygous or wild-type for BRCA2. Mechanistically, RAD52 affects RAD51 recruitment to perturbed replication forks, and its depletion enhances the formation of DNA double-strand breaks (DSBs) and cell death following replication stress induced by hydroxyurea. Thus, these findings suggest divergent requirements for BRCA2 and RAD52 in the regulation of RAD51 during HR versus replication protection. RAD52 is redundant for RAD51-mediated HR in cells that are heterozygous or wild-type for BRCA2, but becomes an essential recombination mediator in cells lacking BRCA2. On the contrary, during replication protection, RAD52 activity is essential for RAD51 regulation regardless of BRCA2 function. Lastly, I describe preliminary results from collaborative experiments deploying electron cryo-microscopy to determine structural mechanisms underlying the regulation of RAD51 filament assembly by BRCA2. A high-resolution structure from a complex of RAD51, ssDNA and the BRC repeats of BRCA2 suggests that BRCA2 BRC repeats may promote conformational changes assisting in homologous DNA strand-pairing. Collectively, the research reported in my thesis provides new insight into the mechanisms by which BRCA2 and RAD52 regulate the RAD51 recombinase during reactions that lead to HR and replication fork protection.
  • ItemOpen Access
    A KLF6-driven transcriptional network links lipid homeostasis and tumour growth in clear cell renal cell carcinoma
    (2019-10-26) Syafruddin, Saiful Effendi
    Clear cell renal cell carcinoma (ccRCC) is characterised by frequent inactivation of the VHL tumour suppressor gene and consequent accumulation of HIF2A that drives tumourigenesis. The current clinically-approved therapies for ccRCC are those targeting the angiogenesis and mTOR signalling pathways, however, the overall patients’ objective response rates are still low, and patients rapidly develop resistance towards the administered therapies. An incomplete understanding of the underlying molecular mechanisms that support ccRCC progression has contributed to the lack of effective diagnostic and/or therapeutic strategies developed, especially for the highly mortal advanced stage ccRCC. Thus, the identification of cellular networks on which ccRCC cells are highly dependent would facilitate the development of better diagnostic and/or therapeutic approaches for ccRCC. Super enhancers have been reported to drive the expression of critical transcription regulators in various biological contexts including the regulation of cancer phenotypes. Previously generated H3K27ac ChIP-Seq data from several ccRCC cell lines has identified KLF6, a zinc finger DNA-binding transcription factor, to be associated with one of the strongest super enhancers in ccRCC, which could signify a biological relevance to KLF6 in supporting ccRCC pathogenesis. Thus, the purpose of this present study was to interrogate the role of KLF6 in ccRCC, and dissect the KLF6-regulated transcriptional networks and how they can contribute in supporting ccRCC pathogenesis. It was discovered that KLF6 expression was supported by a robust super enhancer that integrates signals from multiple pathways, including the ccRCC-initiating VHL-HIF2A pathway. In line with its regulation by the super enhancer, CRISPR-Cas9 and CRISPRi-mediated perturbation of KLF6 led to impaired ccRCC cells growth in vitro and in vivo as well as reducing the cells metastatic lung colonisation capability. KLF6 inhibition led to the deregulation of lipid homeostasis pathways in ccRCC cells. A dual KLF6 role was identified in modulating lipid homeostasis pathways in ccRCC: First, KLF6 directly regulates the expression of several important lipid homeostasis genes. Second, KLF6 promotes PDGFB expression, which activates the mTORC1 signalling pathway and the key lipid metabolism transcriptional regulators SREBF1 and SREBF2. KLF6 and mTORC1 thus co-regulate lipid homeostasis, consequently supporting ccRCC cell growth. Furthermore, findings from this study also reveal a molecular link between the PDGF and mTORC1 signalling pathways, which are the clinically relevant therapeutic targets in ccRCC. In general, the link between super enhancer-driven transcriptional networks and essential metabolic pathways described herein may provide clues to the mechanisms that maintain the stability of cell identity-defining transcriptional programmes in cancer.