No description available.

Cancer is a multi-stage process resulting from accumulation of genetic mutations. Data obtained from assaying a tumor only contains the set of mutations in the tumor and lacks information about their temporal order. Learning the chronological order of the genetic mutations is an important step towards understanding the disease. The probability of introduction of a mutation to a tumor increases if certain mutations that promote it, already happened. Such dependencies induce what we call the monotonicity property in cancer progression. A realistic model of cancer progression should take this property into account. In this thesis, we present two models for cancer progression and algorithms for learning them. In the first model, we propose Progression Networks (PNs), which are a special class of Bayesian networks. In learning PNs the issue of monotonicity is taken into consideration. The problem of learning PNs is reduced to Mixed Integer Linear Programming (MILP), which is a NP-hard problem for which very good heuristics exist. We also developed a program, DiProg, for learning PNs. In the second model, the problem of noise in the biological experiments is addressed by introducing hidden variable. We call this model Hidden variable Oncogenetic Network (HON). In a HON, there are two variables assigned to each node, a hidden variable that represents the progression of cancer to the node and an observable random variable that represents the observation of the mutation corresponding to the node. We devised a structural Expectation Maximization (EM) algorithm for learning HONs. In the M-step of the structural EM algorithm, we need to perform a considerable number of inference tasks. Because exact inference is tractable only on Bayesian networks with bounded treewidth, we also developed an algorithm for learning bounded treewidth Bayesian networks by reducing the problem to a MILP. Our algorithms performed well on synthetic data. We also tested them on cytogenetic data from renal cell carcinoma. The learned progression networks from both algorithms are in agreement with the previously published results. MicroRNAs are short non-coding RNAs that are involved in post transcriptional regulation. A-to-I editing of microRNAs converts adenosine to inosine in the double stranded RNA. We developed a method for determining editing levels in mature microRNAs from the high-throughput RNA sequencing data from the mouse brain. Here, for the first time, we showed that the level of editing increases with development. / <p>QC 20130503</p>

The Receptor Tyrosine Kinase (RTK) Met, overexpressed or mutated in cancer, plays a major role in cancer progression and represents an attractive target for cancer therapy. This study aimed to investigate whether PI3K plays a role in Met oncogenicity. Three cell models were used: (i) NIH3T3 cells expressing WT Met or the constitutively active mutant M1268T Met; (ii) U87MG glioblastoma cells, with endogenous WT Met constitutively activated due to an autocrine loop; (iii) A549 lung cancer cells expressing endogenous WT Met, activated upon binding exogenous HGF. Met dependent Rac1 translocation to the plasma membrane, actin cytoskeleton organisation, cell migration, anchorage independent growth in soft agar and tumour growth were studied in the presence of inhibitors of pan-PI3K / mTOR, various PI3K Class I isoforms, mTOR or Akt, or following siRNA knock-down of PI3K isoforms. We report that PI3K class I (but not class III) regulates Met dependent cell migration. The PI3K class I isoforms required varies among the cell models. Interestingly, the combined inhibition of all p110 Class I isoforms lead to the strongest reduction of Met dependent cell migration. Met dependent phosphorylation of Akt, an effector of PI3K class I, is reduced upon endocytosis inhibition, suggesting that Met signals to PI3K Class I on endosomes. Our results indicate that mTOR is responsible for Met dependent anchorage independent growth and tumour growth in vivo. Surprisingly, PI3K class I (and class III) are not required. Moreover, Rac1 is required for Met dependent mTOR activation, (phosphorylation of mTORC1's effector, p70 S6K) subcellular translocation of mTOR and anchorage independent growth. Finally, our results suggest that this Met-Rac1- mTOR pathway occurs on endosomes. Thus while PI3K class I regulates Met dependent cell migration, mTOR regulates Met driven anchorage independent growth and in vivo tumorigenesis. Thus PI3K Class I / mTOR may be targeted in Met driven cancers.

LIV-1, a zinc transporter, is an effector molecule downstream from soluble growth factors. This protein has been shown to promote epithelial-to-mesenchymal transition (EMT) in human pancreatic, breast, and prostate cancer cells. Despite the implication of LIV-1 in cancer growth and metastasis, there has been no study to determine the role of LIV-1 in prostate cancer progression. Moreover, there is no clear delineation of the molecular mechanism underlying LIV-1 function in cancer cells. In this study, we found increased LIV-1 expression in a progresssive manner in benign, PIN, primary and bone metastatic human prostate cancer. We characterized the mechanism by which LIV-1 drives prostate cancer EMT in an androgen-refractory human prostate cancer cell (ARCaP) bone metastasis model. LIV-1, when overexpressed in ARCaPE cells (derivative cells of ARCaP with epithelial phenotype), promoted EMT irreversibly. LIV-1 overexpressed ARCaPE cells had elevated levels of HB-EGF and matrix metalloproteinase (MMP) 2 and MMP 9 proteolytic enzyme activities, without affecting intracellular zinc concentration. The activation of MMPs resulted in the shedding of heparin binding-epidermal growth factor (HB-EGF) from ARCaPE cells, eliciting constitutive epidermal growth factor receptor (EGFR) phosphorylation and its downstream extracellular signal regulated kinase (ERK) signaling. Further investigation of the HB-EGF promoter revealed that both Stat3 and AP-1 controlled HB-EGF promoter activity. Ectopic LIV-1 overexpression induced AP-1 and Stat3 activation. Blockade of both Stat3 and AP-1 by specific inhibitors or dominant negative expression vectors diminished the HB-EGF promoter activity induced by LIV-1 overexpression. These results suggest that LIV-1 is involved in prostate cancer progression as an intracellular target of growth factor receptor signaling which promotes EMT and cancer metastasis. LIV-1 could be an attractive therapeutic target for the eradication of pre-existing human prostate cancer and bone and soft tissue metastases.

The development of cancer from a population of precancerous cells is a rapid evolutionary process. During progression, cells evolve several new traits for survive and proliferation via a few key `driver' mutations. However, these few driver alterations reside in a cancer genome alongside tens of thousands of additional `passenger' mutations. Passengers are widely believed to have no role in cancer, yet many passengers fall within functional genomic elements that may have potentially deleterious effects on the cancer cells. Here we investigate the potential of moderately deleterious passengers to accumulate and alter neoplastic progression. Evolutionary simulations suggest that moderately-deleterious passengers accumulate during progression and largely evade natural selection. Accumulation is possible because of cancer's unique evolutionary constraints: an initially small population size, an elevated mutation rate, and a need to acquire several driver mutations within a short evolutionary timeframe. Cancer dynamics can be theoretically understood as a tug-of-war between rare, strongly-beneficial drives and frequent mildly-deleterious passengers. In this formalism, passengers present a barrier to cancer progression describable by a critical population size, below which most lesions fail to progress, and a critical mutation rate, above which cancers collapse. In essence, cancer progression can be subverted by its own unique evolutionary constraints. The collective burden of passengers explain several oncological phenomena that are difficult to explain otherwise. Genomics data confirms that many passengers are likely damaging and have largely evaded negative selection, while age-incidence curves and the distribution of mutation totals suggests that drivers and passengers exhibit competing effects. These data also provide estimates of the strength of drivers and passengers. Finally, we use our model to explore cancer treatments. We identify two broad regimes of adaptive evolutionary dynamics and use these regimes to understand outcomes from various treatment strategies. Our theory explains previously paradoxical treatment outcomes and suggest that passengers could serve as a biomarker of response to mutagenic therapies. Deleterious passengers are targetable by either (i) increasing the mutation rate or (ii) exacerbating their deleterious effects. Our results suggest a unique framework for understanding cancer progression as a balance between driver and passenger mutations.

Prognostic models have clinical appeal to aid therapeutic decision making. In the UK, the Nottingham Prognostic Index (NPI) has been used, for over two decades, to inform patient management. However, it has been commented that NPI is not capable of identifying a subgroup of patients with a prognosis so good that adjuvant therapy with potential harmful side effects can be withheld safely. Tissue Microarray Analysis (TMA) now makes possible measurement of biological tissue microarray features of frozen biopsies from breast cancer tumours. These give an insight to the biology of tumour and hence could have the potential to enhance prognostic modelling. I therefore wished to investigate whether biomarkers can add value to clinical predictors to provide improved prognostic stratification in terms of Recurrence Free Survival (RFS). However, there are very many biomarkers that could be measured, they usually exhibit skewed distribution and missing values are common. The statistical issues raised are thus number of variables being tested, form of the association, imputation of missing data, and assessment of the stability and internal validity of the model. Therefore the specific aim of this study was to develop and to demonstrate performance of statistical modelling techniques that will be useful in circumstances where there is a surfeit of explanatory variables and missing data; in particular to achieve useful and parsimonious models while guarding against instability and overfitting. I also sought to identify a subgroup of patients with a prognosis so good that a decision can be made to avoid adjuvant therapy. I aimed to provide statistically robust answers to a set of clinical question and develop strategies to be used in such data sets that would be useful and acceptable to clinicians. A unique data set of 401 Estrogen Receptor positive (ER+) tamoxifen treated breast cancer patients with measurement for a large panel of biomarkers (72 in total) was available. Taking a statistical approach, I applied a multi-faceted screening process to select a limited set of potentially informative variables and to detect the appropriate form of the association, followed by multiple imputations of missing data and bootstrapping. In comparison with the NPI, the final joint model derived assigned patients into more appropriate risk groups (14% of recurred and 4% of non-recurred cases). The actuarial 7-year RFS rate for patients in the lowest risk quartile was 95% (95% C.I.: 89%, 100%). To evaluate an alternative approach, biological knowledge was incorporated into the process of model development. Model building began with the use of biological expertise to divide the variables into substantive biomarker sets on the basis of presumed role in the pathway to cancer progression. For each biomarker family, an informative and parsimonious index was generated by combining family variables, to be offered to the final model as intermediate predictor. In comparison with NPI, patients into more appropriate risk groups (21% of recurred and 11% of non-recurred patients). This model identified a low-risk group with 7-year RFS rate at 98% (95% C.I.: 96%, 100%).

The historic relationship of species and genes are traditionally depicted using trees. However, not all evolutionary histories are adequately captured by bifurcating processes and an increasing amount of research is devoted towards using networks or network-like structures to capture evolutionary history. Lateral gene transfer (LGT) is a previously controversial mechanism responsible for non tree-like evolutionary histories, and is today accepted as a major force of evolution, particularly in the prokaryotic domain. In this thesis, we present models of gene evolution incorporating both LGTs and duplications, together with efficient computational methods for various inference problems. Specifically, we define a biologically sound combinatorial model for reconciliation of species and gene trees that facilitates simultaneous consideration of duplications and LGTs. We prove that finding most parsimonious reconciliations is NP-hard, but that the problem can be solved efficiently if reconciliations are not required to be acyclic—a condition that is satisfied when analyzing most real-world datasets. We also provide a polynomial-time algorithm for parametric tree reconciliation, a problem analogous to parametric sequence alignment, that enables us to study the entire space of optimal reconciliations under all possible cost schemes. Going beyond combinatorial models, we define the first probabilistic model of gene evolution incorporating a birth-death process generating duplications, LGTs, and losses, together with a relaxed molecular clock model of sequence evolution. Algorithms based on Markov chain Monte Carlo (MCMC) techniques, methods from numerical analysis, and dynamic programming are presented for various probability and parameter inference problems. Finally, we develop methods for analysis of cancer progression, a biological process with many similarities to the process of evolution. Cancer progresses by accumulation of harmful genetic aberrations whose patterns of emergence are graph-like. We develop a model of cancer progression based on trees, and mixtures thereof, that admits an efficient structural EM algorithm for finding Maximum Likelihood (ML) solutions from available cross-sectional data. / QC 20100812

Dans les cancers du poumon, une des altérations les plus souvent observées est la perte ou l’atténuation de l’expression du gène FHIT (Fragile Histidine Triad). Nous avons précédemment montré que FHIT est un suppresseur d’invasion tumorale. En effet, FHIT contrôle l’invasion des cellules tumorales bronchiques en régulant négativement l'expression de gènes associés à la transition épithélio-mésenchymateuse (TEM), en particulier la vimentine et la MMP-9 via l’inhibition d’une voie orchestrée par l’EGFR. Un intérêt particulier a donc été porté aux relations entre FHIT et un autre membre de la famille de l’EGFR : HER2. Nous avons non seulement mis en évidence, in vivo et in vitro, une corrélation inverse entre les taux de FHIT et l’activité du récepteur HER2 dans les CBNPC mais également montré que FHIT est capable de réguler l’activité du récepteur HER2 dans les cellules tumorales pulmonaires et ce grâce à sa dimérisation avec HER3. De plus, l’utilisation de deux inhibiteurs spécifiques d’HER2, le Trastuzumab et l’Irbinitinib, nous a permis de mettre en évidence, que l’activation du récepteur HER2 lors de l’inhibition de FHIT, participe à l’acquisition par les cellules tumorales bronchiques de caractéristiques invasives via la régulation de certaines cibles de la TEM, telles la vimentine, la MMP-14 ou encore le facteur de transcription TWIST-1. Ces résultats montrent que FHIT régule l’activité d’HER2 dans les cellules tumorales pulmonaires et que les inhibiteurs d’HER2 sont capables de limiter l’invasion induite par l’inhibition de FHIT. Cette étude laisse envisager de nouvelles perspectives thérapeutiques pour le cancer du poumon. / The lack or decrease of FHIT (fragile histidine triad) expression is a common event in lung cancer. We recently showed that FHIT acts as a suppressor of tumor invasion. Indeed, FHIT controls the invasive phenotype of lung tumor cells by regulating the expression of genes associated with epithelial-mesenchymal transition (EMT) such as vimentin or MMP-9 through an EGFR signaling pathway. Accordingly, we focused on the relationships between FHIT and another member of this tyrosine kinase receptor family: HER2. First, we observed in vivo and in vitro a negative correlation between FHIT expression and the activated form of HER2 in lung tumor cells. Moreover, FHIT controls HER2 activation through its dimerization with HER3. The use of HER2 specific inhibitors, Trastuzumab and Irbinitinib, allowed to demonstrate that the in vitro invasion induced by FHIT inhibition is HER2-dependent. Furthermore, FHIT controls the HER2-dependent invasion by regulating genes associated with EMT such as vimentin, MMP-14 or TWIST-1. In conclusion, we showed that FHIT regulates HER2 activity in lung tumor cells and that HER2 inhibitors reduce invasion induced by FHIT inhibition. This study would allow for the identification of new therapeutic leads for lung cancer.

No description available.

Two areas of particular importance in prostate cancer progression are primary tumour development and metastasis. These processes involve a number of physiological events, the mediators of which are still being discovered and characterised. Serine proteases have been shown to play a major role in cancer invasion and metastasis. The recently discovered phenomenon of their activation of a receptor family known as the protease activated receptors (PARs) has extended their physiological role to that of signaling molecule. Several serine proteases are expressed by malignant prostate cancer cells, including members of the kallikreinrelated peptidase (KLK) serine protease family, and increasingly these are being shown to be associated with prostate cancer progression. KLK4 is highly expressed in the prostate and expression levels increase during prostate cancer progression. Critically, recent studies have implicated KLK4 in processes associated with cancer. For example, the ectopic over-expression of KLK4 in prostate cancer cell lines results in an increased ability of these cells to form colonies, proliferate and migrate. In addition, it has been demonstrated that KLK4 is a potential mediator of cellular interactions between prostate cancer cells and osteoblasts (bone forming cells). The ability of KLK4 to influence cellular behaviour is believed to be through the selective cleavage of specific substrates. Identification of relevant in vivo substrates of KLK4 is critical to understanding the pathophysiological roles of this enzyme. Significantly, recent reports have demonstrated that several members of the KLK family are able to activate PARs. The PARs are relatively new members of the seven transmembrane domain containing G protein coupled receptor (GPCR) family. PARs are activated through proteolytic cleavage of their N-terminus by serine proteases, the resulting nascent N-terminal binds intramolecularly to initiate receptor activation. PARs are involved in a number of patho-physiological processes, including vascular repair and inflammation, and a growing body of evidence suggests roles in cancer. While expression of PAR family members has been documented in several types of cancers, including prostate, the role of these GPCRs in prostate cancer development and progression is yet to be examined. Interestingly, several studies have suggested potential roles in cellular invasion through the induction of cytoskeletal reorganisation and expression of basement membrane-degrading enzymes. Accordingly, this program of research focussed on the activation of the PARs by the prostate cancer associated enzyme KLK4, cellular processing of activated PARs and the expression pattern of receptor and agonist in prostate cancer. For these studies KLK4 was purified from the conditioned media of stably transfected Sf9 insect cells expressing a construct containing the complete human KLK4 coding sequence in frame with a V5 epitope and poly-histidine encoding sequences. The first aspect of this study was the further characterisation of this recombinant zymogen form of KLK4. The recombinant KLK4 zymogen was demonstrated to be activatable by the metalloendopeptidase thermolysin and amino terminal sequencing indicated that thermolysin activated KLK4 had the predicted N-terminus of mature active KLK4 (31IINED). Critically, removal of the pro-region successfully generated a catalytically active enzyme, with comparable activity to a previously published recombinant KLK4 produced from S2 insect cells. The second aspect of this study was the activation of the PARs by KLK4 and the initiation of signal transduction. This study demonstrated that KLK4 can activate PAR-1 and PAR-2 to mobilise intracellular Ca2+, but failed to activate PAR-4. Further, KLK4 activated PAR-1 and PAR-2 over distinct concentration ranges, with KLK4 activation and mobilisation of Ca2+ demonstrating higher efficacy through PAR-2. Thus, the remainder of this study focussed on PAR-2. KLK4 was demonstrated to directly cleave a synthetic peptide that mimicked the PAR-2 Nterminal activation sequence. Further, KLK4 mediated Ca2+ mobilisation through PAR-2 was accompanied by the initiation of the extra-cellular regulated kinase (ERK) cascade. The specificity of intracellular signaling mediated through PAR-2 by KLK4 activation was demonstrated by siRNA mediated protein depletion, with a reduction in PAR-2 protein levels correlating to a reduction in KLK4 mediated Ca2+mobilisation and ERK phosphorylation. The third aspect of this study examined cellular processing of KLK4 activated PAR- 2 in a prostate cancer cell line. PAR-2 was demonstrated to be expressed by five prostate derived cell lines including the prostate cancer cell line PC-3. It was also demonstrated by flow cytometry and confocal microscopy analyses that activation of PC-3 cell surface PAR-2 by KLK4 leads to internalisation of this receptor in a time dependent manner. Critically, in vivo relevance of the interaction between KLK4 and PAR-2 was established by the observation of the co-expression of receptor and agonist in primary prostate cancer and prostate cancer bone lesion samples by immunohistochemical analysis. Based on the results of this study a number of exciting future studies have been proposed, including, delineating differences in KLK4 cellular signaling via PAR-1 and PAR-2 and the role of PAR-1 and PAR-2 activation by KLK4 in prostate cancer cells and bone cells in prostate cancer progression.