Chronic myeloid leukaemia (CML) is a clonal myeloproliferative disorder arising in a haemopoietic stem cell (HSC) as a result of the reciprocal translocation between the long arms of chromosomes 9 and 22 (t9;22), leading to the formation of the fusion oncogene BCR-ABL. BCR-ABL has constitutive tyrosine kinase (TK) activity which drives, at least during the chronic phase (CP) of the disease, myeloid progenitor cells expansion through terminally differentiated cells and is necessary for the transformed phenotype. The introduction at the end of the last century of BCR-ABL TK inhibitors (TKI) has dramatically changed the management of newly diagnosed CP CML patients as the vast majority achieve deep molecular responses while enjoying good quality of life when treated with TKI. However about 20% of patients still show various degree of resistance to all currently available TKI while in those achieving deep responses, there is compelling evidence of persistent minimal residual disease demanding lifelong treatment which has obvious implications in terms of compliance, adverse events and costs. It is now known that the main reason for disease persistence in CML patients treated with TKI lies in the insensitivity of the most primitive CML leukaemia stem cell (LSC). More recent evidence has demonstrated that, in contrast to more mature leukaemic progenitor cells, CML LSC are not addicted to BCR-ABL kinase activity but rather rely on other stem cell intrinsic pathways for their survival. The main focus in the CML field is therefore to identify these pathways while also trying to exploit them therapeutically to achieve CML LSC eradication and as a result disease cure. Growth factor (GF) signals are known to provide survival cues to CML stem and progenitor cells (SPC) and potentially support their survival even in the presence of TKI. Moreover CML SPC are also known to produce higher levels of some GFs via an autocrine loop and support their survival and proliferation through this mechanism. In this thesis, the characterisation of the autocrine GF production by CML SPC was extended while also investigating the role of several GFs and downstream signals in survival, proliferation and self-renewal of CML SPC. Whenever possible, the consequences of therapeutic targeting of these signals on CML SPC survival and proliferation were also assessed in vitro. In particular the role of the intracellular janus kinase (JAK) 2, which relays several myeloid GF signals, such as those from interleukin (IL)-3 and granulocyte macrophage colony-stimulating factor (GM-CSF), in CML SPC survival and proliferation was investigated mainly because higher levels of autocrine expression of GM-CSF by CML SPC relative to normal were demonstrated, while autocrine IL-3 production by CML SPC had already been shown. Moreover the cognate receptor of both GM-CSF and IL-3 (CSF2RB) was also shown to be expressed at higher levels in CML SPC relative to their normal counterparts, further supporting investigations on the role of JAK2 in CML SPC biology. Indeed targeting JAK2 with small molecule inhibitors in CML SPC in vitro, particularly in the presence of maximal BCR-ABL TK inhibition, resulted in increased apoptosis, reduced proliferation and colony output of CML SPC. The JAK2 inhibitor plus TKI combination treatment, compared to either single agent, further reduced survival of the more primitive quiescent LSCs in vitro, while also reducing engraftment of primary CML CD34+ cells in vivo in immunocompromised hosts. Although a degree of toxicity to normal haemopoietic stem and progenitor cells (HSPC) was observed, this was not as great as seen in CML SPC, thus suggesting that a therapeutic window for using JAK2 inhibitors in CML patients might be present when a carefully selected concentration of these compounds is chosen. Tumour necrosis factor (TNF)-α was another GF shown to be produced in an autocrine fashion at higher levels by CML SPC relative to normal HSPC. Moreover its levels of production by CML SPC were not modulated by BCR-ABL TK activity. Using a small molecule TNF-α inhibitor and exogenous TNF-α, it was shown that autocrine TNF-α acts as a survival and proliferative signal in CML SPC. Moreover its role became even more important in the presence of TKI, as combining TNF-α inhibition with TKI led to high levels of apoptosis in CML CD34+ cells, including the more primitive quiescent population, while also causing increased apoptosis in a population enriched for CML LSCs based on its surface marker expression (CD34+ CD38-). Finally given the known importance of quiescence and self-renewal pathways in CML LSC persistence following TKI treatment, the role of transforming growth factor (TGF)-β1 and novel neurotransmitter mediated pathways in CML LSC quiescence and self-renewal was investigated based on the findings of a genome and epigenome-wide screen of primary CML LSCs and normal HSCs carried out in our laboratory. Using in vitro assays the putative role of the neuromediators norepinephrine and acethylcoline in CML LSC self-renewal was demonstrated. Moreover the role of TGF-β1 in inducing primary CML LSC quiescence mainly by modulating the AKT pathway was also demonstrated. Overall the work presented in this thesis has furthered our understanding of the role of both autocrine and paracrine known and novel regulators of haemopoiesis in several aspects of CML SPC biology such as their survival, proliferation and self-renewal. Furthermore the efficacy in eradicating CML SPC of therapeutic strategies targeting some of these GF signals has been explored in vitro, thus providing evidence supporting their subsequent testing in in vivo assays and in due course in clinical studies. It is hoped therefore that the work presented will contribute to devise novel therapeutic strategies to eradicate CML LSC and in turn lead to a cure for CML patients.

The aim of this body of work was to characterise a novel cisplatin drug delivery system and to develop new tools based on biophotonic imaging that could be used to enhance studies of drug delivery in vivo. Cucurbiturils (CB) are macrocycles which are formed by acid catalysed condensation of glycoluril and formaldehyde. The internal cavity of CB[7] encapsulates a single molecule of cisplatin and the hypothesis was that encapsulation would reduce thiol degradation of the drug. Drug sensitivity studies in vitro with the cisplatin-sensitive human ovarian cancer cell line, A2780, and a cisplatin-resistant derivative, A2780/cp70, showed that the CB[7] encapsulated cisplatin retained activity but that this encapsulation drug delivery system was not able to overcome resistance to platinum. However, when these cell lines were grown as subcutaneous xenografts in nu/nu mice, the encapsulated cisplatin was able to reduce the growth of A2780/cp70 tumours which are resistant to the maximum tolerated dose of cisplatin in vivo. One possible explanation of this observation is that encapsulation might alter the pharmacokinetics of cisplatin and a method for the detection of platinum in biological samples by ICP-MS was established and validated. This assay was sufficiently sensitive to detect the low levels of platinum present in mouse plasma 24 hours after administration of either free or encapsulated cisplatin. Plasma and tissue pharmacokinetics show that encapsulation had no effect on the peak plasma concentration of cisplatin but did reduce the rate at which cisplatin was cleared from the plasma. The increased plasma AUC of cisplatin resulted in a non-selective increase in the delivery of cisplatin to both tumour and normal tissues. However, there was no apparent increase in toxicity which could be explained by the fact that encapsulation, unlike an increase in the dose of free cisplatin, had no effect on the peak plasma concentration. Subcutaneous xenografts lack critical features of human tumours. The development of more complex models for use in drug development has been limited due to lack of a method for monitoring tumour growth. Biophotonic imaging was, therefore, investigated to determine whether it is sufficiently sensitive and reproducible to be able to evaluate growth of disseminated tumours in mice. The bioluminescent signal is dependent on the metabolism of luciferin by luciferase. Subcutaneous injection of luciferin was shown to produce a consistent signal in all injected mice. The bioluminescent signal was transient but reached a maximum intensity 6 minutes after injection and remained stable for about 4 minutes which defined the window during which measurements were taken. Sensitivity was shown to be dependent on the level of expression of luciferase by the cells. Injection of commercially available HCT116Luc cells, where the luciferase gene was inserted by a lentiviral system, was shown to allow detection of 10,000 cells in the lungs of mice. This sensitivity was about 10 fold greater than was obtained by lipofectamine based gene transfection. When HCT116Luc cells were grown as subcutaneous xenografts in mice, an exponential growth pattern was easily detected by bioluminescence imaging and the reproducibility between mice was comparable to that routinely obtained by calliper measurements. Activity of encapsulated cisplatin was determined in a model of disseminated ovarian cancer. Rab25, a member of the RAS oncoprotein superfamily, is up-regulated in around 80% of ovarian cancer samples compared to normal ovarian epithelium. Rab25 contributes to tumour progression by enabling the tumour cells to invade the extracellular matrix by altering the trafficking of integrin. Transfection of Rab25 into A2780 cells results in cells that can grow in the peritoneal cavity of mice. A2780-Rab25 cells were 4 fold resistant to cisplatin in vitro which confirms a previous observation that Rab25 expression in A2780 makes them less sensitive to the induction of apoptosis in response to stress. A2780-Rab25 cells that express the luciferase gene (A2780-Rab25Luc) were injected into the peritoneal cavity of mice and growth was measured by biophotonic imaging. Exponential growth was clearly apparent at a stage at which no obvious abdominal distension was apparent. The disseminated A2780-Rab25Luc tumour xenografts were less sensitive to cisplatin than are subcutaneous xenografts of A2780. This is the first study that suggests that Rab25 over-expression results in reduced drug sensitivity in vivo. In contrast, a very significant growth inhibition was observed when mice were treated with an equivalent dose of encapsulated cisplatin regardless of whether it was administered by the intraperitoneal or subcutaneous route. These results are very encouraging since they confirm the enhanced activity of encapsulated cisplatin and also demonstrate the value of biophotonic imaging for measurement of tumour growth in vivo. Pharmacodynamic measures of drug activity in vivo in animal models are often based either on measures of surrogate tissue response or on single measures on tumour tissue removed at the end of the experiment. Biophotonic imaging in vivo allows the translation of reporter assays used in cell lines in vitro to studies of tumour response in vivo. A plasmid was prepared that links the p53 transcriptional response element to the luciferase gene and it was then transfected in to A2780 cells which express wild type p53. Stable transfectants of A2780p53Luc were treated with cisplatin, doxorubicin and paclitaxel and induction of p53 determined by bioluminescence and confirmed by Western blotting. A very low bioluminescent signal was present in untreated cells and a clear dose dependent increase in bioluminescence was seen in response to all three drugs. When A2780p53Luc cells were grown as subcutaneous xenografts the bioluminescent signal was significant in untreated tumours but was markedly increased 24 hours after treatment of the mice with cisplatin. Induction of p53 in the tumours was confirmed by immunohistochemistry and this also confirmed significant expression of p53 in untreated tumours. The possible implications of these findings for the improved delivery of cisplatin are discussed.

Raman spectroscopy is a powerful tool in oncological imaging. Optical biopsies in which an accurate diagnosis of the tumour areas is spectroscopically performed are especially interesting for application to skin cancer treatments. In the first part of this dissertation a study on automated Raman spectral imaging allowed accurate diagnosis and delineation of the borders of a common type of skin cancer, basal cell carcinoma (BCC). Automated detection and imaging of BCC in skin sections excised during surgery was performed by combining Raman micro-spectroscopy with supervised multivariate mathematical algorithms based on linear discriminant analysis (LDA). The model allowed 90±9% sensitivity and 85±9% specificity in BCC detection. Raman spectral images based on the LDA model were created and compared with the gold-standard of the conventional histopathological diagnoses resulting in excellent agreement. Additional studies on the ability of the model in discriminating between BCC and hair follicles produced accurate diagnoses. In this thesis instrumental implementation and design of a Raman spectral imaging prototype aiming to reduce the acquisition time required to build the Raman spectral images was developed. High sensitivity variants of Raman spectroscopy such as surface enhanced Raman spectroscopy (SERS) are known to enable optical detection down to single molecules and can be applied to thin supported lipid research. The combination of SERS with a complementary topographic technique simultaneously synchronised adds to the chemical information the morphology of the sample surface. In the second part of this thesis simultaneous atomic force microscopy (AFM) and SERS characterisation of thin (≈15-20 nm) supported films of arachidic acid and cationic phospholipids on sapphire/silver substrates was successfully achieved. Supports were fabricated with nanosphere lithographic procedures and allowed enhancement of the weak Raman signals from the amphiphilic films by a maximum factor of ×10^8.

Protein tyrosine phosphatases (PTPs) have been identified that mediate a range of physiological and pathological processes, such as proliferation and tumour metastasis. PTPRK and PTPRM belong to the same subfamily of PTPs. This study aims to investigate the role of PTPRK and PTPRM in cancer development and progression. Knockdown of PTPRK expression was performed in PC-3 and DU-145 cells. Functional assays were then carried out on these cells in order to determine any changes in their biological properties. Knockdown of PTPRK significantly reduced the growth and adhesion of both PC-3 and DU-145 cells. The experimental results suggested that reduction of cell growth is potential involvement of p53 and/or caspase-3 and -8 and its up-stream molecule JNK. The decreased expression of PTPRK and PTPRM are associated with poor prognosis and reduced survival. Knockdown of PTPRK resulted in increased adhesive and invasive abilities, and promoted cell proliferation and motility of breast cancer cells. Moreover, PTPRM knockdown resulted in elevated adhesion, invasion, and proliferation of breast cancer cells. Activation of ERK and JNK by tyrosine phosphorylation and consequent elevated MMP9 activity is involved in increased cell migration and invasion by PTPRM knockdown. These results suggested that PTPRK and PTPRM are involved in the disease progression of prostate and breast cancer by regulating a complex network of pathways and molecules. This provides further proof of the importance of the R2B subfamily, a subgroup of PTP superfamily, in cancer. In addition, it sheds some light on the use of PTPs as prognostic indicators of disease, aiding in diagnosis and treatment. The major effect is the promotion of motility and invasiveness of cancer cells via ERK and JNK pathways. However it can also impair the apoptosis mediated by JNK pathways in certain cancer cells, such as prostate cancer cells. Such contrasting effects on survival and motility require further investigation, and should also be considered when treating cancers by targeting these molecules.

CLL is an incurable B-cell malignancy associated with profound tumour cell-mediated immune dysfunction. It therefore represents a challenging disease for the successful application of immunotherapeutic strategies aimed at promoting anti-tumour T-cell responses. In this study, extensive immunophenotypic analysis of T-cells from the blood of CLL patients was performed, in order to better characterise their dysfunctional status within the disease. Analysis of CLL patient blood samples revealed a skewing of T-cells towards a highly differentiated effector memory phenotype as well as the expression of markers associated with exhaustion/senescence (CD28- and CD57+) and immunosuppressive molecules (PD-1 and CD200). In addition this study revealed the expansion of CD8+ T-cells in a subset of CLL patients leading to an inversion in the normal CD4:CD8 ratio. The presence of an inverted CD4:CD8 ratio was subsequently shown to be associated with a shorter time to first treatment and reduced progression-free survival. Characterisation of T-cells identified several molecules that could be targeted therapeutically in order to break T-cell tolerance in CLL patients and potentially restore normal immune responses. Investigation of the immunosuppressive molecules PD-1 and CD200 showed that they are over expressed in CLL patients, suggesting that they may be involved in maintaining T-cell tolerance in the disease. However, blockade of PD-1-PDL-1 and CD200-CD200R signalling pathways failed to enhance T-cell responses from CLL patients in vitro. Investigation of an alternative approach to enhance T-cell responses in CLL involved the use of a bi-specific antibody targeting CD19 and CD3 called blinatumomab. In vitro testing showed that blinatumomab can induce T-cell activation, promoting the release of pro-inflammatory cytokines and granzyme B secretion from both CD4+ and CD8+ T-cells. In addition, blinatumomab was shown to mediate T-cell dependent killing of CLL cells requiring the formation of T-cell:CLL cell conjugates. Finally this study provided clear evidence that blinatumomab can break T-cell tolerance in CLL and strongly advocates the progression of blinatumomab into clinical trials as a novel therapeutic agent in CLL.

CD8+ T-cells recognise pathogens and cancer through a specific interaction between the T-cell receptor (TCR) and a 8-14 amino-acid residue peptide presented by class I major histocompatibility complex (pMHCI) molecules expressed on the target cell surface. The first structures of murine and human TCR/pMHC complexes, published in 1996, revealed a number of important features of the TCR/pMHC interface. Currently, <25 unique human TCR/pMHC complexes are reported in the literature. This is a relatively low number compared with the number of antibody or unligated pMHC structures. The lack of structural information regarding human TCR/pMHC complexes has compromised the determination of a comprehensive and accepted set of rules that govern T-cell antigen recognition. Difficulties in generating TCR/pMHC complex crystals partly explain the low number of these structures. The first part of this thesis reports the development of a new crystallization screen (TOPS) designed specifically for the generation of such protein crystals. I also had access to MART-1-specific TCRs, the MART-1 protein being expressed by virtually all fresh melanoma tumour specimens. Different human leukocyte antigen (HLA)- A*0201-restricted peptides from this protein are presented at the melanoma cell surface. As TCRs are known to bind to cancer-derived “self” peptides with weak affinity, there is considerable interest in designing enhanced affinity TCRs for the recognition of HLA-A*0201-MART-1. My work concentrated on the MART-1- specific TCR MEL5 and its affinity-enhanced variant selected by phage display, α24β17. I analysed the biophysical properties of α24β17 and determined that it bound HLA-A*0201-MART-1 with >30,000-fold enhanced affinity and distinct thermodynamics. Comparison of TCR/HLA-A*0201-MART-1 complex structures solved with TOPS and binding biophysics showed that: (i) TCR affinity can be enhanced by increasing interactions between the TCR and the MHC surface; (ii) soluble α24β17 retains the peptide specificity by a novel mechanism involving interactions with solvent molecules; and, (iii) MEL5 interaction with the physiologically relevant MART-127-35 nonameric antigen led to a peptide anchor residue switch, a TCR-induced modification that has never been observed before. I also initiated a preliminary study on the generation of genetically modified Jurkat cells and CD8+ T-cells expressing a range of affinity-enhanced TCRs directed against melanoma for adoptive cell therapy. These results suggested that melanoma specificity is retained after MEL5 transduction and that there is no need to optimize beyond a TCR affinity threshold to obtain optimal T-cell activation. Collectively, these data shed light on the complex and unpredictable nature of T-cell antigen recognition.

Background: Traditionally, coping and decision making have been viewed as separate concepts. However, analysis of the role of emotions during decision making in healthcare suggests that coping with health threats, and associated emotions, should be viewed as an integral part of deliberation processes. This thesis reports on the development of a framework that merges deliberation and coping processes. Subsequently, this framework is operationalised by adapting it to specifically describe deliberations about risk-reducing salpingo-oophorectomy (RRSO) by women at increased genetic risk of ovarian cancer. Methods: A narrative review of the literature on decision making and coping theories informed the development of a novel framework that integrates theories from both fields. A multi-methods approach, which included a systematic literature search and qualitative methods, examined women’s decision making about RRSO and informed adaptation of the framework to specifically describe such decisions. The adapted framework was then used to develop a decision support intervention for women at increased risk of ovarian cancer in the UK. Incremental prototypes of the intervention were reviewed by a group of stakeholders and usability of the final prototype was tested using cognitive interviews. Results: The Coping in Deliberation (CODE) framework describes deliberations as multi-step appraisal and coping processes. The framework was successfully adapted to decisions about RRSO and used to develop a two part patient decision support intervention, consisting of a brief paper-based tool (Option Grid) and a longer website (the Oophorectomy Decision Explorer, OvDex). Conclusions: Emotions and coping are integral parts of deliberations in healthcare and the CODE framework, which acknowledges the importance of these concepts during deliberations, can be used to guide the development of patient decision support interventions. Future research should apply this framework to other healthcare decisions. The intervention developed in this thesis requires field testing to assess its impact before implementation in clinical practice.

Despite recent advances in the clinic to integrate novel targeted therapeutic agents into standard therapy, colorectal cancer (CRC) remains a significant cause of mortality. The high attrition rate of novel compounds at phase III clinical trials for CRC, has been attributed to limited information from pre-clinical strategies, in particular, the use of inadequate xenograft models. In response, this thesis aimed to utilise robust and relevant genetically engineered mouse models of CRC to evaluate a number of novel therapeutic strategies. Whilst mutations in the tumour suppressor APC are crucial for initiation of CRC, mutations which lead to activation of the PI3K and MAPK signalling pathways, such as through loss of the tumour suppressor protein PTEN and activation of oncogenic KRAS, have been implicated in promoting progression of CRC. As such, combinations of these genetic alterations within the murine intestine, using the Cre-LoxP system, lead to differing mouse models of invasive intestinal adenocarcinoma. This thesis reports therapeutic targeting of the PI3K and MAPK signalling pathways using the dual PI3K and mTOR inhibitor NVP-BEZ235, and the MEK inhibitor MEK162, respectively. For this, compounds were initially evaluated for pharmacodynamic and anti-tumour effects through short term exposure experiments. These analyses yielded a range of effects, some of which appeared predictive of long term efficacy, others which were contradictory and some which revealed novel feedback mechanisms. Furthermore, these agents were assessed in a long term therapeutic setting to evaluate the effect of continuous treatment on longevity and tumour burden of tumour models. Here, whilst dual PI3K/mTOR inhibition significantly increased longevity of all mouse models, MEK inhibition was only effective in the Apc and Apc Kras mutant settings, identifying Pten loss as a marker of non-response to MEK inhibition, independently and also in the Kras mutant setting. Furthermore, analysis of the combination therapy in short term settings identified scheduling of these agents to be key to achieve concomitant pathway inhibition, particularly in the Apc Pten deficient tumour setting. Ultimately, when evaluated in a long term setting, although the combination therapy displayed no further benefit in the Pten deficient setting, this had additive benefits in the Kras mutant setting and synergistic benefits in the Pten Kras mutant setting. Nevertheless despite the promise of targeted therapy, standard chemotherapeutic agents such as 5-flurouracil (5-FU) remain the backbone of therapy for CRC, regardless of only moderate benefits of 10-15% in advanced tumour settings. Furthermore, resistance to 5-FU predominantly through upregulation of the enzyme thymidylate synthase (TS) frequently 2 occurs in human tumours. Investigations reported here aimed to target tumours with upregulated TS using novel analogues of the anti-viral agent Brivudin (BVDU), metabolites of which are converted to anti-cancer metabolites by TS. Initially, In vitro characterisation of a small library of compounds reported here identified a number of potent compounds. Following further in vitro characterisation and short term evaluation in the Apc Pten tumour model of invasive adenocarcinoma, two lead compounds: CPF472 and CPF3172 were taken forward for long term experiments. Subsequently, this study evaluated and identified compounds which showed increased efficacy in the TS upregulated setting. Taken together, the investigations presented in this thesis highlight the utility of appropriate mouse models in evaluating novel therapeutic strategies and generating clinically relevant hypotheses.

The Myelodysplastic syndromes (MDSs) are comprised of a heterogeneous group of clonal disorders characterised by ineffective haematopoiesis. Although 30 to 35% of MDS cases progress to Acute Myeloid Leukaemia (AML), the majority of patients die from blood related ailments caused by progressive bone marrow failure. Large-scale genomic rearrangements are a key feature of MDS, with different aberrations conferring specific risks of progression. Telomere erosion, dysfunction and fusion, creating cycles of anaphase bridging breakage and fusion is a mechanism that has the potential to drive genomic instability in many tumour types including MDS. The key aim of this project was to examine the role that telomere dysfunction may play in the generation of genomic rearrangements observed in MDS/AML. High resolution Single Telomere Length Analysis (STELA) revealed telomere shortening when compared to age-matched individuals in two cohorts of MDS and AML patients; this included large-scale telomeric deletion events observed within the MDS cohort. A PCR based telomere fusion assay detected telomere-telomere fusion events at a frequency that was consistent with sporadic fusion arising as a consequence of large-scale deletion. Telomerase activity was up-regulated in AML which may contribute to the reduction of deletion and fusion events in these cells. Sequence analysis revealed that telomere fusion was associated with microhomology and sub-telomeric deletion; this profile was consistent with error-prone Ku-independent alternative end joining processes. Telomere length at diagnosis irrespective of conventional markers appeared to influence the overall survival of MDS patients, but this was not apparent in AML. More importantly, telomere length was able to refine favourable prognostic markers, specifically good risk cytogenetics, uni-lineage cytopenia and low-risk IPSS (International Prognostic Scoring System) scores of which MDS patients bearing shorter telomeres for their respective age displayed reduced overall survival. This may be a particularly important finding given the heterogeneous clinical outcomes observed within low-risk MDS patients.

Kiss-1 and its receptor (Kiss-1R) are suggested as a novel pair of metastasis suppressors for several human solid tumours. However, the role of Kiss-1 and Kiss-1R in colorectal cancer remains largely unknown. Therefore, the aim of this study was to investigate the role and signal transduction of Kiss-1 and its receptor in colorectal cancer. Colorectal cancer cell lines (HT115, HRT18, RKO and Caco-2) were screened for the mRNA expression levels of Kiss-1 and Kiss-1R. Sublines of cancer cells with differential expression of Kiss-1 and Kiss-1R were created, using ribozyme transgenes to knock down the expression of Kiss-1 and Kiss-1R, respectively. The stabilized transfected cells were used to study the influence of Kiss-1 and Kiss-1R on the function of colorectal cancer cells using by in vitro function assays (growth, adhesion, wounding and invasion assays) and ECIS assay. The influence of Kiss-1 on tumour growth was also tested using an in vivo tumour model. To further explore the receptor activation and signalling pathways downstream of Kiss-1, Kisspeptin-10 was also used in HT115 Kiss-1 knockdown cells. Phosphorylation of Kiss-1 and the effect of Kissthe effect on MMP 1 on MMP 1 on MMP-9 and MMP and MMP -2 were detected using immunoprecipitation and immunoprecipitation and zymographyzymography respectively. The study also investigated Kiss-1 and Kiss-1R expression and their correlation to the clinical outcome in human colorectal cancer, using real-time PCR and immunohistochemistry. Kiss-1 and Kiss-1R played a suppressive role in the invasion and migration of colorectal cancer cells, in that knocking down both Kiss-1 resulted in increased cell-matrix invasion and cellular migration as demonstrated by a series of cell models. Exogenous Kiss-1 (Kisspeptin-10) decreased cellular migration of colorectal cancer cells and required ERK signalling as shown during the ECIS based analyses. The inhibitory influence of Kiss-1 on the motility of cancer cells was via the reduction of MMP-9, shown by zymography. In the in vivo tumour model, tumour growth rate of Kiss-1 knockdown colorectal cancer cells was significantly faster than the control cells. In human colorectal cancer tissues, levels of message expression of Kiss-1 had a negative correlation with Dukes staging, TNM staging, tumour size and lymph node involvement. Kiss -1R expression was significantly decreased in tumour tissues compared with adjacent normal tissues. The present study has demonstrated that Kiss-1 and Kiss-1R play a pivotal tumour suppressor role in colorectal cancer. The inhibitory effect on cancer cells involves the regulation of MMPs and the ERK signaling pathway. The molecule pair is candidate prognostic indicator in patients with colorectal cancer.