Biotechnological inventions during the 20th century have resulted in a wide range of approaches for explorations in the functional genomics field. Microarray technology is one of the recent advances which have provided us with snapshots of which genes are expressed in cells of various tissues and diseases. Methods to obtain reliable microarray data are continuously being developed and improved to meet the demands of biological researchers. In this thesis microarrays have been used to investigate gene expression patterns in cancer research. Four studies in three different areas were carried out covering adrenocortical tumors, p53 target genes and a comparison of RNA amplification methods. Adrenocortical tumours are among the most common tumours with an incidence of 7-9%. Malignancy of these tumors is rare. Distinction between malignant and benign tumours is often difficult to establish which makes an improvement of diagnostic approaches important. To elucidate biological processes in adrenocortical tumour development and to examine if there is a molecular signature associated with malignancy, microarray analysis was performed on 29 adrenocortical tumors and four normal specimens. It was possible to classify malignant and benign samples based on the entire expression profile. A number of potential biomarkers was identified which will be further evaluated. P53 is a gene which is mutated in 50% of all cancers. Functional p53 is a transcription factor which is activated upon cellular stress and DNA damage. Target genes are mainly involved in cell cycle arrest and apoptosis. In solid tumors cells are stressed by hypoxia. To examine which target genes p53 activate under hypoxic conditions a microarray study of the cell lines HCT116p53+/+ and HCT116p53-/- was performed. A set of novel potential p53 target genes was identified while many known target genes were found to be not transcriptionally activated during hypoxia. Follow up which was focused on how p53 affected hypoxia induced apoptosis showed that the death receptor Fas was critical. When small amounts of tissue are available, amplification of the transcript population is necessary for microarray analysis. A new strategy for amplification based on PCR was evaluated and compared to a commercial in vitro transcription protocol. Both protocols produced reliable results. Advantages with the PCR based method are a lower cost and a high flexibility due to compatibility with both sense and antisense strand microarrays. Keywords: adrenocortical tumour, apoptosis, cancer, classification, gene expression, microarray, p53, RNA amplification / QC 20100907

In addition to malignant cells, solid tumors comprise supporting stromal tissue that consists of extra cellular matrix (ECM), connective tissue cells, inflammatory cells and blood vessels. The stromal compartment and the malignant cells together shape the tumor microenvironment that in turn determines tumor progression and efficacy of anti-tumor treatments. In this thesis, studies that investigate the roles of different kinds of interactions between tumor cells and stromal cells were undertaken. Further, growth factors that have important roles in interactions between tumor cells and stromal cells were investigated in a non-tumor environment. Tumor cells were found to modulate the response to the platelet derived growth factor  (PDGF) by microvascular pericytes, a cell type found in the vasculature of solid tumors. The importance of this growth factor in biology of tumors has earlier been shown, but here it was shown that PDGF also modulate the ECM phenotype of solid tumors. The ECM of tumors treated with an inhibitor of PDGF receptor (PDGFR) signaling induced a less fibrotic collagen scaffold, which could explain how PDGFR inhibition in earlier reports lowered tumor interstitial fluid pressure (IFP). Lowering the normally high IFP in tumors increases efficacy of chemotherapy. The integrin αVβ3 is activated downstream of PDGF-B in acute inflammations, and this integrin is important for raising IFP in loose connective tissue in such conditions. However, in tumors we found that lack of the β3 subunit lead to an increased IFP, which were attributed to a more fibrotic ECM phenotype. In addition to PDGF-B, transforming growth factor β (TGFβ) is an important growth factor in the biology of tumors. These two growth factors were separately overexpressed in mouse skin and they both induced an inflammatory response. Expressed in a tumor free context, they evoked a response that was in many ways reminiscent of what can be observed in the tumor microenvironment. This thesis contributes further understanding of how the complex tumor microenvironment affects the phenotype of solid tumors.

Antibody-based targeting techniques play an increasingly important role in cancer research. By targeting a structure that is abundant in tumour cells, but rare in healthy tissues, an antibody can mediate the delivery of radioactivity specifically to tumour cells in the body. This idea is particularly appealing for head and neck squamous cell carcinoma (HNSCC), as the advanced stages have a large fraction of spread disease that is difficult to treat with procedures available today. In this thesis, we have investigated possible radioimmunotargeting structures for HNSCC, and found that CD44v6 is a suitable target for antibody-based radiotherapy and diagnostics in this patient group. We have identified radiohalogens as attractive nuclides for such use, and have investigated the possibility of radiohalogenating the anti CD44v6 chimeric monoclonal antibody (cMAb) U36. Several feasible labelling methods were identified, using both direct and indirect labelling. The cMAb U36 was then successfully labelled with 211At and 131I, and preclinically evaluated for therapeutic use. Results proved the astatinated conjugate to be most efficient in this context, demonstrating a specific and dose-dependent cytotoxicity. The cMAb U36 was then evaluated for diagnostic use in thyroid anaplastic carcinoma, using 124I as the diagnostic nuclide. Results in tumour-bearing mice were promising, with all of the tumours identified in micro-PET studies. These results demonstrate how antibody-based radionuclide targeting can provide more sensitive and specific methods for identifying and treating head and neck cancer, and hopefully help improve long-term survival rates for this patient group in the future.

Targeted radiotherapy (TRT) of cancer is a promising approach that enables selective treatment of tumour cells, while sparing normal tissue. The humanized monoclonal antibody A33 (huA33) is a potential targeting agent for TRT of colorectal cancer, since its antigen is expressed in more than 95 % of all colorectal carcinomas. The aim of this thesis was to evaluate the therapeutic potential of the two huA33-based TRT-conjugates, 177Lu-huA33, and 211At-huA33. The conjugates 177Lu-huA33, and 211At-huA33, bound specifically to colorectal cancer cells, both in vitro and in vivo. A dose dependent cytotoxic effect of 211At-huA33 was also demonstrated in vitro. From a therapeutic perspective, both conjugates had a favourable biodistribution in tumour-bearing nude mice, with high tumour uptake and a low uptake in normal organs (with the exception of an expected thyroid uptake of 211At). After injection of 211At-huA33, the blood absorbed a slightly higher dose than the tumour, but for 177Lu-huA33, the tumour received a 12 times higher dose than blood. Two days after intravenous injection of 177Lu-huA33 in tumour-bearing mice, the tumours could be clearly visualised by gamma camera imaging, with very low interference from normal tissue radioactivity. In an experimental therapy study, also performed in tumour-bearing mice, there was an excellent therapeutic effect of 177Lu-huA33. About 50 % of the treated animals were tumour free 140 days after injection of 177Lu-huA33, while none of the non-radioactive controls survived beyond 20 days after injection of treatment substances. In conclusion, this thesis demonstrates that the therapeutic conjugates 177Lu-huA33, and 211At-huA33, are promising targeting agents that might help improve therapy of colorectal cancer.

<p>Cancer metastasis is the cause of about 90% of cancer patients' deaths. Despite significant improvements in the past three decades in understanding the molecular bases of oncogenic transformation of cancer cells, little is known about the molecular mechanisms underlying tumour cells' alteration of their microenvironment, entrance into the circulation, and colonization of distant organs. In recent years, accumulating evidence has indicated that tumour microenvironment, which consists of a variety of cell types and extracellular matrix components&#65292;plays an important role in regulating the metastatic abilities of carcinoma cells. Co-opted by cancer cells, those stromal cells promote tumour progression via multiple mechanisms, including enhancement of tumour invasiveness, elevation of angiogenesis, and suppression of immune surveillance activity. </p><p>Using a series of human breast cancer cell lines with different metastatic potentials <italic>in vivo</italic>, we performed an unbiased screen examining expression of miRNAs, and found that miR-126 and miR-126*, whose expression are regulated by methylation of the promoter of their host gene Egfl7 inside tumour cells, were significantly negatively correlated with metastatic potential. Using both mouse xenograft models and <italic>in vitro</italic> assays, we showed that this pair of miRNAs suppressed breast cancer metastasis through shaping the tumour microenvironment without changing tumour cell autonomous properties. Specifically, miR-126 and miR-126* act independently to suppress the sequential recruitment of mesenchymal stem cells (MSCs) and inflammatory monocytes into the primary tumour stroma, consequently inhibiting lung metastasis by breast tumour cells. Mechanistically, these miRNAs directly inhibit the production of stromal cell-derived factor-1 alpha (Sdf-1&alpha;, also known as Cxcl12), and indirectly suppress the expression of chemokine (C-C motif) ligand 2 (Ccl2) by the cancer cells within the tumour mass in an Sdf-1&alpha;-dependent manner. In addition, in contrast with the majority of reports which have shown incorporation of only the guiding strand of the miRNA duplex into the mRNA-targeting RNA induced silencing complex (RISC), both strands of the miR-126 RNA duplex are maintained at a similar level and suppress Sdf-1&alpha; expression independently. </p><p>Collectively, we have determined a dynamic process by which the composition of the primary tumour microenvironment could be altered via a change in the expression of two tumour-suppressive miRNAs derived from a single miRNA precursor to favor metastasis by breast cancer cells. Importantly, this work provides a prominent mechanism to explain the clinical correlation between reduced expression of miR-126/126* and poor metastasis-free survival of breast cancer patients.</p> / Dissertation

Background: The limited within-breed genetic heterogeneity and an enrichment of disease-predisposing alleles have made the dog a very suitable model for the identification of genes associated with risk for specific diseases. Canine mammary cancer is an example of such a disease. However, the underlying inherited risk factors for canine mammary tumours (CMTs) are still largely unknown. In this study, 52 single nucleotide polymorphisms (SNPs) in ten human cancer-associated genes were genotyped in two different datasets in order to identify genes/alleles associated with the development of CMTs. The first dataset consisted of English Springer Spaniel (ESS) CMT cases and controls. ESS is a dog breed known to be at increased risk of developing CMTs. In the second dataset, dogs from breeds known to have a high frequency of CMTs were compared to dogs from breeds with a lower occurrence of these tumours. Results: We found significant associations to CMT for SNPs and haplotypes in the estrogen receptor 1 (ESR1) gene in the ESS material (best P-Bonf = 0.021). A large number of SNPs, among them several SNPs in ESR1, showed significantly different allele frequencies between the high and low risk breed groups (best P-Bonf = 8.8E-32, best P-BPerm = 0.076). Conclusions: The identification of CMT-associated SNPs in ESR1 in two independent datasets suggests that this gene might be involved in CMT development. These findings also support that CMT may serve as a good model for human breast cancer research.

Background: Colorectal cancer (CRC) remains one of the leading causes of cancer-related mortality. About 50 % of patients with CRC will develop subsequent liver metastases (CLM). The survival for untreated CLM is only a few months and liver resection provides the only chance for a lasting cure. It is therefore essential to detect CLM early, enabling successful surgical resection and achieving a long-term cure. There are no optimal tumour markers for CRC or CLM. The best marker available is Carcinoembryonic Antigen (CEA), a marker found elevated in about 50-60% of patients with CLM, but also in many other conditions. The main focus of cancer research has been on the malignant cancer cell. However, a tumour consists of more than cancer cells. A major part of all solid tumours is made up by the stroma. The tumour stroma is defined as the non-malignant cells of a tumour such as fibroblasts, the cells of the vascular and immune systems as well as the extracellular matrix (ECM). The basement membrane (BM) is a specialized form of the ECM in which type IV collagen is the major protein component. All epithelial cells need a contact to the BM and the definition of an invasive cancer is the degradation of the BM and the spread of cancer cells beyond this structure. Different metastatic growth patterns of CLM have previously been described, namely the desmoplastic, pushing and replacement type of CLM. These differ in their stromal reaction in the border, which separates the tumour from the normal liver. In this thesis the tumour stroma of CRC and CLM is studied with a special emphasis on stromal collagens. The aim is to investigate whether stromal collagens/ circulating type IV collagen can be used as tumour markers for CRC and CLM, and to compare this to the conventional marker CEA. The circulating type IV collagen level is also measured in liver metastases from other primary tumours than CRC. Furthermore, the differences between the stroma of a primary CRC that metastasizes to the liver when compared to a CRC that never spreads are analysed. Additionally, the metastatic growth pattern of CLM is studied in relation to the primary tumour, stromal components and survival. We also sought out to find whether CRC cell lines possess the trait to produce ECM proteins endogenously, and in response to a normal liver stroma in a novel organotypic model for CLM. Methods: Expression patterns of type I, III and IV collagen were studied by immunofluorescence (IF), chemical staining and immunohistochemistry (IHC) in normal colorectal tissue, normal liver, CRC, CLM, benign liver lesions and in liver metastases of other origin than CRC. Circulating plasma levels of type IV collagen were analysed in healthy controls, patients with CRC (T stage I-III) and in patients with CLM. Samples were analysed at the time of diagnosis, during and after oncological and surgical treatment and at the time of relapsing or progressive disease. Additionally, circulating levels were analysed in patients with benign liver lesions and in liver metastases of other origin than CRC. The metastatic growth pattern of CLM was classified according to earlier descriptions. CRC cell lines were studied regarding their production of type IV collagen. The growth, invasiveness and stromal production in CRC cell lines were also investigated in a new organotypic model for CLM using human liver specimens. Results: Circulating type IV collagen levels are increased in patients with CLM and other epithelial-derived liver metastases, and is found normal in patients with primary CRC (stage I-III), with liver metastases from tumours of non-epithelial origin, benign liver lesions and in healthy controls. The type IV collagen levels in patients with CLM reflect the tumour burden in the liver, decreases in response to therapy and is found increased in progressive or relapsing disease. The combination of circulating type IV collagen and CEA increased the sensitivity and specificity for detecting CLM. Livermetastatic CRC displayed an increased stromal production when compared to non-metastatic CRC, with an increased type IV collagen expression in the direct vicinity of the CRC cells. The earlier described growth patterns of CLM were verified, with the pushing type of CLM associated with a short survival and poor outcome. Furthermore, CRC cell lines possess the trait of endogenously producing type IV collagen. The novel organotypic liver model revealed that CRC cell lines grown in the context of normal liver stroma, devoid of other cells, does not elicit a desmoplastic reaction. Conclusion: Circulating type IV collagen is a promising tumour marker for CLM, where the levels reflect the hepatic tumour burden and can detect disease relapse after liver surgery. The combination of the tumour markers CEA and type IV collagen is superior to CEA alone. The stromal composition of primary CRC predicts the risk of subsequent CLM and the metastatic growth pattern of CLM is related to survival.

In an attempt to identify agents that specifically target neuroblastoma (NB) tumour-initiating cells (TIC) we performed drug screens using libraries of bioactive compounds. Cardiac glycosides (CGs) were the largest class of drugs identified with antitumour activity. At high CG doses inhibitory effects on the Na+/K+-ATPase induce cardiotoxicity; therefore, CG analogues were designed in an attempt to separate the effects on NB cells from cardiotoxicity. We identified RIDK34 as our lead compound from a structure-activity-relationship analysis (IC50 8 nM). RIDK34 contains a unique oxime group and shows increasing potency against NB TICs. The Na+/K+-ATPase is a target for the apoptotic activity of digoxin and RIDK34, whereby a signaling cascade involving Src and ERK may induce apoptosis. Furthermore, we predict that signaling activation does not require inactivation of the Na+/K+-ATPase and subsequent deregulation of [Na+]i and [K+]I gradients. Thus CGs and particularly RIDK34 may be expected to display diminished cardiotoxicity and greater therapeutic potential.

Clear cell renal cell carcinoma (ccRCC) is an aggressive malignancy characterized by inactivation of the von Hippel-Lindau tumour suppressor gene, the protein product of which mediates degradation of the transcription factor hypoxia-inducible factor (HIF). GAL3ST1 is a sulfotransferase which catalyzes the production of sulfatide, a plasma membrane sulfolipid previously implicated in metastasis. We observed GAL3ST1 overexpression in primary ccRCC tumours relative to matched-normal tissue and subsequently asked if GAL3ST1 was a HIF-responsive gene that facilitates ccRCC metastasis. GAL3ST1 expression was suppressed in ccRCC cells by stable reconstitution of wild-type VHL and also siRNA-mediated knockdown of HIF1alpha and HIF2alpha. Dual luciferase assays and chromatin immunoprecipitation revealed a hypoxia-response element in the GAL3ST1 5’-UTR that appeared to be crucial for HIF-mediated upregulation. Finally, stable knockdown of GAL3ST1 significantly impeded ccRCC cell invasion through an in vitro basement membrane mimic. These results suggest GAL3ST1 is a HIF-responsive gene that promotes tumour cell invasion.

In an attempt to identify agents that specifically target neuroblastoma (NB) tumour-initiating cells (TIC) we performed drug screens using libraries of bioactive compounds. Cardiac glycosides (CGs) were the largest class of drugs identified with antitumour activity. At high CG doses inhibitory effects on the Na+/K+-ATPase induce cardiotoxicity; therefore, CG analogues were designed in an attempt to separate the effects on NB cells from cardiotoxicity. We identified RIDK34 as our lead compound from a structure-activity-relationship analysis (IC50 8 nM). RIDK34 contains a unique oxime group and shows increasing potency against NB TICs. The Na+/K+-ATPase is a target for the apoptotic activity of digoxin and RIDK34, whereby a signaling cascade involving Src and ERK may induce apoptosis. Furthermore, we predict that signaling activation does not require inactivation of the Na+/K+-ATPase and subsequent deregulation of [Na+]i and [K+]I gradients. Thus CGs and particularly RIDK34 may be expected to display diminished cardiotoxicity and greater therapeutic potential.