Dioxins are environmental contaminants that raise concern because they are potent and persistent. The most potent dioxin congener, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), causes a wide variety of biochemical and toxic effects in laboratory animals and in humans. Major toxicities of TCDD are initiated by their binding to the AH receptor (AHR), a ligand-activated transcription factor that regulates expression of numerous genes. However, the specific genes whose dysregulation leads to major toxicities such as wasting, hepatotoxicity, and lethality are unknown. The objective of this thesis research was to identify the molecular mechanisms by which dioxins cause lethality. To this end, a powerful genetic rat model was utilized – the Han/Wistar (Kuopio) rat which is highly resistant to dioxin toxicity due to a major deletion in the AHR’s transactivation domain (TAD) leading to 3 potential AHR variant transcripts. We found that insertion-variant transcripts (IVs) are the dominant forms of AHR expressed in H/W rats, constitutively and after TCDD treatment. Gene expression array analysis revealed that the total number of TCDD-responsive genes in liver was significantly lower in H/W rats (that carry the TAD deletion) than in dioxin-sensitive rats (that carry wildtype AHR). Genes that are well-known to be AHR-regulated and dioxin-inducible  such as CYP1 transcripts  remained responsive to TCDD in H/W rats; thus the TAD deletion selectively interferes with expression of a subset of hepatic genes rather than abolishing global AHR-mediated responses. Genes that differed in response to TCDD between dioxin-sensitive rats and dioxin-resistant rats are integral parts of pathways known to be disrupted by dioxin treatment such as protein synthesis/degradation, fatty acid transport/metabolism, and apoptosis. These genes are worthy candidates for further mechanistic studies to test their role in major dioxin toxicities. Numerous differentially-regulated genes were downregulated; however, microRNAs, which downregulate mRNA levels in other systems, likely play no role in downregulation of mRNAs by dioxins in adult liver and are unlikely to be involved in hepatotoxicity. Findings in this research support the hypothesis that H/W rats are resistant to TCDD lethality because the TAD deletion prevents the AHR from dysregulating specific mRNA transcripts but not hepatic miRNAs.

Dioxins are environmental contaminants that raise concern because they are potent and persistent. The most potent dioxin congener, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), causes a wide variety of biochemical and toxic effects in laboratory animals and in humans. Major toxicities of TCDD are initiated by their binding to the AH receptor (AHR), a ligand-activated transcription factor that regulates expression of numerous genes. However, the specific genes whose dysregulation leads to major toxicities such as wasting, hepatotoxicity, and lethality are unknown. The objective of this thesis research was to identify the molecular mechanisms by which dioxins cause lethality. To this end, a powerful genetic rat model was utilized – the Han/Wistar (Kuopio) rat which is highly resistant to dioxin toxicity due to a major deletion in the AHR’s transactivation domain (TAD) leading to 3 potential AHR variant transcripts. We found that insertion-variant transcripts (IVs) are the dominant forms of AHR expressed in H/W rats, constitutively and after TCDD treatment. Gene expression array analysis revealed that the total number of TCDD-responsive genes in liver was significantly lower in H/W rats (that carry the TAD deletion) than in dioxin-sensitive rats (that carry wildtype AHR). Genes that are well-known to be AHR-regulated and dioxin-inducible  such as CYP1 transcripts  remained responsive to TCDD in H/W rats; thus the TAD deletion selectively interferes with expression of a subset of hepatic genes rather than abolishing global AHR-mediated responses. Genes that differed in response to TCDD between dioxin-sensitive rats and dioxin-resistant rats are integral parts of pathways known to be disrupted by dioxin treatment such as protein synthesis/degradation, fatty acid transport/metabolism, and apoptosis. These genes are worthy candidates for further mechanistic studies to test their role in major dioxin toxicities. Numerous differentially-regulated genes were downregulated; however, microRNAs, which downregulate mRNA levels in other systems, likely play no role in downregulation of mRNAs by dioxins in adult liver and are unlikely to be involved in hepatotoxicity. Findings in this research support the hypothesis that H/W rats are resistant to TCDD lethality because the TAD deletion prevents the AHR from dysregulating specific mRNA transcripts but not hepatic miRNAs.

The hunt for biomarkers for anti-VEGF agent bevacizumab is ongoing since last decade with no success. Identifying robust biomarkers for stratifying patients and for monitoring response is important for the future use of bevacizumab in breast cancer. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) analysis and genome wide gene expression analysis are two promising approaches to understand the molecular mechanisms and search for biomarker of anti-angiogenic therapy. Firstly, with the retrospective pilot study, a close link between DCE-MRI findings and the molecular mechanisms assisting cancer survival and metastasis was established. Secondly, the prospective window of opportunity study conducted using single cycle of bevacizumab given before neoadjuvant chemotherapy and by performing detailed pharmacodynamic analyses with DCE-MRI and gene expression before and two weeks after bevacizumab had shown a wide variation in responses to bevacizmab both at genomic and imaging level. A close link between changes in DCE-MRI and the changes in gene expression profile was further established suggesting DCE-MRI has potential to serve as non-invasive biomarkers of antiangiogenic therapy. Tumours with high baseline values of forward transfer constant K^trans showed the maximum response as assessed by DCE-MRI after bevacizumab. By performing biopsy after single cycle of bevacizumab, the changes in genes related to immune response, metabolism and cell signalling were observed that gives a useful insight into mechanisms governing response and resistance to bevacizumab. Also the certain gene expression changes observed with post bevacizumab biopsies, such as down regulation of endothelial cell specific molecule-1 (ESM1), cyclin E1 (CCNE1) and up regulation of pyruvate dehydrogenase kinase 1 (PDK1), cyclic GMP-inhibited phosphodiesterase B (PDE3B) could be helpful in decision-making about future therapy with bevacizumab at an early stage. This study has suggested that using bevacizumab in combination with other targeted agents could overcome resistance.

>Magister Scientiae - MSc / A large volume of gene expression data exists in public repositories like the NCBI’s Gene Expression Omnibus (GEO) and the EBI’s ArrayExpress and a significant opportunity to re-use data in various combinations for novel in-silico analyses that would otherwise be too costly to perform or for which the equivalent sample numbers would be difficult to collects exists. For example, combining and re-analysing large numbers of data sets from the same cancer type would increase statistical power, while the effects of individual study-specific variability is weakened, which would result in more reliable gene expression signatures. Similarly, as the number of normal control samples associated with various cancer datasets are often limiting, datasets can be combined to establish a reliable baseline for accurate differential expression analysis. However, combining different microarray studies is hampered by the fact that different studies use different analysis techniques, microarray platforms and experimental protocols. We have developed and optimised a method which transforms gene expression measurements from continuous to discrete data points by grouping similarly expressed genes into quantiles on a per-sample basis. After cross mapping each probe on each chip to the gene it represents, thereby enabling us to integrate experiments based on genes they have in common across different platforms. We optimised the quantile discretization method on previously published prostate cancer datasets produced on two different array technologies and then applied it to a larger breast cancer dataset of 411 samples from 8 microarray platforms. Statistical analysis of the breast cancer datasets identified 1371 differentially expressed genes. Cluster, gene set enrichment and pathway analysis identified functional groups that were previously described in breast cancer and we also identified a novel module of genes encoding ribosomal proteins that have not been previously reported, but whose overall functions have been implicated in cancer development and progression. The former indicates that our integration method does not destroy the statistical signal in the original data, while the latter is strong evidence that the increased sample size increases the chances of finding novel gene expression signatures. Such signatures are also robust to inter-population variation, and show promise for translational applications like tumour grading, disease subtype classification, informing treatment selection and molecular prognostics.

Gene expression arrays are used to assess expression of exons and genes of orga- nisms. The design of expression arrays is based on a genome of laboratory strains of model organisms. The most frequent summarization algorithms used to pro- cess data from measurements are gcRMA, PLER and IterPLIER. When using expression arrays to research free living species, the measured values are influen- ced by differences in genomes of free living and model organisms. We propose a method to improve the results by removing parts of genomes influenced by known differences between species from the summarization. Removing influenced parts can improve summarization, especially on exon level. 1

Notch signaling plays a key role in development, tissue homeostasis, and cancer. High expression levels of Notch signaling components are associated with aggressive disease and poor patient prognosis in breast cancer. Mesenchymal‐epithelial transition factor (MET) is a receptor tyrosine kinase with an established prognostic significance correlating with poor disease outcome in breast cancer patients as a result of high metastatic rate. We performed expression array analysis to identify candidate Notch target genes; we identified and validated MET as a target of NOTCH1 signaling in breast cancer. We found that NOTCH1 knockdown significantly reduces MET promoter activity, as well as expression levels of MET transcript and protein. The mechanism of NOTCH1 regulation of MET expression will be the focus of future work. To further identify candidate target genes of NOTCH1 signaling, we generated and validated a NOTCH1 antibody for use in chromatin immunoprecipitation experiments.

Notch signaling plays a key role in development, tissue homeostasis, and cancer. High expression levels of Notch signaling components are associated with aggressive disease and poor patient prognosis in breast cancer. Mesenchymal‐epithelial transition factor (MET) is a receptor tyrosine kinase with an established prognostic significance correlating with poor disease outcome in breast cancer patients as a result of high metastatic rate. We performed expression array analysis to identify candidate Notch target genes; we identified and validated MET as a target of NOTCH1 signaling in breast cancer. We found that NOTCH1 knockdown significantly reduces MET promoter activity, as well as expression levels of MET transcript and protein. The mechanism of NOTCH1 regulation of MET expression will be the focus of future work. To further identify candidate target genes of NOTCH1 signaling, we generated and validated a NOTCH1 antibody for use in chromatin immunoprecipitation experiments.

In chronic lymphocytic leukemia (CLL), the presence of multiple subsets expressing ‘stereotyped’ B-cell receptors (BCRs) has implicated antigen(s) in leukemogenesis. These stereotyped subsets display similar immunoglobulin (IG) gene usage, almost identical complementarity determining region 3’s and may share clinical features. For instance, subsets #1 (IGHV1/5/7/IGKV1-39) and #2 (IGHV3-21/IGLV3-21) have inferior outcome compared to non-subset patients, whereas subset #4 (IGHV4-34/IGKV2-30) display a favourable prognosis. The aim of this thesis was to investigate genomic aberrations, gene expression patterns and methylation profiles in stereotyped subsets and compare epigenetic profiles in CLL and mantle cell lymphoma (MCL). In paper I, we investigated genomic aberrations in subsets #2, #4 and #16 and in non-stereotyped samples (n=101) using high-density 250K SNP arrays. Subset #2 and non-subset #2 IGHV3-21 cases displayed a higher frequency of aberrations than subset #4 cases. The high incidence of del(11q) in both subset #2/non-subset #2 may reflect the adverse survival reported for IGHV3-21 patients. In contrast, the lower frequency of genetic events and lack of poor-prognostic aberrations in subset #4 may partially explain their indolent disease. In paper II, we analysed the global RNA expression in subset #4, #16 and non-subset IGHV4-34 CLL patients (n=25). Subsets #4 and 16 showed distinct gene expression profiles, where genes involved in cell regulatory pathways were significantly lower expressed in subset #4, in line with their low-proliferative disease. In paper III, a genome-wide methylation array was applied to investigate methylation profiles in subsets #1, #2 and #4 (n=39). We identified differential methylation patterns for all subsets and found affected genes to be involved in e.g. apoptosis and therapy resistance. When performing functional annotation, a clear enrichment of genes involved in adaptive immunity was observed. These genes were preferentially methylated in subset #1 when compared to either subset #2 or #4, possibly due to different antigen responses. In paper IV, the genome-wide methylation profiles for 30 CLL and 20 MCL patients were investigated. Distinct methylation profiles were observed, where MCL displayed a more homogeneous profile. Homeobox transcription factor genes showed a higher degree of methylation in MCL, while apoptosis-related genes and proliferation-associated genes were methylated in CLL. In summary, this thesis demonstrates that stereotyped CLL subsets display differences in gene expression profiles, genetic aberrations and methylation patterns, underscoring the functional relevance of subgrouping according to BCR stereotypy. The distinct methylation profiles of CLL and MCL suggests that different epigenetic mechanisms are involved in the pathogenesis of these B-cell malignancies.

Copy number variation (CNV) comprises a recently discovered kind of variation involving deletion and duplication of DNA segments of variable size, ranging from a few hundred basepairs to several million. By altering gene dosage levels or disrupting proximal or distant regulatory elements CNVs create human diversity. They represent also an important factor in human evolution and play a role in many disorders including cancer. Array-based comparative genomic hybridization as well as expression arrays are powerful and suitable methods for determination of copy number variations or gene expression changes in the human genome. In paper I we established a 32K clone-based genomic array, covering 99% of the current assembly of the human genome with high resolution and applied it in the profiling of 71 healthy individuals from three ethnic groups. Novel and previously reported CNVs, involving ~3.5% of the genome, were identified. Interestingly, 87% of the detected CNV regions overlapped with known genes indicating that they probably have phenotypic consequences. In papers II through IV we applied this platform to different tumor types, namely two collections of brain tumors, glioblastoma (paper II) and medulloblastoma (paper III), and a set of bladder carcinoma (paper IV) to identify chromosomal alterations at the level of DNA copy number that could be related to tumor initiation/progression. Tumors of the central nervous system represent a heterogeneous group of both benign and malignant neoplasms that affect both children and adults. Glioblastoma and medulloblastoma are two malignant forms. Glioblastoma often affects adults while the embryonal tumor medulloblastoma is the most common malignant brain tumor among children. The detailed profiling of 78 glioblastomas, allowed us to identify a complex pattern of aberrations including frequent and high copy number amplicons (detected in 79% of samples) as well as a number of homozygously deleted loci. These regions encompassed not only previously reported oncogenes and tumor suppressor genes but also numerous novel genes. In paper III, a subset of 26 medulloblastomas was analyzed using the same genomic array. We observed that alterations involving chromosome 17, especially isochromosome 17q, were the most common genomic aberrations in this tumor type, but copy number alterations involving other chromosomes: 1, 7 and 8 were also frequent. Focal amplifications, on chromosome 1 and 3, not previously described, were also detected. These loci may encompass novel genes involved in medulloblastoma development. In paper IV we examined for the presence of DNA copy number alterations and their effect on gene expression in a subset of 21 well-characterized Ta bladder carcinomas, selected for the presence or absence of recurrences. We identified a number of novel genes as well as a significant association between amplifications and high-grade and recurrent tumors which might be clinically useful. The results derived from these studies increase our understanding of the genetic alterations leading to the development of these tumor forms and point out candidate genes that may be used in future as targets for new diagnostic and therapeutic strategies.