Les rapides progrès technologiques en matière de techniques de biologie à grande échelle, comprenant notamment les microarrays, conduisent en 2006 au développement d’une nouvelle génération de puces à très haute résolution, capables de cibler à la fois tous les gènes du transcriptome humain, mais également tous les exons de ces gènes pris individuellement. L’avènement de cette puce, communément appelée puce exon, permit d’obtenir une mesure précise des changements transcriptomiques affectant les cellules cancéreuses, en offrant la possibilité de prendre en compte l’expression relative de différents exons d’un même gène.L’épissage alternatif et la transcription alternative sont les deux principaux mécanismes biologiques à l’origine de l’existence de plusieurs transcrits pour un même gène. Ces processus biologiques ont été mis en évidence depuis longtemps mais leur régulation dans les cellules normales ainsi que leurs dérégulations dans les cancers sont encore mal caractérisées de par la complexité des mécanismes impliqués. Par leur design, les puces exons permettent de mettre en évidence la présence de variations d’expression entre plusieurs transcrits potentiels d’un même gène, ouvrant ainsi la voie à une meilleure compréhension de ces processus biologiques.A partir d’un important jeu de données d’échantillons de cancers de la vessie dont le profil transcriptomique fut obtenu par puces exons, nous nous sommes intéressés à l’étude des changements d’épissage alternatif et à l’utilisation de promoteurs alternatifs dans les tumeurs de vessie. L’utilisation d’outils statistiques et mathématiques dédiés à l’analyse de ces puces nous a permis dans un premier temps d’identifier de nombreux gènes dont l’expression relative des différents transcrits est spécifiquement dérégulée dans les tumeurs de vessie. Ces transcrits constituent une nouvelle source pour l’identification de cibles thérapeutiques spécifiques des tumeurs. Nous avons pu montrer qu’avec une approche ciblée sur les changements d’expression relative de transcrits alternatifs d’un même gène, il était possible de constituer un panel de potentiels marqueurs tumoraux permettant le développement de nouveaux tests urinaires utiles à la détection des cancers de vessie et à la surveillance des patients.Par une analyse non supervisée des profils d’exons potentiellement dérégulés, nous avons pu observer une stratification des tumeurs similaire à celle observée par l’étude des profils géniques issus de puces classiques, confirmant alors l’existence d’un sous groupe de tumeurs de vessie présentant des caractéristiques transcriptomiques propres. Nous avons pu associer à ce sous-groupe de mauvais pronostic, une signature d’inclusion différentielle de certains exons. Cette signature impliquant 19 gènes permet d’identifier précisément ces tumeurs de manière très spécifique et constitue par conséquent un outil puissant utilisable en clinique.L’étude ciblée d’une voie de signalisation fréquemment dérégulée dans les cancers nous a permis de mettre en évidence une dérégulation globale de l’expression relative des transcrits alternatifs de gènes impliqués dans la prolifération cellulaire, et d’en identifier de probables régulateurs. Enfin, L’analyse des données de puces exons à la lumière des données de méthylation de l’ADN nous a permis d’identifier un mécanisme épigénétique régulant l’utilisation de promoteurs alternatifs dans un sous-groupe de tumeurs de vessie.L’ensemble des résultats obtenus par l’analyse de ces puces exons a par conséquent permis de caractériser à l’échelle du transcrit les dérégulations spécifiques des tumeurs de vessie, et d’en identifier certains mécanismes. Ces dérégulations permettent non seulement d’identifier spécifiquement plusieurs sous-groupe de tumeurs dont un de mauvais pronostic, mais offrent également de nouvelles possibilités quant-à la recherche de marqueurs urinaires pour la surveillance des patients. / The development of microarray technology in the late 1990’s served as an essential tool to comprehend the scope of transcriptomic deregulations occurring in cancer cells. Signals generated from the first generation of transcriptomic microarrays gave simultaneous measures of expression from a large number of genes, therefore enabling to identify candidate genes involved in cancer progression and putative therapeutic targets. In 2006, through a fast de- velopment of high-throughput technologies, the available large scale analysis tools became enriched with a new generation of high resolution microarrays measuring expression signals both at the gene-level and at the exon-level of each gene. The advent of this high-resolution microarray, commonly called exon array, provided the opportunity to get a more accurate meas- ure of transcriptomic changes affecting cancer cells by enabling to consider relative expression changes of the exons from a same gene.Alternative splicing and alternative transcription are the two main biological mechanisms accounting for the production of several transcripts from a same gene. Although these bio- logical processes have been known for a long time, their regulation in normal cells and their deregulation in cancer still remain challenging to well-characterize, mainly due to the complex- ity of the involved mechanisms. Through their design, exon arrays enable to identify variable expression patterns within several potential transcripts of a same gene, therefore bringing new insight into these biological processes.Based on a large dataset of bladder cancer samples that were profiled on exon arrays, we focused on the study of alternative splicing changes and alternative promoter usage in bladder tumours. Analysis of these exon arrays through the use of adapted statistical and mathemat- ical tools initially resulted in the identification of numerous genes showing differential relative expression patterns of their transcripts between cancer and normal samples. These transcripts represent a new opportunity to define tumour-specific therapeutic targets. We demonstrated that using an approach targeted on relative expression changes of transcripts from a same gene, it was possible to build up a panel of potential tumour-specific markers enabling the development of new urinary test to detect bladder cancer and monitor its evolution.Through an unsupervised analysis of putatively deregulated exon profiles, we observed that the partitioning of bladder tumours was similar to the classification resulting from the study of classical gene microarray expression profiles, consequently confirming the existence of a bladder subgroup with peculiar transcriptomic properties. For this subgroup of bad prognosis, we established a signature based on the differential alternative inclusion of several exons. This signature relates to 19 genes and enables to accurately identify tumours from this subgroup, therefore providing a powerful tool to be used in clinical practice.By studying a specific pathway often deregulated in cancer, we highlighted an overall dereg- ulation of the relative expression of alternative transcripts from genes involved in cell prolifer- ation, and identified potential actors involved in the underlying regulatory process. Eventually, the analysis of exon arrays in the light of DNA methylation array data enabled us to identify an epigenetic mechanism regulating the use of alternative promoters in a subgroup of bladder tumours.Together, the results obtained from exon array analysis consequently provided a character- ization at the transcript level of bladder tumour specific deregulations and brought insight into the underlying mechanisms. The highlighted deregulations not only allow to accurately identify two subgroups of tumours, of which one has a bad prognosis, but also offer new possibilities regarding the definition of urinary markers for patient monitoring.

Em virtude dos pobres resultados obtidos no tratamento do Câncer de Pulmão, seja em estágios iniciais ou na doença avançada localmente, há a necessidade de se desenvolver marcadores moleculares e imunohistoquímicos que possam prever o comportamento tumoral. Ácido Hialurônico (HA) é um componente da matriz extracelular, responsável pela hidratação e manutenção do equilíbrio osmótico tecidual. Concentrações de HA estão elevadas em vários tipos de cânceres, incluindo pulmão. Hialuronidases (HAases), são uma família de enzimas relacionadas com a propagação de infecções bacterianas, toxinas de venenos e progressão tumoral. A quebra do HA em pequenos fragmentos (3-25 dissacarídeos) promovidos pela ação das HAases tipo Hyal1, Hyal2 e Hyal3, está relacionada à promoção do câncer através da indução da angiogênese e estímulo a proliferação através de ativação da via tirosina quinase. Algumas isoformas de HAases, descritas como produto de splicing alternativo, possuem atividade enzimática diversificada. A heterogeneidade de expressão das HAases foi identificada em alguns tipos de câncer e pode ser correlacionada com o comportamento diferenciado dos tumores. Em uma primeira instância, o perfil de expressão das HYAL foi avaliado em tecidos pulmonares tumorais e normais de 69 tumores ressecados de pacientes com adenocarcinomas (ADC) e carcinomas de células escamosas (CCE) oriundos do Hospital das Clinicas e Hospital do Câncer AC. Camargo. A expressão da HYAL1- selvagem (wt) e variantes 1 a 5, HYAL2-wt, HYAL3-wt e variantes 1 a 3 foi identificada por PCR e seqüenciamento direto. Diferentes proporções de HYAL3-wt e variantes foram expressas em tecidos pulmonares tumorais e controles. HYAL1-wt esteve associada com prognóstico desfavorável e HYAL3-v1 com prognóstico favorável. Diante dos resultados obtidos dos tumores de pacientes do Hospital das Clínicas e Hospital AC. Camargo, prosseguimos a investigação para estudar a imunoexpressão das Hyal 1 e 3 e HAS 1, 2 e 3 nos CCE e ADC. Observamos que a intensidade de expressão de Hyal 3 foi maior pelas células tumorais quando comparada aos controles, porém esta diferença foi marginalmente significante. Já o resultado da análise da freqüência de imunoexpressão das Hyal 1 e 3, e HAS1, 2 e 3 demonstrou expressão na maioria dos espécimes tumorais e controles. A associação entre as variáveis foi testada e evidenciou imunoexpressão concomitante de HYAL e HAS nos tumores. O modelo matemático de sobrevida , controlado para sexo, idade e estadiamento mostrou risco de morte associado com adenocarcinoma sólido e imunoreatividade para HAS2 e HAS3. Para validar os resultados obtidos, sobretudo com a imunoexpressão das Hyal e HAS nos CCE e ADC, estudamos a população de pacientes do Hospital Universitário de Coimbra. Documentamos pela primeira vez uma via pela qual a hiperexpressão de HAS3 e Hyal 3 respectivamente por células epiteliais neoplásicas e mesenquimais, podem favorecer a invasão nos ADC e CCE. Surpreendentemente, demonstramos que a imunoexpressão de HAS1 e 3 pelas células epiteliais neoplásicas confere mais agressividade aos ADC acinares e papilares, mas uma expressão negativa de HAS1 pelas células mesenquimais confere um papel protetor a MEC auxiliando-a a evitar a invasão pelas células tumorais em ambos os tipos subtipos histológicos. A interação entre a expressão das hialuronidades e sintases do àcido hialurônico foi avaliada em relação à expressão de proteínas da transição epitélio-mesênquimal nos tumores de pacientes do Hospital Universitário de Coimbra. Hyal, HAS, E-caderina e TGF- modularam uma via invasiva tumorinduzida nos ADC e CCE de pulmão, e estiveram associados a um espectro diferente de agressividade, uma vez que houve uma relação inversa entre a expressão de biomarcadores epiteliais e mesenquimais. Enquanto a hiperexpressão de HAS1 e HAS3 provê uma agressividade aos CCE e ADC, uma hiperexpressão de TGF- e E-caderina, confere um efeito protetor à MEC ao evitar a invasão por células tumorais em ambos os tipos histológicos. Comparamos os níveis de imunoexpressão das Hyal1 e 3 e HAS 1, 2 e 3 nos tumores ressecados no Hospital das Clínicas e Hospital AC. Camargo com os níveis obtidos em tumores do Hospital Universitário de Coimbra. Verificamos que a imunoexpressão das HAS 1, 2, 3 e Hyal1 foi significativamente maior nos tumores de pacientes do Hospital Universitário de Coimbra, enquanto que a imunoexpressão de Hyal 3 foi significativamente maior nos tumores de pacientes brasileiros. Por todas essas razões, nossos resultados sugerem que estratégias direcionadas à modulação dos níveis de HYAL1-wt e HYAL3-v1, da hiper imuno expressão de HAS3 e Hyal 3 respectivamente por células epiteliais neoplásicas e mesenquimais, da alta síntese de HAS3 e Hyal 1, ou a resposta local baixa de TGF- e E-caderina, poderão ter grande impacto no câncer de pulmão / Given the poor results obtained in the treatment of Lung Cancer, in early stages or locally advanced disease, there is a need to develop molecular markers and immunohistochemical studies that can predict tumor behavior. Hyaluronic Acid (HA) is a component of extracellular matrix is responsible for hydration and maintenance of tissue osmotic equilibrium. Concentrations of HA are elevated in several types of cancers, including lung. Hyaluronidases (HAases) are a family of enzymes involved in the spread of bacterial toxins, poisons and tumor progression. The breakdown of HA into small fragments (3-25 disaccharides) promoted by the action of type HAases Hyal1, Hyal 2 and Hyal 3 is related to the promotion of cancer by inducing angiogenesis and stimulate proliferation through activation of the tyrosine kinase. Some isoforms HAases, described as the product of alternative splicing, have diverse enzymatic activity. The heterogeneity of expression of HAases was identified in some cancers and can be correlated with the different behavior of tumors. In a first instance, the expression profile of Hyal spliced forms was evaluated in tumor and normal lung tissue of 69 tumors resected from patients with adenocarcinomas(ADC) and squamous cell carcinomas (SqCC) from the Hospital das Clínicas and Hospital AC. Camargo. Gene expression of HYAL1 wild-type (wt) and variants 1 to 5 HYAL2-wt, and HYAL3-wt and variants 1 to 3 was identified by PCR and direct sequencing. Different proportions of HYAL3-wt and variants were expressed in tumor and normal lung tissue. HYAL1-wt was associated with unfavorable prognosis and HYAL3-v1 with favorable prognosis. Given the genetic abnormalities found in tumors of patients from Hospital das Clinicas and Hospital AC. Camargo, we continued our research to study the expression of Hyal 1.3 and HAS 1, 2, 3 in squamous cell carcinomas and adenocarcinomas. We observed that the intensity of expression of Hyal 3 was higher in tumor cells compared to controls, but this difference was marginally significant. Since the result of frequency analysis of immunoreactivity of Hyal 1 and 3, and HAS1, 2 e 3 showed expression in the majority of tumor samples and controls. The association between variables was tested and showed concomitant immunoexpression of the HAS and HYAL in tumors. The mathematical model of survival, adjusted for sex, age and staging showed risk of death associated with adenocarcinoma and solid and HAS3 HAS2 immunoreactivity.To validate the results, especially with the immunostaining of Hyal and HAS in squamous cell carcinomas and adenocarcinomas of the lung, the patient population studied at the University Hospital of Coimbra. Documented for the first time a route by which the overexpression of HAS3 and Hyal 3 respectively by neoplastic epithelial and mesenchymal cells may favor the invasion in ADC and SqCC, respectively. Surprisingly, we demonstrated that hyper HAS1 and 3 immunoreactivity by neoplastic epithelial cells confers more aggressiveness to the ADC acinar and papillary, but a negative expression of HAS1 by mesenchymal cells confers a protective role ECM-helping to prevent the invasion by tumor cells in both types histological subtypes.The interaction between the expression of hialuronidades and hyaluronic acid synthases was evaluated for protein expression of epithelial-mesenchymal transition in tumor patients at the University Hospital of Coimbra. Hyaluronidase, hyaluronan synthase, Ecadherin, and TGF- modulated via an invasive tumor-induced in the ADC and SqCC lung, and were associated with a different spectrum of aggressiveness, since there was an inverse relationship between the expression of epithelial biomarkers and mesenchymal cells. While overexpression of HAS1 and HAS3 provides an aggressiveness to SqCC and ADC, an overexpression of TGF- and E-cadherin confers a protective effect by preventing the ECM invasion by tumor cells in both histological types. We compared the levels of immunostaining Hyal 1, 3 and HAS1, 2 and 3 in tumors resected at the Hospital AC. Camargo, and the levels obtained in tumors of the Hospital Universitário de Coimbra. We found that the immunostaining of HAS 1, 2, 3 and Hyal1 was significantly higher in tumors from patients of Coimbra, while Hyal 3 immunoreactivity was significantly. higher in tumors of patients in Brazil. For all these reasons, our results suggest that strategies directed at modulating the levels of HYAL1-wt and HYAL3-v1, the immunohistochemical expression of HAS3 and Hyal 3 respectively by neoplastic epithelial and mesenchymal cells, the synthesis of HAS3 and Hyal1 or the local response of low TGF- and E-cadherin, may have great impact on lung cancer

Our laboratory delineated a role for alternative pre-mRNA splicing (AS) in triple negative breast cancer (TNBC). We found the translational regulator cytosolic polyadenylation element binding protein 2 (CPEB2) which has two isoforms, CPEB2A and CPEB2B, is alternatively spliced during acquisition of anoikis resistance (AnR) and metastasis. The splicing event which determines the CPEB2 isoform is via inclusion/ exclusion of exon four in the mature mRNA transcript. The loss of CPEB2A with a concomitant increase in CPEB2B is required for TNBC cells to metastasize in vivo. We examined RNAseq profiles of TNBC cells which had CPEB2 isoforms specifically downregulated to examine the mechanism by which CPEB2 isoforms mediate opposing effects on cancer-related phenotypes. Downregulation of the CPEB2B isoform inhibited pathways driving the epithelial-to-mesenchymal transition (EMT) and hypoxic response, whereas downregulation of the CPEB2A isoform did not have this effect. Specifically, CPEB2B functioned as a translational activator of TWIST1 and HIF1a. Functional studies showed that specific downregulation of either HIF1α or TWIST1 inhibited the ability of CPEB2B to induce AnR and drive metastasis. The mechanism governing inclusion/ exclusion of exon 4 was determined to be serine/ arginine-rich splicing factor 3 (SRSF3). Binding of SRSF3 to a consensus sequence within CPEB2 exon 4 promoted its inclusion in the mature mRNA, and mutation of this sequence abolished association of SRSF3 with exon 4. SRSF3 expression was upregulated in TNBC cells upon acquisition of AnR correlating with a reduction in the CPEB2A/B ratio. Importantly, downregulation of SRSF3 by siRNA in these cells induced the exclusion of exon 4. Downregulation of SRSF3 also reversed the CPEB2A/B ratio in a wild-type CPEB2 exon 4 minigene construct, but not a mutant CPEB2 minigene with the SRSF3 RNA cis-element ablated. Physiologic studies demonstrated SRSF3 downregulation ablated AnR in TNBC cells, and was “rescued” by ectopic expression of CPEB2B. Importantly, biostatistical analysis of The Cancer Genome Atlas database showed a positive relationship between alterations in SRSF3 expression and lower overall survival in TNBC. Overall, this study demonstrates that SRSF3 modulates CPEB2 AS to induce the expression of the CPEB2B isoform that drives TNBC phenotypes correlating with aggressive human breast cancer.

<p>3’ end cleavage and polyadenylation are important steps in the maturation of eukaryotic mRNAs. Poly(A) polymerase (PAP), the enzyme catalysing the addition of adenosine residues, exists in multiple isoforms. In this study the functional significance of multiple poly(A) polymerases have been investigated. It is concluded (i) that at least three mechanisms generate the multiple isoforms i.e. gene duplication, post-translational modification and alternative mRNA processing and (ii) that the different isoforms of poly(A) polymerases have different catalytic properties. The study highlights regulation of poly(A) polymerase activity through modulation of its affinity for the substrate as visualised by the K_M parameter. We suggest that trans-acting factors modulating the K_M of poly(A) polymerase will play important roles in regulating its activity.</p><p>A new human poly(A) polymerase (PAPγ) encoded by the PAPOLG gene was identified. PAPγ is 65% homologous to the previously identified PAP. In human cells three isoforms of poly(A) polymerases being 90, 100 and 106 kDa in sizes are present. These native isoforms were purified. The PAPOLA gene encoded the 100 and 106 kDa isoforms while the 90 kDa isoform was encoded by the PAPOLG gene. Native PAPγ was found to be more active than 100 kDa PAP while the hyperphosphorylated 106 kDa PAP isoform was comparably inactive due to a 500-fold decrease in affinity for the RNA substrate. </p><p>The PAPOLG gene was shown to encode one unique mRNA while the PAPOLA gene generated five different PAP mRNAs by alternative splicing of the last three exons. The PAPOLA encoded mRNAs were divided into two classes based on the composition of the last three exons. Poly(A) polymerases from the two classes were shown to differ in polyadenylation activities. These differences revealed two novel regulatory motifs in the extreme C-terminal end of PAP, one being inactivating and the other activating for polyadenylation activity.</p>

<p>High-risk type human papillomaviruses (HPVs) are associated with cancer. HPVs are strictly epitheliotropic and infect basal cell layers, establishing a life cycle strongly linked to the differentiation stage of the infected cells. The viral capsid late genes, L2 and L1, are only expressed in terminally differentiated epithelium. Late gene expression involves regulation of most gene processing events including transcription, splicing, polyadenylation, mRNA stability and translation. </p><p>Both L2 and L1 have elements present in the open reading frames (ORFs) negatively affecting mRNA levels and translation. The negative elements in L1 were mapped to the first 514 nucleotides, with the strongest inhibitory effect located in the first 129 nucleotides. The negative elements in the L2 sequence were concentrated in two locations on the gene. Both genes were mutated by changing the nucleotide sequence while retaining the amino acid sequence. Mutating the first 514 nucleotides in L1 deactivated the negative elements while the entire L2 gene had to be mutated to achieve the same result. The L2 protein was found to localise the L1 protein into a punctuated pattern in the nucleus.</p><p>In the HPV-16 genome the negative elements reside in regions important for regulation of polyadenylation and splicing, critical for late gene expression. By exchanging parts of the L2 gene in subgenomic constructs with the corresponding mutant sequence we show that certain features of the L2 elements direct splicing to the L1 splice acceptor, and also regulate the efficiency of the early polyadenylation site. Cumulative binding of hnRNP H to the L2 mRNA gradually increased polyadenylation efficiency. Most interestingly, hnRNP H levels were downregulated in more differentiated epithelial cells. </p><p>Elucidation of how expression of the immunogenic late proteins is regulated would be greatly beneficial in prevention and treatment of HPV infection and thereby cancer.</p>

<p>In this thesis, the four-color fluorescence tag-microarray minisequencing system pioneered by our group was further developed and applied for analysing genetic variation in human DNA and RNA. A SNP marker panel representing different chromosomal regions was established and used for identification of informative SNP markers for monitoring chimerism after stem cell transplantation (SCT). The success of SCT was monitored by measuring the allelic ratios of informative SNPs in follow-up samples from nine patients with leukaemia. The results agreed with data obtained using microsatellite markers. Further the same SNP marker panel was used for evaluation of two whole genome amplification methods, primer extension preamplification (PEP) and multiple displacement amplification (MDA) in comparison with genomic DNA with respect to SNP genotyping success and accuracy in tag-array minisequencing. Identical results were obtained from MDA products and genomic DNA.</p><p>The tag-microarray minisequencing system was also established for multiplexed quantification of imbalanced expression of SNP alleles. Two endothelial cell lines and a panel of ten coding SNPs in five genes were used as model system. Six heterozygous SNPs were genotyped in RNA (cDNA) from the cell lines. Comparison of the relative amounts of the SNPs alleles in cDNA to heterozygote SNPs in genomic DNA displayed four SNPs with significant imbalanced expression between the SNP alleles. Finally, the tag-array minisequencing system was modified for detection of splice variants in mRNA from five leukaemia cell lines. A panel of 20 cancer-related genes with 74 alternatively splice variants was screened. Over half of the splice variants were detected in the cell lines, and similar alternative splicing patterns were observed in each cell line. The results were verified by size analysis of the PCR product subjected to the minisequencing primer extension reaction. The data from both methods agreed well, evidencing for a high sensitivity of our system.</p>

3’ end cleavage and polyadenylation are important steps in the maturation of eukaryotic mRNAs. Poly(A) polymerase (PAP), the enzyme catalysing the addition of adenosine residues, exists in multiple isoforms. In this study the functional significance of multiple poly(A) polymerases have been investigated. It is concluded (i) that at least three mechanisms generate the multiple isoforms i.e. gene duplication, post-translational modification and alternative mRNA processing and (ii) that the different isoforms of poly(A) polymerases have different catalytic properties. The study highlights regulation of poly(A) polymerase activity through modulation of its affinity for the substrate as visualised by the KM parameter. We suggest that trans-acting factors modulating the KM of poly(A) polymerase will play important roles in regulating its activity. A new human poly(A) polymerase (PAPγ) encoded by the PAPOLG gene was identified. PAPγ is 65% homologous to the previously identified PAP. In human cells three isoforms of poly(A) polymerases being 90, 100 and 106 kDa in sizes are present. These native isoforms were purified. The PAPOLA gene encoded the 100 and 106 kDa isoforms while the 90 kDa isoform was encoded by the PAPOLG gene. Native PAPγ was found to be more active than 100 kDa PAP while the hyperphosphorylated 106 kDa PAP isoform was comparably inactive due to a 500-fold decrease in affinity for the RNA substrate. The PAPOLG gene was shown to encode one unique mRNA while the PAPOLA gene generated five different PAP mRNAs by alternative splicing of the last three exons. The PAPOLA encoded mRNAs were divided into two classes based on the composition of the last three exons. Poly(A) polymerases from the two classes were shown to differ in polyadenylation activities. These differences revealed two novel regulatory motifs in the extreme C-terminal end of PAP, one being inactivating and the other activating for polyadenylation activity.

High-risk type human papillomaviruses (HPVs) are associated with cancer. HPVs are strictly epitheliotropic and infect basal cell layers, establishing a life cycle strongly linked to the differentiation stage of the infected cells. The viral capsid late genes, L2 and L1, are only expressed in terminally differentiated epithelium. Late gene expression involves regulation of most gene processing events including transcription, splicing, polyadenylation, mRNA stability and translation. Both L2 and L1 have elements present in the open reading frames (ORFs) negatively affecting mRNA levels and translation. The negative elements in L1 were mapped to the first 514 nucleotides, with the strongest inhibitory effect located in the first 129 nucleotides. The negative elements in the L2 sequence were concentrated in two locations on the gene. Both genes were mutated by changing the nucleotide sequence while retaining the amino acid sequence. Mutating the first 514 nucleotides in L1 deactivated the negative elements while the entire L2 gene had to be mutated to achieve the same result. The L2 protein was found to localise the L1 protein into a punctuated pattern in the nucleus. In the HPV-16 genome the negative elements reside in regions important for regulation of polyadenylation and splicing, critical for late gene expression. By exchanging parts of the L2 gene in subgenomic constructs with the corresponding mutant sequence we show that certain features of the L2 elements direct splicing to the L1 splice acceptor, and also regulate the efficiency of the early polyadenylation site. Cumulative binding of hnRNP H to the L2 mRNA gradually increased polyadenylation efficiency. Most interestingly, hnRNP H levels were downregulated in more differentiated epithelial cells. Elucidation of how expression of the immunogenic late proteins is regulated would be greatly beneficial in prevention and treatment of HPV infection and thereby cancer.

In this thesis, the four-color fluorescence tag-microarray minisequencing system pioneered by our group was further developed and applied for analysing genetic variation in human DNA and RNA. A SNP marker panel representing different chromosomal regions was established and used for identification of informative SNP markers for monitoring chimerism after stem cell transplantation (SCT). The success of SCT was monitored by measuring the allelic ratios of informative SNPs in follow-up samples from nine patients with leukaemia. The results agreed with data obtained using microsatellite markers. Further the same SNP marker panel was used for evaluation of two whole genome amplification methods, primer extension preamplification (PEP) and multiple displacement amplification (MDA) in comparison with genomic DNA with respect to SNP genotyping success and accuracy in tag-array minisequencing. Identical results were obtained from MDA products and genomic DNA. The tag-microarray minisequencing system was also established for multiplexed quantification of imbalanced expression of SNP alleles. Two endothelial cell lines and a panel of ten coding SNPs in five genes were used as model system. Six heterozygous SNPs were genotyped in RNA (cDNA) from the cell lines. Comparison of the relative amounts of the SNPs alleles in cDNA to heterozygote SNPs in genomic DNA displayed four SNPs with significant imbalanced expression between the SNP alleles. Finally, the tag-array minisequencing system was modified for detection of splice variants in mRNA from five leukaemia cell lines. A panel of 20 cancer-related genes with 74 alternatively splice variants was screened. Over half of the splice variants were detected in the cell lines, and similar alternative splicing patterns were observed in each cell line. The results were verified by size analysis of the PCR product subjected to the minisequencing primer extension reaction. The data from both methods agreed well, evidencing for a high sensitivity of our system.

The chimpanzee is humankind’s closest living relative and the two species diverged ~6 million years ago. Comparative studies of the human and chimpanzee genomes and transcriptomes are of great interest to understand the molecular mechanisms of speciation and the development of species-specific traits. The aim of this thesis is to characterize differences between the two species with regard to their genome sequences and the resulting transcript profiles. The first two papers focus on indel divergence and in particular, indels causing premature termination codons (PTCs) in 8% of the chimpanzee genes. The density of PTC genes is correlated with both the distance to the telomere and the indel divergence. Many PTC genes have several associated transcripts and since not all are affected by the PTC we propose that PTCs may affect the pattern of expressed isoforms. In the third paper, we investigate the transcriptome divergence in cerebellum, heart and liver, using high-density exon arrays. The results show that gene expression differs more between tissues than between species. Approximately 15% of the genes are differentially expressed between species, and half of the genes show different splicing patterns. We identify 28 cassette exons which are only included in one of the species, often in a tissue-specific manner. In the fourth paper, we use massive parallel sequencing to study the chimpanzee transcriptome in frontal cortex and liver. We estimate gene expression and search for novel transcribed regions (TRs). The majority of TRs are located close to genes and possibly extend the annotations. A subset of TRs are not found in the human genome. The brain transcriptome differs substantially from that of the liver and we identify a subset of genes enriched with TRs in frontal cortex. In conclusion, this thesis provides evidence of extensive genomic and transcriptomic variability between human and chimpanzee. The findings provide a basis for further studies of the underlying differences affecting phenotypic divergence between human and chimpanzee.