Long Interspersed Element 1(LINE1 or L1), along with the parasitic Short Interspersed Element (SINE), are the only two currently active retrotransposons in the human genome. These genetic elements are capable of causing DNA damage through their mobilization and the enzymatic activities of the L1-encoded Open Reading Frame 2 (ORF2) protein (ORF2p). The L1 ORF2p contains four annotated domains important to retrotransposition. These include the endonuclease (EN) and reverse transcriptase (RT) domains, as well as the Z domain and the cysteine rich domain (Cys). While much is known about the enzymatic activities of the EN and RT domains, and individual amino acids important to retrotransposition have been identified in the Z and Cys domains, more than 50% of the 150kDa ORF2p amino acid sequence serves no known function. I hypothesized that the unannotated areas of the ORF2p, specifically the sequence C-terminal to the EN domain and N-terminal to the Z domain as well as the sequence C-terminal to the Cys domain, contained amino acids important to the retrotransposition process. Specifically, I hypothesized that they contained amino acids involved in the activity of the EN domain, RT domain, or interaction with the L1 ORF1p. To test this hypothesis, I developed a technique termed Bipartile Alu Retrotransposition (BAR) that utilized EN and RT-containing ORF2 fragments combined with the Alu retrotransposition reporter construct. This system allowed me to define a new ORF2p region, which I termed Cryptic. This region contains an essential WD pair important for cDNA syntheses by the ORF2p. This WD pair is also involved in the regulation of the EN domain activity. I also discovered a putative PCNA binding domain that is essential for retrotransposition. Additionally, I identified the region in Cryptic that is involved in the previously reported differences in subcellular localization and cytotoxic potential of EN-containing ORF2p fragments. Using truncated ORF2p fragments generated for use in BAR, I also made several discoveries concerning the extreme C-terminus of the ORF2p. Notably, I discovered that the extreme C-terminal end of the ORF2p is dispensable for retrotransposition. I also identified a human-specific Y residue that is important for Alu retrotransposition driven by the ORF2p. / 1 / Claiborne Magnant Christian

archives@tulane.edu / These studies make strides in better understanding retroelement L1 expression and regulation at the locus-specific level using a combination of sequencing technologies. A picture is painted demonstrating tissue specific patterns of L1 expression when identified stringently and confidently with our developed EL-Seq approach. As it was also determined that expressed L1s significantly correlate with regions of open chromatin, these tissue-specific patterns of L1 expression are then most likely explained by tissue-specific chromatin architecture. Evidence is also presented here that L1s in tissues respond differently with genomic stresses and perturbations as is seen in the case of aging indicating that the risk associated with L1 damage and mutagenesis is related to cell type and tissue. This is particularly notable when considering the genesis and promotion of age-related somatic diseases like epithelial cancers. L1s are commonly referred to as the dark matter of the genome, but here we illuminate its biology and regulation to better understand L1-associated damage and risk to human health. / 0 / Tiffany Kaul

Submitted by Gizele Lima (gizele.lima@inpa.gov.br) on 2016-09-22T14:31:54Z No. of bitstreams: 2 Dissertação Dayane.pdf: 20358939 bytes, checksum: d0ddbe8f452b69b381be809c7baeff64 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2016-09-22T14:31:54Z (GMT). No. of bitstreams: 2 Dissertação Dayane.pdf: 20358939 bytes, checksum: d0ddbe8f452b69b381be809c7baeff64 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2015-06-29 / Fundação de Amparo à Pesquisa do Estado do Amazonas - FAPEAM / Saguinus is the largest and most complex genus of the subfamily Callitrichinae, with 23 species. They are distributed from Southern of Central America to northern of South America. Saguinus bicolor have very limited geographic distribuition, affected by demographic expansion of the city Manaus. In contrast, Saguinus midas have largest geographic distribuition among the Saguinus. They share the same characteristics general and overlap the north of Manaus. Cytogenetics studies with Saguinus described a karyotypic macrostructure conserved, with 2n=46 and patterns of similar bands. However, mapping studies of repetitive sequence are incipient. Repetitive sequence in tandem: telomere and rDNA; and repetitive sequence dispersed include the transposable elements were searched in the work. Analysis were made on S. midas and two populations of S. bicolor. The classical cytogenetics confirmed macrostructure of 2n=46, but differed in morphology of chromosomes, classified into: 8 metacentrics; 10 submetacentrics; 10 subtelocentrics and 6 acrocentrics. The patterns bands were similar, but showed variations among individuals of the same species. The G-bands patterns suggest the fourth pair as cytogenetic markers that show differences among two species and identify natural hybrids in contact zone. The NOR’s were detected in pairs 17 and 18, agreeing with the localization of sequences of rDNA 18S in region pericentromeric of long arms of chromosomes 17, 18 and 19, located in heterochomatic region. LINE–1 was found in regions: euchromatics – having an impact on the organization and function of genome, and; heterochromatics - particularly in centromeric heterochromatin. Accumulation in sex chromosomes are associated with inactivation of one chromosome X in females to promote the gene silencing and ensure gene dosage of sex pair when compared with male. It is possible to observe his presence in regions of negatives G-bands (light bands) implying that deposition in genome of S. bicolor and S. midas is recent in evolutionary time. Differences of sinalization of LINE-1 among populations of S. bicolor were detected, possibly due to isolation the two populations. / Saguinus é o maior e mais complexo gênero da subfamília Callitrichinae, com 23 espécies. Eles estão distribuídos do sul da América Central ao norte da América do Sul. Saguinus bicolor possui uma distribuição geográfica muito limitada, afetada pela expansão demográfica da cidade de Manaus. Ao contrário, Saguinus midas possui a maior distribuição geográfica dentre os Saguinus. São próximas filogeneticamente, compartilham das mesmas características gerais e se sobrepõem ao norte de Manaus. Estudos citogenéticos dos Saguinus descrevem uma macroestrutura cariotípica conservada, com 2n=46 e padrões de bandas similares. Porém, estudos com mapeamento de sequências repetitivas são incipientes. Sequências repetidas em tandem: telômericas e DNAr, e; sequências repetitivas dispersas que englobam os elementos transponíveis são pesquisadas neste trabalho. Análises citogenéticas foram feitas em S. midas e duas populações de S. bicolor. A citogenética clássica confirmou a macroestrutura das duas espécies em 2n=46, porém diferiu na morfologia dos cromossomos quando comparadas com estudos anteriores, sendo aqui, classificados em: 8 M; 10 SM; 20 ST e 6 A. O padrão de banda G apresentou variações entre as espécies, sugerindo o quarto par como marcador citogenético que diferenciaria as duas espécies e identificaria híbridos naturais de 1a geração, em zona de contato. As RON’s foram detectadas nos pares 17 e 18, sendo confirmadas pela localização da sequência de DNAr 18S na região pericentromérica dos braços longos nos cromossomos 17, 18 e 19, localizada em regiões próximas as heterocromatinas. A sequência LINE-1 foi encontrada nas regiões: eucromáticas – podendo influenciar na organização e função do genoma, e; heterocromáticas - particularmente na heterocromatina centromérica. O acúmulo de LINE-1 nos cromossomos sexuais está relacionado com a inativação de um dos cromossomos X nas fêmeas garantindo a dosagem gênica do par sexual quando comparado com os machos. A presença de LINE- 1 nas regiões de bandas G negativas (bandas claras) indica que sua deposição no genoma de S. bicolor e S. midas é recente no tempo evolutivo. Diferenças de sinalização do LINE-1 entre populações de S. bicolor foram detectadas, possivelmente devido ao isolamento das duas populações.

Although several lines of evidence have established the central role of epithelial-to-mesenchymal-transition (EMT) in malignant progression of non-small cell lung cancers (NSCLCs), the molecular events connecting EMT to malignancy remain poorly understood. This study presents evidence that Long Interspersed Nuclear Element-1 (LINE-1) retrotransposon couples EMT programming with malignancy in human bronchial epithelial cells (BEAS-2B). This conclusion is supported by studies showing that: 1) activation of EMT programming by TGF-beta 1 increases LINE-1 mRNAs and protein; 2) the lung carcinogen benzo(a)pyrene coregulates TGF-beta 1 and LINE-1 mRNAs, with LINE-1 positioned downstream of TGF-beta 1 signaling; and, 3) forced expression of LINE-1 in BEAS-2B cells recapitulates EMT programming and induces malignant phenotypes and tumorigenesis in vivo. These findings identify a TGF beta 1-LINE-1 axis as a critical effector pathway that can be targeted for the development of precision therapies during malignant progression of intractable NSCLCs.

There are two classes of transposable elements: DNA transposons and retrotransposons. DNA transposons spread in the genome by “cut and paste” mechanism. In contrast, retrotransposons use copy and paste strategy involving RNA and retrotranscriptase mediated mechanism; these include long interspersed nuclear elements-1 (LINE-1, L1) and short interspersed nuclear elements (SINE). In mammals, in order to maintain genome integrity both types of transposons are tightly repressed. However, some copies of retrotransposons are still active in germ cells contributing to natural variation. Surprisingly, recent reports indicate that also somatic cells support L1 reactivation in early development, in particular in the brain leading to mosaicism. However, whether L1 retrotransposition is a part of other cell lineage developmental programs and its functional significance in the context of cell differentiation remain to be elucidated. To address this question, I investigated whether L1 retrotransposition was occurring during in vitro osteogenic and adipogenic differentiation of bone marrow derived human mesenchymal stem cells (hMSCs). Interestingly, clinical observations have revealed loss of bone density in HIV-infected individuals treated with nucleoside analogs that inhibit HIV retrotranscriptase, as well as the endogenous one encoded by L1s. This observation made us to hypothesize that transposable elements played a positive role in post-natal bone homeostasis. I found that while adipogenesis is “retrotransposition free”, osteogenic differentiation is a “retrotransposition-prone” process and its inhibition blocks its genetic program. Indeed, L1 DNA content does not change during adipogenic differentiation and that of retrotranscriptase does not have any effect on the acquisition of a terminally differentiated phenotype. In contrast, soon after MSCs commitment into pre-osteoblasts, L1 retrotransposable elements increase their expression and actively transpose. Inhibition of retrotransposition and knock down of L1 mRNA strongly impairs matrix deposition. Moreover, I forced L1 expression in in vitro adipogenesis, by directly delivering L1 mRNA to the cells. Interestingly, overexpression of L1 elements was detrimental for in vitro adipogenesis. Then, I performed loss of function experiments in osteogenesis by directly targeting and degrading the L1 endogenous transcript. This experiment confirmed the positive role of L1 reactivation in the osteogenic context, suggesting also a possible role for L1 RNA, distinct from retrotransposition.

LINE-1 retrotransposons, also known as "jumping genes", are repetitive sequences capable of copying, pasting, and reinserting themselves into the genome. These events were documented at high frequency in various types of cancers, including colorectal cancer (CRC). Furthermore, the expression of proteins encoded by these elements, such as L1ORF1p, has been linked to cancer aggressiveness, stemness, and lower patient survival rates. Colorectal cancer stem cells (CCSC), constitute a small subset of cells endowed with pluripotency and self-renewal abilities. They play roles in tumorigenesis, cancer aggressiveness, drug resistance, cancer recurrence, and metastasis. Conventional chemotherapeutics primarily target bulk tumor cells and tend to spare cancer stem cell populations. Consequently, targeting CCSC is expected to significantly increase complete remission and survival rates in CRC patients. Here, I have characterized specific aspects of a novel repurposed drug that effectively targets CCSC, reactivates the expression of transposable elements and, consequently, triggers an innate immune response. Further, I tested an in silico drug screening approach to identify compounds with high predicted affinity for the RNA binding domains of L1ORF1p, a key protein for the LINE1 retrotransposition event to occur, from a virtual drug library. Two lead compounds, both FDA-approved drugs, were identified and evaluated for their capacity to block L1ORF1p nuclear translocation, a needed step to complete the LINE1 lifecycle, as well as their capacity to decrease LINE-1 retrotransposition levels. In addition, I established a protocol for the isolation, culture, propagation, and cryopreservation of patient-derived normal colonic organoids. This protocol is crucial to the establishment of a colonic organoid biobank, representing a powerful resource to assess cancer-selective toxicity of putative CCSC-targeting compounds. Together, this thesis emphasizes the importance of transposable elements in CRC and contributes to the establishment of a gold standard ex vivo disease-modeling system for the discovery of new therapeutic agents.

Le génome humain est constitué d'environ 45% d'éléments transposables, et l'élément LINE-1 (L1), qui représente 17% de la masse totale, est le seul élément actif connu de notre génome capable de rétrotransposer. Les protéines de L1, ORF1p et ORF2p, peuvent mobiliser divers ARN en trans, tel que Alu, le snRNA U6 ou encore d'autres ARNm cellulaires. C'est, en partie pour ces raisons que l'élément L1 est considéré comme un acteur majeur de la plasticité de notre génome.Dans un premier temps, nous avons élaboré un nouveau système, la lignée stable inductible Trex, qui permettra d'étudier la rétrotransposition de L1 à partir d'une localisation chromosomique. Cette lignée stable présente une intégration d'un élément L1 marqué avec T7 et Flag-HA pour ORF1p et ORF2p, respectivement. Ce système nous permettra dans un avenir proche de purifier les partenaires protéiques d'ORF2p selon le protocole de double purification sur colonne d'affinité Flag suivi de HA, et de les identifier par spectrométrie de masse (MS/MS). Dans un deuxième temps, nous nous sommes intéressés aux séquences chimères U6-L1 présentes au niveau du génome humain. Nous avons entamé ainsi une étude in silico, qui nous a conduit à la découverte d'une nouvelle unité répétée présente dans les génomes de primates, que l'on a appelé ‘'RSU6''. / The human génome is composed of about 45% of mobile elements. LINE-1 (L1, which represents 17% of the total mass, is the only active element, capable of retrotransposition in our génome. L1 proteins, ORF1p and ORF2p, can mobilise other RNA in trans, such Alu, snRNA U6 or other cellular mRNA. This is one of the reasons why L1 element is considered as the major actor of genome plasticity. First, we have developed a new system, the stable Trex cell line, which will allow us to study L1 retrotransposition from a chromosomal location. This stable cell line has integrated a marqued L1 (with ORF1-T7 and ORF2- Flag-HA) into the genome. This new system will enable us to purify cellular cofactors of ORF2p, according to the double affinity protocol (Flag and HA), and to identify them by mass spectrometry. Second, we were intersted in the chimeric sequences U6-L1 present in the human genome. We have done an in silico analyses which allowed us to identify a new repeated unit in the primate genome that we called ‘'RSU6''.

Le gène hybride BCR-ABL1, responsable de la leucémie myéloïde chronique (LMC), code une protéine à activité tyrosine kinase constitutive. Lors d’une étude transcriptomique menée au laboratoire sur des patients résistants secondaires à l’imatinib, les deux gènes codant les protéines des rétrotransposons LINE-1 ont été trouvés sous exprimés d’environ 20 fois lorsque les patients rechutent. Le rôle des transposons n’a jamais été clairement défini, ils assurent certainement une fonction importante puisqu’ils sont conservés au cours de l’évolution et présents chez tous les organismes. Le but de ce travail a été d’étudier l’implication de LINE-1 dans la LMC. La sous-expression de LINE-1 est-elle une conséquence de la présence de BCR-ABL1 ou une cause de son apparition ? Différents groupes ont montré que les rétrotransposons LINE-1 possédent la capacité de réparation des cassures double-brin de l’ADN. Nous avons fait l’hypothèse qu’une diminution de l’expression des gènes codés par les rétrotransposons LINE-1 entraînerait l’instabilité génétique observée dans la LMC. Une étude réalisée chez des patients atteints de LMC et des sujets contrôles a montré une correlation inverse entre l’expression de LINE-1 et celle de l’oncogène BCR-ABL1. Parallèlement, une étude sur des lignées cellulaires leucémiques humaines BCR-ABL positives et négatives a été réalisée. Nous avons recherché le lien qui existe entre l’expression de LINE 1, de BCR-ABL1 et la réparation des cassures double-brin de l’ADN. Nous avons montré d’une part qu’une inhibition de l’expression de BCR-ABL1 induit une augmentation de l’expression des transposons LINE-1 D’autre part, une diminution de l’expression de LINE-1 entraîne une apparition du transcrit BCR-ABL1 dans les cellules BCR-ABL negatives. / BCR-ABL1 fusion gene, responsible of the chronic myeloid leukemia (CML) encodes a constitutively activated tyrosine kinase protein. Expression of both LINE-1 retrotransposon ORFs were found decreased at the time of imatinib resistance in a comparative transcriptional study focused on secondary resistant patients. The role of retrotransposons is unclear. They are conserved through evolution. This project focuses on the involvement of LINE-1 in CML. Is LINE-1 under expression a result of BCR-ABL1 expression or is it at the origin of BCR ABL1? Different groups have shown that LINE-1 retrotransposons were able to repair DNA double strands breaks. We suggest that LINE-1 under expression could be responsible of genetic instability observed in CML. We show in a study on CML patients and healthy subjects that LINE-1 expression is inverse correlated to BCR-ABL1 expression. Moreover, study on BCR-ABL+ and BCR-ABL- human leukemic cell lines was carried on. First, we show that decrease of BCR-ABL1 expression induces increase of LINE-1 expression. Then that decrease of LINE-1 expression generates BCR-ABL1 transcript in BCR-ABL negatives cell lines.

L’élément L1 (Long Interspersed Nuclear Element 1 ou L1) est le seul rétrotransposon autonome et actif connu dans notre génome, représentant 17% de celui-ci. Capable de se répliquer grâce à un intermédiaire à ARN et un mécanisme de transcription inverse initiée au site d’intégration, il encode deux protéines ORF1p et ORF2p, qui s’associent à l’ARN L1 pour former une particule ribonucléoprotéique (RNP). L’élément L1 rétrotranspose préférentiellement en cis : un L1 défectif est complémenté en trans par un élément fonctionnel de façon inefficace. Ce travail s'intéresse à deux étapes clefs du cycle réplicatif du L1 : l'assemblage de la RNP L1 en cis ou en trans afin d’explorer le mécanisme de la cis-préférence et la spécificité de l’initiation de la reverse transcription initiée. Nous avons d’abord comparé deux méthodes d’analyse de l’activité RT. Puis, nous avons montré l’importance de la complémentarité entre queue poly(A) de l’ARN L1 et site d’intégration durant l’initiation de la RT, ainsi que l’impact de mésappariements terminaux éventuels. Enfin, nous avons étudié les bases biochimiques de la cis-préférence, à travers la coexpression et la purification de deux éléments distincts étiquetés, ce qui nous a permis de suivre l'assemblage et l'activité de leurs RNPs respectives. Nos données suggèrent que ORF1p et ORF2p peuvent lier en trans l’ARN L1 de façon efficace et que la cis-préférence pourrait nécessiter des quantités limitantes de L1. / The Long Interspersed Nuclear Element 1 (LINE-1 ou L1) is the only known active and autonomous retrotransposon in the human genome and constitutes around 17% of our genomic DNA. The L1 element is able to replicate through an RNA intermediate by a mechanism called target-primed reverse transcription and encodes two proteins ORF1p and ORF2p, which associate with the L1 RNA to form a ribonucleoprotein particle (RNP). L1 preferentially retrotranspose in cis: a defective L1 can only be rescued in trans at low levels by a replication-competent copy. During this work, we focused on two essential steps of the L1 replication cycle: the assembly of the L1 RNP in cis or in trans to explore the mechanism of the cis-preference and the specificity of L1 reverse transcription priming. First, we compared two different methods to detect L1 RT activity. Then, we showed the importance of base-pairing between the poly(A) tail of the L1 RNA and the integration site to prime reverse transcription and the impact of potential mismatches. Finally, we investigated the biochemical basis of the cis-preference through the coexpression and purification of two different tagged L1 elements, which allowed us to follow the assembly and activity of their RNP. Our data suggest that binding of ORF1p and ORF2p in trans is efficient and that the cis-preference might requires limiting L1 levels.

Les éléments LINE-1 (L1) sont une classe abondante de rétrotransposons représentant 17% du génome humain. La région 5’UTR des sous-familles les plus récentes (L1PA1 à 6) contient un promoteur bidirectionnel contenant non seulement un promoteur sens interne mais aussi un promoteur antisens, nommé ASP. Dans les cellules normales, l’un des mécanismes impliqués dans la régulation du promoteur de L1 est la méthylation ADN. Dans les tumeurs, une hypométhylation globale affectant notamment les L1 est observée. Il a été mis en évidence que cette hypométhylation pouvait induire la transcription, à partir de l’ASP, de transcrits chimères ou LCT (L1 Chimeric Transcript). Ces LCT sont composés en 5’ de la séquence L1 et se poursuivent dans la région génomique adjacente. Afin d’étudier l’impact pangénomique de cette dérégulation et son implication dans les processus tumoraux, un outil bio-informatique dédié, nommé CLIFinder, a été développé pour identifier dans des données de RNA-seq paired-end orientés des LCT putatifs. Les RNA-seq de 13 gliomes, qui sont les cancers du cerveau les plus fréquents chez l’adulte, et de 3 tissus cérébraux contrôles ont été étudiés. CLIFinder identifie 2675 chimères dans les gliomes dont 84% impliquent des L1 récents (PA1 à 7) pleine taille supposés posséder un ASP fonctionnel et 50% sont détectées spécifiquement dans les échantillons tumoraux. 78 chimères correspondent à des LCT déjà décrits dans la littérature. De même, l’étude de RNA-seq d’autres types tumoraux (lignée MCF7 et métastases ovariennes) par CLIFinder identifie des chimères en commun suggérant une récurrence de certaines d’entre elles. L’étude d’un groupe de chimères par marche en 5’ par RT-PCR valide que 89% (56/63) des chimères impliquant des L1 récents (L1PA1 à PA7) sont initiées dans la région de l’ASP et correspondent à des LCT alors que toutes les chimères testées impliquant des L1PA8 sont initiées en amont de cette région. Des études de RT-qPCR sur une cohorte plus large de 51 gliomes montrent que les 56 LCT testés, incluant des LCT spécifiques de tumeurs, sont exprimés non seulement dans les tumeurs mais aussi dans les contrôles. Par contre, 70% des LCT spécifiques de tumeurs, montrent alors une surexpression tumorale significative. Ces résultats suggèrent donc une transcription basale provenant de l’ASP dans les tissus normaux et que la dérégulation transcriptionnelle liées aux LCT dans les gliomes passe par une surexpression. Par ailleurs, afin de déterminer le ou les mécanismes sous-jacents impliqués dans l’augmentation de l’activité transcriptionnelle de l’ASP, deux hypothèses ont été testées. La première implique l’hypométhylation du promoteur de L1. Toutefois mes résultats tendent à réfuter cette hypothèse puisqu’aucune diminution de la méthylation ADN n’est retrouvée au niveau de la région promotrice des L1 impliqués dans la transcription de LCT surexprimés. Par contre, les gènes associés à des LCT dont l’expression est dérégulée en contexte tumoral présentent une dérégulation dans le même sens que celle du LCT. De plus, les variations d’expression de gènes corrèlent systématiquement avec celle des LCT correspondants. Ceci suggère qu’une augmentation d’activité transcriptionnelle aux loci des LCT serait responsable de la surexpression de ceux-ci. Enfin 2 LCT candidats surexprimés et ayant un potentiel de biomarqueur prédictif de la survie des patients, pourraient jouer un rôle fonctionnel dans l’initiation, la progression et/ou l’agressivité tumorale. En conclusion, mes travaux ont validé que CLIFinder se positionne comme un outil pertinent permettant d’identifier, de façon pangénomique, les LCT exprimés dans différents types tumoraux à partir de données de RNA-seq paired-end orientées. L’observation d’une récurrence ainsi que d’une surexpression tumorale de certains LCT suggère qu’ils pourraient jouer un rôle fonctionnel dans les processus de tumorigenèse. / LINE-1 (L1) is the most abundant class of retrotransposons which represents 17% of the human genome. The 5’ region of the youngest L1 sub-families (L1PA1 to 6) contains a bidirectional promoter consisting, in addition to the internal sense promoter, of an antisense promoter, called ASP. In normal cells, the main defense mechanism, developed to counteract the deleterious effect of L1 activity, consists in L1 promoter DNA methylation. A hallmark of cancer genomes consists in a global DNA hypomethylation which affects especially L1 promoters. In tumors, evidences suggest that this hypomethylation could result in transcription from ASP of aberrant L1-Chimeric Transcripts (LCTs) composed of L1 5’end and its adjacent sequence. To investigate the pangenomic extent of this transcriptional deregulation and its impact in tumoral processes, a dedicated bioinformatic tool, CLIFinder, was designed to select putative LCTs among RNA-seq oriented paired-end reads. RNA-seq analyses of 13 gliomas, which are the most common brain cancer in adults, and 3 control brains were performed.CLIFinder identifies 2675 chimeras in gliomas, among which 84% involves recent L1 (PA1 to 7) full size, supposed to possess a functional ASP, and 50% are detected specifically in tumors samples. 78 chimeras correspond to LCT already described in literature. In addition, study of additional RNA-seq data from other tumor types (MCF7 and ovarian metastasis) by CLIFinder identifies common chimeras suggesting that some of them can be recurrent. The analysis of a group of chimeras by 5’ walk RT-PCR validate that 89% (56/63) of chimeras implying recent L1 (L1PA1 to 7) are initiated at the ASP region and therefore correspond to LCT; whereas all tested chimeras implying an L1PA8 element are transcribed from an upstream region. RT-qPCR studies on a larger cohort of 51 gliomas show that all 56 tested LCT, even identified by CLIFinder as “tumor specific”, are not only expressed in tumors but also in controls. Nevertheless, 70% of the “tumor specific” LCTs are significantly overexpressed in tumors. My results suggest that, even L1 5’ UTR methylation, some ASP are active in normal tissues and lead to a basal LCT expression in normal tissues. Moreover, a transcriptional deregulation linked to LCTs in tumors exists and implies a LCTs’ overexpression.In order to determine the underlying mechanisms involved in the increase of transcriptional activity of ASP, two hypothesis were tested. The first one implies L1 promoter hypomethylation. My results tend to refute this hypothesis because no decrease of the DNA methylation is found at the promoter region of L1 linked to overexpressed LCTs. On the other hand, the genes associated to LCT presenting an expression deregulation in tumors demonstrate a deregulation in the same way. Moreover, gene expression variations correlates systematically with the one corresponding LCTs. This suggests that an increase of transcriptional activity at the LCTs loci would be responsible of their overexpression. Finally, 2 candidate LCT overexpressed and presenting as potential predictive biomarkers for patient’s survival, could play a functional role in initiation, progression and/or the tumoral aggressiveness.In conclusion, my work has validated CLIFinder as a useful tool to identify, at pangenomic level, LCTs expressed in different tumor types from paired-end stranded RNA-seq data. The observation of the recurrence and tumoral overexpression for some LCTs suggests that they may play a functional role in tumoral processes.