L’existance de cellules souches leucémiques (CSL) dans la Leucémie Myéloïde Chronique (LMC) prédit que que seule la destruction des CSL conduirait à une guérison. Une proportion importante de patients atteints de LMC développe une résistance aux drogues, environ ~ 30% des cas, Les mécanismes de résistance aux inhibiteurs de tyrosine kinase (TKI) dans la leucémie myéloïde chronique (LMC) restent souvent obscures. Les cellules souches leucémiques de la LMC pourraient rester viables et en repos, malgré la présence de facteurs de croissance ou de médicaments qui semblent les protéger de l'apoptose. Nous avons montré dans la première parti de cette étude que certains transporteurs telles ABCG2, hOCT-1 pourraient jouer un rôle avec le micro environnement dans la résistance des lignées LMC en adhésion au stoma médullaire. De plus, dans deuxième parti nous avons montré que le gène TWIST-1 (est un acteur clé de l'embryogenèse) est déréglementé dans les cellules de LMC innée des résistants à l'imatinib, et que la sur-expression de l’oncogène TWIST-1 pourrait représenter un nouveau facteur clé de pronostiques potentiellement utiles pour améliorer la guérison de LMC aux TKI. De plus, nous avons pu également montrer que le gène TP73 est impliqué dans la résistance des CSL de LMC. Ce gène pourrait être un facteur prédictif pour identifier une résistance potentielle des patients de LMC au moment du diagnostic. Nous avons montré également que ce gène est régulé par le micro-environnement. Nous avons montrés une sur-expression des isoformes tronquées dans les lignées LMC avec l’adhésion au stroma. Les résultats suggèrent que les molécules intrinsèques comme TWIST-1, les transporteurs et les isoformes de p73 sont dérégulées dans les CSL par des mécanismes extrinsèques qui interviennent avec le micro environnement leucémique par le mécanisme d’adhésion dans le phénomène de résistance aux traitements. Ce mécanisme spécial de résistance due à la conservation de ces CSL dans l’état immature en adhésion avec son micro environnement. / The existence of Leukemia stem cells (CSL) in chronic myelogenous leukemia (CML) predicts that only the destruction of CSL lead to a cure. A significant proportion of CML patients develop resistance to drugs, ~ 30% cases, mechanisms of resistance to tyrosine kinase inhibitors (TKI) in chronic myeloid leukemia (CML) often remain obscure. Leukemic stem cells of CML could remain viable and quiet, despite the presence of growth factors or drugs that seem to protect them from apoptosis. We have shown in the first part of this study that some carriers such as ABCG2, hOCT could to be play a role with the microenvironment in the resistance among CML adhesion of stoma Bone marrow. Furthermore, in the second part we showed that the gene TWIST-1 (is a key player of the embryogenesis) is deregulated in cells of CML innately resistant to imatinib, and that overexpression of the oncogene TWIST-1 could represent a new prognostic factor key potentially useful for improving the querison CML to TKI. In addition, we also could show that the TP73 gene is involved in the resistance of CSL CML. This gene could be a predictor to identify potential resistance of CML patients at diagnosis. We have also shown that this gene is regulated by the microenvironment. We have shown an overexpression of truncated isoforms in CML cell lines with the accession to the stroma. The results suggest that intrinsic molecules such as TWIST-1 carriers and p73 isoforms are deregulated in CSL by extrinsic mechanisms involved with the leukemia microenvironment by the mechanism for participation in the phenomenon of drug resistance. This mechanism with its microenvironment.

Philadelphia (Ph) chromosome, as a result of reciprocal translocation, is in majority of cases connected to two types of leukaemia - chronic myelogenous (CML) and acute lymphoblastic (ALL). The translocation occurs within large intronic sequences of BCR and ABL genes. The breakpoints are specific for individual patient and may be used as a target for monitoring of leukemic burden (MRD, minimal residual disease) during the treatment. In general, MRD is an important prognostic factor, which influences the treatment intensity. Two standardized methods are currently used for its monitoring. The first one is based on the detection of clonal specific Immunoglobulin and/or T-cell receptor genes rearrangements (and thus cannot be used for CML cases) at the DNA level, the second one utilizes detection of the BCR/ABL fusion gene at the mRNA level. Our aim was to optimize and standardize the process to find individual patient breakpoints on Ph chromosome and to use it for MRD quantification. We found the breakpoint in 80 % cases. The MRD data from 15 patients obtained by our method were compared to the levels obtained by standard methods (Ig/TCR and BCR/ABL transcript quantification). In all but 1 patient we found significant discrepancies, raising the questions about leukemic origin and the most accurate method for...

Dissertation (Ph.D.)--University of South Florida, 2009. / Includes vita. Includes bibliographical references. Also available online.

Dissertation (Ph.D.)--University of South Florida, 2009. / Title from PDF of title page. Document formatted into pages; contains 149 pages. Includes vita. Includes bibliographical references.

Chronic myeloid leukemia (CML) is a clonal hematopoietic stem cell disorder associated with the Philadelphia chromosome (Ph) that arises from a reciprocal translocation between chromosomes 9 and 22, thereby resulting in the formation of the chimeric BCR-ABL oncogene encoding a constitutively activated tyrosine kinase. BCR-ABL tyrosine kinase inhibitors (TKIs) induce a complete hematologic and cytogenetic response in the majority of chronic phrase CML patients. However, TKIs cannot efficiently eradicate leukemia stem cells (LSCs) because of the insensitivity of LSCs to TKIs. Therefore, developing new strategies to target LSCs is necessary and critical for curing CML, and success of this approach depends on further understanding the molecular mechanisms by which LSCs survive and are maintained. In Chapter I, I briefly introduce CML disease, BCR-ABL oncoprotein, and TKIs. I also describe the identification and features of LSCs. Several key pathways in LSCs including Wnt/ß-catenin, hedgehog, FoxO, Bcl6 and HIF1, are discussed. I also propose our strategy to identify unique molecular pathways that are important for LSCs but not their normal stem cell counterparts. In Chapter II, I describe our finding about the function of the positive regulator, HIF1α, in CML development and LSC survival. I show that loss of HIF1α impairs the maintenance of CML through impairing cell cycle progression and inducing apoptosis of LSCs, and I also report that p16Ink4a and p19Arf mediate the effect of HIF1α on LSCs, as knockdown of p16Ink4a and p19Arf rescues the defective colony-forming ability of HIF1α-/- LSCs. As detailed in Chapter III and IV, through comparing the global gene expression profiles of LSCs and HSCs, I find two novel regulators, Blk and Scd1, which act as tumor suppressors in CML development. In Chapter III, I show that Blk is markedly down-regulated by BCR-ABL in LSCs, and that c-Myc and Pax5 mediate this down-regulation. Deletion of Blk accelerates CML development; conversely, Blk overexpression significantly delays the development of CML and impairs the function of LSCs. I also demonstrate that p27, as a downstream effector, is involved in the function of Blk in LSCs. Blk also functions as a tumor suppressor in human CML stem cells, and inhibits the colony-forming ability of human CML cells. In Chapter IV, I investigate the function of another negative regulator, Scd1, in CML LSCs, and find that expression of Scd1 is down-regulated in mouse LSCs and human CML cells. We report that Scd1 acts as a tumor suppressor in CML, as loss of Scd1 causes acceleration of CML development and overexpression of Scd1 delays CML development. Using a colony-forming assay, I demonstrate that Scd1 impairs the maintenance of LSCs due to the change of expression of Pten, p53 and Bcl2. Importantly, I find that both Blk and Scd1 do not affect normal hematopoietic stem cells (HSCs) or hematopoiesis. Taken together, our findings demonstrate that HIF1α is required for the maintenance of CML LSCs, and conversely that Blk and Scd1 suppress the function of LSCs, suggesting that combining TKI treatment with specific targeting of LSCs will be necessary for curing CML.

Une des principales causes d’échec dans le traitement des cancers est le développement de résistances aux drogues par les cellules tumorales. Les cellules souches cancéreuses (CSC) sont suspectées d’être responsables de ces rechutes, conduisant à la récurrence de la maladie et bien souvent au décès des patients. En clinique, il est donc nécessaire de développer des stratégies thérapeutiques capables de cibler ces CSC résistantes et aboutir à la guérison des patients. Les CSC sont régulées par un ensemble de signaux aussi bien biologiques que physiques au sein de la niche tumorale. Mon projet a pour objectif de déterminer l’implication du microenvironnement tumoral (voie de signalisation BMP et contraintes mécaniques) dans le maintien et la résistance des cellules souches leucémiques (CSLs) de la leucémie myéloïde chronique (LMC). Pour cela, nous avons combiné tests fonctionnels et moléculaires ainsi que l’analyse de la niche tumorale sur plus de 200 échantillons de patients atteints de LMC. Nous avons ainsi démontré que l’altération de la voie BMP intrinsèque aux cellules immatures de la LMC corrompt et amplifie la réponse à BMP2 et BMP4, présents en quantités anormalement abondantes au sein de la niche tumorale. Ces résultats récemment publiés dans Blood nous ont amenés à évaluer le rôle de la voie BMP dans le maintien des CSLs sous traitement par les ITK. La microscopie à force atomique nous a permis de démontrer que l’expression de BCR-ABL est suffisante pour induire une augmentation de la rigidité des cellules immatures de LMC par rapport à des cellules saines. Enfin, l’utilisation d’un système de confinement cellulaire nous a permis de démontrer que le stress mécanique contrôle la prolifération des cellules leucémiques immatures en régulant l’expression de gènes mécano-sensibles comme Twist-1. Ces résultats pourraient expliquer comment des CSLs tirent profit des contraintes mécaniques issues de leur microenvironnement afin d’acquérir un avantage prolifératif par rapport aux cellules saines. Ultimement, nous espérons que cette approche transdisciplinaire permettra d’identifier les molécules clés de la transduction de signaux mécaniques potentiellement impliqués dans le maintien et la résistance des CSC et ainsi proposer de nouvelles cibles pour contrer ces effets. / One of the main causes of treatment failure in cancers is the development of drug resistance by cancer cells. The persistence of cancer stem cells (CSCs) might explain cancer relapses as they could allow reactivation of cancer cells proliferation following therapy, leading to disease persistence and ultimately to patients’ death. Clinically, it is crucial to develop therapeutic strategies able to target resistant CSCs in order to cure the patients. CSCs are controlled by a variety of biochemical and biomechanical signals from the leukemic niche. My project aims to determine the involvement of the tumor microenvironment (BMP signaling pathway and mechanical stress) in the maintenance and resistance of Leukemic Stem Cells (LSCs) in Chronic Myelogenous Leukemia (CML). For this, we combined functional and molecular assays to the analysis of tumor microenvironment on more than 200 CML patients’ samples. We demonstrated that alterations of intracellular BMP signaling pathway in CP-CML primary samples corrupt and amplify the response to exogenous BMP2 and BMP4, which are abnormally abundant in the tumor microenvironment. These results, recently published in Blood led us to evaluate the role of the BMP pathway in LSC maintenance under TKI treatment. Atomic force microscopy allowed us to demonstrate that BCR-ABL expression alone is sufficient to increases the rigidity of immature CML cells compared to healthy ones. Finally, using a unique cell confining system, we were able to demonstrate that mechanical stress controls the proliferation of immature leukemic cells by regulating the expression of mechano-sensitive genes such as Twist-1. These results could explain how LSCs can benefit from a mechanical stress exerted by their microenvironment to acquire a proliferative advantage over normal cells. Ultimately, we hope that this transdisciplinary approach will help to identify key molecules in the transduction of mechanical signals potentially involved in maintenance and resistance of CSCs and thus offer new targets to counter these effects.

The human Philadelphia chromosome (Ph) arises from a translocation between chromosomes 9 and 22 [t(9;22)(q34;q11)]. The resulting chimeric BCR-ABLoncogene encodes a constitutively activated, oncogenic tyrosine kinase that induces chronic myeloid leukemia (CML) and B-cell acute lymphoblastic leukemia (B-ALL). The BCR-ABL tyrosine kinase inhibitor (TKI), imatinib mesylate, induces a complete hematologic and cytogenetic response in the majority of CML patients, but is unable to completely eradicate BCR-ABL–expressing leukemic cells, suggesting that leukemia stem cells are not eliminated. Over time, patients frequently become drug resistant and develop progressive disease despite continued treatment. Two major reasons cause the imatinib resistance. The first one is the BCR-ABL kinase domain mutations which inhibit the interaction of BCR-ABL kinase domain with imatinib; the second one is the residual leukemia stem cells (LSCs) in the patients who are administrated with imatinib. To overcome these two major obstacles in CML treatment, new strategies need further investigation. As detailed in Chapter II, we evaluated the therapeutic effect of Hsp90 inhibition by using a novel water-soluble Hsp90 inhibitor, IPI-504, in our BCR-ABL retroviral transplantation mouse model. We found that BCR-ABL mutants relied more on the HSP90 function than WT BCR-ABL in CML. More interestingly, inhibition of HSP90 in CML leukemia stem cells with IPI-504 significantly decreases the survival and proliferation of CML leukemia stem cells in vitro and in vivo. Consistent with these findings, IPI-504 treatment achieved significant prolonged survival of CML and B-ALL mice. IPI-504 represents a novel therapeutic approach whereby inhibition of Hsp90 in CML patients and Ph+ ALL may significantly advance efforts to develop a cure for these diseases. The rationale underlying the use of IPI-504 for kinase inhibitor–resistant CML has implications for other cancers that display oncogene addiction to kinases that are Hsp90 client proteins. Although we proved that inhibition of Hsp90 could restrain LSCs in vitro and in vivo, it is still unclear how to define specific targets in LSCs and eradicate LSCs. In Chapter III, we took advantage of our CML mouse model and compared the global gene expression signature between normal HSCs and LSCs to identify the downregulation of Pten in CML LSCs. CML develops faster when Pten is deleted in Ptenfl/fl mice. On the other hand, Pten overexpression significantly delays the CML development and impairs leukemia stem cell function. mTOR is a major downstream of Pten-Akt pathway and it is always activated or overepxressed when Pten is mutated or deleted in human cancers. In our study, we found that inhibition of mTOR by rapamycin inhibited proliferation and induced apoptosis of LSCs. Notably, our study also confirmed a recent clinical report that Pten has been downregulated in human CML patient LSCs. In summary, our results proved the tumor suppressor role of Pten in CML mouse model. Although the mechanisms of Pten in leukemia stem cells still need further study, Pten and its downstream, such as Akt and mTOR, should be more attractive in LSCs study.

A leucemia mielóide crônica (LMC) é caracterizada pela translocação (9;22) que dá origem ao gene quimérico BCR-ABL. Este gene codifica uma proteína com atividade tirosina quinase, p210, constitutivamente ativa. O três mecanismos envolvidos na patogênese da LMC são o aumento da proliferação celular, alteração da adesão celular ao estroma e matriz medular e inibição da apoptose. A introdução do mesilato de imatinibe (MI), um inibidor de tirosina quinase, revolucionou o tratamento da LMC levando pacientes em fase crônica a remissões duráveis, porém uma parcela destes não responde ou perde a resposta ao longo do tratamento. Os mecanismos de resistência ao MI podem ser classificados como independentes de BCR-ABL (a1- glicoproteína ácida e genes de resistência a múltiplas drogas) ou dependentes de BCR-ABL (superexpressão de BCR-ABL e mutações do domínio quinase do gene ABL). Objetivo: avaliar a presença de mutações no domínio quinase do gene ABL e a expressão dos genes de resistência a múltiplas drogas MDR1 e BCRP em amostras pré-tratamento com MI, acompanhar estes pacientes mensalmente através da quantificação de transcritos BCR-ABL e quando ocorrer resistência reavaliar a presença de mutações do domínio quinase do ABL e a expressão dos genes de resistência a múltiplas drogas. Material e Métodos: Foram avaliados 61 pacientes com LMC em fase crônica. A pesquisa de mutações do domínio quinase foi realizada pela técnica de seqüenciamento direto e a expressão relativa dos genes de resistência a múltiplas drogas foi avaliada por PCR em tempo real. A quantificação absoluta do número de transcritos BCR-ABL foi realizada pela técnica de PCR em tempo real utilizando-se o sistema Taqman de sondas de hibridização. Resultados: Nas amostras pré-tratamento dos 61 pacientes estudados não foram detectadas mutações. Quando relacionamos o aumento da expressão dos genes MDR1 e BCRP à resposta citogenética completa aos 12 meses de tratamento não houve diferença estatística significativa (p>0,05). Quanto ao número de transcritos BCR-ABL, observamos que os pacientes que apresentaram menos de 1% pela escala internacional aos 3 meses de tratamento atingiram a RMM em período menor (7 meses) do que os que apresentaram mais de 1% (12 meses) com diferença estatística significativa (p = 0,03). Conclusões: As mutações do domínio quinase do gene BCR-ABL nas amostras pré-tratamento não foram detectadas ou pela sensibilidade da técnica de seqüenciamento direto (10%) ou porque tais mutações são mais freqüentes nas fases acelerada e blástica. A expressão dos genes de resistência a múltiplas drogas (MDR1) e BCRP) em pacientes com LMC-FC ao diagnóstico não apresentou correlação com o aparecimento de resistência secundária ao MI. Além disso a quantificação mensal dos transcritos BCR-ABL aos 3 meses pode ser considerada um marcador com valor prognóstico. / Chronic myeloid leukemia is characterized by t(9;22) translocation. The chimeric gene BCR-ABL encodes a p210BCRABL protein with constitutive tyrosine kinase activity which is directly related to CML pathogenesis. The imatinib mesylate, a tyrosine kinase inhibitor, is the first-choice treatment for patients in chronic phase but some patients show primary resistance or relapse after initial response. The mechanisms of resistance to the imatinib mesylate treatment are BCR-ABL dependent (amplification of BCR-ABL and mutation of kinase domain of BCR-ABL) or independent of BCR-ABL (1-acid glycoprotein and expression of multidrug resistance genes). Objective: The objective of this work was to evaluate the mechanisms of resistance (kinase domain mutation and MDR1 and BCRP genes expression) to imatinib mesylate in pretreatment samples, quantify of BCR-ABL transcript on a monthly follow up plan, and to re-evaluate the mechanisms of resistance in the absence or loss of treatment response. Patients and Methods: We have evaluated 61 pretreatment samples derived from chronic phase CML patients. The number of BCR-ABL transcripts was quantified by RTQ-PCR with taqman probes and MDR1 and BCRP expression were evaluated by RTQ-PCR with Syber Green. Mutations within the BCR-ABL kinase domain were screened by direct sequencing and we also have screened the T315I mutation in pretreatment samples by allele-specific PCR. Results:We detected no mutations in the 61 pretreatment samples. The correlation analysis between the expression of MDR1/BCRP genes and the cytogenetic response at 12 months of treatment revealed no significant statistical difference (p = > 0.05). The results of BCR-ABL quantification in the follow up of our cohort indicated that patients who had transcripts <1% by the international scale at 3 months of therapy are more likely to achieve rapid MMR (median of 7 months) than those who had >1% (median of 12 months) (p = 0,03). Conclusions: As expected, the kinase domain mutations of BCR-ABL in pretreatment samples of CML chronic phase patients are not detectable by direct sequencing because of the sensitivity of the assay (10%) and also because these mutations are more common in accelerated phase and blast crisis. About the expression of multidrug resistance genes MDR1 and BCRP, they showed no correlation with secondary resistance to imatinib mesylate. And finally the number of BCR-ABL transcripts at 3 months of treatment can be considered a marker with prognostic value.

A leucemia mielóide crônica (LMC) é caracterizada pela translocação (9;22) que dá origem ao gene quimérico BCR-ABL. Este gene codifica uma proteína com atividade tirosina quinase, p210, constitutivamente ativa. O três mecanismos envolvidos na patogênese da LMC são o aumento da proliferação celular, alteração da adesão celular ao estroma e matriz medular e inibição da apoptose. A introdução do mesilato de imatinibe (MI), um inibidor de tirosina quinase, revolucionou o tratamento da LMC levando pacientes em fase crônica a remissões duráveis, porém uma parcela destes não responde ou perde a resposta ao longo do tratamento. Os mecanismos de resistência ao MI podem ser classificados como independentes de BCR-ABL (a1- glicoproteína ácida e genes de resistência a múltiplas drogas) ou dependentes de BCR-ABL (superexpressão de BCR-ABL e mutações do domínio quinase do gene ABL). Objetivo: avaliar a presença de mutações no domínio quinase do gene ABL e a expressão dos genes de resistência a múltiplas drogas MDR1 e BCRP em amostras pré-tratamento com MI, acompanhar estes pacientes mensalmente através da quantificação de transcritos BCR-ABL e quando ocorrer resistência reavaliar a presença de mutações do domínio quinase do ABL e a expressão dos genes de resistência a múltiplas drogas. Material e Métodos: Foram avaliados 61 pacientes com LMC em fase crônica. A pesquisa de mutações do domínio quinase foi realizada pela técnica de seqüenciamento direto e a expressão relativa dos genes de resistência a múltiplas drogas foi avaliada por PCR em tempo real. A quantificação absoluta do número de transcritos BCR-ABL foi realizada pela técnica de PCR em tempo real utilizando-se o sistema Taqman de sondas de hibridização. Resultados: Nas amostras pré-tratamento dos 61 pacientes estudados não foram detectadas mutações. Quando relacionamos o aumento da expressão dos genes MDR1 e BCRP à resposta citogenética completa aos 12 meses de tratamento não houve diferença estatística significativa (p>0,05). Quanto ao número de transcritos BCR-ABL, observamos que os pacientes que apresentaram menos de 1% pela escala internacional aos 3 meses de tratamento atingiram a RMM em período menor (7 meses) do que os que apresentaram mais de 1% (12 meses) com diferença estatística significativa (p = 0,03). Conclusões: As mutações do domínio quinase do gene BCR-ABL nas amostras pré-tratamento não foram detectadas ou pela sensibilidade da técnica de seqüenciamento direto (10%) ou porque tais mutações são mais freqüentes nas fases acelerada e blástica. A expressão dos genes de resistência a múltiplas drogas (MDR1) e BCRP) em pacientes com LMC-FC ao diagnóstico não apresentou correlação com o aparecimento de resistência secundária ao MI. Além disso a quantificação mensal dos transcritos BCR-ABL aos 3 meses pode ser considerada um marcador com valor prognóstico. / Chronic myeloid leukemia is characterized by t(9;22) translocation. The chimeric gene BCR-ABL encodes a p210BCRABL protein with constitutive tyrosine kinase activity which is directly related to CML pathogenesis. The imatinib mesylate, a tyrosine kinase inhibitor, is the first-choice treatment for patients in chronic phase but some patients show primary resistance or relapse after initial response. The mechanisms of resistance to the imatinib mesylate treatment are BCR-ABL dependent (amplification of BCR-ABL and mutation of kinase domain of BCR-ABL) or independent of BCR-ABL (1-acid glycoprotein and expression of multidrug resistance genes). Objective: The objective of this work was to evaluate the mechanisms of resistance (kinase domain mutation and MDR1 and BCRP genes expression) to imatinib mesylate in pretreatment samples, quantify of BCR-ABL transcript on a monthly follow up plan, and to re-evaluate the mechanisms of resistance in the absence or loss of treatment response. Patients and Methods: We have evaluated 61 pretreatment samples derived from chronic phase CML patients. The number of BCR-ABL transcripts was quantified by RTQ-PCR with taqman probes and MDR1 and BCRP expression were evaluated by RTQ-PCR with Syber Green. Mutations within the BCR-ABL kinase domain were screened by direct sequencing and we also have screened the T315I mutation in pretreatment samples by allele-specific PCR. Results:We detected no mutations in the 61 pretreatment samples. The correlation analysis between the expression of MDR1/BCRP genes and the cytogenetic response at 12 months of treatment revealed no significant statistical difference (p = > 0.05). The results of BCR-ABL quantification in the follow up of our cohort indicated that patients who had transcripts <1% by the international scale at 3 months of therapy are more likely to achieve rapid MMR (median of 7 months) than those who had >1% (median of 12 months) (p = 0,03). Conclusions: As expected, the kinase domain mutations of BCR-ABL in pretreatment samples of CML chronic phase patients are not detectable by direct sequencing because of the sensitivity of the assay (10%) and also because these mutations are more common in accelerated phase and blast crisis. About the expression of multidrug resistance genes MDR1 and BCRP, they showed no correlation with secondary resistance to imatinib mesylate. And finally the number of BCR-ABL transcripts at 3 months of treatment can be considered a marker with prognostic value.