Diffuse large B-cell lymphoma (DLBCL) is the most common form of non-Hodgkin lymphoma (NHL). The current standard of care is the combination of rituximab with cyclophosphamide, doxorubicin, vincristine and prednisone (R-CHOP), but this only results in a 60% over-all 5-year survival rate, thus highlighting a need for new therapeutic approaches. Histone deacetylase inhibitors (HDACi) are novel therapeutics that is being clinically evaluated for combination therapy. Rational selection of companion therapeutics for HDACi is difficult due to their poorly understood, cell-type specific mechanisms of action. To understand these mechanisms better, we developed a pre-clinical model system of response to the HDACi belinostat. Using this model system, we identified two major responses. Resistance, consisting of a reversible G1 cell cycle arrest with little induction of apoptosis; or sensitivity, consisting of mitotic arrest and high levels of apoptosis. In this dissertation, we determine that the induction of G1 cell cycle arrest is due to the increased expression of cyclin dependent kinase inhibitors (CDKi) that bind to and inhibit the cyclin E/CDK2 complex thereby blocking the final repressive phosphorylation steps of Rb protein. Repression of transcriptional elongation blocked CDKi upregulation and prevented G1 cell cycle arrest in belinostat-resistant cells. Additionally, we identified that belinostat arrests sensitive cells prior to metaphase and belinostat-resistant cells slow-down in mitosis but complete the process prior to arresting in G1. The combination of belinostat with the microtubule-targeting agent, vincristine resulted in strong synergistic induction of apoptosis by targeting mitotic progression. Furthermore, this combination prevents polyploidy, a key mechanism of resistance to microtubule targeting agents. Finally, we utilized selective class one HDAC inhibitors to identify the individual contributions of HDACs in the eliciting the responses observed with belinostat treatment. HDAC1&2 inhibition recapitulated the belinostat-resistant phenotype of G1 cell cycle arrest with little apoptosis, in both belinostat-resistant and sensitive cell lines. HDAC3 inhibition resulted in the induction of DNA damage, increased S phase and the induction of apoptosis in belinostat-resistant cells. Belinostat-resistant cells did not have observable effects to HDAC3 inhibitor alone but when combined with vincristine had significantly increased G2/M population at early time points. This suggests that HDAC3 maintains roles in DNA replication and also in mitotic progression. HDAC3 inhibition combined with vincristine resulted in a significant increase in polyploidy, suggesting that HDAC3 might not regulate the expression of apoptotic regulating factors as belinostat does.

Les agents pharmacologiques ciblant la dynamique des microtubules (MTs) sont très utilisés en chimiothérapie des cancers agressifs. Le paclitaxel (PTX) est utilisé depuis des décennies et donne de bons résultats pour le traitement des tumeurs solides. Plusieurs inconvénients, notamment ses effets secondaires et la résistance de certains cancers limitent cependant l'efficacité de ce médicament. Dans le but d'identifier de nouveaux composés pharmacologiques qui sensibilisent les cellules au PTX, nous avons recherché, parmi une collection de 8000 molécules, celles capables de sensibiliser des cellules cancéreuses à une dose non toxique de PTX. Nous avons ainsi sélectionné un composé de la famille des carbazoles : Carba1. Dans les cellules, l’association carba1/PTX a un effet cytotoxique supérieur à la somme des effets de carba1 et de PTX, quand ces molécules sont appliquées séparément, indiquant un effet synergique. De plus, des analyses approfondies de différents phénotypes ont permis de montrer que l'administration de carba1 avait pour conséquence d'amplifier des effets du PTX.À fortes doses, carba1 entraine un blocage des cellules en prométaphase mais n’altère pas le réseau microtubulaire, ni en interphase ni en mitose. En revanche, in vitro, carba1 cible la tubuline en se fixant sur le site colchicine, provoquant un retard et une diminution de la polymérisation des MTs. En plus de la tubuline, des études génétiques réalisées sur la levure suggèrent que carba1 a d'autres cibles dont CENP-E, kinésine essentielle à l’alignement des chromosomes au cours de la mitose.Des études menées sur un modèle de cancer mammaire murin agressif (allogreffes) ont révélé que carba1 seul et carba1/PTX ne présentaient aucune toxicité. De plus, les effets anti-tumoraux et anti-métastatiques de la combinaison carba1/PTX sur ces modèles se sont montrés encourageants, bien que des mises au point, notamment sur la posologie sont encore à prévoir. Carba1 est une molécule nouvelle, avec des applications jusque-là inconnues. C’est pourquoi une déclaration d’invention, en vue d'un dépôt de brevet, a été soumise au CNRS. / Microtubules (MTs) targeting agents are a powerful weapon in the war against aggressive cancers. Paclitaxel (PTX) has been used successfully for the treatment of solid tumors for decades. Several features, including side-effects and resistance of some cancers make this drug not always effective. With the aim to identify new chemical compounds that sensitize cells to paclitaxel we screened a library of 8,000 compounds, to select those not toxic for cell cultures when applied alone, that become toxic when applied in combination with a non-toxic dose of paclitaxel. This lead to the selection of a carbazole derivative: carba1. In cells, the carba1/PTX combination has a greater cytotoxic effect than the addition of the effects of each drug assayed separately, indicating a synergistic effect. In addition, in-depth phenotypic analyzes indicate that the administration of carba1 amplify the effects of PTX.High doses of carba1 induce a cell blockade in prometaphase, but do not alter the MT network in interphase or mitosis. In contrast, in vitro, carba1 targets the tubulin colchicine binding site, causing a delay and a decrease in MT polymerization. Genetic studies conducted on yeast indicated other potential additional targets including CENP-E, an essential kinesin for chromosome alignment during mitosis.Studies conducted on a preclinical mouse model of aggressive breast cancer (orthotopic grafts) revealed that carba1 alone and carba1/PTX showed no toxicity. In addition, the anti-tumor and anti-metastatic effects of the carba1/PTX combination on these models have been encouraging, but an optimization of the posology is still needed. Carba1 is a new molecule, with previously unknown applications. This is why a declaration of invention, with a view to filing a patent, has been submitted to the CNRS.

Oesophageal cancer is a malignant and aggressive disease with very poor survival. The aim of this thesis was to evaluate novel therapeutic targets in oesophageal cancer. In paper I, Hsp90 was expressed in all 81 oesophageal cancer tissues and also in nine oesophageal cancer cell lines. A specific Hsp90 inhibitor, 17-AAG, could efficiently inhibit cell proliferation, cell survival and sensitise oesophageal cancer cells to gamma photon irradiation. By inhibition of Hsp90 using 17-AAG, EGFR- and IGF-1R-mediated signalling was downregulated. In paper II, tumour samples from 80 oesophageal cancer patients were investigated for the expression of EGFR and LRIG1-3. Based on a total score of intensity and expression fraction a trend towards survival differences was found for LRIG2 (p=0.18) and EGFR (p=0.09). Correlation analysis revealed a correlation between expression of EGFR and LRIG3 (p=0.0007). Significant correlations were found between LRIG1 mRNA expression levels and sensitivity to cisplatin (r = –0.74), docetaxel (r = –0.69), and vinorelbine (r = –0.82). In paper III, microtubule targeting drugs podophyllotoxin (PPT), vincristine and docetaxel inhibited survival and proliferation of oesophageal cancer cells. Unexpectedly, experiments showed that microtubule destabilising agents inhibited EGFR phosphorylation and signalling. A tyrosine phosphatase inhibitor, sodium orthovanadate, was able to reverse the EGFR dephosphorylation. In paper IV, imetelstat, a telomerase antagonist, inhibited telomerase activity, colony formation ability and decreased proliferation of oesophageal cancer cells. Inhibition of telomerase activity by imetelstat led to an increase of 53BP1 foci indicating induction of DSBs. Furthermore, the fraction and size of radiation-induced 53BP1 foci were increased by imetelstat pre-treatment. In conclusion, Hsp90 and telomerase represent potential therapeutic targets in oesophageal cancer. And, the implication of EGFR and LRIG as prognostic factors is limited. Furthermore, disruption of the microtubule network may activate a protein tyrosine phosphatase that can regulate EGFR phosphorylation.