Cancer is the second leading cause of death in the United States (more than half a million people each year), and even with billions of dollars in medical effort patients are rarely cured. This dissertation research is devoted to meeting this medical need by providing new cancer therapeutics that are more potent and safer than current chemotherapies. This is achieved by using two state of the art anticancer “warheads”: 1) gold nanoparticle (AuNP) technology and 2) a new class of epigenetic anticancer small molecules, histone deacetylase inhibitors (HDACi). These warheads are then selectively delivered to cancer cells via “homing devices” targeted to receptors that are overexpressed in the cancers. This work primarily focuses on the androgen receptor (AR) to target prostate cancer. The 1st chapter sets the stage, providing scientific rationale and background for the central hypothesis: small molecules that engage the AR can, upon conjugation to a therapeutic agent, enable selective delivery of that agent to prostate cancer cells. Chapter 2 delves into the structural molecular biology of the androgen receptor. There is a survey of the crystallographic data for all nuclear receptors, providing structural information which is used to build AR homology models for antagonist and inverse agonist modes of ligand binding. These models are used to design AR targeting ligands (Chapters 3, 5, 6 and 7). The application of the targeting technology is illustrated by attaching them to AuNPs for selective delivery to prostate cancer cells (Chapter 3). Next, in order to appreciate the importance of using targeting agents in HDACi cancer therapeutics, we reviewed this recently emerged field in Chapter 4. In this chapter we argue that the failure of HDACi in solid tumors, despite more than 500 clinical trials in the last decade, is primarily due to an inability of these small molecules to accumulate at effective concentrations in the cancer. We provide an analysis of the paradigms being pursued to overcome this barrier, including HDAC isoform selectivity, localized administration, and targeting cap groups to achieve selective tissue and cell type distribution. In Chapter 5, this last approach (targeting cap groups, or a “homing device”) is illustrated with HDACi targeted to prostate cancer via antiandrogens that bind the AR. The second generation of improved “homing devices” is disclosed in Chapter 6 (for both AuNPs and HDACi), in addition to preliminary ADMET data and safety studies in mice. Excitingly, our three dimensional understanding of binding to the AR allowed design and structure-activity-relationship studies that lead to the first reported examples of AR inverse agonists (Chapter 7) Several points of significance: • AuNP targeted to AR ∙ have the strongest binding affinity ever reported (IC50 ~14 picomolar) ∙ are actively recruited to prostate cancer cells ∙ overcome treatment resistance in advanced prostate cancer cells ∙ exhibit nanomolar anticancer potency ∙ resolved the identity of the “membrane AR” as the GPRC6A • HDACi targeted to AR ∙ have HDACi activity and AR binding affinity superior to their clinical precursors ∙ exhibit potent AR antagonist activity ∙ induce AR translocation to the nucleus in a HDACi dependent fashion ∙ selectively and potently kill prostate cancer cells that express AR ∙ are safer than Tylenol®, as tested in small animals • Pure AR binding ligand studies ∙ resulted in the discovery of the first examples of AR inverse agonists, which are vastly more potent that clinically available antiandrogens for prostate cancer ∙ work via a never-before-seen mechanism of action, by localizing to the nucleus and recruiting corepressors to actively shut off AR genes

Platinum-based anticancer drugs, such as cisplatin, carboplatin, and oxaliplatin, have been approved for clinical use worldwide for decades. Despite their enormous success, their widespread application is hindered by either cross-resistance or toxic side effects, including nephrotoxicity and neurotoxicity. The need to overcome these drawbacks has stimulated the search for new platinum-based drugs. This dissertation will start with the accidental discovery of cisplatin, followed by an introduction of other platinum-based anticancer agents, including the action mechanism, general structures, and development history. Picoplatin is a newer generation of platinum-based anticancer agent. The bulky 2-methylpyridine as a non-leaving group on picoplatin could reduce the detoxification effect caused by thiol-containing species, such as glutathione and metallothionein, thus may grant picoplatin the ability to overcome cisplatin resistance. A convenient synthesis route for picoplatin derivatives has been developed. 11 new picoplatin derivatives have been designed by varying the bulkiness of the non-leaving amine group. All complexes have been characterized by different instrumentations, including MS, 1H NMR, 13C NMR, 195Pt NMR, HMQC, X-ray crystallography, and elemental analysis. Different bioassays, such as DNA binding, cell viability, and cellular accumulation, have been applied to evaluate their efficacy on cisplatin-sensitive ovarian cancer cell line A2780 and cisplatin-resistant ovarian cancer cell line A2780cis. The newly designed picoplatin derivatives show comparable efficacy with that of picoplatin and less resistance compared with cisplatin. The study of picoplatin derivatives laid the foundation toward the research of bifunctional platinum-based anticancer agents by incorporating histone deacetylase (HDAC) inhibition. Histone acetyltransferase (HAT) and histone deacetylase (HDAC) are a pair of important enzymes in epigenetic regulation. They work in harmony to acetylate and deacetylate histone lysine residues, resulting in a more relaxed or more condensed chromatin structure, respectively. HDAC has been found to be overexpressed in some cancer cells. Since 2006, 5 HDAC inhibitors (HDACi) have entered clinical use for cancer treatment. 19 new HDACi with additional coordination sites on the phenyl cap have been designed, synthesized, and evaluated. A few of the new HDACi show comparable or even better HDAC inhibition than that of Vorinostat (SAHA, the first FDA approved HDACi). A logical design would involve the installation of HDACi on the platinum center as a non-leaving group ligand. When the bifunctional drug reaches the cancer cell, the synergistic effect could be maintained as the relaxed chromatin structure makes DNA more susceptible to be attacked by the platinum centers, thus increase the anticancer activity and possibly selectivity toward cancer cells. 6 Pt-HADCi conjugates have been designed and synthesized. Dual functions of the new Pt-HDACi have been confirmed by DNA electrophoresis assay and HDAC inhibition assay. One of the Pt-HDACi (CF-101) shows comparable cytotoxicity with cisplatin and less resistance, which could be used as the lead compound for further structural modification and in vivo studies.

The enhanced expression of histone deacetylases (HDACs) in a variety of malignancies drew attention to investigate a new category of anti-cancer drugs that are based on the inhibition of those enzymes. Valproic acid (VPA) is a well-known antiepileptic drug that exhibits antitumor activities through inhibition of HDACs class I and IIa. Cancer stem cells (CSCs) have been recognized to drive the tumor growth and progression hence; attention has been given to target this small subpopulation of CSCs rather than the whole bulk tumor cells. CD133 is considered to be a CSC marker in several tumors and its transcription is strongly influenced by epigenetic changes that will be altered upon administration of histone deacetylase inhibitors (HDACi) in cancer treatment. Therefore, we evaluated the epigenetic and cytotoxic effects of treatment with 1 mM VPA in combination with other chemotherapeutics and its influence on the expression of CD133 in human neuroblastoma (NB) cell lines. Our results revealed that addition of VPA to DNA-damaging chemotherapeutics induced a synergistic anti-tumor effect that was associated with caspase-3 dependent induction of apoptosis in UKF-NB-4 cells. This synergism was related to the increase of the acetylation status of histones H3 and H4 and was only produced either by...

Diffuse large B-cell lymphoma (DLBCL) is a genetically heterogeneous disease with multiple distinct molecular subtypes. Increased NF-κB activity and expression of the microRNA miR-155 (product of the BIC gene) are associated with one subtype, called the activated B-cell (ABC) subtype. It is shown here that induction of NF-κB activity leads to increased miR-155 expression, the levels of miR-155 in a panel of B-lymphoma cell lines correlate with increased NF-κB activity, and the NF-κB p50/p65 heterodimer binds to a specific DNA site in the BIC promoter. Also described is a regulatory network wherein NF-κB-dependent up-regulation of miR-155 leads to reduced PU.1 transcription factor expression and consequently reduced PU.1-driven expression of B-lymphoma marker CD10 in the human B-lymphoma cell line BJAB. Genetic variation in DLBCL can be used to explain the response of individual patients to chemotherapy. One cancer therapeutic approach currently in clinical trials uses histone deacetylase inhibitors (HDACi's) as a monotherapy or in combination with other vi agents. It is shown here that two pan-HDACi's, trichostatin A and vorinostat, induce apoptosis in seven of eight human DLBCL cell lines. Ectopic over-expression of antiapoptotic proteins BCL-2 and BCL-XL or the pro-apoptotic protein BIM in select DLBCL cell lines can confer further resistance or sensitivity, respectively, to HDACi treatment. Additionally, the BCL-2 family antagonist ABT-737 can increase the sensitivity of several DLBCL cell lines to vorinostat-induced apoptosis, including the HDACi-resistant SUDHL6 cell line. Moreover, one vorinostat-resistant variant of the HDACi-sensitive cell line SUDHL4 has increased expression of anti-apoptotic proteins BCL-XL and MCL-1 and decreased sensitivity to ABT-737, and a second such variant cell line has increased expression of anti-apoptotic protein MCL-1. These results suggest that the balance of anti- to pro-apoptotic BCL-2 family protein expression is important in determining the sensitivity of DLBCL cell lines to HDACi-induced apoptosis. Thus, the sensitivity of DLBCL cell lines to treatment with HDACi's appears to depend on the complex regulation of BCL-2 family members, suggesting that the response of a subset of DLBCL patients to HDACi treatment may benefit from co-treatment with BCL-2 antagonists.

Nous avons mis en place au cours de ce travail de thèse différentes stratégies permettant d'améliorer l'efficacité thérapeutique des adénovirus (Ad) oncolytiques contre différents types de tumeurs. Une première stratégie a été de combiner un inhibiteur d'histone-désacétylase, l'acide valproique (VPA) avec un Ad oncolytique à capside sauvage E1Δ24 (CRAd) dans le traitement de carcinomes côliques. Nous avons dans un premier temps démontré que la combinaison du CRAd et du VPA permettait une diminution plus importante de la survie des cellules cancéreuses côliques comparé au simple traitement basé sur le CRAd ou le VPA in vitro mais aussi in vivo.De plus, nous avons observé que cet effet n'était pas lié à une meilleure réplication du CRAd par le VPA. En effet, le VPA provoquait un ralentissement de la réplication virale à des temps précoces mais ne modifiait pas la production virale. Nous avons également découvert que le co-traitement CRAd+VPA conduisait à une forte inhibition de la croissance cellulaire mais aussi à une mort cellulaire non apoptotique. Par ailleurs, nous avons mis en évidence que les cellules co-traitées par le CRAd et le VPA affichaient une forte polyploïdie accompagnée d'une augmentation de la phosphorylation de l'histone H2AX, un marqueur de dommages à l'ADN. Une deuxième stratégie a été de fournir aux Ad oncolytiques de nouvelles voies d'entrée afin d'infecter et de détruire plus efficacement des cellules de carcinomes rénaux réfractaires à l'infection adénovirale. Nous avons démontré que les CRAd à hexon modifié porteurs d'un ligand CKS-17 (Ad-HCKS-17-E1Δ24) ou à fibre modifiée de sérotype 3 (AdF3-E1Δ24) étaient capables d'infecter et de tuer plus efficacement ces cellules qu'un CRAd à capside sauvage in vitro. Malheureusement in vivo, les modifications de capside n'ont permis ni d'améliorer l'entrée des CRAd dans les tumeurs rénales, ni d'améliorer leur efficacité anti-tumorale. Cependant, nous avons observé qu'après administration intra-tumorale, les Ad à capside modifiée présentaient un plus faible tropisme hépatique comparé à un Ad à capside sauvage.

No description available.