In the field of multidrug resistance mediated by the multidrug transporter, P glycoprotein, which is encoded by the MDR-1 gene, our efforts continue to have a major focus on translational research, while trying to pursue basic investigations that have the potential for future clinical correlations. We have identified gene rearrangements as the mechanism responsible for the activation of MDR-1 in a large number of cell lines, and in patient samples. These rearrangements occur randomly and are characterized by the juxtaposition of a transcriptionally active gene 5' to MDR-1, thus avoiding disruption of MDR-1 structure. Our current efforts are directed at identifying the sites and mechanisms of gene rearrangement and we are currently directing efforts at identifying the breakpoints in these rearrangements, and have nearly completed this work having identifed seven sites of rearrangements. Our understanding of this process should be very valuable in furthering our understanding of the acquisition of drug resistance. While the occurrence of this phenomenon in clinical samples remains to be expanded, its demonstration in several samples from patients with refractory ALL, indicates this may be important in a defined group of patients, and our efforts are increasingly focused in this direction. Our efforts in this regard will be directed not only at identifying the frequency with which this phenomenon occurs clinically, but also efforts at understanding how this occurs and how it might be prevented. With regard to the latter we have completed studies examining the frequency with which this occurs as a function of the mode of drug administration. Specifically, we sought to answer whether administration as a bolus or as a continuous infusion can significantly affect the occurrence of chromosomal aberrations. These studies were conducted in a primate model by looking at the frequency of chromosomal damage in normal bone marrow following the administration of either bolus or infusional drug. The drugs selected include VP-16, thiotepa and paclitaxel. The data gathered using these three drugs shows a significant difference with less chromosomal damage seen following infusional therapy than following bolus administration. Our current investigations with MDR-1 arose out of studies which revealed a low frequency of acquired mutations in MDR-1. This prompted us to look at another mechanisms of drug resistance for comparison. We began to actively pursue studies aimed at identifying non-Pgp mechanisms of paclitaxel resistance. Selections performed with paclitaxel in the presence of verapamil succeeded in isolating cell lines with acquired resistance to paclitaxel that did not over-express MDR 1. Characterization of these cells led to the identification of mutations in the predominant tubulin isotype, M40. Similar studies performed using two additional tubulin stabilizing agents, epothilone A and epothilone B, led to the isolation of drug resistant cells which were also shown to harbor mutations in beta tubulin. Using a refined molecular model of tubulin, and guided by the mutations identified in our drug resistant cell lines and the cross-resistance profile of these cells, we were able to dock both paclitaxel and the epothilones in the putative binding pocket and propose a common pharmacophore for the taxanes, and the epothilones. The identification of the sites as distinct from those of other microtubule active agents, supports abundant pre-existing data, and provides excellent models to study the mechanism of paclitaxel and epothilone resistance, and the potential synergism of other agents in drug activity. Subsequent to that we have expanded our studies to the hemiasterlins, another class of microtubule active agents that look very promising in early studies. With these agents we have again succeeded in identifying acquired mutations, that in this case appear to function not by altering the binding of drug to tubulin, but rather by changing the physical properties of tubulin, so that it is intrinsically more stable, and therefore more resistant to depolymerization. We have also used these models to examine the role of p53 mutations in the acquisition of paclitaxel and epothilone resistance. Our studies indicate that the occurrence of a non-functional p53 early in the selection process is a universal finding, although the mechanism by which this occurs differ. We have also shown that p53 traffics on microtubules and that nuclear accumulation following DNA damage requires both an intact microtubule network and motor proteins of the dynein family. Our current efforts are dirested at identifying the interaction of p53 with dyneien, and we are making progress in this direction. We have identified cels with naturally occurring p53 mutations in which p53 does not bind dynein and is impaired in its nuclear accumulation, despite the presence of a normal nuclear localization signal. Studies in further pursuit of this are underway. In the clinic we have been involved in the conduct of clinical trials with a novel epothilone B analog, BMS-247550, and have conducted numerous translational studies in conjunction with these phase I and phase II trials. In these studies we have been able to demonstrate drug interaction with its target tubulin. The latter has been observed even in patients without clinical benefit from the drug, prompting studies to idenfity potentially distal mechanisms of resistance, as well as investigations designed to understand the kinetics and dynamics of drug/tubulin/microtubule interaction. Finally, we have been actively studying the expression of cell specific genes in endocrine neoplasms and how this might be potentially exploited in a gene therapy strategy. Studies with adrenal cancers and thyroid cancers indicate that expression of specific genes occurs in these cells, and that the expression of these genes can be modulated by the addition of differentiating agents which target the cAMP pathway, and novel agents that target histone deacetylases. These studies which arose out of our clinical trials in patients with adrenocortical cancer, are designed to find more specific and effective alternatives for the treatment of endocrine cancers, most of which are composed of very unique cells. As part of these studies we also came to recognize that histone deacetylase inhibitors could effectively regulate gene expression. In the treatment of thyroid cancer, we have shown that these agents can increase the expression of the sodium/iodide symporter resulting in increased iodine accumulation. We were also able to demonstrate that these agents could also be used to upregulate the expression of the coxsackie/adenovirus receptor, and in turn increase the efficiency of adenoviral infection. This latter observation has the potential to improve the delivery of adenovirus to cells, and may increase both the efficacy and specificity of this therapeutic approach. Studies to investigate this are ongoing, but to date have shown in several in vivo models that teatment with the histone deacetylase inhibitor, depsipeptide (FR901228, FK228) can increase CAR expression selectively in tumors, without having any effect on normal tissues. These studies have demonstrated remarkable specificity as well very marked increases in CAR expression in the tumor, raising hopes that indeed we may be able to selectively enhance adenovirus targeting of tumors. Studies are underway to further establish this.
Showing the most recent 10 out of 15 publications
