The objective of this project is the research and development of suitable bioanalytical methods to: (1) establish the structure and purity of potential anti-AIDS agents and new antiviral drugs, (2) determine the physical, chemical and biochemical properties of these compounds and their metabolites, and (3) measure these drugs and their metabolites in biological samples to elucidate pharmacology and to determine plasma and intracellular pharmacokinetics. High-performance liquid chromatography (HPLC), capillary electrophoresis (CE) and mass spectrometry are the major analaytical tools that are employed. The orally active DNA methyltransferase inhibitor 2(1H)-pyrimidinone riboside (zebularine) and its analogues are currently the major compounds of interest. A range of bioanalytical methods have been and are being devised for the measurement of zebularine and its intracellular metabolites in biological samples. A collaborative effort has resulted in the development and validation of rapid and sensitive HPLC methods for the measurement of this agent in pharmaceutical media and biological samples. Zebularine exhibits impressive hydrolytic stability at acid and neutral pH and can be administered orally to rodents for extended periods in drinking water. The small sample size required for the measurement of zebularine in plasma allows pharmacokinetics to be determined in an individual rodent. This latter method is suitable for both current preclinical pharmacology studies and for future toxicology and clinical studies. Collaborative pharmacokinetic studies have been carried out to define the plasma kinetics and oral bioavailability of zebularine in individual rats. These preliminary studies showed rapid plasma clearance after intravenous bolus doses of 10 - 100 mg/kg zebularine. Oral administration of zebularine at equivalent doses indicated variable bioavailablity, ranging from low (1%) to moderate (31%). Additional studies to further assess zebularine bioavailability and disposition are planned so that a previously developed species-scalable physiological pharmacokinetic model for nucleoside-based prodrugs can be refined and extended. This model is being used to investigate the effects of various physiological and biochemical processes on drug disposition and activation, with emphasis on gastrointestinal absorption, blood-brain-barrier penetration into the CNS, and metabolic activation. Collaborative studies on the metabolic activation of zebularine have been conducted in selected human and murine cell lines. In T-24 bladder carcinoma cells, zebularine readily undergoes intracellular phosphorylation to form the corresponding 5'-mono-, di- and triphosphates in a dose- and time-dependent manner. In addition to these expected metabolites, a major phosphorylated conjugate containing the intact zebularine base is observed in all cell lines. Based on chromatographic, enzymatic and spectrospcopic evidence, we have tentatively identified this new metabolite as a 3',5'-cyclic diphosphoethanolamine adduct and postulate that it arises from coupling of zebularine-5'-triphosphate with ethanolamine. Zebularine is incorporated into both DNA and RNA with RNA incorporation predominating by 7- to 30-fold depending on the cell line. It is thought that incorporation of zebularine into DNA is required before the drug can function as an inhibitor of the methyltranferase by formation of a tight complex between it and the enzyme. The very limited DNA incorporation that we have observed suggests that this is the reason for the equivalent activity but lesser potency relative to other inhibitors of DNA methylation. The development of methods using capillary electrophoresis to measure intracellular nucleotide pools and metabolites continues with emphasis on the rapid, nonradiometric evaluation of the intracellular metabolic activation of zebularine. CE has been used to characterize the postulated diphosphoethanolamine adduct of zebularine and show that it is a potential depot source of zebularine-5'-monophosphate. Our previous work has demonstrated that a 100- to 160-fold signal enhancement can be obtained for the CE analysis of mixtures of synthetic nucleotides, but that a marked peak width broadening and loss of resolution is noted during sample stacking of biological samples. This deterioration in resolution is partially related to sample ionic strength. Sample and/or run buffer ionic strength has been controlled on an individual basis to enhance the determination of minor components of various synthetic nucleotide mixtures and to characterize nucleotide drug metabolites in cellular matrices such as cultured MOLT-4 cells. Sample preparation methods and analysis strategies to overcome this effect remain under investigation. CE with sample stacking also dramatically increases the speed and sensitivity of the determination of the oligonucleotide products generated in a palindromic oligonucleotide-directed enzymatic assay being developed for the measurement of intracellular deoxy- and dideoxynucleotides in order to more fully characterize various antiretroviral therapies. Ongoing research is currently directed toward the application of CE for the determination of intracellular nucleoside drug metabolism and to interfacing CE with mass spectrometry for structural analysis.
