In a model system employing Chinese hamster V-79 cells, the DNA synthesis inhibitor 3'-azido-3'-deoxythymidine (BW A509U, AZT) was shown to induce genome-wide DNA hypermethylation, low-frequency silencing of thymidine kinase (TK; EC 2.7.1.21) gene expression, and resistance to AZT. Twenty-four hours of exposure of V-79 cells to 150 ,uM AZT led to >2-fold enhancement of genomic 5-methylcytosine levels and produced TK-epimutants at a rate -43-fold above background. Such AZT-induced TK-epimutants were shown to be severely reduced in their capacity to activate AZT to its proximate antiviral form, AZT 5'-monophosphate, as compared with the TK+ parental cell line from which they were derived. TK-clones isolated under these conditions were shown to be 9-to 24-fold more resistant to the cytotoxic effects of AZT than the parental TK+ cell line and showed collateral resistance to 5-fluoro-2'-deoxyuridine. Three of four TK-epimutants could be reactivated at very high frequency (8-73%) to the TK+ AZT-sensitive phenotype by 24 hr of exposure to the demethylating agent 5-azadeoxycytidine (5-azadC), implying that drug-induced DNA hypermethylation, rather than classical mutation, was involved in the original gene-silencing event in these three clones. These 5-azadC-induced TK+ revertants concomitantly regained the ability to metabolize AZT to its 5'-monophosphate. RNA slot blot analyses indicated that the four AZT-induced TK-clones expressed 8.9%, 15.6%, 17.8%, and 11.1% of the parental level of TK mRNA. The three clones that were reactivatable by 5-azadC showed reexpression of TK mRNA to levels 84.4%, 51.1%, and 80.0% that of the TK+ parental cell line. These experiments show that one potential mechanism of drug resistance involves drug-induced DNA hypermethylation and resulting transcriptional inactivation of cellular genes whose products are required for drug activation.

EPI-2010 is a respirable antisense oligonucleotide (RASON), which selectively attenuates discordantly overexpressed adenosine A(1) receptors in allergic lung (Nature 1997;385:721). In the present study, aerosolized [(35)S]-labeled EPI-2010 (5 mg exposure; specific activity 0.055 Ci/mmol) was administered to normal rabbits by endotracheal tube to assess biodistribution, route of elimination, and potential cardiovascular toxicity. The animals were killed at 0, 6, 24, 48, and 72 h after inhalation of EPI-2010. Duplicate aliquots from different tissues and samples were solubilized and assessed for radioactivity. Approximately 1.4% of the total aerosolized EPI-2010 was deposited into the lung. The concentration of the drug in the lung at 0, 6, 24, 48, and 72 h was 64.0 +/- 1.5, 67.0 +/- 4.4, 32.0 +/- 3.7, 23.4 +/- 1.4, and 2.1 +/- 0.5 microg equivalents, respectively. Only a small amount of the radioactivity was detected in extrapulmonary tissues. By 72 h, 67.5% of the administered dose was excreted in the urine, which represented the major pathway of elimination. In postlabeling studies, intact full-length EPI-2010 could only be detected in the lung. Autoradiographic analysis after inhalation of [(35)S]-labeled EPI-2010 showed a relatively uniform deposition of drug throughout the lung. The aerosolized EPI-2010 did not have any significant systemic effects on the cardiovascular system as determined by Cardiomax-II analysis. This pattern of distribution and the lack of effect on cardiovascular function support the concept that RASONs offer the potential to safely address respiratory targets for which systemic distribution and systemic bioavailability may be contraindicated.

Respirable antisense oligonucleotides (RASONs) represent a novel class of respiratory therapeutic molecules with the potential to specifically address the challenges posed by the successes of the Human Genome Program, namely, the need to rapidly identify the critical pulmonary disease-relevant drugable targets from the vast pool of 30,000-40,000 human genes and to discover and develop drugs that specifically attack these targets. We have shown that EPI-2010, a RASON targeting the adenosine A1 receptor, a G-protein coupled receptor that has been implicated in the regulation of three major determinants of asthma, can be delivered directly to the target disease tissue as an aerosol formulation. In vivo efficacy, absorption, distribution, metabolism, and excretion (ADME), and safety studies of inhaled EPI-2010 employing animal models of human asthma suggest that the RASON approach enables the specific delivery of efficacious, safe, and long-acting doses of phosphorothioate oligonucleotides to the respiratory tract. Moreover, these data indicate that RASONs truly have the potential to address the respiratory drug discovery bottleneck of the postgenomic era, that is, the ability to rapidly validate disease targets and develop pulmonary disease therapeutics for these validated targets.

No abstract