Several studies have shown improved efficacy of cholesteryl-conjugated phosphorothioate antisense oligodeoxynucleotides. To gain insight into the mechanisms of the improved efficacy in vivo, we investigated the disposition of ISIS-9388, the 3'-cholesterol analog of the ICAM-1-specific phosphorothioate oligodeoxynucleotide ISIS-3082, in rats. Intravenously injected [(3)H]ISIS-9388 was cleared from the circulation with a half-life of 49.9 +/- 2.2 min (ISIS-3082, 23.3 +/- 3.8 min). At 3 h after injection, the liver contained 63.7 +/- 3. 3% of the dose. Compared to ISIS-3082, the hepatic uptake of ISIS-9388 is approximately 2-fold higher. Endothelial, Kupffer and parenchymal cells accounted for 45.7 +/- 5.7, 33.0 +/- 5.9 and 21.3 +/- 2.6% of the liver uptake of [(3)H]ISIS-9388, respectively, and intracellular concentrations of approximately 2, 75 and 50 microM, respectively, could be reached in these cells (1 mg/kg dose). Preinjection with polyinosinic acid or poly-adenylic acid reduced the hepatic uptake of [(3)H]ISIS-9388, which suggests the involvement of (multiple) scavenger receptors. Size exclusion chromatography of mixtures of the oligonucleotides and rat plasma indicated that ISIS-9388 binds to a larger extent to high molecular weight proteins than ISIS-3082. Analysis by agarose gel electrophoresis indicated that ISIS-9388 binds more tightly to plasma proteins than ISIS-3082. The different interaction of the oligonucleotides with plasma proteins possibly explains their different dispositions. We conclude that cholesterol conjugation results in high accumulation of phosphorothioate oligodeoxynucleotides in various liver cell types, which is likely to be beneficial for antisense therapy of liver-associated diseases.

We previously modulated, by conjugating a single cholesterol, plasma protein binding and liver cell uptake of a phosphorothioate oligodeoxynucleotide (PS-ODN). In this study, we investigated the biological fate of a PS-ODN, denoted ISIS-9389 (3Ј,5Ј-bis-cholesteryl-conjugated ISIS 3082), provided with two cholesteryl moieties. After intravenous injection of into rats, [ 3 H]ISIS-9389 was cleared from plasma with a half-life of 23.6 Ϯ 0.3 min. After 90 min (approximately 95% cleared), the liver contained 83.0 Ϯ 0.8% of the dose. Spleen and bone (marrow), which constitute with the liver the reticuloendothelial system, contained 3.1 Ϯ 0.3 and 4.3 Ϯ 0.2%, respectively. All other tissues accumulated together Ͻ5% of the dose. The hepatic uptake of [ 3 H]ISIS-9389 occurred mainly by endothelial cells (51.9 Ϯ 6.4% of the liver uptake). Parenchymal and Kupffer cells were responsible for 24.9 Ϯ 7.7 and 23.3 Ϯ 2.5%, respectively. Preinjected polyinosinic acid and polyadenylic acid reduced hepatic uptake, albeit the latter was less effective. This finding suggests implication of (multiple) scavenger receptors in liver uptake of ISIS-9389. The interaction of ISIS-9389 with plasma proteins, analyzed by size exclusion chromatography, differs from that of unconjugated PS-ODN and PS-ODN with a single cholesterol. Plasma-incubated ISIS-9389 was mainly recovered as a high molecular weight complex. In conclusion, conjugation of PS-ODNs with two cholesteryl moieties results in almost quantitative uptake by the liver. The liver targeting exceeds the already impressive gain in liver uptake achieved by conjugation of a single cholesterol, and is expected to increase the therapeutic activity against liver-associated targets and reduce side effects in nonhepatic tissues.

No abstract