Hepatic and intrahepatic targeting of an anti-inflammatory agent with human serum albumin and neoglycoproteins as carrier molecules.

The anti-inflammatory agent naproxen (Nap) was covalently coupled to human serum albumin (HSA) and to the neoglycoproteins, galactose and mannose terminated HSA, to deliver this drug selectively to different cell types of the liver. Disposition of Nap20-HSA was studied in rats and compared to that of equivalent doses of mixtures of uncoupled drug and protein. The liver to kidney ratios of the drug (L/K-Nap) and the protein (L/K-prot.) were increased, indicating an improved delivery of both protein and drug to the target site. After injection of 10 micrograms Nap20-HSA the L/K-prot. was increased 15.0 +/- 0.21-fold as measured 1 hr after injection. Even after injection of 5 mg of the conjugate, the L/K-prot. was enhanced 5.6 +/- 0.34-fold and the L/K-Nap 4.6 +/- 0.23-fold as measured 1 hr after injection. Immunohistochemical staining of liver slices revealed that the endothelial cells were the main sites for hepatic uptake. Further pharmacokinetic studies of Nap20-HSA in isolated perfused rat livers showed a saturable uptake process (Vmax = 2.46 micrograms/min/10.0 g liver and Km = 4.27 x 10(-6) M). The uptake in the liver could be inhibited by various polyanionic probes, indicating the major involvement of a scavenger receptor system in the internalization mechanism of Nap20-HSA. This endothelial uptake via the scavenger receptor system is likely to be related to the increased negative charge of the Nap-albumin conjugate as was revealed by anion exchange chromatography. Studies in the intact organ and in purified liver lysosomal lysates indicate that after internalization of Nap20-HSA the conjugate is proteolytically degraded leading to the formation of the lysine conjugate of Nap. This amino acid conjugate of Nap was shown in a previous study by us to be equipotent to Nap itself with regard to prostaglandin-E2 synthesis inhibition. A pronounced altered intrahepatic distribution was observed when Nap was coupled to lactosaminated and mannosylated HSA (Lact-HSA and Man-HSA, respectively). Coupling of Nap to Lact27-HSA and Man10-HSA resulted in a major shift in intrahepatic distribution from endothelial cells to the hepatocytes and Kupffer cells, respectively. We conclude that conjugation of Nap to HSA itself results in a selective delivery to endothelial cells and that the local proteolysis of the conjugate produces an active catabolite. Selective delivery to other cell types of the liver can be achieved by attaching naproxen to neoglycoproteins with an appropriate type and number of sugar groups.