Tumor targeting based on the effect of enhanced permeability and retention (EPR) and the mechanism of receptor-mediated endocytosis (RME).

This review is focused on the macromolecular drug carrier systems by the effect of enhanced permeability and retention (EPR) and the mechanism of receptor-mediated endocytosis (RME). The effect of EPR is thought to be useful for the targeting of the macromolecular drugs to the tumor tissues on a vasculolymphatic level. The RME reveals the selective recognition, high affinity binding, and immediate internalization for the ligand on a cellular level. In the receptor, recognizing transferrin, a level of expression on the tumor cells is higher than that on the normal cells. We have used serum albumin and transferrin as drug carriers to deliver mitomycin C (MMC) to the tumor tissues and into the tumor cells. The properties of the conjugates of MMC to serum albumin and transferrin were examined in vitro and in vivo. We concluded that MMC could be delivered to the tumor tissue and cells by the use of albumin and transferrin as drug carriers.

Biodisposition of FITC-labeled aloemannan in mice.

Biodisposition of FITC-labeled aloemannan (F-AM) with the homogenate from some organs in mice was demonstrated. F-AM was metabolized only by the mucosa from the large intestine into smaller molecules that were effectively absorbed in mice. The homogenate from the other tissues did not affect the metabolism of F-AM. The degraded product (1) of F-AM after incubation with 10% feces homogenate for 24 h was chromatographed on a highly porous polymer and a Sephadex LH-20 column to provide an FITC-degraded fraction (2), which was shown to have a molecular weight of 800 D on Sephadex G-25 gel permeation. Metabolite 2 was examined by physicochemical methods and shown to be a mixture of FITC-hexose and -2 hexose on FAB-MS. An FITC-degraded fraction (3) with a molecular weight of 3 KD was obtained by 6-h incubation with 10% feces homogenate on Sephadex G-25 column chromatography and was shown to be a mixture of FITC-9 and 12 x hexose on TOF-MS.

Receptor mediated endocytosis and cytotoxicity of transferrin-mitomycin C conjugate in the HepG2 cell and primary cultured rat hepatocyte.

Intracellular disposition and cytotoxicity of macromolecular conjugate of mitomycin C (MMC) with transferrin (TF) were examined in the human hepatoma cell line HepG2 cell and normal cultured rat hepatocyte. The conjugate (TF-MMC) was specifically bound to the HepG2 cell as well as TF. The number of the binding site and the association constant of TF-MMC in the HepG2 cell were 396000+/-31000 molecules/cell and 3.24 x 10(7)+/-0.58 x 10(7) M(-1), respectively. No difference in the binding parameters of TF-MMC and TF can be detected in the HepG2 cell. The association constant for the TF receptor was almost identical between HepG2 cell and hepatocyte, however, the numbers of the binding site of TF-MMC and TF in the HepG2 cell were from 40-times to 50-times greater than those in the hepatocyte. Furthermore, TF-MMC was internalized into the HepG2 cell and the hepatocyte as well as TF. The rates of internalization of TF-MMC and TF into the HepG2 cell were nearly identical to those into the hepatocyte. However, the levels of the internalization into the HepG2 cell were remarkably higher than those into the hepatocyte because the number of receptors in the HepG2 cell was larger than that in the hepatocyte, and the rate of release from the HepG2 cell was slower than that from the hepatocyte. TF-MMC inhibited the growth of the HepG2 cells. The 50% growth inhibition (GI50) of TF-MMC against the HepG2 cell was 0.9 microg MMC/ml, which was a little higher than that of MMC (GI50=0.5 microg/ml). These results indicated that the TF-MMC might be useful for delivery of MMC to the HepG2 cell.

Evidence for receptor-mediated hepatic uptake of pullulan in rats.

Fluorescein-labeled pullulan (FP-60; MW 58,200) was prepared by reaction with FITC according to the method of de Belder and Granath. The hepatic distribution of FP-60 was examined using a specific high-performance size-exclusion chromatography. Intravenously administered FP-60 was rapidly eliminated from the blood circulation followed by an appreciable distribution to the liver. A marked dose-dependency was seen in the hepatic uptake of FP-60 which was markedly reduced by the coadministration of both asialofetuin and arabinogalactan. Measurement of the hepatocellular localization demonstrated the overwhelming distribution of FP-60 in the parenchymal liver cell fraction. Furthermore, microscopic examination revealed that FP-60 was effectively endocytosed by the parenchymal liver cells. Radiolabeled pullulan ([(125)I]P-60) was prepared by (125)I-labeling the tyramine derivative of pullulan which was synthesized by the cyano-transfer method. [(125)I]P-60 was predominantly accumulated in sliced rat liver tissue at 37 degrees C, which was drastically inhibited by the addition of both asialofetuin and arabinogalactan. The kinetic parameters of the specific binding of [(125)I]P-60 to monolayered hepatocytes at 0 degrees C were almost identical to those for asialofetuin. The binding of [(125)I]P-60 to isolated parenchymal cells was significantly inhibited by arabinogalactan and asialofetuin, however dextran, the same glucan as pullulan, did not affect the binding of [(125)I]P-60. It was found that pullulan, which is bound to the asialoglycoprotein receptor with high affinity, is subsequently internalized to the hepatocyte via receptor-mediated endocytosis.

Pharmacokinetics and biodisposition of fluorescein-labeled arabinogalactan in rats.

Fluorescein-labeled arabinogalactan (FA) was prepared by the reaction with FITC in methyl sulphoxide according to the method of deBelder and Granath. A systemic kinetic analysis of FA in rats was carried out by using a specific high-performance size-exclusion chromatography. Intravenously administered FA was rapidly eliminated from the blood circulation followed by an appreciable distribution to the liver and kidney. FA was accumulated in these organs over a long period whereas negligible levels of FA were detected in the other organs. A marked dose-dependency was seen in the hepatic uptake of FA which was markedly reduced by coinjected asialofetuin whereas the renal uptake of FA was not altered. Measurement of the hepatocellular localization demonstrated the overwhelming distribution of FA in the parenchymal liver cell fraction. Furthermore, the microscopic examination revealed FA that was effectively endocytosed by the parenchymal liver cells. These results suggested that FA which is bound to the asialoglycoprotein receptor with a high affinity is subsequently internalized to the hepatocyte via receptor-mediated endocytosis. FA was partially activated by periodate oxidation in order to acquire aldehyde groups to which guest molecules can be bound. A 12.5% oxidized arabinogalactan keeping a hepatocellular targetability showed a good conjugating reactivity to guest molecules via Schiff-base formation or by reductive amination. It was suggested that arabinogalactan can serve as a potential carrier for the delivery of enzymes and drugs to the parenchymal liver cells via the asialoglycoprotein receptor.

In vivo metabolism of aloemannan.

The metabolism of fluoresceinyl isothiocyanate labeled aloemannan (FITC-AM) was examined by p.o. and i.v. administration in mice at a dose of 120 mg/kg. Analysis of FITC-AM in urine and feces showed that FITC-AM (MW 500 KD) was metabolized into smaller molecules that mainly accumulated in the kidneys. AM was catabolized by the human intestinal microflora to catabolites 1 and 2 with molecular weights of 30 and 10 KD, respectively. Hydrolysis of AM showed hexosamine peaks on HPAE. The findings suggest that the immunomodulation of AM may come from not only neutral polysaccharides but also contaminated hexosamine in AM.

Intracellular disposition and cytotoxicity of transferrin-mitomycin C conjugate in HL60 cells as a receptor-mediated drug targeting system.

A macromolecular conjugate of mitomycin C (MMC) with transferrin (TF) which possessed binding ability for TF receptor was synthesized. The conjugate (TF-MMC) was internalized into the human leukemia cell line HL60 cells and distributed into intracellular fractions, then exocytosed into an incubation medium. Although these phenomena were similar to those of TF, part of the internalized TF-MMC was degraded to a trichloroacetic acid (TCA)-soluble fraction. Therefore, the intracellular disposition of the conjugate was analyzed kinetically. The mean time of internalization of TF-MMC (7.14 min) was longer than that of TF (5.46 min). The mean exocytosis time of TF-MMC (22.1 min) was also longer than that of TF (13.0 min). Although elongation of both the internalization and exocytosis steps was responsible for the increase in recycling time of the conjugate, the binding process to the TF receptor in the internalization stage was found to be markedly retarded. The recycling times of TF-MMC and TF were 29.2 and 18.5 min, respectively. The mean decomposition time of TF-MMC was 76.3 min. Proliferation of HL60 cells was inhibited by TF-MMC in vitro. These results indicate that the TF-MMC was internalized via a TF receptor and a part of the internalized TF-MMC was degraded, so the released MMC might represent antitumor activity. TF-MMC was demonstrated to be a useful hybrid as a receptor-mediated targeting system.

Polysaccharides as drug carriers: biodisposition of fluorescein-labeled dextrans in mice.

The biodisposition of fluorescein-labeled dextrans (FDs) with different molecular weights (MW = 4-500 kDa) was systematically examined in mice. After intravenous injection of FDs at a dose of 120 mg/kg, the levels of FDs in the blood circulation and in the various organs were measured fluorometrically. FDs with a molecular weight lower than 20 kDa showed poor hepatic distribution (2.1-3.7% of dose/g tissue) due to their rapid elimination from the blood circulation. FDs with higher molecular weights were appreciably distributed in the liver (18.9-24.0% of dose/g tissue) and accumulated there over a long period, whereas the FD levels in the other organs were almost negligible a few days after injection. The hepatic mean residence time of FDs ranged from 22.5 to 28.1 d. Partial depolymerization of FDs which accumulated in the liver was observed within 10 d after administration. The hepatic uptake clearance of FDs was decreased with an increase in molecular weight. A marked molecular weight dependency was also seen in the urinary and fecal excretions of FDs. An appreciable dose-dependency was demonstrated in the hepatic uptake of FDs (MW = 40 kDa), as well. The amount of hepatic uptake as a function of dose showed saturation kinetics and was analyzed by a Michaelis-Menten type equation. The apparent values of K(m) (dose) and Vmax (hepatic level) estimated were 116 +/- 5 mg/kg and 1.10 +/- 0.05 mg/g tissue, respectively.

Synthesis of transferrin-mitomycin C conjugate as a receptor-mediated drug targeting system.

Macromolecular conjugates of mitomycin C (MMC) were synthesized by binding an active ester of glutarylated MMC (MMC-G-OSu) to human holo-transferrin (TF). Water-soluble TF-MMC conjugates (TF-G-MMC) were obtained in a good yield (> 95%) by this method. The MMC content of the conjugate increased (0.82-9.49 MMC/w%) with increasing amounts of MMC-G-OSu added to the conjugation mixture. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis showed no aggregation in these conjugates. 125I-TF-G-MMC was bound specifically to the TF receptor on Sarcoma 180 cells; the measurement of equilibrium binding of the 125I-labeled conjugate resulted in a saturation isotherm. The amount of conjugate specifically bound to the TF receptor decreased as the MMC content of the conjugate increased. However, it was found that the conjugate with an MMC content below 10 mol MMC/mol TF still retains a binding activity of more than half that of TF. Therefore, when an optimal chemical modification was chosen, TF could be used as a tumor specific drug carrier.

Properties of water-insoluble mitomycin C-albumin conjugate as a sustained release drug delivery system in mice inoculated with Sarcoma 180.

In order to improve the deposition of mitomycin C (MMC) in the administered site, water-insoluble mitomycin C-albumin conjugate (MMC-G-BSA) was prepared. MMC was covalently attached to the glutarylated bovine serum albumin (G-BSA) in the presence of 1-ethyl-3-(3-dimethyl-aminopropyl) carbodiimide hydrochloride (EDC) to give MMC-G-BSA. The MMC content of the conjugate (16.3% (w/w)) was higher than that of water-soluble mitomycin C-albumin conjugates. MMC was liberated from MMC-G-BSA suspended in a phosphate buffer (pH 7.4, 37 degrees C) with a half-life of 155.3 h. In the same buffer system containing alpha-chymotrypsin, MMC-G-BSA was dissolved perfectly within 24 h due to enzymatic degradation, and the liberation of MMC from the conjugate was significantly accelerated (t1/2 = 24.5 h). After intraperitoneal injection in mice, most of the MMC-G-BSA was retained in the abdominal cavity. Furthermore, the survival time of mice inoculated with Sarcoma 180 was significantly increased by the intraperitoneal injection of MMC-G-BSA. These findings suggest that MMC-G-BSA is a biodegradable macromolecular hybrid which acts as a sustained release delivery system of MMC.

Pharmacokinetics of glutathione-dextran macromolecular conjugate in mice.

A fluorescein-labeled dextran-glutathione conjugate (FD-GSH) was synthesized in order to examine its disposition in the body. GSH was covalently attached to the FITC-labeled dextran by the cyanogen bromide activation method. Mice were injected with FD-GSH through the tail vein, and the levels of FD-GSH in the blood and various organs were measured fluorometrically. A substantial level of FD-GSH was found in the liver and this reached a maximum at 6-8 h after the injection. The hepatic uptake clearance was estimated to be 0.541 +/- 0.014 ml/h/g tissue or 42.4 +/- 9.8 ml/h/kg body weight. FD--GSH accumulated in the liver for a long period, while the half-life of the conjugate in the blood circulation was 1.45h. The cumulative urinary and fecal excretions of FD-GSH were 14% and 4% of dose at 72h after the injection, respectively. A molecular design of the conjugate was discussed on the basis of the results.

A protective effect of glutathione-dextran macromolecular conjugates on acetaminophen-induced hepatotoxicity dependent on molecular size.

Glutathion (GSH) was covalently attached to dextrans with various molecular weights of 2, 5, 10, 40, and 70 kDa by the cyanogen bromide activation method. The conjugates obtained synthetically were white or pale yellowish powders containing 6-10% (w/w) of GSH. The average molecular weights of the conjugates were estimated to be larger and the molecular weight distribution was a little broader than that of each original dextran. The conjugates significantly stabilized GSH and liberated it gradually under physiological conditions (t1/2 = 0.99-1.6h). Mice depleted of GSH by treatment with buthionine sulfoximine, a potent inhibitor of gamma-glutamylcysteine synthetase, exhibited a significant increase in hepatic GSH level after intravenous injection of the conjugates. In mice given a hepatotoxic dose of acetaminophen, the survival rate increased progressively with coadministration of the conjugates, whereas a small improvement was found when free GSH was given. The conjugate of GSH attached to dextran with the molecular weight of 40 kDa exhibited the highest prophylactic effect on acetaminophen-induced hepatotoxicity in mice. The prolonged retention of the conjugates of larger molecular weight in the circulation would cause a higher hepatic accumulation. These results suggested that molecular size would be the most critical factor in the delivery of GSH, as a dextran conjugate, into the liver.

The disposition of serum proteins as drug-carriers in mice bearing Sarcoma 180.

The tumor distribution and the disposition of serum proteins, such as albumin, fetuin, transferrin, and IgG, were investigated in mice bearing Sarcoma 180. Serum proteins labeled with fluoresceinisothiocyanate (FITC) were administered to the mice. The FITC-labeled proteins acylated with glutaric anhydride were also administered to the mice in order to investigate the effect of chemical modification. The plasma concentration of each glutarylated serum protein was significantly lower, about 15 to 46-fold, in comparison to that of the non-acylated protein at 24 h after administration. The tissue distributions of the glutarylated serum proteins were also decreased compared to those of the non-acylated proteins. Especially, the hepatic distribution of albumin and IgG was significantly reduced with glutarylation. The urinary excretion of albumin and transferrin, and fecal excretion of IgG, were significantly increased with glutarylation. The serum proteins were accumulated effectively in the tumor tissue in mice bearing Sarcoma 180. It was found that the tumor distributions were not impaired by the glutarylation, except involving fetuin. It was suggested, therefore, that the glutarylated serum proteins were valuable for relative tumor-selectivity and might be utilized in a macromolecular carrier system for antitumor drugs.

Properties of mitomycin C-albumin conjugates in vitro and in vivo.

Mitomycin C, an antineoplastic agent, was covalently attached to bovine serum albumin through a spacer of the glutaryl group. Two different synthetic methods were adopted; one was by the prior glutarylation of albumin followed by binding to mitomycin C, and the other was by the synthesis of glutarylated mitomycin C followed by binding to albumin. Physicochemical properties of the conjugates, such as Stokes radius, molecular weight, and helical content, were comparatively examined. The glutarylation of albumin resulted in an increase in Stokes radius of the protein due to the conformational change. The conjugates significantly stabilized mitomycin C and liberated it gradually under the physiological condition (t1/2 = 66-84 h). Both conjugates accumulated effectively in the tumor tissues. However, the distribution behavior of the conjugates depended on physicochemical properties such as molecular size. Treatment with the conjugates suppressed the tumor growth and increased the survival rate in the tumor-bearing mice.

Targeting of mitomycin C to the liver by the use of asialofetuin as a carrier.

Desialylated fetuin (asialofetuin) was adopted as a carrier for introducing drugs in parenchymal liver cells. Mitomycin C, as a model guest-compound, was covalently attached to asialofetuin through a spacer of the succinyl group. The asialofetuin-mitomycin C conjugate contained 4.4 w/w% of mitomycin C and liberated it gradually at physiological conditions (t1/2 = 37 h). The survival time of the conjugate in rat blood circulation was significantly smaller than that of the non-desialylated fetuin conjugate; the elimination half-life was 7.3 min after intravenous injection. At 30 min after injection of the conjugate in rats, 40% of the dose was present in the liver. Parenchymal cells in the liver selectively took up the conjugate, which was highly distributed to the lysosomal fraction in the cell. The greater uptake of the conjugate by hepatocytes reflected the increased excretion in the bile; totally 10.4% of the dose was recovered.

Preparation and properties of a mitomycin C-albumin conjugate.

Mitomycin C, an anti-neoplastic agent, was covalently attached to bovine serum albumin through various kinds of spacers such as glutaryl, succinyl, trans-aconityl, methylsuccinyl and the trimellityl group. The prior acylation of albumin not only prevented protein polymerization in the presence of carbodiimide, but also increased the extent of conjugation of the drug. The conjugate of mitomycin C-glutarylated albumin showed the best properties among the conjugates prepared in meeting the requirements for a high yield of nonpolymerized product with an adequately high mitomycin C content and stability as a macromolecular prodrug.

Intrahepatic delivery of glutathione by conjugation to dextran.

Glutathione was covalently attached to dextran (T-40) by the CNBr activation method. The compound obtained was a water-soluble powder containing 10 (w/w%) glutathione, which was gradually released from the conjugate in aqueous media. Mice depleted of glutathione by treatment with buthionine sulfoximine, a potent inhibitor of gamma-glutamylcysteine synthetase, exhibited a significant increase in hepatic glutathione level after intravenous injection of the conjugate. In mice given a lethal dose of acetaminophen, the survival rate increased progressively with coadministration of the conjugate, whereas little improvement was found when free glutathione was given. The conjugate maintained the serum transaminase activities at lower level after acetaminophen administration. These findings suggest that the dextran conjugate of glutathione is transported into hepatic cells and is intracellularly hydrolyzed to free form, which protects mice from hepatotoxicity due to acetaminophen.

Effects of glutathione, as the dextran conjugate, on acetaminophen-induced hepatotoxicity.

Glutathione was covalently attached to dextran (T-40) by the CNBr activation method. In mice given a lethal dose of acetaminophen, the 30-d survival rate increased progressively with coadministration of the conjugate, whereas little improvement was found when free glutathione was given. The dextran conjugate of glutathione maintained the serum transaminase activities at lower levels after acetaminophen administration, giving effective protection against acetaminophen hepatotoxicity.

Biohydroxylation of aminopyrine: quantitative studies of 3-hydroxymethyl metabolite in rats.

Significance of the formation of 4-dimethylamino-3-hydroxymethyl-2-methyl-1-phenyl-3-pyrazolin-5-on e (AM-3-CH2OH) for the metabolism of AM was examined quantitatively in rats. Although urinary excretion of AM-3-CH2OH accounted for only 0.7% to the dose, incubation of AM in the isolated hepatocyte system resulted in the formation of 9% of AM-3-CH2OH. Furthermore, metabolic disappearance of AM-3-CH2OH in the same system was fast, indicating the properties of an intermediate metabolite. These results suggested that the metabolic pathways via AM-3-CH2OH are very important in the metabolism of AM.