There is an increasing interest in improving treatment of liver diseases. Normal drug administration by intravenous and oral route have difficulty in achieving a local site of action in the liver. Although the direct way of application to the liver surface should yield local drug distribution, the drug absorption from the liver surface has not been reported in literature.
In the present study, we have analyzed the absorption mechanism of organic anions as a model after application to the rat liver surface in vivo employing a cylindrical glass cell. Every drug appeared gradually in the plasma, followed by excretion into the bile. This indicates the possibility of drug absorption from the liver surface. The pharmacokinetics of phenol red, of which absorption rate was fastest, after application to the rat liver surface was explained by the model incorporating the first-order absorption. Because the effect of dose and transport inhibitors on the absorption of phenol red could not be recognized, a specific transport system might not be involved in the process of absorption from the liver surface. Moreover, protein binding tended to suppress the absorption of organic anions, and molecular weight was found to be a determinant factor of absorption from the liver surface, by utilizing dextrans with different molecular weights. Also, the targeting efficacy to the liver was enhanced by application to the liver surface, as high liver/plasma concentration ratio as compared with i.v. administration. Furthermore, we confirmed that the change of the drug absorption behavior from the liver surface in the basic diseased state (CCl₄- or D-galactosamine-treated rat) was not significant.
Consequently, the present results on the mechanism of drug absorption from the liver surface provide an useful information in the development of new administration route for drug delivery to the target site in liver.

View full abstract

Several organic anions, including the glutathione- and glucuronide-conjugates, are excreted into the bile via a canalicular multispecific organic anion transporter (cMOAT), which is hereditarily defective in mutant rats such as Eisai hyperbilirubinemic rat (EHBR). In the present study, we cloned cMOAT from the SD rat liver cDNA library using the ATP-binding cassette region as a probe which was amplified by PCR with the degenerated primers for human multidrug resistance associated protein (hMRP). cMOAT was encoded by 4623 by cDNA with a homology of 53.0 % and 46.3 % with hMRP at the cDNA and deduced amino acid level, respectively. By screening the library, three kinds of cDNA species for cMOAT with the same open reading frame and different 3′-untranslated region lengths were isolated. The Northern blot analysis of poly A⁺ RNA from the liver with the entire open reading frame of cMOAT as a probe revealed that the expression of plural bands (approximately 5, 6 and 8 kb) was defective in EHBR and this may be due to the presence of these cDNA species. RT-PCR and subsequent sequence analysis were performed using RNA specimens from EHBR liver. The analysis revealed that one base pair replacement within the open reading frame resulted in the introduction of the premature stop codon. The impaired expression of this particular protein is related to the pathogenesis of hyperbilirubinemia in the mutant rats. Although the expression of cMOAT is defective in EHBR, our recent kinetic studies suggested the presence of ATP-dependent organic anion transporters other than cMOAT in EHBR. Using the degenerated PCR primers described previously, we could amplify two kinds of PCR product (MRP like protein (MLP-1 and MLP-2)) from EHBR liver. The nucleotide homology of ABC region between MLP-1, 2, MRP and cMOAT was approximately 60-70%. Northern blot analysis indicated that MLP-1 hybridized with a single 6 kb transcript in the liver of SD rats and EHBR. In contrast, MLP-2 hybridized to a single 5 kb transcript in the liver and the expression was rather induced in EHBR. These results may suggest the possibility that MLP-1 and MLP-2 are novel transporters for organic anions.

View full abstract

In this study, we constructed the physiological model linked with cell kill kinetic model via the concentration of antitumor drugs in extracellular space. We studied the antitumor effect of liposomal antitumor drugs with alteration on uptake rate constant of liposomes by reticuloendothelial system and release rate constant of drug from liposomes. This simulation study was performed to clarify the relationship between antitumor effect and pharmacokinetic or physicochemical parameters of liposomes as well as pharmacological or physiological parameters of tumor tissues. Thus, we attempted to estimate the antitumor effect of liposomal antitumor drugs quantitatively with the tumor cells numbers. The existence of optimum release rate of each liposomal doxorubicin and vincristine was indicated constantly in this study.

View full abstract

Many patients of diabetes mellitus fall into renal impairment requiring dialysis. To quantitatively describe the peritoneal transport of drugs without any approximation for the changes in dialysate volume and osmolality, a kinetic model based on the hydrodynamic pore theory of transcapillary exchange was proposed. To further validate the present model, the concentration-time profile of cefazolin in the dialysate after intravenous administration in a patient with end-stage renal failure was successfully predicted using the apparent capillary membrane permeability (P_d) value extrapolated from those of the experimental animals. There was a good correlation between the P_d values of unbound drugs in rats and the partition coefficients for acidic drugs such as barbiturates and quinolonecarboxylic acids. This relationship can be useful to predict the peritoneal pharmacokinetics of acidic drugs. But P_d of benzoic acid varied in a dose dependent manner. It is suggested that certain facilitated transport systems constitute the mechanism of enhanced P_d of benzoic acid.

View full abstract

A complementary DNA (cDNA) encoding a novel member of organic anion transoporter, OAT-K1, expressed specifically in the kidney of rats was isolated, and the transport characteristics were assessed by LLC-PK, cells stably expressing OAT-K1 (LLC-OAT-K1). The OAT-K1 cDNA (2, 788 bp) consists of a 669-amino acid protein (calculated molecular mass of 74 kDa) which shows 72% identy with the rat liver Na⁺-independent organic anion transporting polypeptide (oatp). Northern blot analysis and reverse transcription-coupled PCR revealed that the rat OAT-K1 mRNA is expressed predominantly in the kidney medulla, but not in the liver. LLC-OAT-K1 monolayers showed that the OAT-KI mediates basolateral uptake of methotrexate, but not taurocholate, p-aminohippurate, prostaglandin E₂, and leukotriene C₄. Folate, sulfobromophthalein and DIDS inhibited the methotrexate accumulation markedly. Furthermore, some steroides and nonsteroidal anti-inflammatory drugs showed the potent inhibitory effects on methotrexate uptake by LLC-OAT-K1 monolayers. These findings suggest that the rat OAT-K1 is a new organic anion transporter involved in detoxification process in the kidney, mediating basolateral uptake of anionic xenobiotics.

View full abstract

The possibility of oral delivery of recombinant human insulin-like growth factor-I was investigated in rats. The degradation of ¹²⁵I-rhlGF-I in the gastrointestinal (GI) contents was in the order of jejunum = ileum >> large intestine, and little degradation was found in the stomach. Some adjuvants could inhibit the degradation, especially aprotinin and casein inhibited by 70-95%. On the other hand, rhlGF-I was relatively stable in the subcellular fraction of mucosal cells (BBM, cytosol, lysosome), and there was no site difference through the intestine. The bioavailability (BA) of rhlGF-I after oral administration was 9.3%, which was much greater than that of insulin. The BA was further increased by coadministration with aprotinin and casein, indicating that the stabilization of rhlGF-I in GI tract could enhance its absorption. From the results of gel filtration chromatography of rat plasma following oral administration of ¹²⁵l-rhIGF-I, the peaks in high molecular region were detected and they were thought to be macromolecular complexes which consist of ¹²⁵I-rhlGF-I and IGFBP-2 and/or -3. These results strongly suggest the possibility of oral delivery of rhlGF-I.
We also examined the absorption kin etics of rhlGF-I by the in situ single-pass perfusion method. The absorption clearance (CLa) of rhlGF-I was dependent on its concentration. Using acid-washing technique, rhlGF-I was found to be adsorbed acid-sensitively to the mucosal surface and the incorporation of immunoreactive rhIGF-I to the mucosal tissue was recognized at the same time. These results suggest that rhlGF-I is absorbed partly by a specialized absorption mechanism.

View full abstract

A predicting method for the intestinal absorption of drugs by measuring their physicochemical properties such as organic solvent / buffer partition coefficient, molecular volume (molecular weight), hydrogen bonding and diffusion rate across silicone membrane was investigated. A poor correlation was observed between absorption rate from intestinal lumen and organic solvent / buffer partition coefficient of tested drugs. On the contrary, good regression coefficient (R = 0.801) was obtained when the three factors (molecular weight, hydrogen bonding, and permeation rate across silicone membrane) were used. On the other hand, the excellent regression was observed when the drugs were classified into anionic, cationic, and neutral groups. These results suggest that the permeation rate across silicone membrane, molecular weight, and hydrogen bonding are valuable measures for predicting the absorption behavior of drugs. Moreover, an additional parameter which reflects the effect of the molecular charge will be necessary.

View full abstract

A serotonin antagonist azasetron, which is relatively hydrophilic and cationic compound, has been shown to exhibit a relatively high and a dose-dependent increase of intestinal absorbability. In the present study, the significant contribution of a specific secretory mechanism to the nonlinear intestinal absorption of azasetron and other drugs was studied. The initial uptake rate of azasetron by Caco-2 cells was saturable and temperature-dependent, demonstrating a participation of a carrier-mediated uptake process. Net transport of [¹⁴C]azasetron from basolateral-to-apical-side across Caco-2 cells was greater than that from apical to basolateral side. When 0.5 mM unlabeled azasetron was included in the apical side medium, apical-to-basolateral transport of [¹⁴C]azasetron was enhanced, whereas basolateral-to-apical transport was decreased. However, when higher concentrations of unlabeled azasetron were added, a decreased apical-to-basolateral and an increased basolateral-to-apical transport of [¹⁴C]azasetron were observed.
Furthermore, first-order intestinal absorption rate constants of azasetron, which were evaluated by Doluisio's technique in rats, exhibited a complicated nonlinear feature. Namely, the rate constant was increased with increasing concentration of azasetron from 0.01 to 10 mM, whereas the value was decreased at higher concentrations up to 60 mM. These results may demonstrate that an intestinal absorption of azasetron includes two nonlinear processes, including carrier-mediated uptake and specific secretory mechanisms. Interestingly, since cyclosporin A inhibited secretory process of azasetron in Caco-2 cells, P-glycoprotein may be involved as the secretory mechanism. Similar nonlinear increase of intestinal absorption was observed in cyclosporin A after intra-ileal administration in rats. In addition, intestinal absorption of cyclosporin A was increased with a coadministration of P-glycoprotein inhibitor.
These results demonstrate that the specific intestinal secretory mechanism is a significant factor, resulting in nonlinear increase of oral bioavailability.

View full abstract

A distributed model was employed to clarify the mechanism for the restricted distribution of the quinolone antibiotics (norfloxacin, AM-1155, ofloxacin, fleroxacin, sparfloxacin and pefloxacin) in the rat central nervous system (CNS). The permeability-surface area (PS) product for the symmetrical transport across the blood-brain barrier (BBB; PS_{BBB, sym}) and across the blood-cerebrospinal fluid barrier (BCSFB; PS_{BCSFB, sym}), and the efflux clearance across the BBB (PS_{BBB, act}) were obtained by fitting in vivo data to this kinetic model by using a nonlinear least squares regression analysis combined with the fast inverse Laplace transforming program. The values of PS_{BBB, act} were 10- to 260-fold greater than those of PS_{BBB, sym}, providing the kinetic evidence to support the hypothesis that a significant efflux transport at the BBB is responsible for the limited distribution of quinolones in brain tissue. A good correlation between the apparent brain tissue-to-unbound serum concentration ratio at steady-state and the ratio of PS_{BBB, sym} to the sum of PS_{BBB, sym} and PS_{BBB, act} suggests the importance of the efflux transport across the BBB in determining the brain distribution of quinolones.

View full abstract

The formation and maintenance of blood-brain barrier (BBB) have been ascribed to endothelial cell responses to inductive signal(s) from astrocytes that encircle microvessels in the central nervous system. Here we describe an in vitro BBB model, in which rat fetal cerebral astrocytes are cultivated on one surface of a porous membrane and human umbilical vein endothelial cells on the opposite surface. Electron microscopic examination revealed that astrocytes passed their endfeet through the pores, making contact with endothelial cells. Genes for the representative BBB markers, including γ-glutamyl transpeptidase, transferrin receptor, P-glycoprotein and brain-type glucose transporter, were found to be up-regulated in endothelial cells by contacting astrocytes, but not when the former cells were maintained apart from the latter or when the feeder layer was substituted for by COS cells. Astrocyte induction of close membrane appositions resembling zonulae occludentes between endothelial cells and of an increase in the number of mitochondrion was also noted. Further, an increased barrier activity against inulin was conferred on endothelial cells, when lined with astrocytes, with due polarity. The results thus indicate that, through contact with their feet, astrocytes are capable of transdifferentiating non-neural endothelial cells into the brain-type, endowing them with the BBB properties. The present BBB model would seem useful for studying the drug delivery to the brain.

View full abstract