The absorption of drugs from the gastrointestinal tract is one of the important determinants for oral bioavailability. Development of in vitro experimental techniques such as isolated membrane vesicles and cell culture systems has allowed us to elucidate the transport mechanisms of various drugs across the plasma membrane. Recent introduction of molecular biological techniques resulted in the successful identification of drug transporters responsible for the intestinal absorption of a wide variety of drugs. Each transporter exhibits its own substrate specificity, though it usually shows broad substrate specificity. In this review, we first summarize the recent advances in the characterization of drug transporters in the small intestine, classified into peptide transporters, organic cation transporters and organic anion transporters. In particular, peptide transporter (PEPT1) is the best-characterized drug transporter in the small intestine, and therefore its utilization to improve the oral absorption of poorly absorbed drugs is briefly described. In addition, regulation of the activity and expression levels of drug transporters seems to be an important aspect, because alterations in the functional characteristics and/or expression levels of drug transporters in the small intestine could be responsible for the intra- and interindividual variability of oral bioavailability of drugs. As an example, regulation of the activity and expression of PEPT1 is summarized.
The active transport of solutes mediated by the bile salt export pump (BSEP/ABCB11) and multidrug resistance associated protein-2 (MRP2/ABCC2) are thought to involve bile acid-dependent and -independent bile formation, respectively. To evaluate the potential of therapeutic agents as inhibitors of such transporters on bile canalicular membranes, we examined the inhibition of the primary active transport of typical substrates by 15 drugs, clinically known to cause cholestasis in canalicular membrane vesicles. The inhibition by most of the compounds in rat canalicular membrane vesicles (CMVs) was minimal or observed at much higher concentrations than obtained in clinical situations. However, cloxacillin, cyclosporin A and midecamycin inhibited BSEP, and cyclosporin A and midecamycin inhibited MRP2 with an inhibition constant close to the clinical concentration. By comparing the inhibition potential between rat and human CMVs, the inhibition of BSEP- and MRP2-mediated transport by midecamycin and cyclosporin A was relatively similar whereas the inhibitory effect on BSEP-mediated transport by cloxacillin and glibenclamide was more marked in humans than in rats. These results suggest that the majority of cholestasis-inducing drugs have a minimal inhibitory effect on rat BSEP and MRP2 although species differences in inhibitory potential should be considered, especially in the case of BSEP.
We synthesized four novel lipophilic derivatives of phenylalanyl-glycine (Phe-Gly), C4-Phe-Gly, Phe-Gly-C4, C6-Phe-Gly and C8-Phe-Gly by chemical modification with butyric acid (C4), caproic acid (C6) and octanoic acid (C8). The effect of the acylation on the stability, permeability and accumulation of Phe-Gly in the intestine was investigated by in vitro studies. The stability of Phe-Gly in homogenates of duodenal and colonic membranes was low, but was significantly improved by the acylation except for Phe-Gly-C4. In the transport studies, a modified Ussing chamber was used for the intestinal permeability experiments with Phe-Gly and its acyl derivatives. The permeability of native Phe-Gly and Phe-Gly-C4 across the intestinal membrane was not observed during the transport studies. However, the permeability of Phe-Gly was much improved by chemical modification with various fatty acids to its N-terminal portion. The permeability of acyl-Phe-Gly derivatives across the intestinal membrane decreased with increasing the chain length of fatty acids. In addition, the intestinal tissue accumulation of acyl-Phe-Gly derivatives at the end of the transport studies was much higher than that of native Phe-Gly. The intestinal tissue accumulation of acyl-Phe-Gly in the duodenum increased as the chain length of fatty acids increased. Furthermore, intestinal permeability of C4-Phe-Gly was slightly inhibited in the presence of 5 mM ceftibuten and was significantly reduced under low temperature condition. We observed a directional difference in the transport of C4-Phe-Gly (the mucosal to serosal transport of C4-Phe-Gly was higher than its serosal to mucosal transport) suggesting that C4-Phe-Gly might be transported by a carrier-mediated process as well as other dipeptides. These findings indicate that acylation might be useful approach to enhance the transport of Phe-Gly, a model dipeptide, transported by a carrier-mediated process.
The uptake properties of taurocholate (TC) and estradiol 17β-D-glucuronide (E217βG) were examined in freshly isolated and cryopreserved human hepatocytes to discover if active transport is retained in cryopreserved human hepatocytes. Firstly, the uptake of TC and E217βG was measured before and after cryopreservation. The uptake of TC was found to be Na+-dependent both in fresh and cryopreserved hepatocytes. The uptake activity in cryopreserved hepatocytes was found to range from 10 to 200% of that observed in freshly isolated cells. A kinetic analysis was performed to evaluate the transport activity of TC and E217βG and revealed that the Michaelis constant (Km) for these compounds in cryopreserved human hepatocytes was 2-8 and 3-18μM, respectively. This was within the range of Km values previously found in human Na+-taurocholate cotransporting polypeptides (NTCP) and organic anion transporting polypeptides (OATP) 2 and 8, respectively. The kinetic analyses also showed that the species difference between human and rat hepatocytes was more marked for the maximal uptake rate (Vmax) (>22 and >22 times higher for TC and E217βG in rats than in humans, respectively) than that for Km (2-12 and 0.7-4 times higher, respectively), compared with earlier data we obtained in primary cultured rat hepatocytes. Hence, we conclude that cryopreserved human hepatocytes, at least in part, retain their transporter functions and, therefore, can be a useful experimental system for examining the mechanism of the hepatic uptake of drugs.
To produce a large amount of CYP3A4, we applied a jarfermenter (ABLE, BMJ-PI or BMS-PI) to culture the genetically engineered E. coli cells harboring CYP3A4 along with NADPH-cytochrome P450 reductase (OR). The jarfermenter is a stirred bacterial culture vessel in which the pH, the dissolved oxygen (DO) and the temperature of a culture medium can be controlled. The expression of CYP3A4 in E. coli cells in the 500 mL of culture medium contained in the BMJ-PI (1 L vessel) (JFM-1) was examined by altering the parameters mentioned above. The highest expression of CYP3A4 in E. coli cells was attained when cultured at pH 6.0, at 30°C under the DO of 0.1 ppm. The incubation was performed 18 hr after the addition of 1.5 mM isopropyl β-D(—)-thiogalactopyranoside. The expression levels of CYP3A4 and the OR in the membrane fraction of E. coli cells were 267 nmol/L culture and 552 units/L culture, respectively. The CYP3A4 level was about three times higher than that obtained by incubation in a 500 mL flask (100 mL of medium) (84 nmol/L culture). The testosterone 6β-hydroxylase activity of CYP3A4 expressed in the membrane fraction of E. coli obtained with the JFM-1 was examined. The apparent Km and Vmax values were 66.4μM and 57.8 nmol/min/nmol CYP, respectively. Expecting the mass production of the CYP3A4 by a culture of E. coli, the possibility of a scale up of the culture with the BMS-PI (10 L vessel) (JFM-10) was examined. The optimal culture condition to achieve the highest expression of CYP3A4 with JFM-1 was employed. The expression levels of CYP3A4 and the OR obtained with JFM-1 and JFM-10 were almost equal. The total level of CYP3A4 obtained by using JFM-10 (5 L of medium) was calculated to be about 1.4 μmol. Based on these results, we confirm that the jarfermenter is a useful tool to produce large amounts of CYP3A4.
The enzyme activities of CYP2D6 and CYP2C19 show a genetic polymorphism, and the frequency of poor metabolizers (PMs) on these enzymes depends on races. In the present study, the frequencies of mutant alleles and PMs in each race were analyzed based on information from published studies, considering the genetic polymorphisms of CYP2D6 and CYP2C19 as the causal factors of racial and inter-individual differences in pharmacokinetics. As a result, it was shown that there were racial differences in the frequencies of each mutant allele and PMs. The frequencies of PMs on CYP2D6 are 1.9% of Asians and 7.7% of Caucasians, and those of PMs on CYP2C19 are 15.8% of Asians and 2.2% of Caucasians. Based on the results, it was suggested that there would be racial differences in the frequencies of PM subjects whose blood concentrations might be higher for drugs metabolized by these enzymes. Additionally, it was suggested that enzyme activities would vary according to the number of functional alleles even in subjects judged to be extensive metabolizers (EMs). In the bridging study, genetic information regarding CYP2D6 and CYP2C19 of the subjects will help extrapolate foreign clinical data to a domestic population.
The enzyme activities of CYP2D6 and CYP2C19 show a genetic polymorphism, and the frequency of poor metabolizers (PMs) on these enzymes depends on races. We have analyzed frequencies of mutant alleles and PMs based on the published data in previous study (Shimizu, T. et al.: Bioinformatics research on inter-racial difference in drug metabolism, I. Analysis on frequencies of mutant alleles and poor metabolizers on CYP2D6 and CYP2C19.). The study shows that there were racial differences in the frequencies of each mutant allele and PMs. In the present study, the correlation between genotypes and drug-metabolizing enzyme activities was investigated. The result showed that enzyme activities varied according to the genotypes of subjects even in the same race. On the other hand, if subjects had the same genotypes, almost no racial differences were observed in drug-metabolizing enzyme activities. From these results, it was supposed that the racial differences in activities of these enzymes could be explained by the differences in distribution of genotypes. It would be possible to explain the racial differences in drug-metabolizing enzyme activities based on the differences on individual pharmacogenetic background information, not merely by comparison of frameworks such as races and nations.
The effects of the substitution of glycine at position 42 with various other amino acid residues on the functions of CYP2D6 were studied using debrisoquine (DB) and bunitrolol (BTL) 4-hydroxylations as indices of drug-metabolizing enzymes. The substitution with hydrophobic amino acid residues such as valine and phenylalanine did not affect the enzymatic properties such as reduced CO-difference spectra, microsomal CYP contents and oxidation activities towards DB and BTL. The substitution of glycine-42 with a polar but noncharged amino acid residue (serine) exhibited a similar reduced CO-different spectrum, but the substitution with a charged basic (lysine and arginine) or acidic (glutamic acid) amino acid residue commonly produced a peak at 420 nm in addition to a Soret peak at 450 nm. Cytochrome P450 contents and microsomal contents of G42S, G42K, G42R and G42E estimated spectrophotometrically and estimated by Western blot analysis, respectively, were lower than those of the wild-type. Kinetic analysis revealed that the substitution of glycine-42 with charged amino acid residues such as lysine, arginine and glutamic acid markedly increased the apparent Km values for DB and BTL oxidations without remarkable changes in the Vmax values. The subsitution with noncharged amino acid residues such as serine, valine and phenylalanine did not cause such a marked change in the Km values. Efficiencies (Vmax/Km) as DB and BTL 4-hydroxylases of CYP2D6 mutant proteins having charged amino acid residues were found to be decreased mainly by increasing their Km values. These results indicate that the properties of amino acid residues at position 42 affect the behavior of CYP2D6 proteins such as anchoring into ER membranes, conversion of P450 to P420 and incorporation of heme into apoproteins.
We sequenced all nine exons, exon-intron junctions including a part of introns, 5'-flanking and 3'-untranslated regions of the cytochrome P450 (CYP) 2A13 gene from 192 Japanese individuals. We found eighteen novel genetic polymorphisms including five single nucleotide polymorphisms (SNP) and one three base pair insertion causing amino acid substitution and one amino acid insertion, respectively, one silent SNP in exon 4, four SNPs in a 5'-flanking region, and seven SNPs in introns. The five SNPs (74G>A in exon 1, 579G>A in exon 2, 1706C>G in exon 3, and 7343T>A and 7465C>T in exon 9) causing amino acid substitutions (Arg25Gln, Arg101Gln, Asp158Glu, Phe453Tyr, and Arg494Cys), respectively. The one three base pair insertion (1634_1635 ACC insertion in exon 3) caused one amino acid insertion (133_134Thr ins). These sequences are as follows: SNP, 021125Fujieda005; GENE NAME, CYP2A13; ACCESSION NUMBER, NG_000008; LENGTH, 25 base; 5'-TGTCAGTCTGGCG/AGCAGAGGAAGAG-3'. SNP, 021125Fujieda007; GENE NAME, CYP2A13; ACCESSION NUMBER, NG_000008; LENGTH, 25 base; 5'-AGTTCAGCGGGCG/AAGGCGAGCAGGC-3'. SNP, 021125Fujieda009; GENE NAME, CYP2A13; ACCESSION NUMBER, NG_000008; LENGTH, 25 base; 5'-CTTCCTCATCGAC/GGCCCTCCGGGGC-3'. SNP, 021125Fujieda017; GENE NAME, CYP2A13; ACCESSION NUMBER, NG_000008; LENGTH, 25 base; 5'-TCTTTCTCTTCTT/ACACCACCATCAT-3'. SNP, 021125Fujieda018; GENE NAME, CYP2A13; ACCESSION NUMBER, NG_000008; LENGTH, 25 base; 5'-AGCTTCCTGCCCC/TGCTGAGCGAGGG-3'. SNP, 021125Fujieda008; GENE NAME, CYP2A13; ACCESSION NUMBER, NG_000008; LENGTH, 25 base; 5'-CTCCATCGCCACC-/ACCCTAAGGGGTTTT-3'.