Teruo Murakami

Intestinal efflux transporters and drug absorption

Background: The intestinal epithelial membrane expresses ATP-binding cassette (ABC) transporters such as P-glycoprotein (P-gp), multi-drug resistance-associated proteins (MRPs) and breast cancer resistance protein (BCRP), in addition to various solute carrier (SLC) transporters. These ABC transporters affect the oral bioavailability of their substrate drugs. Objective: To review the contribution of ABC efflux transporters such as P-gp, MRP2, MRP3, and BCRP in the intestinal absorption of substrate drugs. Methods: Discussion was made by focusing on the site-specific expression and function of these ABC transporters, and the solubility and permeability of their substrate compounds. Results/conclusion: The increase in the solubility and permeability of orally administered drugs could be the key to escape barrier function of ABC transporters, especially P-gp.

Role of intestinal efflux transporters in the intestinal absorption of methotrexate in rats

The role of intestinal efflux transporters such as P‐glycoprotein (P‐gp), breast cancer resistance protein (BCRP) and multidrug resistance‐associated proteins (MRPs) in intestinal absorption of methotrexate was examined in rats. In everted intestine, the mucosal efflux of methotrexate after application to serosal side was higher in jejunum than ileum, and the efflux in jejunum was suppressed by pantoprazole, a BCRP inhibitor, and probenecid, an MRP inhibitor, but not by verapamil, a P‐gp inhibitor. The mucosal methotrexate efflux in ileum was suppressed by pantoprazole, but not by other inhibitors. On the other hand, the serosal efflux of methotrexate after application to mucosal side was greater in ileum than jejunum, and was suppressed by probenecid. In in‐vivo rat studies, the intestinal absorption of methotrexate was significantly higher when methotrexate was administered to ileum than jejunum. Pantoprazole increased methotrexate absorption from jejunum and ileum. Probenecid increased the absorption of methotrexate from jejunum but decreased the absorption from ileum, as evaluated by peak plasma methotrexate levels. In conclusion, BCRP and MRPs are involved in the regional difference in absorption of methotrexate along the intestine, depending on their expression sites.

Characterization of intestinal absorption of mizoribine mediated by concentrative nucleoside transporters in rats.

Mizoribine, an imidazole nucleoside, is an inhibitor of purine synthesis and has been used as an orally available immunosuppressive agent in human renal transplantation. In the present study, the intestinal absorption of mizoribine was characterized by examining the contribution of concentrative nucleoside transporters (CNT1, CNT2) in rats. When mizoribine was administered orally in conscious rats, the bioavailability of mizoribine estimated by urinary excretion percentage of unchanged mizoribine was a dose dependent: 53.1+/-6.0% at 5 mg/kg and 24.0+/-5.1% at 20 mg/kg. In in-situ loop studies, the disappearance rate, or absorption rate, of mizoribine from the intestinal lumen was comparable between 1 and 5 mg/kg, but significantly lower at 25 mg/kg. Coadministration of adenosine (a substrate of both CNT1 and CNT2), thymidine (a CNT1 substrate) and inosine (a CNT2 substrate) significantly suppressed the intestinal mizoribine absorption, depending on the nucleoside concentrations coadministered. Gemcitabine (a pyrimidine nucleoside analogue, a CNT1 substrate) and ribavirin (a purine nucleoside analog, a CNT2 substrate) also significantly suppressed the mizoribine intestinal absorption. Bile salts such as sodium cholate and sodium glycocholate (10 mM) also significantly suppressed the intestinal mizoribine absorption, but not ribavirin absorption. Mizoribine is an amphoteric compound, however, the suppression of intestinal absorption by bile salts was not ascribed to the electrostatic interaction or micellar formation between mizoribine and bile salts. In conclusion, the intestinal absorption of mizoribine is mediated by CNT1 and CNT2, and nucleoside-derived drugs such as gemcitabine and ribavirin can suppress the intestinal absorption of mizoribine. Bile salts such as sodium glycocholate were also found to cause interaction with mizoribine.

Site‐specific bidirectional efflux of 2,4‐dinitrophenyl‐S‐glutathione, a substrate of multidrug resistance‐associated proteins, in rat intestine and Caco‐2 cells

The site‐specific function of multidrug‐resistance‐associated proteins (MRPs), especially MRP2 and MRP3, was examined in rat intestine and human colon adenocarcinoma (Caco‐2) cells. The MRP function was evaluated pharmacokinetically by measuring the efflux transport of 2,4‐dinitrophenyl‐S‐glutathione (DNP‐SG), an MRP substrate, after application of 1‐chloro‐2,4‐dinitrobenzene (CDNB), a precursor of DNP‐SG. The expression of rat and human MRP2 and MRP3 was analysed by Western blotting. The rat jejunum exhibited a higher apical MRP2 and a lower basolateral MRP3 expression than ileum. In accordance with the expression level, DNP‐SG efflux to the mucosal surface was significantly greater in jejunum, while serosal efflux was greater in ileum. Site‐specific bidirectional efflux of DNP‐SG was also observed in in‐vivo studies, in which portal and femoral plasma levels and biliary excretion rate of DNP‐SG were significantly higher when CDNB was administered to ileum. Caco‐2 cells also showed a bidirectional efflux of DNP‐SG. Probenecid, an MRP inhibitor, significantly suppressed the mucosal efflux in jejunum and serosal efflux in ileum. In contrast, probenecid significantly suppressed both apical and basolateral efflux of DNP‐SG in Caco‐2 cells, though the inhibition was of small magnitude. In conclusion, the efflux of DNP‐SG from enterocytes mediated by MRPs exhibited a significant regional difference in rat intestine, indicating possible variability in intestinal bioavailabilities of MRP substrates, depending on their absorption sites along the intestine.

Role of phosphatidylserine binding in tissue distribution of amine-containing basic compounds.

Introduction: Amine-containing basic compounds with pKa values of ≥ 7 for at least one functional group, or moderately strong amines, typically exhibit large distribution volumes with a marked interorgan variation. Understanding the tissue distribution mechanisms of various basic compounds is important for drug design. Areas covered: This review focuses on the role of tissue phosphatidylserine (PS) binding in the characteristic tissue distribution of amine-containing basic compounds. The mechanisms underlying tissue distribution, including intercellular and intracellular distribution, and effects of pKa values and lipophilicity on the extent of tissue distribution of various basic compounds are clarified. Expert opinion: The extent of tissue distribution of membrane-permeable moderately strong amines depends on tissue PS concentrations and binding affinities to PS. A linear relationship is observed between the extent of tissue distribution of each compound and tissue PS concentrations in various tissues. In contrast, very weakly basic compounds with pKa values of < 6.5 distribute to various tissues at almost comparable magnitudes, independent of tissue PS concentrations. Understanding the tissue distribution mechanisms and predicting Kp values for each tissue is essential in controlling pharmacological and/or toxic effects in pharmacotherapy with various amine-containing basic compounds, as well as in drug design.

Site-specific contribution of proton-coupled folate transporter/haem carrier protein 1 in the intestinal absorption of methotrexate in rats.

Objectives Methotrexate is reportedly a substrate for proton‐coupled folate transporter/haem carrier protein 1 (PCFT/HCP1) and reduced folate carrier 1 (RFC1). In this study, we examined the contribution of PCFT/HCP1 and RFC1 in the intestinal absorption of methotrexate in rats.

Effect of genistein, a natural soy isoflavone, on the pharmacokinetics and intestinal toxicity of irinotecan hydrochloride in rats.

The effect of genistein, a natural soy isoflavone, on pharmacokinetics and intestinal toxicity, or late‐onset diarrhoea, of irinotecan hydrochloride (CPT‐11) was examined in rats.

Involvement of Multiple Transporters-mediated Transports in Mizoribine and Methotrexate Pharmacokinetics

Mizoribine is administered orally and excreted into urine without being metabolized. Many research groups have reported a linear relationship between the dose and peak serum concentration, between the dose and AUC, and between AUC and cumulative urinary excretion of mizoribine. In contrast, a significant interindividual variability, with a small intraindividual variability, in oral bioavailability of mizoribine is also reported. The interindividual variability is mostly considered to be due to the polymophisms of transporter genes. Methotrexate (MTX) is administered orally and/or by parenteral routes, depending on the dose. Metabolic enzymes and multiple transporters are involved in the pharmacokinetics of MTX. The oral bioavailability of MTX exhibits a marked interindividual variability and saturation with increase in the dose of MTX, with a small intraindividual variability, where the contribution of gene polymophisms of transporters and enzymes is suggested. Therapeutic drug monitoring of both mizoribine and MTX is expected to improve their clinical efficacy in the treatment of rheumatoid arthritis.

Modulated function of tissue efflux transporters under hyperbilirubinemia in rats.

The effect of hyperbilirubinemia on the function of tissue efflux transporters such as multidrug resistance-associated proteins (Mrps) and organic anion transporting polypeptides (Oatps) was examined by measuring tissue accumulation of 2,4-dinitrophenyl-S-glutathione (DNP-SG) after intravenous administration of 1-chloro-2,4-dinitrobenzene (CDNB), a precursor of DNP-SG, in rats. DNP-SG is known as a substrate of both Mrps and Oatps. Hyperbilirubinemia was induced by a bolus intravenous administration of bilirubin. Treatment with probenecid, an inhibitor for both Mrps and Oatps, significantly increased DNP-SG concentrations in the brain, heart, liver, kidney, jejunum, spleen and skeletal muscle as compared with those in control rats, suggesting the expression of some probenecid-sensitive efflux transporters in these tissues. Rats with more than 70 microM of unconjugated/conjugated bilirubin in plasma exhibited significantly higher DNP-SG concentrations in the brain, liver, jejunum, and skeletal muscle. These results suggested that probenecid-sensitive efflux transporters in tissues were suppressed functionally under hyperbilirubinemia. In conclusion, hyperbilirubinemia accompanied by obstructive jaundice is caused by various disease states, which may increase harmful toxicities of exogenously administered Mrps and/or Oatps substrate drugs at various tissues, by suppressing the efflux transporters function systemically.

Study on intestinal absorption sites of mizoribine and ribavirin, substrates for concentrative nucleoside transporter(s), in rats.

The absorption sites of mizoribine (an imidazole nucleoside) and ribavirin (a purine nucleoside) in the small intestine were evaluated in rats. The intestinal absorption of mizoribine is known to be mediated by rat concentrative nucleoside transporter (CNT)1 and CNT2. In contrast, the absorption mechanism of ribavirin in rats is not yet fully understood. Thus, the intestinal absorption of ribavirin was characterized firstly. In in-situ jejunum loop studies, the absorption percentage of ribavirin at a dose of 25mg/kg was significantly lower than those after 1mg/kg and 5mg/kg doses. Coadministration of adenosine, inosine and mizoribine, but not thymidine and gemcitabine, significantly suppressed the intestinal absorption of ribavirin, indicating that ribavirin absorption is mediated by CNT2 in rats. In in-situ loop studies, mizoribine and ribavirin were absorbed to the same extents both in the proximal and distal small intestine. In vivo study was carried out using mizoribine, in which the gastric emptying rates altered by a subcutaneous injection of metoclopramide or scopolamine butylbromide exerted no significant effects on the values of peak plasma level (Cmax), area under the plasma concentration-time profile from 0 to 6h (AUC(0-6)), and urinary excretion percentage of mizoribine given orally, though the time to reach Cmax (Tmax) of mizoribine was altered by each treatment. In conclusion, mizoribine and ribavirin were found to be absorbed efficiently to the same extents from the whole small intestine. Also, the altered gastric emptying rates exerted no significant effects on the oral bioavailabilities of mizoribine and ribavirin.