- Email: [email protected]
The Apical Uptake Transporter of Levofloxacin is Distinct from the Peptide Transporter in Human Intestinal Epithelial Caco-2 Cells
Drug Metab. Pharmacokinet. 23 (5): 373–378 (2008).
Note The Apical Uptake Transporter of Levofloxacin is Distinct from the Peptide Transporter in Human Intestinal Epithelial Caco-2 Cells Shiro FUKUMORI, Toshiya MURATA, Mari TAKAAI, Katsutoshi TAHARA, Masato TAGUCHI and Yukiya HASHIMOTO* Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama, Toyama, Japan Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk
Summary: The aim of this study was to investigate the involvement of the peptide transporter for absorption of levofloxacin in Caco-2 cells. To evaluate the activity of apical and basolateral peptide transport, we first performed pharmacokinetic analysis of transcellular transport of glycylsarcosine (Gly-Sar) in cell monolayers grown on porous membrane filters. Transcellular transport of Gly-Sar at the medium pH 6 was greater in the apical-to-basolateral direction than in the opposite direction. Influx clearance of Gly-Sar at the apical membrane was much greater than basolateral influx and efflux clearance, indicating that the apical peptide transporter plays an important role in directional transcellular transport of the dipeptide across Caco-2 cell monolayers. We then evaluated the effect of various compounds on the uptake of Gly-Sar and levofloxacin at the apical membrane of Caco-2 cells. The apical uptake of [3H]Gly-Sar was significantly inhibited by Ala-Ala, Gly-Sar, and also levofloxacin, whereas that of [14C]levofloxacin was not inhibited by Ala-Ala and Gly-Sar. On the other hand, the apical uptake of [14C]levofloxacin was inhibited by nicotine, enalapril, fexofenadine, and L-carnitine. These findings indicated that the apical uptake transporter of levofloxacin is distinct from the peptide transporter in Caco-2 cells. Keywords: levofloxacin, Caco-2 cells, peptide transporter, OATP, glycylsarcosine
levofloxacin at medium pH 6 was much greater than the basolateral influx clearance value in Caco-2 cells.3) These results suggested that levofloxacin uptake across the apical membrane in Caco-2 cells was mediated by a specific transport system. Maeda et al. evaluated the levofloxacin uptake activity of Caco-2 subclones, and selected candidate transporter genes functioning for influx transport of levofloxacin.1) Based on functional analysis of each transporter gene for which a good correlation was found between the expression level and levofloxacin transport activity in Caco-2 subclones, organic anion transporting polypeptide 1A2 (OATP1A2) was concluded to transport levofloxacin. When OATP1A2 was expressed in Xenopus oocytes, levofloxacin transport was essentially pH-independent. OATP1A2-mediated uptake of levofloxacin showed a Km value of 136 mM. Apparent uptake of levofloxacin by Caco-2 cells showed high- and low-affinity components with Km values of 0.489 and 14.6 mM, respectively. Accordingly, plural transporters are functional for the
Introduction Levofloxacin is well absorbed from the intestine, and the bioavailability following oral administration is approximately 100% in humans.1) Since levofloxacin is a zwitterionic compound, a passive diffusion mechanism may not fully explain the high intestinal absorption.1) Yamaguchi et al. evaluated the transport characteristics of levofloxacin using human intestinal epithelial Caco-2 cells grown on plastic dishes.2) Apical uptake of levofloxacin in Caco-2 cells was rapid, and almost reached a steady state at 15 min after the start of incubation. The uptake of levofloxacin was markedly decreased by lowering the temperature (49C) and showed concentrationdependent saturation.2) In order to evaluate the membrane transport responsible for rapid intestinal absorption of levofloxacin, we performed pharmacokinetic analysis of the transcellular transport of levofloxacin across Caco-2 cell monolayers grown on porous membrane filters.3) The apical influx clearance value of
Received; February 25, 2008, Accepted; March 24, 2008 *To whom correspondence should be addressed: Yukiya HASHIMOTO, Ph.D., 2630 Sugitani, Toyama 930–0194, Japan. Tel. ＋81-76-434-7585, Fax. ＋81-76-434-7587, E-mail: yukiya＠pha.u-toyama.ac.jp This work was supported in part by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Sciences (JSPS).
373
374
Shiro FUKUMORI, et al.
transport of levofloxacin in Caco-2 cells, and OATP1A2 is likely to function as a high-affinity transporter.1) However, it is still unclear which transporter is responsible for the lower-affinity (high-capacity) uptake of levofloxacin in Caco-2 cells. The peptide transporter (PEPT1) is one of the wellknown transporters, which are expressed in brush-border membranes of intestinal epithelial cells and are involved in the absorption of zwitterionic compounds from intestinal lumen into cells.4) Rabbaa et al. evaluated the absorption characteristics of S-ofloxacin (levofloxacin) in the rat intestine to clarify the mechanism involved in intestinal transport of the drug.5) Dipeptides and b-lactam antibiotics did not modify levofloxacin absorption, suggesting that levofloxacin absorption is probably independent of the dipeptide transport system in the rat intestine.5) In contrast, Yamaguchi et al. reported the absorption characteristics of sparfloxacin in rats using the in situ ligated loop method and the everted jejunum uptake method in vitro.6) The absorption of sparfloxacin in rats with a jejunum loop was reduced to 60% of the control by the addition of di-, tripeptides, and cyclacillin. The in vitro uptake of sparfloxacin into the rat everted jejunum was inhibited by the addition of di-, tripeptides, and blactam antibiotics. In addition, glycylglycine competitively inhibited the uptake of sparfloxacin. These results indicated that approximately 40% of sparfloxacin dose can be transported by a common dipeptide carrier-mediated system.6) The primary aim of the present study was to systematically investigate the involvement of peptide transporter in the transport of levofloxacin across human intestinal Caco-2 cell monolayers. In order to evaluate the activity of apical and basolateral peptide transport in Caco-2 cells, we first performed pharmacokinetic analysis of transcellular transport of glycylsarcosine (Gly-Sar) in cell monolayers grown on porous membrane filters. We then evaluated the effect of unlabeled levofloxacin, lomefloxacin, Ala-Ala, and Gly-Sar on the uptake of radio-labeled Gly-Sar and levofloxacin at the apical membrane of the cells. Furthermore, we also investigated the effect of peptide-like drugs, nucleoside derivatives, organic cations and anion, and zwitterionic compounds on the apical uptake of levofloxacin in Caco-2 cells.
Materials and Methods Materials: [3H]glycylsarcosine (Gly-Sar) (18.5 GBq/ mmol) and [14C]mannitol (1.96 GBq/mmol) were purchased from Moravek Biochemicals Inc. (Brea, CA, U.S.A.). [14C]levofloxacin (2.43 MBq/mg) was obtained from Daiichi Pure Chemicals Co. (Ibaraki, Japan). [3H]mannitol (740 GBq/mmol) was purchased from American Radiolabeled Chemicals Inc. (St. Louis, MO, U.S.A.). Gly-Sar was purchased from Tokyo Kasei Kogyo Co. (Tokyo, Japan). Dulbecco's modified Eagle's medium
(DMEM) was acquired from Nacalai Tesque (Kyoto, Japan). Fetal bovine serum (FBS) was acquired from Biowest Inc. (Nuaille, France). All other chemicals were of the finest grade available.3,7) Cell culture and preparation of cell monolayers: Caco-2 cells at passage number 43 were obtained from the Riken Bioresource Center (Tsukuba, Japan), and all experiments were conducted with Caco-2 cells between passages 62 to 70. The cells were maintained by serial passage in plastic dishes with DMEM supplemented with 10% heat-inactivated FBS, 100 mM nonessential amino acids, 100 units/ml penicillin G, and 100 mg/ml streptomycin in an atmosphere of 5% CO2–95% air at 379 C. When the cells reached 80–90% confluence, they were subcultured using a 0.05% trypsin/0.53 mM EDTA solution. Caco-2 cells were seeded at a density 5×105 cells/cm2 on a 0.9 cm2 porous membrane (0.4 mm pore size) in a Falcon} cell culture insert (BD Bioscience, Bedford, MA, U.S.A.) to evaluate the transcellular transport of Gly-Sar. The seeded cells were maintained for 21 days to prepare differentiated cell monolayers. The volume of the medium was 1.0 and 2.3 ml for inside (apical side) and outside (basolateral side) the insert, respectively. Caco-2 cell monolayers whose TEER was above 500 Q･cm2 were used to assess the transcellular transport of Gly-Sar. On the other hand, the cells were seeded at a density 5×105 cells/cm2 on a 3.8 cm2 plastic dish using a Falcon} multiwellTM plate (BD Bioscience), and were maintained for 21 days to characterize the uptake of levofloxacin at the apical membrane.3,7) Pharmacokinetic analysis of transcellular transport of Gly-Sar: The transcellular transport of Gly-Sar in Caco-2 cell monolayers prepared on a porous membrane was examined as described previously.3,7) In brief, the monolayer was pre-incubated for 1 h at 379C with culture medium (pH 6–8) containing an unlabeled GlySar (50 or 500 mM) to equilibrate the drug concentration. After the 1-h equilibration period, [3H]Gly-Sar (0.4–0.6 mCi/well) was applied to the apical chamber (1.0 ml) to examine apical-to-basolateral transcellular transport. [14C]mannitol was used to estimate the paracellular transport and extracellular trapping of the radio-labeled drug. A volume (50 mL) of medium in the basolateral chamber (2.3 mL) was then collected after 1, 2, and 3 h. Cells on the porous membrane were collected following the last collection of medium. Radioactivity in the medium and cells was determined using a liquid scintillation counter, and normalized against the initially applied doses. The time course of the transport of Gly-Sar in the opposite (basolateral-to-apical) direction was examined in a similar manner. The transcellular transport of Gly-Sar was analyzed in a model-dependent manner using NONMEM software running on a mainframe UNIX machine at the Kyoto
Peptide and Levofloxacin Uptake in Caco-2 Cells
University Data Processing Center.3,7) That is, the following mass balance equations were prepared for the pharmacokinetic analysis: dXA CLAªC CLCªA CLAB CLAB ＝－ ･ X A＋ ･ X C－ ･XA＋ ･XB (1) dt VA VC VA VB dXB CLBªC CLCªB CLAB CLAB ＝－ ･ X B＋ ･ X C＋ ･XA－ ･XB (2) dt VB VC VA VB dXC CLAªC CLBªC CLCªA CLCªB ＝ ･ X A＋ ･XB－ ･ X C＋ ･XC (3) dt VA VB VC VC where XA, XB, and XC are the amount of [3H]Gly-Sar in the apical chamber, basolateral chamber, and monolayer determined at time t, respectively. VA and VB indicate the volume of the apical medium (1.0 ml) and basolateral medium (2.3 ml), respectively. VC indicates the cell volume (2.50 ml/cm2), measured with sulfanilamide as described previously.3,7) The influx and efflux clearance of [3H]Gly-Sar at the apical membrane of the cells was designated as CLAªC and CLCªA, respectively. The influx and efflux clearance of [3H]Gly-Sar at the basolateral membrane of the cells was designated as CLBªC and CLCªB, respectively. Paracellular transport clearance (CLAB) was estimated by analyzing the transport profile of [14C]mannitol using the following mass balance equations:
dXA CLAB CLAB ＝－ ･XA＋ ･ XB dt VA VB dXB CLAB CLAB ＝＋ ･ X A－ ･XB dt VA VB
(4) (5)
Inhibition of cellular uptake of Gly-Sar and levofloxacin by various compounds: The cellular uptake of [3H]Gly-Sar was examined using Caco-2 cell monolayers grown on porous membrane of a multiwell plate. The composition of the incubation medium was as follows: 125 mM NaCl, 4.8 mM KCl, 5.6 mM D-glucose, 1.2 mM CaCl2, 1.2 mM KH2PO4, 1.2 mM MgSO4･7H2O, 25 mM HEPES, and 50 mM Gly-Sar (pH 6.0). The monolayers were first pre-incubated for 60 min at 379C with 0.5 ml of incubation medium. In order to evaluate the effect of zwitterionic drugs and organic cations on the cellular uptake of [3H]Gly-Sar, the incubation medium was replaced with fresh incubation medium supplemented with various compounds at 5 min before the addition of [3H]Gly-Sar (0.4 mCi/well). [14C]mannitol was used to estimate extracellular trapping of the radio-labeled drug. After the cells were incubated with [3H]GlySar for another 15 min at 379C, they were immediately washed with ice-cold phosphate buffer, and collected. The amount of [3H]Gly-Sar in the cells was determined using a liquid scintillation counter, and normalized against the initially applied doses. The effect of various compounds on the apical uptake of [14C]levofloxacin in Caco-2 cell monolayers prepared
375
on plastic dishes of a multiwell plate was also evaluated at 379C in the presence of 100 mM levofloxacin. The incubation medium was replaced with fresh incubation medium supplemented with various compounds at 5 min before the addition of [14C]levofloxacin (0.2 mCi/well). [3H]mannitol was used to estimate extracellular trapping of the radio-labeled drug. The cells were incubated with [14C]levofloxacin for 15 min at 379C, and the amount of [14C]levofloxacin in the cells was determined as described above. Statistical analysis: Values are expressed as the mean±S.E. In all figures, when error bars are not shown, they are smaller than the symbol. The statistical significance of the difference between mean values was calculated using a non-paired t-test. Multiple comparisons were performed using Scheffeá 's test following oneway ANOVA provided that the variances of groups were similar. If this was not the case, Scheffeá -type test was applied following Kruskal-Wallis analysis. Pº0.05 was considered to be statistically significant.
Results and Discussion Transcellular transport and cellular accumulation of Gly-Sar in Caco-2 cell monolayers: We first evaluated the transcellular transport and cellular accumulation of Gly-Sar in Caco-2 cell monolayers grown on porous membrane filters. Fig. 1 shows the profiles of transcellular transport and cellular accumulation of GlySar in Caco-2 cells. Transcellular transport of Gly-Sar in the apical-to-basolateral direction was similar to that in the opposite direction at apical and basolateral medium pH 8 (Fig. 1A). On the other hand, the amount of GlySar transported in the apical-to-basolateral direction was greater than that in the opposite direction at apical and basolateral medium pH 7 and 6 (Figs. 1B and 1C). When unlabeled Gly-Sar concentration was increased from 50 mM to 500 mM, the fraction of transport of the radio-labeled drug was decreased only slightly (Figs. 1C and 1D). In addition, the amount of Gly-Sar accumulated in cells from the apical side within 3 h was greater than that from the basolateral side in any experimental condition used (Fig. 1). These results confirmed that Gly-Sar was transported by a pH-dependent transport system at the apical membrane in Caco-2 cells.7) Influx and efflux clearance of Gly-Sar in Caco-2 cell monolayers: Irie et al. reported that small peptides and some pharmacologically active compounds are absorbed from the small intestine by the apical H＋-coupled PEPT1 and the basolateral peptide transporter.8) In order to evaluate the characteristics of apical and basolateral peptide transport in the Caco-2 cells, we performed pharmacokinetic analysis of the data on transcellular transport and cellular accumulation of Gly-Sar (Fig. 1). Table 1 shows membrane transport clearance of GlySar in Caco-2 cells. Influx clearance value at the apical
376
Shiro FUKUMORI, et al.
Fig. 1. Transcellular transport and cellular accumulation of Gly-Sar in Caco–2 cell monolayers A: 50 mM at pH 8, B: 50 mM at pH 7, C: 50 mM at pH 6, D: 500 mM at pH 6. Open and closed circles represent apical-to-basolateral (AªB) and basolateral-to-apical (BªA) transport, respectively. Solid and broken lines are simulation curves obtained by pharmacokinetic analysis of Gly-Sar and mannitol transport, respectively. Open and closed columns represent cellular accumulation from the apical (AªC) and basolateral (BªC) side, respectively. Data are expressed as the mean±S.E. for 7–10 measurements.
Table 1. Influx and efflux clearance values (nl/min/cm2) of GlySar in Caco-2 cell monolayers pH
Conc. ( mM)
CLAªC
CLCªA
CLCªB
CLBªC
CLA B
8 7 6 6
50 50 50 500
74.5±3.0 110±4 131±6 117±3
31.5±2.1 28.4±1.0 39.1±2.0 33.9±1.0
11.3±1.0 14.8±1.2 18.2±1.7 13.6±1.2
29.2±0.9 43.8±1.6 37.9±1.6 32.8±1.1
15.6±0.8 12.8±0.5 14.1±0.5 13.9±0.5
CLAªC and CLCªA represent the influx and efflux clearance at the apical membrane, respectively. CLCªB and CLBªC represent the efflux and influx clearance at the basolateral membrane, respectively. CLAB represents the paracellular transport clearance. Values are expressed as the mean±S.E. for 7–10 measurements.
membrane (CLAªC) was much greater than basolateral influx (CLBªC) and efflux (CLCªB) clearance values in any experimental condition used; for example, the values of CLAªC and CLCªB for 50 mM Gly-Sar at the medium pH 6 were 131 and 18.2 nl/min/cm2, respectively (Table 1). This finding indicated that the apical peptide transporter plays an important role in the directional transcellular transport of Gly-Sar across Caco-2 cell monolayers. We previously performed pharmacokinetic analysis of the transcellular transport of levofloxacin across Caco-2
cell monolayers grown on porous membrane filters.3) The apical influx clearance value of levofloxacin at medium pH 6 (547 nl/min/cm2) was much greater than the basolateral influx clearance value (301 nl/min/cm2) in Caco-2 cells, which suggested that levofloxacin uptake at the apical membrane was mediated by a specific transport system.3) However, the apical influx clearance value of Gly-Sar at medium pH 6 (131 nl/min/cm2) in the present study (Table 1) was much lower than that of levofloxacin (547 nl/min/cm2). Inhibitory effect of fluoroquinolones and dipeptides on cellular uptake of Gly-Sar and levofloxacin in Caco-2 cell monolayers: To investigate the involvement of the peptide transporter in the apical membrane transport of levofloxacin in Caco-2 cells, we evaluated the inhibitory effect of fluoroquinolones and dipeptides on the apical uptake of [3H]Gly-Sar and [14C]levofloxacin at medium pH 6. The apical uptake of [3H]Gly-Sar in Caco-2 cell monolayers significantly decreased to 2% and 40% of control by treatment with 5 mM levofloxacin and lomefloxacin, respectively (Fig. 2A). Similarly, 5 mM Ala-Ala and Gly-Sar diminished the apical uptake of [3H]Gly-Sar (Fig. 2A). In our previous study, the apparent apical uptake of [14C]levofloxacin was
Peptide and Levofloxacin Uptake in Caco-2 Cells
Fig. 2. Inhibitory effect of fluoroquinolones and dipeptides on cellular uptake of Gly-Sar and levofloxacin in Caco-2 cell monolayers Cells were incubated with [3H]Gly-Sar for 15 min in the presence of 50 mM unlabeled Gly-Sar and 5 mM various compounds at medium pH 6 (A). Cells were incubated with [14C]levofloxacin for 15 min in the presence of 100 mM unlabeled levofloxacin and 5 mM various compounds at medium pH 6 (B). Closed column expresses the mean±S.E. for 15 measurements. Open column expresses the mean±S.E. for 5–8 measurements. *pº0.05: significantly different from the control.
not affected by 5 mM levofloxacin at medium pH 7.4, which suggested that unlabeled levofloxacin inhibited not only influx but also efflux of the radio-labeled drug at the apical membrane of Caco-2 cells.3) In contrast, the apical uptake of [14C]levofloxacin in Caco-2 cell monolayers was significantly inhibited by treatment with 5 mM levofloxacin and lomefloxacin at medium pH 6 (Fig. 2B). On the other hand, the uptake of [14C]levofloxacin was not changed by Ala-Ala, Gly-Sar (Fig. 2B). These results indicated that the peptide transporter is not involved in the apical uptake of levofloxacin in Caco-2 cells. Effect of various compounds on apical uptake of levofloxacin in Caco-2 cell nonolayers: We further evaluated the effect of peptide-like drugs, nucleoside derivatives, OATP inhibitors, organic cations and anion, and zwitterionic compounds on the apical uptake of levofloxacin at medium pH 6. Substrates of the peptide transporter, benzylpenicillin, cephalexin, and valacyclovir, had no effect on the apical uptake of levofloxacin (Fig. 3), confirming that the peptide transporter is not responsible for the apical uptake of levofloxacin into
377
Fig. 3. Effect of various compounds on cellular uptake of levofloxacin in Caco-2 cell monolayers grown on plastic dishes Cells were incubated with [14C]levofloxacin for 15 min in the presence of 100 mM unlabeled levofloxacin and various compounds at medium pH 6. Final concentrations of valacyclovir, erythromycin, rifamycin SV, and fexofenadine were 1 mM, and those of other compounds were 5 mM. Closed column expresses the mean±S.E. for 15 measurements. Open column expresses the mean±S.E. for 5–8 measurements. *pº0.05: significantly different from the control.
Caco-2 cells. Nucleoside derivatives, inosine, mizoribine, ribavirin, and AZT, and OATP inhibitors, erythromycin and rifamycin SV, had no significant effect on the uptake of levofloxacin (Fig. 3). In our previous study, the apical uptake of levofloxacin was significantly inhibited by an organic cation, imipramine, at medium pH 7.4.3) In contrast, imipramine had no significant effect on the apparent uptake of levofloxacin at medium pH 6 (Fig. 3), suggesting that imipramine inhibited not only influx but also efflux of levofloxacin at the apical membrane. On the other hand, another organic cation, nicotine, significantly decreased the uptake of levofloxacin to 10% of the control value (Fig. 3). An organic anion, enalapril, also decreased the uptake of levofloxacin to 14% of the control value. Furthermore, zwitterionic compounds, fexofenadine and L-carnitine, decreased the uptake of levofloxacin to 63% and 19% of the control value, respectively (Fig. 3). These results suggested that the apical uptake of levofloxacin was mediated by a specific transport system other than the peptide transporter. Hirano et al. reported that levofloxacin inhibits the Na＋-coupled apical uptake of L-carnitine in Caco-2 cells;
378
Shiro FUKUMORI, et al.
however, it is still unclear whether levofloxacin is transported by the carnitine transporter (OCTN2).9) Nicotine, enalapril, and fexofenadine may be useful inhibitors to investigate influx transporter(s) of levofloxacin. Further studies are needed to clarify whether these inhibitors affect the function of OCTN2 as well as the other apical transporters.
Conclusion The apical uptake of levofloxacin was not changed by dipeptides and substrates of the peptide transporter, but was significantly inhibited by several structurally unrelated compounds, such as nicotine, enalapril, fexofenadine, and L-carnitine. These findings indicated that the apical uptake transporter of levofloxacin is distinct from the peptide transporter in human intestinal epithelial Caco-2 cells.
3)
4)
5)
6)
7)
8)
References 1)
2)
Maeda, T., Takahashi, K., Ohtsu, N., Oguma, T., Ohnishi, T., Atsumi, R. and Tamai, I.: Identification of influx transporter for the quinolone antibacterial agent levofloxacin. Mol. Pharm., 4: 85–94 (2007). Yamaguchi, H., Yano, I., Saito, H. and Inui, K.: Transport characteristics of grepafloxacin and levofloxacin in the human intes-
9)
tinal cell line Caco-2. Eur. J. Pharmacol., 431: 297–303 (2001). Takaai, M., Suzuki, H., Ishida, K., Tahara, K. and Hashimoto, Y.: Pharmacokinetic analysis of transcellular transport of levofloxacin across LLC-PK1 and Caco-2 cell monolayers. Biol. Pharm. Bull., 30: 2167–2172 (2007). Irie, M., Terada, T., Tsuda, M., Katsura, T. and Inui, K.: Prediction of glycylsarcosine transport in Caco-2 cell lines expressing PEPT1 at different levels. Pflugers. Arch., 452: 64–70 (2006). Rabbaa, L., Dautrey, S., Colas-Linhart, N., Carbon, C. and Farinotti, R.: Absorption of ofloxacin isomers in the rat small intestine. Antimicrob. Agents Chemother., 41: 2274–2277 (1997). Yamaguchi, T., Yokogawa, M., Sekine, Y. and Hashimoto, M.: Intestinal absorption characteristics of sparfloxacin. Xenobiot. Metab. Dispos., 6: 53–59 (1991). Ishida, K., Takaai, M. and Hashimoto, Y.: Pharmacokinetic analysis of transcellular transport of quinidine across monolayers of human intestinal epithelial Caco-2 cells. Biol. Pharm. Bull., 29: 522–526 (2006). Irie, M., Terada, T., Okuda, M. and Inui, K.: Efflux properties of basolateral peptide transporter in human intestinal cell line Caco-2. Pflugers. Arch., 449: 186–194 (2004). Hirano, T., Yasuda, S., Osaka, Y., Kobayashi, M., Itagaki, S. and Iseki, K.: Mechanism of the inhibitory effect of zwitterionic drugs (levofloxacin and grepafloxacin) on carnitine transporter (OCTN2) in Caco-2 cells. Biochim. Biophys. Acta., 1758: 1743–1750 (2006).
Note The Apical Uptake Transporter of Levofloxacin is Distinct from the Peptide Transporter in Human Intestinal Epithelial Caco-2 Cells Shiro FUKUMORI, Toshiya MURATA, Mari TAKAAI, Katsutoshi TAHARA, Masato TAGUCHI and Yukiya HASHIMOTO* Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama, Toyama, Japan Full text of this paper is available at http://www.jstage.jst.go.jp/browse/dmpk
Summary: The aim of this study was to investigate the involvement of the peptide transporter for absorption of levofloxacin in Caco-2 cells. To evaluate the activity of apical and basolateral peptide transport, we first performed pharmacokinetic analysis of transcellular transport of glycylsarcosine (Gly-Sar) in cell monolayers grown on porous membrane filters. Transcellular transport of Gly-Sar at the medium pH 6 was greater in the apical-to-basolateral direction than in the opposite direction. Influx clearance of Gly-Sar at the apical membrane was much greater than basolateral influx and efflux clearance, indicating that the apical peptide transporter plays an important role in directional transcellular transport of the dipeptide across Caco-2 cell monolayers. We then evaluated the effect of various compounds on the uptake of Gly-Sar and levofloxacin at the apical membrane of Caco-2 cells. The apical uptake of [3H]Gly-Sar was significantly inhibited by Ala-Ala, Gly-Sar, and also levofloxacin, whereas that of [14C]levofloxacin was not inhibited by Ala-Ala and Gly-Sar. On the other hand, the apical uptake of [14C]levofloxacin was inhibited by nicotine, enalapril, fexofenadine, and L-carnitine. These findings indicated that the apical uptake transporter of levofloxacin is distinct from the peptide transporter in Caco-2 cells. Keywords: levofloxacin, Caco-2 cells, peptide transporter, OATP, glycylsarcosine
levofloxacin at medium pH 6 was much greater than the basolateral influx clearance value in Caco-2 cells.3) These results suggested that levofloxacin uptake across the apical membrane in Caco-2 cells was mediated by a specific transport system. Maeda et al. evaluated the levofloxacin uptake activity of Caco-2 subclones, and selected candidate transporter genes functioning for influx transport of levofloxacin.1) Based on functional analysis of each transporter gene for which a good correlation was found between the expression level and levofloxacin transport activity in Caco-2 subclones, organic anion transporting polypeptide 1A2 (OATP1A2) was concluded to transport levofloxacin. When OATP1A2 was expressed in Xenopus oocytes, levofloxacin transport was essentially pH-independent. OATP1A2-mediated uptake of levofloxacin showed a Km value of 136 mM. Apparent uptake of levofloxacin by Caco-2 cells showed high- and low-affinity components with Km values of 0.489 and 14.6 mM, respectively. Accordingly, plural transporters are functional for the
Introduction Levofloxacin is well absorbed from the intestine, and the bioavailability following oral administration is approximately 100% in humans.1) Since levofloxacin is a zwitterionic compound, a passive diffusion mechanism may not fully explain the high intestinal absorption.1) Yamaguchi et al. evaluated the transport characteristics of levofloxacin using human intestinal epithelial Caco-2 cells grown on plastic dishes.2) Apical uptake of levofloxacin in Caco-2 cells was rapid, and almost reached a steady state at 15 min after the start of incubation. The uptake of levofloxacin was markedly decreased by lowering the temperature (49C) and showed concentrationdependent saturation.2) In order to evaluate the membrane transport responsible for rapid intestinal absorption of levofloxacin, we performed pharmacokinetic analysis of the transcellular transport of levofloxacin across Caco-2 cell monolayers grown on porous membrane filters.3) The apical influx clearance value of
Received; February 25, 2008, Accepted; March 24, 2008 *To whom correspondence should be addressed: Yukiya HASHIMOTO, Ph.D., 2630 Sugitani, Toyama 930–0194, Japan. Tel. ＋81-76-434-7585, Fax. ＋81-76-434-7587, E-mail: yukiya＠pha.u-toyama.ac.jp This work was supported in part by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Sciences (JSPS).
373
374
Shiro FUKUMORI, et al.
transport of levofloxacin in Caco-2 cells, and OATP1A2 is likely to function as a high-affinity transporter.1) However, it is still unclear which transporter is responsible for the lower-affinity (high-capacity) uptake of levofloxacin in Caco-2 cells. The peptide transporter (PEPT1) is one of the wellknown transporters, which are expressed in brush-border membranes of intestinal epithelial cells and are involved in the absorption of zwitterionic compounds from intestinal lumen into cells.4) Rabbaa et al. evaluated the absorption characteristics of S-ofloxacin (levofloxacin) in the rat intestine to clarify the mechanism involved in intestinal transport of the drug.5) Dipeptides and b-lactam antibiotics did not modify levofloxacin absorption, suggesting that levofloxacin absorption is probably independent of the dipeptide transport system in the rat intestine.5) In contrast, Yamaguchi et al. reported the absorption characteristics of sparfloxacin in rats using the in situ ligated loop method and the everted jejunum uptake method in vitro.6) The absorption of sparfloxacin in rats with a jejunum loop was reduced to 60% of the control by the addition of di-, tripeptides, and cyclacillin. The in vitro uptake of sparfloxacin into the rat everted jejunum was inhibited by the addition of di-, tripeptides, and blactam antibiotics. In addition, glycylglycine competitively inhibited the uptake of sparfloxacin. These results indicated that approximately 40% of sparfloxacin dose can be transported by a common dipeptide carrier-mediated system.6) The primary aim of the present study was to systematically investigate the involvement of peptide transporter in the transport of levofloxacin across human intestinal Caco-2 cell monolayers. In order to evaluate the activity of apical and basolateral peptide transport in Caco-2 cells, we first performed pharmacokinetic analysis of transcellular transport of glycylsarcosine (Gly-Sar) in cell monolayers grown on porous membrane filters. We then evaluated the effect of unlabeled levofloxacin, lomefloxacin, Ala-Ala, and Gly-Sar on the uptake of radio-labeled Gly-Sar and levofloxacin at the apical membrane of the cells. Furthermore, we also investigated the effect of peptide-like drugs, nucleoside derivatives, organic cations and anion, and zwitterionic compounds on the apical uptake of levofloxacin in Caco-2 cells.
Materials and Methods Materials: [3H]glycylsarcosine (Gly-Sar) (18.5 GBq/ mmol) and [14C]mannitol (1.96 GBq/mmol) were purchased from Moravek Biochemicals Inc. (Brea, CA, U.S.A.). [14C]levofloxacin (2.43 MBq/mg) was obtained from Daiichi Pure Chemicals Co. (Ibaraki, Japan). [3H]mannitol (740 GBq/mmol) was purchased from American Radiolabeled Chemicals Inc. (St. Louis, MO, U.S.A.). Gly-Sar was purchased from Tokyo Kasei Kogyo Co. (Tokyo, Japan). Dulbecco's modified Eagle's medium
(DMEM) was acquired from Nacalai Tesque (Kyoto, Japan). Fetal bovine serum (FBS) was acquired from Biowest Inc. (Nuaille, France). All other chemicals were of the finest grade available.3,7) Cell culture and preparation of cell monolayers: Caco-2 cells at passage number 43 were obtained from the Riken Bioresource Center (Tsukuba, Japan), and all experiments were conducted with Caco-2 cells between passages 62 to 70. The cells were maintained by serial passage in plastic dishes with DMEM supplemented with 10% heat-inactivated FBS, 100 mM nonessential amino acids, 100 units/ml penicillin G, and 100 mg/ml streptomycin in an atmosphere of 5% CO2–95% air at 379 C. When the cells reached 80–90% confluence, they were subcultured using a 0.05% trypsin/0.53 mM EDTA solution. Caco-2 cells were seeded at a density 5×105 cells/cm2 on a 0.9 cm2 porous membrane (0.4 mm pore size) in a Falcon} cell culture insert (BD Bioscience, Bedford, MA, U.S.A.) to evaluate the transcellular transport of Gly-Sar. The seeded cells were maintained for 21 days to prepare differentiated cell monolayers. The volume of the medium was 1.0 and 2.3 ml for inside (apical side) and outside (basolateral side) the insert, respectively. Caco-2 cell monolayers whose TEER was above 500 Q･cm2 were used to assess the transcellular transport of Gly-Sar. On the other hand, the cells were seeded at a density 5×105 cells/cm2 on a 3.8 cm2 plastic dish using a Falcon} multiwellTM plate (BD Bioscience), and were maintained for 21 days to characterize the uptake of levofloxacin at the apical membrane.3,7) Pharmacokinetic analysis of transcellular transport of Gly-Sar: The transcellular transport of Gly-Sar in Caco-2 cell monolayers prepared on a porous membrane was examined as described previously.3,7) In brief, the monolayer was pre-incubated for 1 h at 379C with culture medium (pH 6–8) containing an unlabeled GlySar (50 or 500 mM) to equilibrate the drug concentration. After the 1-h equilibration period, [3H]Gly-Sar (0.4–0.6 mCi/well) was applied to the apical chamber (1.0 ml) to examine apical-to-basolateral transcellular transport. [14C]mannitol was used to estimate the paracellular transport and extracellular trapping of the radio-labeled drug. A volume (50 mL) of medium in the basolateral chamber (2.3 mL) was then collected after 1, 2, and 3 h. Cells on the porous membrane were collected following the last collection of medium. Radioactivity in the medium and cells was determined using a liquid scintillation counter, and normalized against the initially applied doses. The time course of the transport of Gly-Sar in the opposite (basolateral-to-apical) direction was examined in a similar manner. The transcellular transport of Gly-Sar was analyzed in a model-dependent manner using NONMEM software running on a mainframe UNIX machine at the Kyoto
Peptide and Levofloxacin Uptake in Caco-2 Cells
University Data Processing Center.3,7) That is, the following mass balance equations were prepared for the pharmacokinetic analysis: dXA CLAªC CLCªA CLAB CLAB ＝－ ･ X A＋ ･ X C－ ･XA＋ ･XB (1) dt VA VC VA VB dXB CLBªC CLCªB CLAB CLAB ＝－ ･ X B＋ ･ X C＋ ･XA－ ･XB (2) dt VB VC VA VB dXC CLAªC CLBªC CLCªA CLCªB ＝ ･ X A＋ ･XB－ ･ X C＋ ･XC (3) dt VA VB VC VC where XA, XB, and XC are the amount of [3H]Gly-Sar in the apical chamber, basolateral chamber, and monolayer determined at time t, respectively. VA and VB indicate the volume of the apical medium (1.0 ml) and basolateral medium (2.3 ml), respectively. VC indicates the cell volume (2.50 ml/cm2), measured with sulfanilamide as described previously.3,7) The influx and efflux clearance of [3H]Gly-Sar at the apical membrane of the cells was designated as CLAªC and CLCªA, respectively. The influx and efflux clearance of [3H]Gly-Sar at the basolateral membrane of the cells was designated as CLBªC and CLCªB, respectively. Paracellular transport clearance (CLAB) was estimated by analyzing the transport profile of [14C]mannitol using the following mass balance equations:
dXA CLAB CLAB ＝－ ･XA＋ ･ XB dt VA VB dXB CLAB CLAB ＝＋ ･ X A－ ･XB dt VA VB
(4) (5)
Inhibition of cellular uptake of Gly-Sar and levofloxacin by various compounds: The cellular uptake of [3H]Gly-Sar was examined using Caco-2 cell monolayers grown on porous membrane of a multiwell plate. The composition of the incubation medium was as follows: 125 mM NaCl, 4.8 mM KCl, 5.6 mM D-glucose, 1.2 mM CaCl2, 1.2 mM KH2PO4, 1.2 mM MgSO4･7H2O, 25 mM HEPES, and 50 mM Gly-Sar (pH 6.0). The monolayers were first pre-incubated for 60 min at 379C with 0.5 ml of incubation medium. In order to evaluate the effect of zwitterionic drugs and organic cations on the cellular uptake of [3H]Gly-Sar, the incubation medium was replaced with fresh incubation medium supplemented with various compounds at 5 min before the addition of [3H]Gly-Sar (0.4 mCi/well). [14C]mannitol was used to estimate extracellular trapping of the radio-labeled drug. After the cells were incubated with [3H]GlySar for another 15 min at 379C, they were immediately washed with ice-cold phosphate buffer, and collected. The amount of [3H]Gly-Sar in the cells was determined using a liquid scintillation counter, and normalized against the initially applied doses. The effect of various compounds on the apical uptake of [14C]levofloxacin in Caco-2 cell monolayers prepared
375
on plastic dishes of a multiwell plate was also evaluated at 379C in the presence of 100 mM levofloxacin. The incubation medium was replaced with fresh incubation medium supplemented with various compounds at 5 min before the addition of [14C]levofloxacin (0.2 mCi/well). [3H]mannitol was used to estimate extracellular trapping of the radio-labeled drug. The cells were incubated with [14C]levofloxacin for 15 min at 379C, and the amount of [14C]levofloxacin in the cells was determined as described above. Statistical analysis: Values are expressed as the mean±S.E. In all figures, when error bars are not shown, they are smaller than the symbol. The statistical significance of the difference between mean values was calculated using a non-paired t-test. Multiple comparisons were performed using Scheffeá 's test following oneway ANOVA provided that the variances of groups were similar. If this was not the case, Scheffeá -type test was applied following Kruskal-Wallis analysis. Pº0.05 was considered to be statistically significant.
Results and Discussion Transcellular transport and cellular accumulation of Gly-Sar in Caco-2 cell monolayers: We first evaluated the transcellular transport and cellular accumulation of Gly-Sar in Caco-2 cell monolayers grown on porous membrane filters. Fig. 1 shows the profiles of transcellular transport and cellular accumulation of GlySar in Caco-2 cells. Transcellular transport of Gly-Sar in the apical-to-basolateral direction was similar to that in the opposite direction at apical and basolateral medium pH 8 (Fig. 1A). On the other hand, the amount of GlySar transported in the apical-to-basolateral direction was greater than that in the opposite direction at apical and basolateral medium pH 7 and 6 (Figs. 1B and 1C). When unlabeled Gly-Sar concentration was increased from 50 mM to 500 mM, the fraction of transport of the radio-labeled drug was decreased only slightly (Figs. 1C and 1D). In addition, the amount of Gly-Sar accumulated in cells from the apical side within 3 h was greater than that from the basolateral side in any experimental condition used (Fig. 1). These results confirmed that Gly-Sar was transported by a pH-dependent transport system at the apical membrane in Caco-2 cells.7) Influx and efflux clearance of Gly-Sar in Caco-2 cell monolayers: Irie et al. reported that small peptides and some pharmacologically active compounds are absorbed from the small intestine by the apical H＋-coupled PEPT1 and the basolateral peptide transporter.8) In order to evaluate the characteristics of apical and basolateral peptide transport in the Caco-2 cells, we performed pharmacokinetic analysis of the data on transcellular transport and cellular accumulation of Gly-Sar (Fig. 1). Table 1 shows membrane transport clearance of GlySar in Caco-2 cells. Influx clearance value at the apical
376
Shiro FUKUMORI, et al.
Fig. 1. Transcellular transport and cellular accumulation of Gly-Sar in Caco–2 cell monolayers A: 50 mM at pH 8, B: 50 mM at pH 7, C: 50 mM at pH 6, D: 500 mM at pH 6. Open and closed circles represent apical-to-basolateral (AªB) and basolateral-to-apical (BªA) transport, respectively. Solid and broken lines are simulation curves obtained by pharmacokinetic analysis of Gly-Sar and mannitol transport, respectively. Open and closed columns represent cellular accumulation from the apical (AªC) and basolateral (BªC) side, respectively. Data are expressed as the mean±S.E. for 7–10 measurements.
Table 1. Influx and efflux clearance values (nl/min/cm2) of GlySar in Caco-2 cell monolayers pH
Conc. ( mM)
CLAªC
CLCªA
CLCªB
CLBªC
CLA B
8 7 6 6
50 50 50 500
74.5±3.0 110±4 131±6 117±3
31.5±2.1 28.4±1.0 39.1±2.0 33.9±1.0
11.3±1.0 14.8±1.2 18.2±1.7 13.6±1.2
29.2±0.9 43.8±1.6 37.9±1.6 32.8±1.1
15.6±0.8 12.8±0.5 14.1±0.5 13.9±0.5
CLAªC and CLCªA represent the influx and efflux clearance at the apical membrane, respectively. CLCªB and CLBªC represent the efflux and influx clearance at the basolateral membrane, respectively. CLAB represents the paracellular transport clearance. Values are expressed as the mean±S.E. for 7–10 measurements.
membrane (CLAªC) was much greater than basolateral influx (CLBªC) and efflux (CLCªB) clearance values in any experimental condition used; for example, the values of CLAªC and CLCªB for 50 mM Gly-Sar at the medium pH 6 were 131 and 18.2 nl/min/cm2, respectively (Table 1). This finding indicated that the apical peptide transporter plays an important role in the directional transcellular transport of Gly-Sar across Caco-2 cell monolayers. We previously performed pharmacokinetic analysis of the transcellular transport of levofloxacin across Caco-2
cell monolayers grown on porous membrane filters.3) The apical influx clearance value of levofloxacin at medium pH 6 (547 nl/min/cm2) was much greater than the basolateral influx clearance value (301 nl/min/cm2) in Caco-2 cells, which suggested that levofloxacin uptake at the apical membrane was mediated by a specific transport system.3) However, the apical influx clearance value of Gly-Sar at medium pH 6 (131 nl/min/cm2) in the present study (Table 1) was much lower than that of levofloxacin (547 nl/min/cm2). Inhibitory effect of fluoroquinolones and dipeptides on cellular uptake of Gly-Sar and levofloxacin in Caco-2 cell monolayers: To investigate the involvement of the peptide transporter in the apical membrane transport of levofloxacin in Caco-2 cells, we evaluated the inhibitory effect of fluoroquinolones and dipeptides on the apical uptake of [3H]Gly-Sar and [14C]levofloxacin at medium pH 6. The apical uptake of [3H]Gly-Sar in Caco-2 cell monolayers significantly decreased to 2% and 40% of control by treatment with 5 mM levofloxacin and lomefloxacin, respectively (Fig. 2A). Similarly, 5 mM Ala-Ala and Gly-Sar diminished the apical uptake of [3H]Gly-Sar (Fig. 2A). In our previous study, the apparent apical uptake of [14C]levofloxacin was
Peptide and Levofloxacin Uptake in Caco-2 Cells
Fig. 2. Inhibitory effect of fluoroquinolones and dipeptides on cellular uptake of Gly-Sar and levofloxacin in Caco-2 cell monolayers Cells were incubated with [3H]Gly-Sar for 15 min in the presence of 50 mM unlabeled Gly-Sar and 5 mM various compounds at medium pH 6 (A). Cells were incubated with [14C]levofloxacin for 15 min in the presence of 100 mM unlabeled levofloxacin and 5 mM various compounds at medium pH 6 (B). Closed column expresses the mean±S.E. for 15 measurements. Open column expresses the mean±S.E. for 5–8 measurements. *pº0.05: significantly different from the control.
not affected by 5 mM levofloxacin at medium pH 7.4, which suggested that unlabeled levofloxacin inhibited not only influx but also efflux of the radio-labeled drug at the apical membrane of Caco-2 cells.3) In contrast, the apical uptake of [14C]levofloxacin in Caco-2 cell monolayers was significantly inhibited by treatment with 5 mM levofloxacin and lomefloxacin at medium pH 6 (Fig. 2B). On the other hand, the uptake of [14C]levofloxacin was not changed by Ala-Ala, Gly-Sar (Fig. 2B). These results indicated that the peptide transporter is not involved in the apical uptake of levofloxacin in Caco-2 cells. Effect of various compounds on apical uptake of levofloxacin in Caco-2 cell nonolayers: We further evaluated the effect of peptide-like drugs, nucleoside derivatives, OATP inhibitors, organic cations and anion, and zwitterionic compounds on the apical uptake of levofloxacin at medium pH 6. Substrates of the peptide transporter, benzylpenicillin, cephalexin, and valacyclovir, had no effect on the apical uptake of levofloxacin (Fig. 3), confirming that the peptide transporter is not responsible for the apical uptake of levofloxacin into
377
Fig. 3. Effect of various compounds on cellular uptake of levofloxacin in Caco-2 cell monolayers grown on plastic dishes Cells were incubated with [14C]levofloxacin for 15 min in the presence of 100 mM unlabeled levofloxacin and various compounds at medium pH 6. Final concentrations of valacyclovir, erythromycin, rifamycin SV, and fexofenadine were 1 mM, and those of other compounds were 5 mM. Closed column expresses the mean±S.E. for 15 measurements. Open column expresses the mean±S.E. for 5–8 measurements. *pº0.05: significantly different from the control.
Caco-2 cells. Nucleoside derivatives, inosine, mizoribine, ribavirin, and AZT, and OATP inhibitors, erythromycin and rifamycin SV, had no significant effect on the uptake of levofloxacin (Fig. 3). In our previous study, the apical uptake of levofloxacin was significantly inhibited by an organic cation, imipramine, at medium pH 7.4.3) In contrast, imipramine had no significant effect on the apparent uptake of levofloxacin at medium pH 6 (Fig. 3), suggesting that imipramine inhibited not only influx but also efflux of levofloxacin at the apical membrane. On the other hand, another organic cation, nicotine, significantly decreased the uptake of levofloxacin to 10% of the control value (Fig. 3). An organic anion, enalapril, also decreased the uptake of levofloxacin to 14% of the control value. Furthermore, zwitterionic compounds, fexofenadine and L-carnitine, decreased the uptake of levofloxacin to 63% and 19% of the control value, respectively (Fig. 3). These results suggested that the apical uptake of levofloxacin was mediated by a specific transport system other than the peptide transporter. Hirano et al. reported that levofloxacin inhibits the Na＋-coupled apical uptake of L-carnitine in Caco-2 cells;
378
Shiro FUKUMORI, et al.
however, it is still unclear whether levofloxacin is transported by the carnitine transporter (OCTN2).9) Nicotine, enalapril, and fexofenadine may be useful inhibitors to investigate influx transporter(s) of levofloxacin. Further studies are needed to clarify whether these inhibitors affect the function of OCTN2 as well as the other apical transporters.
Conclusion The apical uptake of levofloxacin was not changed by dipeptides and substrates of the peptide transporter, but was significantly inhibited by several structurally unrelated compounds, such as nicotine, enalapril, fexofenadine, and L-carnitine. These findings indicated that the apical uptake transporter of levofloxacin is distinct from the peptide transporter in human intestinal epithelial Caco-2 cells.
3)
4)
5)
6)
7)
8)
References 1)
2)
Maeda, T., Takahashi, K., Ohtsu, N., Oguma, T., Ohnishi, T., Atsumi, R. and Tamai, I.: Identification of influx transporter for the quinolone antibacterial agent levofloxacin. Mol. Pharm., 4: 85–94 (2007). Yamaguchi, H., Yano, I., Saito, H. and Inui, K.: Transport characteristics of grepafloxacin and levofloxacin in the human intes-
9)
tinal cell line Caco-2. Eur. J. Pharmacol., 431: 297–303 (2001). Takaai, M., Suzuki, H., Ishida, K., Tahara, K. and Hashimoto, Y.: Pharmacokinetic analysis of transcellular transport of levofloxacin across LLC-PK1 and Caco-2 cell monolayers. Biol. Pharm. Bull., 30: 2167–2172 (2007). Irie, M., Terada, T., Tsuda, M., Katsura, T. and Inui, K.: Prediction of glycylsarcosine transport in Caco-2 cell lines expressing PEPT1 at different levels. Pflugers. Arch., 452: 64–70 (2006). Rabbaa, L., Dautrey, S., Colas-Linhart, N., Carbon, C. and Farinotti, R.: Absorption of ofloxacin isomers in the rat small intestine. Antimicrob. Agents Chemother., 41: 2274–2277 (1997). Yamaguchi, T., Yokogawa, M., Sekine, Y. and Hashimoto, M.: Intestinal absorption characteristics of sparfloxacin. Xenobiot. Metab. Dispos., 6: 53–59 (1991). Ishida, K., Takaai, M. and Hashimoto, Y.: Pharmacokinetic analysis of transcellular transport of quinidine across monolayers of human intestinal epithelial Caco-2 cells. Biol. Pharm. Bull., 29: 522–526 (2006). Irie, M., Terada, T., Okuda, M. and Inui, K.: Efflux properties of basolateral peptide transporter in human intestinal cell line Caco-2. Pflugers. Arch., 449: 186–194 (2004). Hirano, T., Yasuda, S., Osaka, Y., Kobayashi, M., Itagaki, S. and Iseki, K.: Mechanism of the inhibitory effect of zwitterionic drugs (levofloxacin and grepafloxacin) on carnitine transporter (OCTN2) in Caco-2 cells. Biochim. Biophys. Acta., 1758: 1743–1750 (2006).