Multiplicity of the H+-Dependent Transport Mechanism of Dipeptide and Anionic β-Lactam Antibiotic Ceftibuten in Rat Intestinal Brush-Border Membrane
- Ken Iseki,
- Mitsuru Sugawara,
- Kaori Sato,
- Imad Naasani1,
- Tomohisa Hayakawa,
- Michiya Kobayashi and
- Katsumi Miyazaki
- Department of Pharmacy, Hokkaido University Hospital, School of Medicine, Hokkaido University, Kita-ku, Sapporo, Japan
Abstract
To elucidate the transport characteristics of the H+/dipeptide carrier that recognizes the orally active β-lactam antibiotic ceftibuten, the uptake behaviors were compared of ceftibuten and Gly-Sar by rat intestinal brush-border membrane vesicles. The results show that 1) both the uptake of ceftibuten and that of Gly-Sar were dependent on an inwardly directed H+gradient; 2) anionic compounds such as hippurylphenyllactic acid competitively inhibited ceftibuten uptake in the presence of H+ gradient, whereas this anion did not inhibit Gly-Sar uptake; and 3) the carrier-mediated uptake of ceftibuten did not disappear even in the presence of 20 mM Gly-Sar. The results provide an evidence that several transporters with different features are potentially responsible for the uptake of β-lactam antibiotics into the intestinal cells. It is suggested that the dianionic β-lactam antibiotics that carry a net negative charge such as ceftibuten use multiple H+-dependent transport systems for absorption.
The excellent oral availability of certain β-lactam antibiotics is explained by the fact that they serve as substrates for intestinal oligopeptide transporter or transporters (Okano et al., 1986; Tsuji et al., 1986; Naasani et al., 1995). An intestinal peptide transporter (PepT1) has recently been cloned from rabbit (Boll et al., 1994; Fei et al., 1994), rat (Saito et al., 1995), and human (Liang et al., 1995) small intestinal cDNA libraries. Xenopus laevis oocytes injected with the PepT1 cDNA express a transport activity that is characterized as Na+, K+, and Cl− independent but electrogenic as a consequence of peptide/H+ cotransport. PepT1 appears to have a broad substrate specificity accepting dipeptides and tripeptides (Boll et al., 1994; Fei et al., 1994; Liang et al., 1995), as well as a variety of peptide mimetics including some β-lactam antibiotics (Saito et al., 1995; Wenzel and Thwaites, 1995; Wenzel et al., 1996; Terada et al., 1997) and selected angiotensin-converting enzyme inhibitors (Boll et al., 1994; Swaan et al., 1995). Although some functional information on the intestinal peptide transporters are available (Fei et al., 1994; Erickson et al., 1995; Miyamoto et al., 1996), the operational mode of PepT1 has not been established.
On the other hand, it has been suggested that the dianionic β-lactam antibiotics carrying net negative charge like ceftibuten use a different transport system for absorption (Sugawara et al., 1992;Kramer, 1993), although these compounds interact with the H+-coupled transporter for dipeptide. This additional transport system in the intestine has not yet been characterized at the molecular level, and the concept of the presence of multiple transporters for β-lactam antibiotics is not supported by kinetic studies (Naasani et al., 1995). Recently, the two protein fractions that bind to immobilized ceftibuten ligand were purified from rat intestinal brush-border membrane (BBM) (Iseki et al., 1998). The uptake of dianionic β-lactams into the intestinal BBM vesicles or the human Caco-2 cells also has shown a distinct pH dependence compared with the uptake of zwitterionic compounds (Inui et al., 1988; Kramer et al., 1993; Muranushi et al., 1994).
The present study was undertaken to characterize the uptake of anionic β-lactam antibiotics in the intestinal BBM. In particular, the transport characteristics of ceftibuten, an orally active dianionic β-lactam antibiotic, was compared with that of Gly-Sar, a dipeptide with α-amino group.
Experimental Procedures
Materials.
β-Lactam antibiotics, ceftibuten, and compound V were kindly donated by Shionogi Co. (Osaka, Japan), and cefixime was kindly donated by Fujisawa Pharmaceutical Co. (Osaka, Japan). [14C]Gly-Sar (110 mCi/mmol) was purchased from Moravek Biochemicals, Inc. (Brea, CA). All other chemicals were of the highest grade available and were used without further purification.
Preparation of BBM Vesicles.
The experimental protocols were reviewed and approved by the Hokkaido University Animal Care Committee in accordance with the “Guide for the Care and Use of Laboratory Animals” as adopted by the National Institutes of Health. BBM vesicles were prepared from the intestine of the male Wistar rat (200–300 g) by the magnesium/EGTA precipitation method according toHauser et al. (1980) and Booth and Kenny (1974) with some modifications described previously (Sugawara et al., 1998). Unless otherwise specified, the suspending buffer was 20 mM HEPES/Tris, pH 7.5, containing 100 mM d-mannitol and 100 mM KCl. Enrichment of the BBM fraction was routinely more than 10-fold compared with the homogenate as revealed from the assessment of the specific activity of the membrane enzyme marker alkaline phosphatase. Measurement of the Na+-dependent uptake of alanine, a typical substrate for the amino acid transport system, demonstrated the functional integrity of the membrane.
Uptake Experiments.
The uptake of substrates by the freshly isolated membrane vesicles was performed at 25°C using to the method of Sugawara et al. (1992). The reaction was initiated by mixing 40 μl of membrane vesicle suspension (10∼15 mg protein/ml) with 200 μl of the transport buffer [unless otherwise, the transport buffer was composed of 100 mM d-mannitol, 100 mM KCl, 20 mM 2-(N-morpholino)ethanesulfonic acid/Tris, pH 5.5] containing substrates. Then, after a determined time, the reaction was terminated by diluting the reaction mixture with 4 ml of the ice-cold stop buffer (150 mM NaCl, 20 mM HEPES/Tris, pH 7.5) followed by filtration through a Millipore filter (0.45 μm, 2.5-cm diameter; HAWP). The filter was then washed once with 8 ml of the ice-cold stop buffer. Substrate trapped on the filter was extracted with 300 μl of the stop buffer.
Analytical Procedures and Statistical Analyses.
The detection of ceftibuten in BBM vesicles was carried out by the use of HPLC as described previously (Sugawara et al., 1991). Separation of ceftibuten was achieved on a reversed phase column (5 μm, 4 mm i.d. × 250 mm; ODS Hitachi 3053) using a mobile phase consisting of acetonitrile/0.05 M citric acid/0.1 M KCl buffer, pH 2.5 (1:9). Samples were eluted at a flow rate of 0.7 ml/min, and the detection was set at 262 nm. [14C]Gly-Sar was measured by conventional liquid scintillation counting. Uptake was expressed relative to membrane protein. Protein was measured by the method ofLowry et al. (1951) with BSA as a standard. As a filter-adsorption blank, a membrane vesicle-free incubation medium was handled in an identical manner. All experiments were carried out in duplicate or triplicate with at least three preparations, and results presented as mean with S.E. The significance of differences between the uptake values with or without inhibitors was determined by a nonpairedt test. The nonlinear regression analysis and least-squares fitting for Eadie-Hofstee plot of ceftibuten uptake were calculated by Mac Curve Fit (version 1.0.8) on a Macintosh computer.
Results
Inhibitory Effect and Substrate Specificity of Ceftibuten Uptake.
To further characterize the substrate specificity of the transporter or transporters for ceftibuten in the intestinal BBM, the inhibitory effects of dipeptides, β-lactam antibiotics, and several anionic compounds were examined (Table1). The uptake of both ceftibuten and Gly-Sar was dependent on an inward H+ gradient, and the stimulatory effect of the H+ gradient was diminished by the presence of the protonophore FCCP (table I). These results agreed well with the uptake profile of H+/peptide-transporting substrates in human Caco-2 cells (Matsumoto et al., 1994) and intestinal epithelial cells (Tomita et al., 1995).
Effects of Inhibitors on ceftibuten and Gly-Sar uptake by rat intestinal BBM vesicles in the presence of an inward H+gradient
In the presence of the H+ gradient, ceftibuten uptake was significantly inhibited by the analogs (cefixime, compound V), dipeptides (l-Asp-l-Phe,l-Phe-l-Pro,l-Ala-l-Ala), and hippuryl phenyllactic acid, an organic acid with a peptide bond, whereas the interaction ofl-carnosine and Gly-Sar was markedly weak (Table 1). In contrast, neither hippurylphenyllactic acid norl-Asp-l-Phe inhibited Gly-Sar uptake, whereasl-carnosine and ceftibuten inhibited Gly-Sar uptake.
Dixon Plot Analysis of Ceftibuten Uptake in Presence of Inhibitors.
Dixon plot analysis of ceftibuten uptake in the presence of hippurylphenyllactic acid and an inwardly directed H+ gradient is shown in Fig.1. Hippurylphenyllactic acid inhibited the uptake of ceftibuten competitively. The regression line obtained from the replot of the slopes of Dixon plot almost coincided with the origin (Fig. 1, inset), indicating that hippurylphenyllactic acid transport is mediated by a common H+/cotransport system with ceftibuten. In contrast, bothl-Ala-l-Pro and N-CBz-l-Ala-l-Pro demonstrated noncompetitive or partially competitive effects on the uptake of ceftibuten (Fig.2) (e.g., the regression line obtained from the replot of the slopes of Dixon plot was not coincident with the origin; Fig. 2, inset). The apparentKi values calculated from Dixon plots for l-Ala-l-Pro and N-CBz-l-Ala-l-Pro were 1.21 and 17.1 mM, respectively, suggesting that masking the N-terminus weakens substrate affinity but not specificity of the H+-coupled transporter.
Dixon plot of ceftibuten uptake into intestinal BBM vesicles in the presence of hippurylphenyllactic acid. Uptake of 0.2 mM (○), 0.5 mM (●), and 0.75 mM (■) ceftibuten was measured for 20 s at pH 5.5 out/7.5 in the presence of 0, 2.0, and 5.0 mM hippurylphenyllactic acid, respectively. Inset, replot of the slopes of Dixon plot. The apparentKi value was determined to be 2.84 mM by linear regression analysis from the Dixon plot.
Dixon plot of ceftibuten uptake by the intestinal BBM vesicles in the presence of N-CBz-l-Ala-l-Pro (A) or l-Ala-l-Pro (B). Uptake of (▴) 0.1 mM, (▵) 0.2 mM, (○) 0.25 mM, (●) 0.5 mM, (▪) 0.75 mM, or (■) 1.0 mM ceftibuten was measured for 20 s in the presence of an inward H+ gradient with N-CBz-l-Ala-l-Pro or l-Ala-l-Pro. Inset, replot of the slopes of Dixon plot. Apparent Ki values of N-CBz-l-Ala-l-Pro inhibition (A; 17.1 mM) andl-Ala-l-Pro inhibition (B; 1.21 mM) were calculated by linear regression analysis.
Ceftibuten/H+ and Gly-Sar/H+Cotransporters.
To determine whether the ceftibuten/H+ cotransport system was identical to that of Gly-Sar/H+, the inhibitory effect of ceftibuten on the uptake of [14C]Gly-Sar was determined. The 20-s uptake values of [14C]Gly-Sar at a range of concentrations from 0.2 to 1.0 mM were plotted as Dixon analysis in the presence of increasing concentrations of ceftibuten (Fig.3). The kinetics of inhibition were consistent with a competitive type of inhibition (Ki = 1.20 mM). However, as shown in Fig. 4, the inhibitory effect of Gly-Sar on ceftibuten uptake was incomplete, and the overshooting uptake of ceftibuten did not disappear even at a Gly-Sar concentration of 20 mM. On the contrary,l-Ala-l-Ala (20 mM) inhibited ceftibuten uptake (Table 1) and completely suppressed the H+/ceftibuten cotransport under identical conditions. In addition, ceftibuten uptake was still saturable in the presence of 20 mM Gly-Sar, and Michaelis-Menten analysis demonstrated a noncompetitive inhibition (Fig. 5).
Dixon plot of Gly-Sar uptake into intestinal BBM vesicles in the presence of ceftibuten. Uptake of 0.2 mM (○), 0.5 mM (●), and 1.0 mM (▴) Gly-Sar was measured for 20 s at pH 6.0 out/7.5 in the presence of 0, 0.2, 0.5, 1.0, and 1.0 mM ceftibuten, respectively. Inset, replot of the slopes of Dixon plot. ApparentKi value was determined to be 1.20 mM by linear regression analysis from the Dixon plot.
Resistance of the proton-dependent overshooting uptake of ceftibuten by rat intestinal BBM vesicles to the presence of Gly-Sar. ░, control (ceftibuten only); ▨, with Gly-Sar (20 mM); and ▧, with l-Ala-l-Ala (20 mM). Uptake of 1.0 mM ceftibuten was determined at pH 5.5 out/7.5 in, respectively. *p < .05, **p < .01, significantly different compared with the uptake value.
Lineweaver-Burk plot of ceftibuten uptake with (●) or without (○) 20 mM Gly-Sar by the intestinal BBM vesicles. Uptake of ceftibuten was measured for 20 s in the presence of an inward H+ gradient (pH 5.5 out/7.5 in), respectively.
Effects of Inhibitors on H+ Gradient-Sensitive Component of Ceftibuten Uptake by Intestinal BBM Vesicles.
The effect of increasing concentrations of ceftibuten on its uptake in the presence of an inward H+ gradient together with compound V, a typical and competitive inhibitor for ceftibuten uptake (Sugawara et al., 1993), was investigated. The 20-s uptake rates were determined by measuring ceftibuten uptake with a range of concentrations from 100 μM to 20 mM.
Eadie-Hofstee plots of ceftibuten uptake with or without either 5 mM compound V or 20 mM Gly-Sar in the presence of an inward H+ gradient (Fig. 5) showed that the profile of ceftibuten uptake component in the presence of the competitive inhibitor compound V was similar to that seen in the absence of H+ gradient, indicating that both compounds are taken up by common transporter or transporters into the intestinal BBM vesicles.
In contrast, the effect of Gly-Sar on the uptake of ceftibuten was incomplete, and the component of ceftibuten uptake remaining after Gly-Sar inhibition was inconsistent with the uptake component without H+ gradient (Fig.6B). The inhibition by Gly-Sar appeared to be noncompetitive or a mixed-type inhibition. The uptake system for ceftibuten in the BBM appears to be partially identical with that of Gly-Sar, namely, the PepT1 transporter.
Eadie-Hofstee plot of ceftibuten initial (20 s) uptake by rat intestinal BBM vesicles with or without 5 mM compound V (A) and 20 mM Gly-Sar (B). Uptake of ceftibuten was measured in the uptake medium at various concentrations (0.2–10 mM) for 20 s in the presence (●) or absence (○) of H+ gradient (pH 5.5 out/7.5 in). Ceftibuten uptake with compound V ▪) and Gly-Sar (■) were measured in the presence of an inward H+ gradient, respectively. Some errors bar are inside the symbols. Dotted line represents the Eadie-Hofstee plot with its saturable component.
Discussion
Peptide transporters can serve as carriers for exogenous compounds with structural resemblance to their physiologically occurring peptide substrates. Ganapathy et al. (1997) demonstrated unequivocal evidence for the interaction of anionic β-lactam antibiotics such as ceftibuten and cefixime with the intestinal H+/peptide cotransporter (PepT1) using four different approaches. We also reported that the intestinal transport of ceftibuten is mediated by one transport system that is insensitive to the cationic dipeptide carnosine (Naasani et al., 1995).
It has been reported by Wenzel et al. (1996) that a single peptide transporter is capable of transporting neutral as well as anionic/dianionic β-lactam antibiotics because differently charged β-lactam antibiotics are taken up into human Caco-2 cells by a transporter that is very similar to PepT1. These authors proposed that only the zwitterionic species of dipeptide analogs are transported efficiently by intestinal transporters. However, this proposed mechanism cannot explain the differences between the transport rates of ceftibuten and cefixime because these anionic β-lactam antibiotics have very similar structures.
The inhibitory constant (Ki) value of N-CBz-l-Ala-l-Pro (17.1 mM) on ceftibuten uptake was larger than that ofl-Ala-l-Pro (1.21 mM), suggesting that the inhibitory potency of dipeptides is weakened by masking the amino group. Oh et al. (1993) investigated the structural requirements of the peptide transport system with regard to the importance of the α-amino group on the side chain of β-lactam antibiotics and reported that higher permeabilities across the intestinal wall are observed for α-amino β-lactam antibiotics.
On the other hand, Terada et al. (1997) determined the inhibition constant (Ki) values of various β-lactam antibiotics on Gly-Sar uptake into LLC-PK1 cells stably transfected with PepT1 cDNA and set their different affinities to the transporter in the order: ceftibuten (dianionic β-lactams without an α-amino group) > cephalexin = cephradine (amino β-lactams). However, theKi value of cefixime, which has a very similar structure to ceftibuten, was found to be approximately equal to the Ki values of amino β-lactams such as cephalexin, cephradine, and cefadroxil. Therefore, it is still controversial as to how amino and dianionic β-lactams (lacking a-amino group) are recognized by the PepT1 transporter in the intestine.
The present study demonstrated that carnosine exerts a significant inhibitory effect on the H+-driven Gly-Sar uptake by the intestinal BBM vesicles (Table 1). By observing the uptake and mutual inhibition of oral β-lactam antibiotics, Muranushi et al. (1994, 1995) suggested the presence of several transport systems for oligopeptides, and the uptake differences seem to be attributable to differences in the structure of the N-terminal amino acid.
Fei et al. (1994) reported that hybrid depletion of rabbit intestinal mRNA, before injection into X. laevis oocytes, with an antisense oligopeptide corresponding to the 5′-end region of PepT1 cDNA resulted in a complete suppression of the transport activity of the dipeptide Gly-Sar. In the present study, although ceftibuten inhibited completely and competitively H+-dependent Gly-Sar uptake by intestinal BBM vesicles, the opposite inhibitory effect of Gly-Sar on ceftibuten uptake could not be observed even at a concentration of 20 mM Gly-Sar. Moreover, a Eadie-Hofstee plot indicated that the inhibitory effect of Gly-Sar on the H+-driven uptake of ceftibuten was incomplete compared with that of compound V. Because theKm value of Gly-Sar for the rat PepT1 transporter is approximately 1.0 mM, a 20 mM concentration of Gly-Sar should be sufficient to completely suppress the initial uptake of ceftibuten if transport of the latter is mediated solely via the dipeptide transporter (PepT1). These results suggest that ceftibuten, a dianionic cefem, is recognized by not only the dipeptide/H+ cotransporter (PepT1) but also by another H+ coupled transporter.
Recently, two proteins have been isloated that have the ability to recognize ceftibuten in the intestinal BBM (Iseki et al., 1998). The protein with a molecular mass of 117 kDa was different from that of PepT1, which has an apparent molecular mass of 75 kDa (Saito et al., 1995). Reconsitution of this new protein component obtained from the solubilized BBM fraction into asolectin liposomes resulted in proteoliposomes that exhibited a strong uptake activity for ceftibuten.
The discrepancy of the present findings with our previous results that indicated kinetically the presence of one transport system in the intestinal BBM (Naasani et al., 1995) cannot be explained clearly. We reexamined the nonlinear regression analysis and least-squares fitting for the ceftibuten uptake value data. A double Michaelis-Menten model with nonsaturable component (simple diffusion) resulted in the best fit (Km1 = 0.19 mM,Vmax1 = 480.6 pmol/mg protein/20 s;Km2 = 2.37 mM,Vmax2 = 1077.0 pmol/mg protein/20 s). On the other hand, the best fit for ceftibuten uptake in the presence of 20 mM Gly-Sar was achieved with a single Michaelis-Menten model with nonsaturable component (Km = 0.37 mM,Vmax = 415.0 pmol/mg protein/20 s). Gly-Sar inhibited only the low-affinity component of ceftibuten transport system.
The inhibition actions of l-Ala-l-Pro and N-CBz-l-Ala-l-Pro were shown to be noncompetitive or partially competitive, although hippurylphenyllactic acid and compound V competitively inhibited ceftibuten uptake. We confirmed by Dixon analysis that l-Val-l-Pro also exerted a noncompetitive or partially competitive inhibition for ceftibuten uptake (data not shown). Presumably, Gly-Sar may inhibit ceftibuten uptake in a partially competitive manner, although a Dixon plot analysis was not performed. Instead, a Eadie-Hofstee plot of the uptake revealed that the high-affinity component of ceftibuten transport remained in the presence of Gly-Sar.
In conclusion, ceftibuten is absorbed in the intestinal BBM via at least two H+-driven transport systems: the PepT1 transporter and an another H+-driven transporter. Further studies using gene cloning of this new transporter are in progress.
Footnotes
-
Send reprint requests to: Dr. Katsumi Miyazaki, Department of Pharmacy, Hokkaido University Hospital, School of Medicine, Hokkaido University, Kita-14-jo, Nishi-5-chome, Kita-ku, Sapporo 060, Japan. E-mail ken-i{at}med.hokudai.ac.jp
-
↵1 Present address: Cancer Chemotherapy Center, Japanese Foundation for Cancer Research, Kami-Ikebukuro, 1–37-1, Toshima-ku, Tokyo 170 Japan.
- Abbreviation:
- BBM
- brush-border membrane
-
- Received May 26, 1998.
- Accepted October 27, 1998.
- The American Society for Pharmacology and Experimental Therapeutics









