Synthesis and characterization of a new and radiolabeled high-affinity substrate for H + /peptide cotransporters Ilka Knu ¨ tter 1 , Bianka Hartrodt 2 ,Ge ´ za To ´ th 3 , Attila Keresztes 3 , Gabor Kottra 4 , Carmen Mrestani-Klaus 2 , Ilona Born 2 , Hannelore Daniel 4 , Klaus Neubert 2 and Matthias Brandsch 1 1 Biozentrum of the Martin-Luther-University Halle-Wittenberg, Halle, Germany 2 Institute of Biochemistry ⁄ Biotechnology, Faculty of Sciences I, Martin-Luther-University Halle-Wittenberg, Halle, Germany 3 Institute of Biochemistry, Biological Research Center of the Hungarian Academy of Sciences, Szeged, Hungary 4 Molecular Nutrition Unit, Technical University of Munich, Freising-Weihenstephan, Germany The peptide transporters peptide cotransporter 1 (PEPT1) (SLC15A1) and peptide cotransporter 2 (PEPT2) (SLC15A2) are presently under intense inves- tigation because of their physiological importance and their pharmaceutical relevance as drug carriers [1–6]. Both transporters catalyse the uptake of most dipep- tides and tripeptides and a variety of peptidomimetic drugs, such as selected b-lactam antibiotics, some angiotensin-converting enzyme inhibitors and pro- drugs such as valaciclovir. H + -coupled peptide and drug transport across cell membranes by PEPT1 and PEPT2, respectively, have been demonstrated at Keywords Caco-2; H + ⁄ peptide cotransporter 1; H + ⁄ peptide cotransporter 2; SKPT; Xenopus laevis oocytes Correspondence M. Brandsch, Biozentrum of the Martin- Luther-University Halle-Wittenberg, Membrane Transport Group, Weinbergweg 22, D-06120 Halle, Germany Fax: +49 345 5527258 Tel: +49 345 5521630 E-mail: matthias.brandsch@ biozentrum.uni-halle.de (Received 15 August 2007, revised 19 Sep- tember 2007, accepted 20 September 2007) doi:10.1111/j.1742-4658.2007.06113.x In this study we described the design, rational synthesis and functional characterization of a novel radiolabeled hydrolysis-resistant high-affinity substrate for H + ⁄ peptide cotransporters. l-4,4¢-Biphenylalanyl–l-Proline (Bip-Pro) was synthesized according to standard procedures in peptide chemistry. The interaction of Bip-Pro with H + ⁄ peptide cotransporters was determined in intestinal Caco-2 cells constitutively expressing human H + ⁄ peptide cotransporter 1 (PEPT1) and in renal SKPT cells constitutively expressing rat H + ⁄ peptide cotransporter 2 (PEPT2). Bip-Pro inhibited the [ 14 C]Gly-Sar uptake via PEPT1 and PEPT2 with exceptional high affinity (K i ¼ 24 lm and 3.4 lm, respectively) in a competitive manner. By employ- ing the two-electrode voltage clamp technique in Xenopus laevis oocytes expressing PEPT1 or PEPT2 it was found that Bip-Pro was transported by both peptide transporters although to a much lower extent than the refer- ence substrate, Gly-Gln. Bip-Pro remained intact to > 98% for at least 8 h when incubated with intact cell monolayers. Bip-[ 3 H]Pro uptake into SKPT cells was linear for up to 30 min and pH dependent with a maximum at extracellular pH 6.0. Uptake was strongly inhibited, not only by unlabeled Bip-Pro but also by known peptide transporter substrates such as dipep- tides, cefadroxil, Ala-4-nitroanilide and d-aminolevulinic acid, but not by glycine. Bip-Pro uptake in SKPT cells was saturable with a Michaelis– Menten constant (K t ) of 7.6 lm and a maximal velocity (V max ) of 1.1 nmo- lÆ30 min )1 Æmg of protein )1 . Hence, the uptake of Bip-Pro by PEPT2 is a high-affinity, low-capacity process in comparison to the uptake of Gly-Sar. We conclude that Bip-[ 3 H]Pro is a valuable substrate for both mechanistic and structural studies of H + ⁄ peptide transporter proteins. Abbreviations Boc, tert. butyloxycarbonyl; Bip, L-4,4¢-biphenylalanine; Bip-Pro, L-4,4¢-biphenylalanyl–L-Proline; PEPT1, H + ⁄ peptide cotransporter 1; PEPT2, H + ⁄ peptide cotransporter 2; DPro, L-3,4-dehydro-Proline. FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS 5905 intestinal and renal cells [1–3] but also in lung [7], extrahepatic biliary duct [8], choroid plexus [9] and other tissues [1–3]. Essentially nothing is known about the location and structure of the substrate binding domains within the carrier proteins. Available data are restricted to results obtained in experiments with chimeric mammalian peptide transporters derived from the intestinal and renal isoforms [10,11], site-directed mutagenesis experi- ments [12–14] and from extensive studies on substrate specificity combined with molecular modeling [4,5,15,16]. The most commonly used and best known reference substrate of H + ⁄ peptide cotransporters is [ 14 C]glycine- sarcosine ([ 14 C]Gly-Sar). This substrate is relatively stable against intracellular and extracellular enzymatic hydrolysis, but its affinity constants for peptide trans- porters are only in the medium range, with K t values of  1.3 mm for PEPT1 [17] and  108 lm for PEPT2 [18]. New high-affinity labeled probes are required for further studies on the mechanism of transport func- tion and the structure of the carrier proteins. For example, the rate limiting step of peptide transporters has still not yet been determined, and the number of transporters per cell and their turnover rate in epithe- lial cells is not known. With regard to transporter structure, despite the fact that techniques such as intrinsic tryptophan fluorescence measurement have been shown to be useful to study the expression and conformation of recombinant membrane transporters [19], labeled substrates and inhibitors with a broad range of affinity to the respective protein are also essential tools. In the course of our work on high-affinity inhibitors of PEPT1 and PEPT2, on the structural modifications that convert a transported compound into a nontranslocated inhibitor as well as studies on the structural requirements for a high affin- ity of substrates [17,18,20], it became evident that a large aromatic hydrophobic group in the side chain of the N-terminal amino acid of dipeptides could enhance the affinity of many derivatives for binding to the transporters. Besides high affinity, a second important requirement for any peptide transporter substrate is a sufficient stability against enzymatic hydrolysis. Hence, we decided to synthesize a dipep- tide where l-4,4¢-biphenylalanine (Bip), with its large aromatic side chain in a short intramolecular distance from the a-carbon atom, is the N-terminal amino acid and l-Proline (l-Pro) is the C-terminal amino acid. The resulting compound, Bip-Pro, was tested with regard to its interaction with peptide transporters, its affinity and its stability in a biological system. More- over, after radioactive labeling we determined the kinetic parameters and transport characteristics of Bip-[ 3 H]Pro. Results and Discussion Synthesis, chemical characterization and stability of Bip-Pro Figure 1 shows the structure (Fig. 1A) and the synthe- sis strategy (Fig. 1B) of Bip-Pro. The purity of the compound was assessed by TLC, analytical RP- HPLC, MS and NMR, and was found to exceed 98%. As expected for an Xaa-Pro peptide derivative, Bip-Pro exists as a mixture of cis and trans conform- ers in aqueous solution (pH 6.0). In the 1 H NMR spectrum, Bip-Pro exhibited two sets of NMR signals indicating the existence of two conformations. Fig. 1. Structure and synthesis of Bip-Pro. (A) Bip-Pro structure. (B) Bip-Pro synthesis: mixed anhydride (MA) method. (C) Synthesis of Bip-[ 3 H]Pro. EDC, N-ethyl-N¢-(3-dimethyl aminopropyl)-carbodiimide; HOSu, N-hydroxysuccinimide. Labeled high-affinity substrate for peptide transporters I. Knu ¨ tter et al. 5906 FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS Subsequent analysis of the ROESY spectrum revealed characteristic strong ROEs between Bip-C a H and both Pro-C d H A and Pro-C d H B of the major isomer identi- fied as a trans isomer. As strong ROEs (or NOEs) between a protons of adjacent residues C a H(i)-C a H(i +1) allow the resonance assignment of populations containing cis amide linkages [21], the strong ROE between Bip-C a H and Pro-C a H of the minor isomer was used as evidence for its cis conformation. The rel- ative populations of the cis ⁄ trans isomers were deter- mined by integration of well-resolved signals in the 1D proton spectrum such as the two Bip-C a H signals [21,22]. In equilibrium, Bip-Pro shows a trans content of 22%, whereas 78% were in cis conformation. These values are in agreement with the cis ⁄ trans ratios of Xaa-Pro dipeptides containing aromatic amino acids (24–34% trans content) obtained in a previous study of our group [23]. To determine the stability of Bip-Pro, buffer samples were analyzed after incubating Caco-2 and SKPT cell monolayers (surface area 9.62 cm 2 ) with the compound (1 mL, 1 mm) for 10 min up to 8 h. After a 2 h incu- bation of SKPT monolayers with Bip-Pro containing buffer, 100% Bip-Pro was found. After 8 h, 99.9% of Bip-Pro was intact, with the remaining 0.1% being Bip. At all time points, Bip-Pro was recovered from monolayers of both cell types intact to > 99% (HPLC data not shown). Interaction of Bip-Pro with PEPT1 and PEPT2 We next determined the interaction of Bip-Pro with PEPT1 and PEPT2 in competition assays, with [ 14 C]Gly-Sar serving as a reference compound. The intestinal cell line Caco-2, constitutively expressing PEPT1 [5,17,22], and the renal cell line SKPT, con- stitutively expressing PEPT2 [18,24], were used as models. Bip-Pro competed with [ 14 C]Gly-Sar uptake in a dose-dependent manner (Fig. 2). The apparent K i values for substrate uptake inhibition were 24 ± 0.6 lm in Caco-2 cells (PEPT1, Fig. 2A) and 3.4 ± 0.1 lm in SKPT cells (PEPT2, Fig. 2B). It has been shown that peptide transporters are specific for the trans conformers of their substrates [22,23]. Tak- ing the cis⁄ trans content of Bip-Pro (78% ⁄ 22%) into account, we obtained K i trans values for Bip-Pro of 5.2 lm in Caco-2 cells and of 0.75 lm in SKPT cells. We also determined the inhibition constants (K i ) for Bip-Pro by measuring Gly-Sar uptake at two differ- ent Gly-Sar concentrations (50 and 500 lm in Caco-2 cells and 10 and 100 lm in SKPT cells) in the presence of increasing concentrations of Bip-Pro (0– 100 lm and 0–50 lm, respectively). The results are presented as Dixon plots in Fig. 2 (insets). The plots reveal linearity at both Gly-Sar concentrations with lines intersecting above the abscissas in the fourth quadrant, as expected for a competitive inhibitor. Apparent K i values of 34.1 lm (K i trans ¼ 7.5 lm) and 1.3 lm (K i trans ¼ 0.29 lm) were calculated from the points of intersection of data obtained in Caco-2 cells (Fig. 2A, inset) and SKPT cells (Fig. 2B, inset), respectively. Fig. 2. Interaction of Bip-Pro with PEPT1 and PEPT2. Uptake of [ 14 C]Gly-Sar was measured in Caco-2 cells (A) (10 lM [ 14 C]Gly-Sar, pH 6.0, 10 min, n ¼ 4) and in SKPT cells (B) (10 l M [ 14 C]Gly-Sar, pH 6.0, 10 min, n ¼ 4) in the presence of increasing concentrations of Bip-Pro (0–0.316 m M). Uptake rates measured in the absence of Bip-Pro were taken as 100%. Insets: determination of the inhibition constants by Dixon type experiments. Uptake of Gly-Sar was mea- sured at pH 6.0 for 10 min at two Gly-Sar concentrations and at increasing Bip-Pro concentrations. The diffusional component of [ 14 C]Gly-Sar uptake, of 8% in Caco-2 cells and of 4% in SKPT cells, measured in the presence of excess of Gly-Sar (30 m M and 20 mM, respectively), was subtracted from the total uptake to calculate the carrier-mediated uptake (n ¼ 4, v ¼ uptake rate in nmolÆ 10 min )1 Æmg protein )1 ). I. Knu ¨ tter et al. Labeled high-affinity substrate for peptide transporters FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS 5907 Interaction of Bip-Pro with PEPT1 and PEPT2 expressed in Xenopus laevis oocytes Interaction with Gly-Sar uptake does not necessarily allow the conclusion that Bip-Pro is indeed trans- ported. Therefore, the two-electrode voltage clamp technique that determines transport currents was applied in X. laevis oocytes expressing either PEPT1 or PEPT2 [17,18,20,25]. In contrast to the reference dipeptide glycine-glutamine (Gly-Gln), for Bip-Pro only a low substrate-evoked inward transport current was recorded (Fig. 3). At a membrane potential of )60 mV, PEPT1-mediated transport currents were 21 ± 6% of that generated by saturating Gly-Gln con- centrations (Fig. 3A). In the case of PEPT2 at a mem- brane potential of )160 mV, the maximal current was 11 ± 1% of that generated by Gly-Gln (Fig. 3C). However, Bip-Pro at a concentration of 0.5 mm was able to inhibit the inward current evoked by 0.5 mm Gly-Gln at PEPT1 by 44 ± 1% (Fig. 3B). In the case of PEPT2, a concentration of 0.1 mm Bip-Pro was able to inhibit the inward current evoked by 0.1 mm Gly- Gln remarkably by 94 ± 6% (Fig. 3D). The inhibition was found to be dose dependent and reversible, sug- gesting a competitive mode of action. Uptake of Bip-[ 3 H]Pro by SKPT cells After characterization of Bip-Pro as a very high-affin- ity and enzymatically stable substrate of PEPT1 and PEPT2, the compound was synthesized in radiolabeled form according to Fig. 1C [26]. We then characterized the Bip-[ 3 H]Pro uptake across the apical membrane of SKPT cells. Time-dependent uptake of Bip-[ 3 H]Pro Fig. 3. Characterization of the interaction of Bip-Pro with PEPT1 and PEPT2 in Xenopus laevis oocytes by electrophysiology. Steady-state I–V relationships were measured by the two-electrode voltage clamp technique in oocytes expressing PEPT1 (A, B) or PEPT2 (C, D) superfused with modified Barth solution at pH 6.5 and 0.5, or with 0.1 m M Gly–Gln, in the absence or the presence of increasing concentrations (PEPT1, 0–1 m M; PEPT2, 0–0.1 mM) of Bip-Pro. The membrane potential was stepped symmetrically to the test potentials shown, and substrate- dependent currents were recorded as the difference measured in the absence and in the presence of substrates. Labeled high-affinity substrate for peptide transporters I. Knu ¨ tter et al. 5908 FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS (18 nm) at pH 6.0 was linear for up to 1 h and reached a plateau after 2 h of incubation (Fig. 4). The uptake was found to be saturable: unlabeled Bip-Pro at a con- centration of 1 mm strongly inhibited Bip-[ 3 H]Pro uptake at all time points. Bip-[ 3 H]Pro (4 nm) uptake was also strongly pH dependent. Maximal uptake was observed at an extracellular pH of 6.0 (Fig. 4, inset). The same pH optimum has been observed for the uptake of [ 14 C]Gly-Sar, both in SKPT cells [24] and in Caco-2 cells [27]. We also studied the time and pH dependency of Bip-[ 3 H]Pro uptake in Caco-2 cells. Sur- prisingly, in this cell line the uptake was found to be stimulated by external pH 6.0 only modestly, by 26% in comparison to pH 7.5. Moreover, unlabeled Bip-Pro in an excess concentration of 3 mm inhibited the uptake of the tracer Bip-[ 3 H]Pro (4 nm) during 30 min of incubation by only 21% (data not shown). We con- clude that the nonspecific binding of the hydrophobic Bip-[ 3 H]Pro to Caco-2 cells is very much higher than in SKPT cells. Alternatively, a specific intestinal apical efflux system might mediate strong outward-directed Bip-[ 3 H]Pro transport after its uptake into the cells. For further functional characterization of Bip-[ 3 H]Pro uptake we therefore used SKPT cells. Saturation kinetics of Bip-Pro uptake at PEPT2 Bip-Pro uptake as a function of substrate concentra- tion was measured to determine the kinetic parameters of the transport process. Uptake rates of Bip-Pro were determined over a substrate concentration range of 4nm to 100 lm (Fig. 5A) and compared with the uptake rates of Gly-Sar at a concentration range of 5 lm to 5 mm (Fig. 5B). For each compound, the non- specific, linear uptake component, which represents simple diffusion plus binding, was determined by measuring the uptake in the presence of excess Fig. 5. Substrate saturation kinetics of Bip-Pro and Gly-Sar transport in SKPT cells. (A) Uptake of Bip-[ 3 H]Pro (4 nM, 30 min, pH 6.0) was measured over a Bip-Pro concentration range of 0–0.1 m M. Unspe- cific uptake ⁄ binding was determined by measuring uptake in the presence of an excess amount (1 m M) of unlabeled Bip-Pro. This component (24%) was subtracted from the total uptake to calculate the specific uptake. Inset: Eadie–Hofstee transformation of the spe- cific Bip-Pro uptake data [S, Bip-Pro concentration (l M); v, uptake (nmolÆ30 min )1 Æmg of protein )1 )]. (B) Uptake of [ 14 C]Gly-Sar (5– 20 l M, 10 min, pH 6.0) was measured over a Gly-Sar concentration range of 0–5 m M. Nonspecific uptake ⁄ binding was determined by measuring uptake in the presence of an excess amount (20 m M) of unlabeled Gly-Sar. This component (4%) was subtracted from the total uptake to calculate the carrier-mediated uptake. Inset: Eadie–Hofstee transformation of the specific Gly-Sar uptake data [S, Gly-Sar concentration (m M); v, uptake rate (nmolÆ10 min )1 Æmg protein )1 )]. Values represent the means ± standard error (SE) for four determinations. Fig. 4. Time and pH dependence of the uptake of Bip-[ 3 H]Pro in SKPT cells. Uptake of Bip-[ 3 H]Pro (18 nM, n ¼ 4) in SKPT cells was measured at pH 6.0 for 10 min to 4 h in the absence (d) or in the presence (s) of unlabeled Bip-Pro (1 m M). Inset: uptake of Bip- [ 3 H]Pro (4 nM, 2 h) measured at different pH values (n ¼ 4). I. Knu ¨ tter et al. Labeled high-affinity substrate for peptide transporters FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS 5909 amounts of substrate (1 or 20 mm, respectively) and subtracted from the total uptake rates. For both substrates, the relationship between carrier-mediated uptake and substrate concentration was found to fol- low Michaelis–Menten kinetics (Fig. 5). Eadie–Hofstee transformation (uptake rate versus uptake rate ⁄ sub- strate concentration) revealed linearity with a single component (Fig. 5 insets). The apparent K t for Gly-Sar uptake was 91.3 ± 4.1 lm and the V max was 5.6 ± 0.1 nmolÆ10 min )1 Æmg of protein )1 . These parameters agree very well with those of previous reports [18]. For Bip-Pro uptake, an apparent K t of 7.6 ± 1.8 lm and a V max of 1.1 ± 0.1 nmolÆ30 min )1 Æmg of protein )1 was determined. Hence, the maximal velocity of Bip-Pro uptake is 16-fold lower than the maximal velocity of Gly-Sar uptake, whereas the affinity constant of Bip-Pro uptake is 12-fold lower. Bip-Pro uptake represents a high-affinity, low- capacity process, whereas the Gly-Sar uptake occurs with low affinity but high transport capacity. The lower V max of Bip-Pro uptake, and the higher V max of Gly-Sar uptake, correspond well with the currents obtained at PEPT2-expressing X. laevis oocytes. The mean value of the apparent Michaelis–Menten con- stant calculated from the currents measured at )160 mV with Bip-Pro concentrations between 20 and 500 lm was 26 lm and the maximal current amounted to 8% of the current evoked by Gly-Sar at saturating concentration. In comparison, the inward current elic- ited by Gly-Sar is 90% of that generated by Gly-Gln and the affinity of PEPT2 was slightly lower for Gly- Sar (K t ¼ 0.3 mm) than for Gly-Gln (K t ¼ 0.1 m m). The situation is very similar for PEPT1, where Bip-Pro elicited 21% and Gly-Sar elicited 101% of the Gly-Gln current. Thus, the transport of Bip-Pro was also, in PEPT2-expressing oocytes, a high-affinity, low-capacity process. These findings suggest that the conformational change of the carrier protein following H + binding and substrate binding represents the rate limiting step in the substrate translocation cycle. Differences in the maximal transport currents of peptide transporters under saturating substrate concentrations have been reported before, suggesting that not only apparent K t values but also turnover rates may differ between sub- strates [28]. Substrate specificity of Bip-[ 3 H]Pro uptake In the next series of experiments, the specificity of Bip- [ 3 H]Pro uptake was investigated using fixed concentra- tions of competitors. The uptake of Bip-[ 3 H]Pro (4 nm, pH 6.0) into SKPT cells was inhibited not only by unlabeled Bip-Pro itself, but also by well known sub- strates of H + ⁄ peptide cotransporters, such as Gly-Sar, Ala-Ala, Lys-Lys, Ala-Asp, d-Phe-Ala, Ala-Ala-Ala, d-aminolevulinic acid, cefadroxil and Ala-4-nitroanilide (all 100 lm, Table 1). Glycine, which is not a substrate of peptide transporters, did not inhibit Bip-[ 3 H]Pro uptake. The PEPT1 and PEPT2 inhibitor Lys(4-nitrob- enzyloxycarbonyl)-Val [18,20], which is not transported itself but interacts with both transporters with very high affinity, displayed the strongest inhibitory effect of all compounds tested in this study (Table 1). Pro- Ala, at a concentration of 100 lm, did not inhibit Bip- [ 3 H]Pro uptake because it is a low-affinity substrate of PEPT2 with an apparent K i value of 2.6 mm [4]. In contrast, cefadroxil strongly inhibited Bip-[ 3 H]Pro uptake by 85%, corresponding very well with its apparent K i value for PEPT2 of 3 lm [4]. Finally, 8- aminooctanoic acid, which is no substrate for PEPT2 [2–4], also did not inhibit uptake. We then determined the apparent K i values of five compounds that tested positive for inhibition of Bip- [ 3 H]Pro uptake. The apparent K i values (Table 2) were calculated by nonlinear regression from data obtained in competition experiments such as those shown in Fig. 2. For Bip-Pro, the self-inhibition K i was 7.8 ± 0.1 lm (K i trans ¼ 1.7). Cefadroxil (K i ¼ 5.2 ± 0.4 lm) displayed the highest affinity for inhibition followed by Gly-Sar, Lys-Lys and d-aminolevulinic acid with apparent K i values between 75 and 230 lm. For comparison, in Table 2 we also present the respective inhibition constants (K i ) of these five substrates for the inhibition of [ 14 C]Gly-Sar uptake in SKPT cells. This Table 1. Specificity of Bip-[ 3 H]Pro uptake. Uptake of Bip-[ 3 H]Pro (4 n M) into SKPT cells was measured at pH 6.0 for 2 h at room temperature in the absence (control) or presence of inhibitors (all 100 l M). Data are shown as means ± standard error, n ¼ 4. Lys[Z(NO 2 )]-Val, Lys(4-nitrobenzyloxycarbonyl)-Val. Compound Bip-[ 3 H]Pro uptake (%) Control 100 ± 2 Gly 118 ± 8 Gly-Sar 64 ± 5 Bip-Pro 14 ± 1 Ala-Ala 15 ± 1 Pro-Ala 105 ± 3 Lys-Lys 46 ± 2 Ala-Asp 19 ± 1 D-Phe-Ala 65 ± 2 Ala-Ala-Ala 21 ± 1 d-Aminolevulinic acid 78 ± 3 Cefadroxil 15 ± 1 Lys[Z(NO 2 )]-Val 10 ± 1 8-Aminooctanoic acid 111 ± 3 Ala-4-nitroanilide 38 ± 1 Labeled high-affinity substrate for peptide transporters I. Knu ¨ tter et al. 5910 FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS so-called ABC test shows that the affinity constants are very similar. Bip-Pro (A) and Gly-Sar (B) were inhibited to the same extent by the other compounds (C). Hence, Bip-Pro and Gly-Sar are transported by the same system. In conclusion, the results of the present study on the mechanism and specificity of Bip-Pro uptake in SKPT cells, together with the electrophysiological data obtained in X. laevis oocytes expressing PEPT2, pro- vide unequivocal evidence that Bip-Pro is transported by PEPT2. Its enzymatic stability allows it to be used in complex biological systems and its very high affinity should make it particularly useful as a probe for the analysis of the structure of the PEPT2 protein. More- over, via detailed kinetic analyses with the now avail- able two labeled transporter substrates, Bip-Pro and Gly-Sar, which differ markedly in maximal transport rates, the identification of the rate limiting step in the transport cycle of PEPT1 and PEPT2 became feasible. Experimental procedures Materials The renal cell line SKPT-0193 CL.2, established from iso- lated cells of rat proximal tubules [24], was provided by U. Hopfer (Case Western Reserve University, Cleveland, OH, USA). The human colon carcinoma cell line Caco-2 was obtained from the German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany). [Gly- 1- 14 C]Gly-Sar (specific radioactivity 53 mCiÆ mmol )1 ) was custom synthesized by Amersham International (Little Chal- font, UK). Dexamethasone, apotransferrin, Gly-Gln, Ala- Ala, Ala-Ala-Ala, Lys-Lys, d-aminolevulinic acid, cefadroxil, Gly, Pro-Ala, 8-aminooctanoic acid and Gly-Sar were from Sigma-Aldrich (Deisenhofen, Germany). Tert. butyloxycar- bonyl (Boc)–Bip, l-3,4-dehydro-Proline (DPro), d-Phe-Ala and Ala-Asp were purchased from Bachem (Heidelberg, Germany). Culture media, media supplements and trypsin solution were purchased from Invitrogen (Karlsruhe, Germany) or PAA (Pasching, Austria). Fetal bovine serum was from Biochrom (Berlin, Germany) and collagenase A from Roche (Mannheim, Germany). Ala-4-nitroanilide and Lys(4-nitrobenzyloxycarbonyl)-Val were synthesized accord- ing to peptide synthesis standard procedures [18,29]. All other chemicals were of analytical grade. Synthesis of Bip-Pro and Bip-[ 3 H]Pro Boc-Bip-Pro-OtBu was prepared from Boc–Bip-OH and H- Pro–OtBuÆHCl using the mixed anhydride coupling method with isobutylchloroformiate. After purification of the crude product by flash chromatography (ethyl acetate ⁄ petroleum ether: 1 : 2, v ⁄ v) the oily, protected dipeptide was depro- tected with trifluoroacetic acid for 3 h to obtain the dipep- tide as trifluoroacetate. Purity was measured with TLC, RP-HPLC and MS and was found to exceed 98%. H-Bip- Pro–OHÆtrifluoroacetic acid was a cis ⁄ trans isomere mixture according to the HPLC chromatograms. At room tempera- ture two peaks were observed, whereas there was only one peak at temperatures of ‡ 45 °C. The precursor peptide H-Bip–l-3,4-dehydro-Pro-OH (H-Bip–DPro-OH) for 3 H-labeling was synthesized as follows. Boc–Bip-OH was converted to Boc–Bip– N-hy- droxysuccinimide ester using the water-soluble N-ethyl- N¢-(3-dimethyl aminopropyl)-carbodiimide as a coupling reagent. The resulting active ester derivative then reacted with DPro and triethylamine in acetonitrile to give Boc–Bip– DPro-OH. After purification of the crude product by flash chromotography with ethyl acetic acid (5 : 0.1, v ⁄ v) the Boc- Protected dipeptide was recrystallized from ethyl acetate. Deprotection was carried out with 4 m HCl ⁄ dioxane to give H-Bip–DPro-OH as its hydrochloride. Precipitation from isopropanol ⁄ ethyl ether gave the H-Bip–DPro–OHÆHCl in high purity (‡ 98%, checked by TLC, RP-HPLC and MS). The tritium labeling was carried out by catalytic satura- tion of 2 mg of the precursor peptide in N,N-dimethylfor- mamide (room temperature, 30 min) using Pd ⁄ C as the catalyst and carrier-free tritium gas [26]. After tritiation, the crude peptide product was purified by HPLC (Jasco, Budapest, Hungary) on a Vydac (Budapest, Hungary) 218 TP 54 column (250 · 4.6 mm) using linear gradient elution (from 15 to 40%) of acetonitrile (0.08% trifluoroacetic acid) in water (0.1% trifluoroacetic acid) within 25 min at a flow rate of 1 mLÆmin )1 with UV detection at 265 nm. H-Bip-[ 3 H]Pro-OH existed as a mixture of cis ⁄ trans con- formers, according to the chromatograms. Radioactive pur- ity of the final product was > 98% according to TLC [silicagel 60 F254 plate, Merck, Darmstadt, Germany; sol- vent system n-butanol-acetic acid-water (4 : 1 : 1, v ⁄ v ⁄ v) – retention factor 0.41] and analytical HPLC (retention time 17.27 min, k¢ ¼ 4.57). Specific radioactivity of Bip-[ 3 H]Pro, Table 2. Inhibition constants (K i ) of different substrates for the inhibition of Bip-[ 3 H]Pro and [ 14 C]Gly-Sar uptake in SKPT cells. Uptake of Bip-[ 3 H]Pro (4 nM, 2 h) or of [ 14 C]Gly-Sar (10 lM, 10 min) was measured at pH 6.0 at increasing concentrations of unlabeled substrates or inhibitors of PEPT2. Constants were derived from competition curves such as those shown in Fig. 2 for Bip-Pro. Para- meters are shown ± standard error (n ¼ 4). Compound K i (lM) Bip-[ 3 H]Pro uptake [ 14 C]Gly-Sar uptake Gly-Sar 102 ± 9 61 ± 8 [24] Bip-Pro 7.8 ± 0.1 3.4 ± 0.1 Cefadroxil 5.2 ± 0.4 3 ± 1 [4] d-Aminolevulinic acid 230 ± 20 231 ± 90 [4] Lys-Lys 75 ± 9 12 ± 0.3 [4] I. Knu ¨ tter et al. Labeled high-affinity substrate for peptide transporters FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS 5911 estimated by a calibration curve prepared with a standard dipeptide, was 1.853 TBqÆmmol )1 (50.1 Ci mmol )1 ). NMR analysis The relative populations of the cis ⁄ trans isomers were determined by NMR measurements [21,22]. 1 H NMR spec- tra of 5.3 mg of Bip-Pro dissolved in 0.7 mL of H 2 O ⁄ D 2 O (90 : 10, v ⁄ v) were recorded on a Bruker Avance 400 spec- trometer (Rheinstetten, Germany). All measurements were carried out at pH 6.0 and 300 K. The pH of the solution was adjusted by the addition of diluted solutions of DCl and NaOD. Chemical shifts were calibrated with respect to internal DSS. Selective water resonance suppression was achieved by using presaturation during relaxation delay or by using the 3-9-19 pulse sequence with gradients. Stan- dard methods were used to perform 1D and 2D experi- ments, pulse programs being taken from the Bruker software library. Resonance assignments were made by the combined analysis of H,H-COSY, ROESY and 13 C-HSQC spectra. The ROESY spectra were recorded at a mixing time of 300 ms in the phase-sensitive mode using baseline correction in both dimensions. HPLC analysis The stability of Bip-Pro in the extracellular medium was analyzed over incubation periods from 10 min up to 8 h. The amount of Bip-Pro in the extracellular uptake medium was quantified according to the laboratory standard HPLC (La-Chrom Ò ; Merck-Hitachi, Darmstadt, Germany) with a diode array detector and a Polar-RP-80-A Synergi column (150 · 4.6 mm; 4 lm; Phenomenex, Aschaffenburg, Ger- many). The eluent was 30% acetonitrile ⁄ 0.1% trifluoroace- tic acid in water. UV detection was performed at 220 nm. The injection volume was 20 lL and the flow rate was 1mLÆmin )1 . Cell culture and uptake studies SKPT cells were cultured in Dulbecco’s modified Eagle’s medium ⁄ F12 Nutrient Mixture (Ham) (1 : 1, v ⁄ v) and 2mml-glutamine, 10% fetal bovine serum, recombinant insulin (4 lgÆmL )1 ), epidermal growth factor (10 ng ÆmL )1 ), apotransferrin (5 lgÆmL )1 ), dexamethasone (5 lgÆ mL )1 ) and gentamicin (45 lgÆmL )1 ), as described previously [18,24]. The human colon carcinoma cell line Caco-2 was routinely cultured with Minimum Essential Medium with Earle’s salts and l-glutamine (2 mm) supplemented with 10% fetal bovine serum, 1% nonessential amino acid solution and gentamicin (45 lgÆmL )1 ) [17,20]. Both cell lines were subcultured in 35-mm disposable Petri dishes (Sarstedt, Nu ¨ mbrecht, Germany) at a seeding density of 0.8 · 10 6 cells per dish. The cultures of both cell types reached confluence within 20 h. Uptake of [ 14 C]Gly-Sar or Bip-[ 3 H]Pro was measured 4 days (SKPT) or 7 days (Caco-2) after seeding at 22 °C, as described previously [17,18,20]. The uptake buffer was 25 mm Mes ⁄ Tris (pH 6.0) or 25 mm Hepes ⁄ Tris (pH 7.5) containing 140 mm NaCl, 5.4 mm KCl, 1.8 mm CaCl 2 , 0.8 mm MgSO 4 and 5 mm glucose. Uptake was initiated after washing the cells for 30 s in uptake buffer by adding 1 mL of uptake medium containing [ 14 C]Gly-Sar (10 lm)or Bip-[ 3 H]Pro (4 nm) with increasing concentrations of the test compounds (0–31.6 mm). If necessary, the pH of the solutions was corrected before preparing the required dilu- tions. After incubation for the desired time periods, the cells were quickly washed four times with ice-cold buffer, solubilized in 1 mL of Igepal Ò Ca-630 (0.5% v ⁄ v; Sigma Aldrich, Deisenhofen, Germany) in buffer (50 mm Tris ⁄ HCl, pH 9.0, 140 mm NaCl, 1.5 mm MgSO 4 ) and pre- pared for liquid scintillation spectrometry. For each experi- ment, the samples for the protein measurements were prepared and measured as described previously [20]. X. laevis oocytes expressing PEPT1 and PEPT2 and electrophysiology Female X. laevis were purchased from the African Xeno- pus Facility (Kynsa, South Africa). Surgically removed oo- cytes were separated by collagenase treatment and handled as described previously [17,18,20,25]. Individual oocytes were injected with 30 nL of RNA solution containing 30 ng of rabbit PEPT1 or rabbit PEPT2 cRNA. All elec- trophysiological measurements were performed after 3– 6 days by incubation of oocytes in a buffer composed of 88 mm NaCl, 1 mm KCl, 0.82 mm CaCl 2 , 0.41 mm MgCl 2 , 0.33 mm Ca(NO 3 ) 2 , 2.4 mm NaHCO 3 and 10 mm Mes ⁄ Tris at pH 6.5 (modified Barth solution). The two-electrode voltage clamp technique was applied to characterize responses in current (I) and transmembrane potential (V m ) to substrate addition in oocytes expressing PEPT1 or PEPT2 [17,18,20,25]. In short, oocytes were placed in an open chamber in a volume of 0.5 mL and continuously superfused with modified Barth solution or with solutions of Gly-Gln, Gly-Sar and ⁄ or Bip-Pro. Elec- trodes with resistances between 1 and 10 MW were connected to a TEC-05 amplifier (NPI Electronic, Tamm, Germany). Current–voltage (I–V m ) relationships were mea- sured using short (100 ms) pulses separated by 200 ms pauses in the potential range from )160 to +80 mV. I–V m measurements were made immediately before and 30 s after substrate application when current flow reached steady state. Currents evoked by PEPT1 or PEPT2 at a given membrane potential were calculated as the difference of the currents measured in the presence and the absence of substrate. Labeled high-affinity substrate for peptide transporters I. Knu ¨ tter et al. 5912 FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS Calculations and statistics All data are given as the mean ± standard error of three to four independent experiments. The kinetic parameters were calculated by nonlinear regression methods (sigma- plot program; Systat, Erkrath, Germany) and confirmed by linear regression of the respective Eadie–Hofstee Plots. The concentration of the unlabeled compound necessary to inhibit 50% of radiolabeled dipeptide carrier-mediated uptake (IC 50 ) was determined by nonlinear regression using the logistical equation for an asymmetric sigmoid (allosteric Hill kinetics): y ¼ Min + (Max–Min) ⁄ (1 + (X ⁄ IC 50 ) –P ), where Max is the initial Y-value, Min the final Y-value and the power P represents Hills’ coefficient. Inhibition con- stants (K i ) were calculated from IC 50 values. Acknowledgements This work was supported by the State Saxony-Anhalt Life Sciences Excellence Cluster (MB). References 1 Ganapathy V, Ganapathy ME & Leibach FH (2001) Intestinal transport of peptides and amino acids. In Current Topics in Membranes (Barrett KE & Donowitz M, eds), Vol. 50, pp. 379–412. Academic Press, San Diego, CA. 2 Daniel H & Kottra G (2004) The proton oligopeptide cotransporter family SLC15 in physiology and pharma- cology. Pflugers Arch 447, 610–618. 3 Daniel H (2004) Molecular and integrative physiology of intestinal peptide transport. Annu Rev Physiol 66, 361–384. 4 Biegel A, Knu ¨ tter I, Hartrodt B, Gebauer S, Theis S, Luckner P, Kottra G, Rastetter M, Zebisch K, Thon- dorf I et al. (2006) The renal type H + ⁄ peptide symport- er PEPT2: structure-affinity relationships. Amino Acids 31, 137–156. 5 Brandsch M, Knu ¨ tter I & Leibach FH (2004) The intes- tinal H + ⁄ peptide symporter PEPT1: structure-affinity relationships. Eur J Pharm Sci 21, 53–60. 6 Steffansen B, Nielsen CU, Brodin B, Eriksson AH, Andersen R & Frokjaer S (2004) Intestinal solute carri- ers: an overview of trends and strategies for improving oral drug absorption. Eur J Pharm Sci 21, 3–16. 7 Groneberg DA, Fischer A, Chung KF & Daniel H (2004) Molecular mechanisms of pulmonary peptidomi- metic drug and peptide transport. Am J Respir Cell Mol Biol 30, 251–260. 8 Knu ¨ tter I, Rubio-Aliaga I, Boll M, Hause G, Daniel H, Neubert K & Brandsch M (2002) H + -peptide cotrans- port in the human bile duct epithelium cell line SK- ChA-1. Am J Physiol 283, G222–G229. 9 Teuscher NS, Novotny A, Keep RF & Smith DE (2000) Functional evidence for presence of PEPT2 in rat cho- roid plexus: studies with glycylsarcosine. J Pharmacol Exp Ther 294, 494–499. 10 Do ¨ ring F, Dorn D, Bachfischer U, Amasheh S, Herget M & Daniel H (1996) Functional analysis of a chimeric mammalian peptide transporter derived from the intesti- nal and renal isoforms. J Physiol 497, 773–779. 11 Fei YJ, Liu JC, Fujita T, Liang R, Ganapathy V & Lei- bach FH (1998) Identification of a potential substrate binding domain in the mammalian peptide transporters PEPT1 and PEPT2 using PEPT1-PEPT2 and PEPT2- PEPT1 chimeras. Biochem Biophys Res Commun 246, 39–44. 12 Fei YJ, Liu W, Prasad PD, Kekuda R, Oblak TG, Ganapathy V & Leibach FH (1997) Identification of the histidyl residue obligatory for the catalytic activity of the human H + ⁄ peptide cotransporters PEPT1 and PEPT2. Biochemistry 36, 452–460. 13 Bolger MB, Haworth IS, Yeung AK, Ann D, von Gra- fenstein H, Hamm-Alvarez S, Okamoto CT, Kim KJ, Basu SK, Wu S et al. (1998) Structure, function, and molecular modeling approaches to the study of the intestinal dipeptide transporter PepT1. J Pharm Sci 87, 1286–1291. 14 Terada T, Irie M, Okuda M & Inui K (2004) Genetic variant Arg57His in human H + ⁄ peptide cotransporter 2 causes a complete loss of transport function. Biochem Biophys Res Commun 316, 416–420. 15 Biegel A, Gebauer S, Hartrodt B, Brandsch M, Neubert K & Thondorf I (2005) Three-dimensional quantitative structure-activity relationship analyses of b-lactam anti- biotics and tripeptides as substrates of the mammalian H + ⁄ peptide cotransporter PEPT1. J Med Chem 48, 4410–4419. 16 Biegel A, Gebauer S, Brandsch M, Neubert K & Thon- dorf I (2006) Structural requirements for the substrates of the H + ⁄ peptide cotransporter PEPT2 determined by three-dimensional quantitative structure-activity rela- tionship analysis. J Med Chem 49, 4286–42961. 17 Knu ¨ tter I, Theis S, Hartrodt B, Born I, Brandsch M, Daniel H & Neubert K (2001) A novel inhibitor of the mammalian peptide transporter PEPT1. Biochemistry 40, 4454–4458. 18 Theis S, Knu ¨ tter I, Hartrodt B, Brandsch M, Kottra G, Neubert K & Daniel H (2002) Synthesis and character- ization of high-affinity inhibitors of the H + ⁄ peptide transporter PEPT2. J Biol Chem 277, 7287–7292. 19 Tyagi NK, Goyal P, Kumar A, Pandey D, Siess W & Kinne RK (2005) High-yield functional expression of human sodium ⁄ d-glucose cotransporter1 in Pichia pastoris and characterization of ligand-induced confor- mational changes as studied by tryptophan fluorescence. Biochemistry 44, 15514–15524. I. Knu ¨ tter et al. Labeled high-affinity substrate for peptide transporters FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS 5913 20 Knu ¨ tter I, Hartrodt B, Theis S, Foltz M, Rastetter M, Daniel H, Neubert K & Brandsch M (2004) Analysis of the transport properties of side chain modified dipep- tides at the mammalian peptide transporter PEPT1. Eur J Pharm Sci 21, 61–67. 21 Wu ¨ thrich K, Billeter M & Braun W (1984) Polypeptide secondary structure determination by nuclear magnetic resonance observation of short proton-Proton distances. J Mol Biol 180, 715–740. 22 Brandsch M, Thunecke F, Ku ¨ llertz G, Schutkowski M, Fischer G & Neubert K (1998) Evidence for the abso- lute conformational specificity of the intestinal H + ⁄ pep- tide symporter, PEPT1. J Biol Chem 273, 3861–3864. 23 Brandsch M, Knu ¨ tter I, Thunecke F, Hartrodt B, Born I, Bo ¨ rner V, Hirche F, Fischer G & Neubert K (1999) Decisive structural determinants for the interaction of proline derivatives with the intestinal H + ⁄ peptide sym- porter. Eur J Biochem 266, 502–508. 24 Brandsch M, Brandsch C, Prasad PD, Ganapathy V, Hopfer U & Leibach FH (1995) Identification of a renal cell line that constitutively expresses the kidney-specific high-affinity H + ⁄ peptide cotransporter. FASEB J 9, 1489–1496. 25 Boll M, Herget M, Wagener M, Weber WM, Marko- vich D, Biber J, Clauss W, Murer H & Daniel H (1996) Expression cloning and functional characterization of the kidney cortex high-affinity proton-coupled peptide transporter. Proc Natl Acad Sci USA 93, 284–289. 26 To ´ th G, Lovas S & Otvo ¨ s F (1997) Tritium labeling of neuropeptides. Methods Mol Biol 73, 219–230. 27 Brandsch M, Miyamoto Y, Ganapathy V & Leibach FH (1994) Expression and protein kinase C-dependent regulation of peptide ⁄ H + cotransport system in the Caco-2 human colon carcinoma cell line. Biochem J 299, 253–260. 28 Sala-Rabanal M, Loo DD, Hirayama BA, Turk E & Wright EM (2006) Molecular interactions between di- peptides, drugs and the human intestinal H + -oligopep- tide cotransporter hPEPT1. J Physiol 574, 149–166. 29 Goodman M, Felix A, Moroder L & Toniolo C (2002) Houben-Weyl Methods of Organic Chemistry, Vol. E22a, 4th edn. Georg Thieme Verlag, Stuttgart, New York. Labeled high-affinity substrate for peptide transporters I. Knu ¨ tter et al. 5914 FEBS Journal 274 (2007) 5905–5914 ª 2007 The Authors Journal compilation ª 2007 FEBS . peptide cotransporters, such as Gly-Sar, Ala-Ala, Lys-Lys, Ala-Asp, d-Phe-Ala, Ala-Ala-Ala, d-aminolevulinic acid, cefadroxil and Ala-4-nitroanilide (all 100. Synthesis and characterization of a new and radiolabeled high-affinity substrate for H + /peptide cotransporters Ilka Knu ¨ tter 1 , Bianka Hartrodt 2 ,Ge ´ za