INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

PEPTIDE TRANSPORTERS located in the brush-border membrane of kidney tubular cells play a pivotal role in preserving amino acid nitrogen by reabsorption of di- and tripeptides filtered or generated by enzymatic hydrolysis of larger filtered oligopeptides. Two peptide transport systems with different substrate affinities have been described to exist in the brush-border membrane of the tubular cells (9, 10, 23). Both systems were found to operate in an electrogenic mode by coupling of substrate influx to an inwardly directed H⁺ gradient (9, 23, 26, 27). Besides short-chain peptides, a number of peptidomimetics carrying a peptide backbone, such as -lactam antibiotics of the aminocephalosporin class or the anti-cancer agent bestatin, serve as substrates (8, 15, 23). The cDNAs encoding the two distinctly different H⁺-peptide cotransporters have been cloned from intestine (PEPT1) and kidney (PEPT2) of various species (3, 4, 13, 19, 20, 22, 25). Although the high-affinity transporter PEPT2 is expressed mainly in the kidney but not in the gut (3, 20, 25), PEPT1 mRNA is also expressed at low levels in renal tissue (19, 20, 22).

Although the characterization of the renal high-affinity transporter has been performed after heterologous expression (1, 3), a detailed analysis of its function in renal epithelial cells with regard to kinetics and acid-loading mechanisms has not been performed. This lack of information results mainly from the unavailability of suitable cell lines. Until now, the only cell line described to express the kidney-specific high-affinity H⁺-peptide cotransporter endogenously is SKPT-0193 Cl.2 obtained by SV40 transformation of rat proximal tubular cells (6). In the present study we describe for the first time the endogenous expression of a high-affinity peptide transporter in the porcine kidney cell line LLC-PK₁. Because LLC-PK₁ cells have been shown to differentiate into epithelial cells that have been proven to be useful in the study of selected proximal cell processes (21, 31), they might provide a valuable model for studies on the characteristics and regulation of the renal high-affinity peptide transporter.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Materials. Custom-synthesized D-[³H]Phe-L-Ala (9 Ci/mmol) and unlabeled D-Phe-L-Ala were obtained from Zeneca (Cheshire, UK) and Bachem (Heidelberg, Germany), respectively. All other peptides and -lactams were purchased from Sigma Chemical. The pH-sensitive fluorescent dye 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein-AM (BCECF-AM) was obtained from Bioprobes (Leiden, Netherlands). Fluorescent N-hydroxy-succinimidyl 7-amino-4-methylcoumarin-3-acetic acid (AMCA-NHS) was obtained from Pierce (Rockford, IL). All the materials needed for cell culture were either from GIBCO (Eggenstein, Germany) or Renner (Dannstadt, Germany). Rat tail collagen R was purchased from Serva. All reagents for RNA preparation and RT-PCR were from MBI Fermentas (Heidelberg, Germany), and the primers were custom synthesized by Eurogentec (Seraing, Belgium). Tertiary butyl-oxy-carbonyl-D-Ala-L-Lys-tertiary butylester (Boc-D-Ala-L-Lys-OtBu) was a generous gift from Prof. H. Brückner (Giessen, Germany).

Cell culture. LLC-PK₁ cells (American Type Culture Collection, CRL 1392, passage 195) were cultured and passaged in Dulbecco's modified Eagle's medium (GIBCO 41965) supplemented with 10% fetal calf serum, 2 mM glutamine, 1% MEM nonessential amino acids (GIBCO 01140), 10 mM HEPES, 1 mM sodium pyruvate, 100 IU/ml penicillin, and 100 µg/ml streptomycin in a humidified incubator at 37°C under an atmosphere of 5% CO₂. Cells between passages 200 and 215 were seeded at a density of 5 × 10⁵ cells/well on Renner 6-well plastic cell culture plates or 2.2 × 10⁵ cells/well on 12-well plates subsequent to collagen coating of the wells with rat tail collagen.

Transport studies. Flux studies in LLC-PK₁ cells were performed in a buffer containing (in mM) 145 NaCl, 5.4 Cl, 1.8 CaCl₂, 1.8 MgSO₄, 20 glucose, and 25 HEPES/Tris (pH 7.4) or MES/Tris (pH 6.5), respectively. For uptake, cell monolayers grown in six-well plates were washed free of serum-containing medium and incubated with substrates or inhibitors for 15 min at 37°C. After the incubation period the cells were washed three times with ice-cold incubation buffer, scraped off with a rubber policeman after addition of 600 µl TEN buffer/well (in mM: 150 NaCl, 40 Tris, 1 EDTA), and digested with 20 µl of tissue solubilizer. Cellular accumulation of D-[³H]Phe-L-Ala was measured subsequent to the addition of scintillation cocktail by liquid scintillation spectroscopy. Binding of tracer to the cells was determined as the residual radioactivity associated with the cells in the presence of excess nonlabeled (20 mM) Gly-Gln. Uptake of D-[³H]Phe-L-Ala over 15 min was linear for all pH values and substrate concentrations tested.

Synthesis of D-Ala-L-Lys-AMCA and fluorescence microscopy. Conjugation of AMCA with the -amino group of lysine has been carried out using Boc-D-Ala-L-Lys-OtBu and AMCA-NHS as starter molecules (2). After removal of protective groups, D-Ala-L-Lys-AMCA was purified by two-dimensional preparative thin-layer chromatography. Determination of the compound's concentration was based on its molar extinction coefficient (absorption maximum at 340 nm) and fluorescent properties (emission maximum at 450 nm when excited at 340 nm).

RT-PCR from RNA of LLC-PK₁ cells. RNA from LLC-PK₁ cells was isolated by using the Tristar RNA-clean kit from MBI Fermentas (Heidelberg, Germany). RT-PCR was performed with 5 µg of isolated RNA. First-strand cDNA synthesis was accomplished with a primer representing nucleotides 1899-1879 (back primer: 5'-CCTGTGACAGAGAACATGACC-3') of the protein-coding region of rabbit PEPT2. PCR amplification of a 732-bp product was achieved with a forward primer, representing nucleotides 1167-1188 (5'-CTAGCATGCCTG GCATTTGCAG-3') of the rabbit PEPT2 protein-coding region, and the back primer. Amplification was performed with 35 cycles (95°C denaturation for 1 min, 55°C hybridization for 2 min, 72°C extensions for 2 min; Personal Cycler; Biometra, Göttingen, Germany). RT-PCR products were separated on a 1% agrose gel and visualized by ethidium bromide.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Uptake of D-[³H]Phe-L-Ala and D-Ala-L-Lys-AMCA into LLC-PK₁ cells in the postconfluent state. Peptide transport into LLC-PK₁ cells was measured by uptake of radiolabeled D-Phe-L-Ala at various time points after the cells had reached confluency. Although LLC-PK₁ cells have been described not to possess substantial peptide transport activity on the day of reaching confluency (7), uptake of D-[³H]Phe-L-Ala at pH 6.0 increased almost linearly for up to 11 days in the postconfluent state (Fig. 1). Moreover, transport rates were suppressed to rates similar to that of confluent cells (0 days) by the addition of 1 mM Gly-Gln (Fig. 1). All further uptake experiments were performed at day 9 of postconfluency, since cell adherence to the culture wells decreased subsequent to this time point. The uptake of dipeptides or dipeptide mimetics into LLC-PK₁ cells in the postconfluent state was also demonstrated by transport studies, using the fluorescent dipeptide D-Ala-L-Lys-AMCA as a substrate. Although cells displayed a bright blue fluorescence when incubated with the fluorophore-conjugated dipeptide, simultaneous application of 5 mM Gly-Gln reduced staining of the cells to background levels (Fig. 2).

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Uptake of 5 µM D-[3H]Phe-L-Ala into LLC-PK1 cells as a function of time after confluence; 5 × 105 cells/well were seeded in collagen-coated plastic 6-well culture plates. Cells reached confluency on the day postseeding. Fresh medium was given every second day until day 4 postconfluently, after which medium was renewed daily. Uptake of [3H]D-Phe-L-Ala is shown at indicated time intervals postconfluently in absence () or presence () of 1 mM Gly-Gln.

View larger version (167K): [in this window] [in a new window]   Fig. 2.   Uptake of 5 µM D-Ala-Lys-AMCA into LLC-PK1 cells. Uptake of the fluorescent dipeptide was determined 9 days postconfluently at pH 6.0 and is shown in absence (A, C) or presence (B, D) of 5 mM Gly-Gln. Shown are photomicrographs using Nomarski optics (A, B) or conventional fluorescence microscopy (C, D).

Characteristics of D-[³H]Phe-L-Ala influx into LLC-PK₁ cells. Transport of D-Phe-L-Ala into LLC-PK₁ cells as a function of apical pH increased fourfold when buffer pH values were reduced from 7.4 to 6.0 (Fig. 3, inset). At pH values <5.5, transport rates were moderately reduced when compared with the transport optimum at pH 6.0-5.5. Uptake of D-Phe-L-Ala as a function of substrate concentration followed Michaelis-Menten kinetics with an apparent Michaelis-Menten constant (K_m) of 25.8 ± 3.6 µM and a maximum velocity (V_max) of 33.4 ± 1.7 pmol · cm² · 15 min¹ (Fig. 3). The kinetic characteristics of D-Phe-Ala influx into LLC-PK₁ cells therefore closely resemble those found for the same substrate when assessed in Xenopus oocytes expressing the cloned renal PEPT2 transporter (11).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Uptake of D-[3H]Phe-L-Ala into LLC-PK1 cells as a function of substrate concentration and incubation pH. Transport was measured at pH 6.0 at day 9 postconfluently in the presence of increasing concentrations of D-Phe-L-Ala. Radioactivity associated with the cells in the presence of 20 mM Gly-Gln was assumed to represent substrate diffused or bound, and the amount was subtracted from total counts. Curves were fitted to a Michaelis-Menten equation by nonlinear regression analysis, and the apparent Km value was derived by the least-squares method. Inset: transport of 5 µM D-[3H]Phe-L-Ala when measured in a range between pH 5.0 and 8.0.

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Ki values for differently charged dipeptides and the -lactam, cefadroxil, with regard to D-[3H]Phe-L-Ala uptake into LLC-PK1 cells. Uptake of 5 µM D-Phe-L-Ala at pH 6.0 was measured in the presence of increasing concentrations of cefadroxil (), Gly-Asp (), Gly-Gln (), or Gly-Lys (), respectively. Data represent means ± SE from 3-6 wells per concentration tested.

Intracellular acidification of LLC-PK₁ cells as a consequence of H⁺-peptide cotransport. H⁺-coupled peptide transport with a concomitant decline in pH_i has so far been demonstrated for the renal transporter only for the acidic substrate Ala-Asp (17).

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Intracellular acidification of LLC-PK1 cells as a consequence of apical addition of cefadroxil, Gly-Asp, Gly-Gln, or Gly-Lys (all 1 mM). Cells were loaded with intracellular pH (pHi) indicator BCECF to allow determination of pHi. Fluorescence of cells was measured at 1-min intervals. At the beginning of the experiment, cells were incubated with buffer pH 7.4, which was changed for buffer pH 6.0 after 5 min (arrows). When cells reached an equilibrium of pHi, buffer was changed for cefadroxil (A), Gly-Asp (B), Gly-Gln (C), or Gly-Lys (D) containing buffer pH 6.0 (1). Substrates were washed out by superfusion with buffer pH 7.4 (2). Data represent means ± SE from 4 independent wells. Of each well 5 fluorescence spots were measured over the time period indicated and given as the mean per well.

Expression of a PEPT2 isoform in LLC-PK₁ cells. Although the functional data obtained suggested that the transporter expressed in LLC-PK₁ cells is PEPT2-like, the porcine transporter has not been cloned, and therefore it is not known whether the transporter is expressed in these cells. We therefore performed RT-PCR analysis with specific primers derived from highly conserved regions of cloned PEPT2. It becomes evident from Fig. 6 that a product of 732 bp (specific for PEPT2 but not PEPT1) was amplified from LLC-PK₁ RNA samples. In addition, the amplified product revealed the same EcoR V restriction site as rabbit PEPT2, suggesting that at least a very similar gene product is expressed in LLC-PK₁ cells.

View larger version (98K): [in this window] [in a new window]   Fig. 6.   RT-PCR with PEPT2-specific primers using RNA from LLC-PK1 cells. RT-PCR products were amplified from total RNA using specific forward and backward primers from the PEPT2 coding region to yield a 732-bp fragment. The product was digested with EcoR V to 449- and 283-bp fragments, indicating the same restriction site as the cloned PEPT2 from rabbit renal proximal tubule.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Reabsorption of short-chain peptides in the mammalian renal tubule has been described as mediated by two different H⁺-coupled transport systems that differ considerably in substrate affinities (9, 10, 23). Although the high-affinity-type transport system PEPT2 is prominent on the mRNA level and the functional level, Northern blot analysis suggested that the mRNA of the low-affinity-type transporter PEPT1 is also expressed in kidney but at low levels (19, 20, 22). Brandsch et al. (6) suggested that PEPT1 and PEPT2 may be located in different sections of the nephron. According to their hypothesis, the concentration of small peptides increases from proximal to distal parts of the nephron, due to progressive hydrolysis of filtered oligopeptides by brush-border membrane peptidases. Therefore, the presence of the high-affinity-type PEPT2 in more proximal and of the low-affinity-type PEPT1 in more distal parts of the tubule would be advantageous with regard to the most efficient conservation of amino acid nitrogen. However, so far the proposed different localizations of both transporter isoforms along the tubule, e.g., by in situ hybridization techniques or immunolocalization, have not been reported.

Madin-Darby canine kidney cells, which display characteristics of cells of the distal tubule (18) or collecting duct (24), have been demonstrated to express a low-affinity-type peptide (PEPT1-like) transport activity, suggesting that distal parts of the tubule express solely functional PEPT1 carriers. Moreover, it was shown that this transporter activity is regulated by calmodulin-dependent processes (7).

Although, recently, expression of the high-affinity PEPT2 transporter in rat proximal tubular cells immortalized by SV40 transformation (6) has been demonstrated, these cells are not readily available. The well-established renal proximal cell line LLC-PK₁ (16, 21, 28), on the other hand, was reported not to possess endogenous peptide transport activity (7, 29, 30), a feature that was exploited to use LLC-PK₁ as a host for heterologous expression of PEPT1 and PEPT2 (29-31). Confirming the results of these studies, it was shown here that, before reaching the confluent state, LLC-PK₁ cells possess indeed only very low peptide transport activity. However, we also clearly demonstrate that endogenous peptide transport activity resembling the PEPT2 type is expressed in LLC-PK₁ after cells have reached confluency. Phenotypical characteristics of PEPT2, such as the pH optimum for D-[³H]Phe-L-Ala influx, the kinetics of dipeptide influx, and the existence of a PEPT2-specific mRNA present in LLC-PK₁ cells, justify this conclusion. In addition, the apparent affinities for dipeptides and cefadroxil determined in LLC-PK₁ are almost identical to those reported in Xenopus laevis oocytes (3, 11) or Pichia pastoris (12) expressing the cloned rabbit renal PEPT2 or in Hela cells expressing the human PEPT2 (14). Affinities of the same substrates for interaction with the intestinal transporter are >20-fold lower when determined in oocytes expressing PEPT1 (4, 11). In the postconfluent state LLC-PK₁ cells not only transport D-[³H]Phe-L-Ala but also the dipeptide mimetic D-Ala-L-Lys-AMCA. This fluorescent dipeptide has recently been demonstrated to display influx characteristics in PEPT2-expressing P. pastoris that were almost identical to those obtained by use of radiolabeled D-Phe-L-Ala (12). The coumarin-conjugated dipeptide therefore might allow investigators to study peptide transport in LLC-PK₁ cells independently of expensive custom synthesis of radiolabeled substrates.

Although the pH dependency of dipeptide transport and the rheogenic character of PEPT2-mediated influx of neutral dipeptides already suggested that H⁺ is the cotransported ion species, this had been verified experimentally in renal cells only for Ala-Asp (17). Here we show that translocation of peptide substrates is associated with H⁺ influx that reduces pH_i markedly, irrespective of the net charge of the substrates. By using the pH_i indicator BCECF, we demonstrate that intracellular acidification rates following perfusion with Gly-Asp, Gly-Gln, and Gly-Lys are very similar, whereas those generated by the -lactam cefadroxil are significantly lower. When currents associated with transport of Gly-Asp, Gly-Gln, and Gly-Lys were determined in voltage-clamped oocytes expressing the rabbit PEPT2, we observed that the same three substrates generated the same maximal current responses, independently of the net charge of the substrates at extracellular pH 6.5 (1). Although pH_i could not be measured in oocytes expressing PEPT2, we suggested that the different dipeptides were transported by the same peptide-H⁺ flux-coupling ratio and that this is the consequence of transport of only the zwitterionic form of the substrates. Our present findings in LLC-PK₁ cells confirm this hypothesis by almost identical intracellular acidification rates in the presence of the three differently charged peptides, which may also result from similar if not identical flux-coupling ratios and maximal transport rates.

That pH_i is more reduced by dipeptides than by cefadroxil suggests a higher maximal transport capacity for dipeptides. This may be a consequence of the configuration, e.g., when peptides consisting of L-amino acids are compared with substrates with a D-configuration in the amino-terminal position, such as in cefadroxil or D-Phe-L-Ala. This hypothesis is supported by the fact that pH_i changes induced by D-Phe-L-Ala are comparable with those of cefadroxil but smaller than those induced by dipeptides consisting of L-amino acids only (data not shown). However, it needs to be emphasized that a rapid intracellular hydrolysis of the dipeptides consisting of L--amino acids could also contribute to the more pronounced decrease in pH_i observed for the natural dipeptides.

The demonstration that pH_i is markedly reduced when dipeptides are taken up by the renal peptide transporter addresses the physiological importance of these transport-mediated pH_i changes. Because di- and tripeptides are present in plasma and are continuously filtered in the glomerulus, the renal peptide transporter operates as a constant acid loader in tubular cells. This is important for both the pH_i recovery systems and their regulation, as well as for other metabolic events such as increased renal ammoniagenesis in response to a low pH_i. Because a number of protein kinase recognition sites have been identified in the coding sequence of PEPT2, regulation of transport activity, in particular in relation to changes in pH_i, needs to be investigated. For this purpose LLC-PK₁ might provide a very useful cellular model.