INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

The rapid uptake of intact small peptides across the brush-border membrane of the small intestinal epithelium is the major route for absorption of dietary protein -amino nitrogen (1). Hitherto, it has been thought that a number of chemical features, for example free amino and carboxyl termini, are essential in contributing to substrate interaction with, and translocation through, the intestinal peptide transporter. These features include the presence of a peptide bond within the substrate molecules. Indeed a major review (1) states that "it is the presence of peptide bonds which make di- and tripeptide acceptable to the peptide transport systems." Although previous work (e.g. Ref. 2) has shown that molecules lacking this feature can inhibit transport of peptides (presumably by substrate binding), we describe here for the first time rapid transport of a small totally non-peptidic substrate through the intestinal peptide transporter. The substrate, 4-aminophenylacetic acid (4-APAA),¹ was selected on the basis of its chemical structure, it being a potential mimic of a dipeptide (D-Phe-L-Ala) (Fig. 1) which previously we have shown to be an excellent substrate for epithelial peptide transport (3, 4).

View larger version (5K): [in this window] [in a new window]   Fig. 1.   Chemical structure of a dipeptide (where R1 and R2 are amino acid side chains) (A) and 4-aminophenylacetic acid (4-APAA) (B). 4-APAA was designed to mimic the spatial configuration of a dipeptide. It has the terminal amino and carboxyl groups in the transconfiguration and is planar, but it does not have a peptide bond.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Rat renal brush-border membrane vesicles were prepared as described previously (5), and initial rates of labeled peptide transport (influx, efflux) were determined by rapid filtration (4, 6). Rat intestinal loops in vitro and vascularly perfused small intestine in situ were used to measure transepithelial fluxes in the intact small intestine as described previously (3, 7). Luminal pH was changed using a previously published protocol (8). Isolated murine enterocytes were prepared by enzymatic digestion using haluronidase, and intracellular pH was determined fluorimetrically using carboxy-SNARF (9). Isolated cells were exposed to 20 mM HEPES-buffered Krebs solution containing either the dipeptide (Phe-Ala) or 4-APAA. The intracellular ratiometric signal (540 nm excitation, 590 nm/640 nm emission) was calibrated after the experiments by superfusing solutions of 140 mM KCl containing 10 µM nigericin and buffered to pH 5.5, 7.5, and 9.5 with PIPES, HEPES, and CAPS, respectively. Xenopus oocyte expression of PepT1 cRNA was as described (10, 11) with subsequent HPLC detection (3) of transported substrate and correction for transport in water-injected controls. For efflux assays 4.6 nl (110 fmol) of radiolabeled dipeptide (D-Phe-L-Gln, 0.1 µCi/µl) was injected into each oocyte. The assay was started after the oocytes were washed, 15 min after injection.

Inhibition constants (K_i) were determined as described previously (6). Briefly radiolabeled dipeptide influx was corrected for non-mediated transport by subtracting the transport of radiolabel seen in the presence of saturating unlabeled peptide (10 mM). The ENZFITTER program was then used to calculate the K_i for mediated transport only.

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

As part of a screening assay for potential substrates for epithelial peptide transport we used a preparation from rat renal cortex of apical brush-border membrane vesicles (5). Fig. 2 shows, using this assay, that 4-APAA is capable both of cis-inhibition and trans-stimulation of labeled D-Phe-L-Glu influx and efflux, respectively. Because in the efflux assay there is no proton electrochemical gradient (the membrane being voltage-clamped using valinomycin and high K⁺ in both intra- and extravesicular compartments and there being no pH gradient across the membrane) the ability of 4-APAA to trans-stimulate efflux of peptide must mean that it is translocated as a substrate (12). Based on this initial observation, three types of experiments were performed which functionally measure transport of this molecule in relation to intestinal absorption.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   A, cis inhibition of mediated D-Phe-L-Glu (0.34 µM) influx, under initial rate, into rat renal cortex brush-border membrane vesicles by 4-APAA. 4-APAA inhibited peptide entry in a concentration-dependent manner with a Ki of 6.5 mM. B, trans-stimulation of labeled D-Phe-L-Glu (0.34 µM) efflux from rat renal cortex brush-border membrane vesicles in the presence and absence of external substrates. The concentrations of external D-Phe-L-Glu (0.48 mM) and of 4-APAA (13 mM) were twice the estimated Ki for cis-inhibition of influx. Efflux was measured (with external and internal pH 5.5 and membrane potential clamped to zero) over 4 s (i.e. under initial rate conditions). Data points are shown for control (no external substrate) and in the presence of peptide or 4-APAA. Note that peptide efflux in both cases is significantly greater than control (p < 0.02, paired Student's t-test).

Using isolated rat small intestine in vitro, 4-APAA was shown using HPLC detection to be transported rapidly across the intact epithelium at a rate which was very substantially increased by external acidification (pH of lumen decreased from 7.4 to 6.8) characteristic of proton-coupled peptide transport (Table I). Moreover associated with this stimulation of transepithelial transport there was "active accumulation" of 4-APAA within the epithelium to a concentration 3.7 times greater than that in the lumen (Table I). In the intact small intestine of the rat vascularly perfused in situ there was also rapid transepithelial transport (at a rate faster than that seen for the peptide D-Phe-L-Gln); in this preparation transport of 4-APAA was inhibited significantly by concomitant addition of 10 mM D-Phe-L-Gln. Conversely, transport of D-Phe-L-Gln was inhibited significantly by the addition of 10 mM 4-APAA (Table I).

When isolated intestinal enterocytes were superfused with a solution containing Phe-Ala the peptide caused intracellular acidification (as expected for a substrate for a proton-coupled transporter (10)) (Fig. 3); applied separately, the isostere 4-APAA (2 mM) had the same effect. However in the presence of a maximal concentration of peptide (20 mM), 4-APAA was unable to produce further intracellular acidification; similarly in the presence of a high concentration of 4-APAA (20 mM), the initial rate and extent of intracellular acidification in response to the peptide was markedly reduced. These findings strongly suggest that the peptide and 4-APAA share the same route for influx and that this pathway is proton-coupled. (Note that at high concentrations 4-APAA acted as a weak acid (13) producing transient intracellular acidification and alkalinization following, respectively, its addition to and removal from the superfusate.)

View larger version (12K): [in this window] [in a new window]   Fig. 3.   Effect of dipeptide and 4-APAA separately and in combination on intracellular pH measured in mouse isolated enterocytes. Intracellular pH was measured using carboxy-SNARF. In the presence of supramaximal concentration (20 mM) of dipeptide no further acidification was induced by the addition of (2 mM) 4-APAA, although on its own 4-APAA caused (reversible) acidification. At higher concentrations (20 mM) 4-APAA acted as a weak acid producing transient intracellular acidification and alkalinization following, respectively, its addition to and removal from the superfusate. In the presence of 20 mM 4-APAA, dipeptide failed to elicit additional acidification.

Fig. 4 directly confirms that 4-APAA is a substrate for the intestinal peptide transporter PepT1 when this protein is expressed in Xenopus oocytes. This figure shows that 4-APAA (measured by HPLC) is transported into PepT1-expressing oocytes, that 4-APAA inhibits peptide (D-Phe-L-Gln) uptake into PepT1-expressing oocytes, and that radiolabeled peptide (D-Phe-L-Gln) efflux from PepT1-expressing (but not from water-injected control) oocytes is trans-stimulated by 4-APAA as well as by peptide.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Transport of peptide and 4-APAA into PepT1-expressing Xenopus oocytes. A, the time course of 4-APAA (2 mM) influx at pH0 = 6.0 as determined by HPLC of individual oocytes. Data points are mean ± S.E. (n = 5). Transport into control (water-injected) oocytes has been subtracted. B, 4-APAA inhibition of mediated D-Phe-L-Gln influx into PepT1-expressing oocytes. Transport measurements were determined under initial rate conditions (1 h) at pH0 = 6.0. The Ki for 4-APAA is approximately 10 mM. C, trans-stimulation of labeled D-Phe-L-Gln efflux from Xenopus oocytes expressing PepT1 in the presence and absence of external substrate (D-Phe-L-Gln or 4-APAA). The external pH was 6.0. Data points are mean ± S.E. (n = 5). The data are significantly different from control *, p < 0.002 for the peptide and for the 4-APAA in PepT1-injected (cross-hatched bars) but not in water-injected (open bars) controls. Injected oocytes were studied 3 days after cRNA injection. The concentrations of peptide (10 mM) and 4-APAA (20 mM) were chosen to be saturating (peptide) and approximately twice the Ki (4-APAA) for cis-inhibition of influx (cf. panel B).

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

The results from these transport assays demonstrate unambiguously that the presence of a peptide bond is not necessary for the rapid translocation of a substrate through the intestinal peptide transporter (PepT1). Our study leaves open the question as to the minimum requirements for a molecule to be a substrate. In the case of 4-APAA the fact that this molecule has both amino- and carboxyl-terminals separated by 5 carbon atoms as well as being planar (through its possession of an aromatic ring) may be pertinent. Nevertheless the findings reported here offer scope for chemical design of drugs which, despite lacking a peptide bond, will be absorbed rapidly from the small intestine via this transporter.