Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The intestinal (PepT1) and renal (PepT2) peptide transporters have been cloned from mammals (Fei et al., 1994; Liang et al., 1995; Saito et al., 1995). These transporters are expressed predominantly in the brush border membrane (BBM) of intestinal and renal epithelial cells (Ogihara et al., 1996) and are responsible for the uptake of small peptides and zwitterionic cephalosporins (Ganapathy et al., 1995). PepT1 is also capable of transporting anionic -lactam antibiotics, including penicillins and cephalosporins (Doring et al., 1996). These transport systems are energy-dependent and driven by a transmembrane H⁺ gradient (Okano et al., 1986; Tsuji et al., 1987).

The organic anion transporter Npt1 also contributes to the transport of -lactam antibiotics, including both zwitterionic and anionic derivatives, in rat hepatocytes (Yabuuchi et al., 1998), choroid plexus (Suzuki et al., 1987), and renal BBM (Tamai et al., 1988). This transporter mediates Cl-dependent, but Na⁺- and pH-independent transport of anionic drugs (Jacquemin et al., 1994). The observation that the Cl conductance induced by expression of the Na⁺/Pi cotransporter NaPi-1 in Xenopus oocytes was inhibited by benzylpenicillin (BP) and probenecid (Busch et al., 1996) also suggests that NaPi-1 mediates the transport of BP in addition to Na⁺- and pH-dependent Pi transport. NaPi-1 shares 65% amino acid sequence identity with rat and human Npt1 (Yabuuchi et al., 1998).

The activities of many types of transporters are regulated by the interaction of neurotransmitters with their respective receptors. For example, epinephrine increases the activity of the Na⁺/H⁺ antiporter by acting at ₂-adrenoceptors in glioma cells (Isom et al., 1987) and increases that of the Na⁺/K⁺/2Cl cotransporter through -adrenoceptor activation in tracheal epithelial cells (Haas et al., 1995). In contrast, epinephrine inhibits the Na⁺/K⁺/2Cl cotransporter via -adrenoceptor activation in lymphocytes (Feldman, 1992). This neurotransmitter also increases the amount of the Na⁺/glucose cotransporter (SGLT1) in the BBM of intestinal epithelial cells by interacting with -adrenoceptors in the basolateral membrane (BLM) and thereby stimulates glucose absorption from rat small intestine (Ishikawa et al., 1997). Moreover, acetylcholine acting at M₃ muscarinic receptors induces the translocation of aquaporin-5 water channel to the cell membrane in rat parotid gland (Ishikawa et al., 1998).

However, it remains to be determined whether transporters responsible for the absorption of -lactam antibiotics are functionally coupled to neurotransmitter receptors in rat small intestine. BP, one of the anionic penicillins, is transported by the H⁺ gradient-dependent, carrier-mediated transport system shared by zwitterionic and anionic -lactam antibiotics and peptides (Doring et al., 1996; Poschet et al., 1996; Wenzel et al., 1996). With the use of BP as a model substrate, we investigated whether the absorption of this compound by rat small intestine is regulated by neurotransmitters. We found that epinephrine markedly increases the absorption of BP from rat small intestine by interacting with both ₂-adrenoceptors and either ₁- or ₂-adrenoceptors, and we further characterized the mechanism of this effect.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. [³H]BP (718 GBq/mmol) was obtained from New England Nuclear (Wilmington, DE). EDTA and epinephrine were obtained from Dojindo Laboratories (Kumamoto, Japan) and Mann Research Laboratories (New York, NY), respectively. Carbonyl cyanide p-trifluoromethoxy phenylhydrazone (FCCP), BP, l-phenylephrine hydrochloride, clonidine hydrochloride, dobutamine hydrochloride, prazosin hydrochloride, yohimbine hydrochloride, and tolazoline hydrochloride were purchased from Sigma Chemical Co. (St. Louis, MO). Acetylcholine hydrochloride and dl-isoproterenol hydrochloride were purchased from Aldrich Chemical Co. (Milwaukee, WI). Salbutamol (albuterol hemisulfate) and N-[2-(methylamino)ethyl]-5-isoquinolinesulfonamide dihydrochloride (H-8) were obtained from Research Biochemicals Inc. (Natick, MA) and Seikagaku Co. (Tokyo, Japan), respectively.

Animals and Diet. Male Wistar rats (10 weeks old) were used for the experiments. They were provided with a standard laboratory diet (MF; Oriental Yeast, Tokyo, Japan) and water ad libitum and maintained in a temperature-controlled environment (22 ± 2°C) with a 12-h light/dark cycle (lights on at 6:00 AM) for at least 2 weeks before the experiments. All procedures were approved by the Animal Care Committee of Tokushima University.

Perfusion of Intestinal Loops and Measurement of Transmural BP Transport. Rats were sacrificed by a blow to the head, and the abdomen was opened by a midline incision. A 15-cm loop of jejunum, starting at a point 10 cm below the ligament of Treitz, was cut from the small intestine and then cleaned by flushing with Ringer's solution (118 mM NaCl, 4.7 mM KCl, 1.2 mM KH₂PO₄, 1.25 mM CaCl₂, 1.2 mM MgCl₂, and 24.9 mM NaHCO₃ gassed with a mixture of 95% O₂, 5% CO₂, pH 7.4) as described previously (Ishikawa et al., 1997). The loop was cannulated and connected to a perfusion apparatus with a peristaltic pump. Perfusion of the lumen of the loop with Ringer's solution containing 1 mM BP and a trace amount of [phenyl-4(n)-³H]BP ([³H]BP) was initiated immediately at a rate of 1 ml/min and was continued for 15 min at 37°C. The perfusate was then changed to fresh solution with or without the addition of various agents.

Preparation of BBM and BLM Vesicles. After perfusion of the loop of jejunum, mucosa was scraped from the tissue with a glass slide and homogenized in 10 volumes of buffer 1 [5 mM HEPES-HCl (pH 7.4), 50 mM mannitol, and 0.25 mM MgCl₂] with a glass homogenizer and Teflon pestle. BBM and BLM vesicles were then prepared as described by Longbottom and Heyningen (1989). In brief, the homogenate was filtered through a nylon bolting cloth (150 mesh), and the filtrate was centrifuged at 300g for 10 min at 4°C. The resulting supernatant was then centrifuged consecutively at 2000g for 10 min, 9750g for 10 min, and 35,000g for 30 min, with removal of the pellet after each step. The final pellet was suspended in buffer 1, after which 1 M MgCl₂ was added to a final concentration of 10 mM and the suspension was maintained on ice for 30 min. The suspension was then centrifuged at 3000g for 15 min to separate BLMs. After washing the resultant pellet with 1 mM EDTA (pH 7.4) by centrifugation to remove any excess MgCl_2, the final pellet was suspended in buffer 2 [20 mM HEPES-Tris (pH 7.4), 300 mM mannitol, 0.1 mM MgSO₄, and 0.02% NaN₃] and used as the source of BLM vesicles (fraction 1). The 3000g supernatant was again centrifuged at 35,000g for 30 min to separate BBMs. The resultant pellet was washed with 1 mM EDTA (pH 7.4), and the final pellet was suspended in buffer 3 [10 mM HEPES-Tris (pH 7.4), 100 mM mannitol, and 100 mM KCl] (Okano et al., 1986) and used as the source of BBM vesicles (fraction 2).

Uptake of [³H]BP into BBM and BLM Vesicles. The amount of BP taken up into BBM or BLM vesicles was quantified with the use of [³H]BP. BBM vesicles (0.25 mg protein) suspended in buffer 3 or BLM vesicles suspended in buffer 2 were incubated in a final volume of 1 ml of the same buffer containing various concentrations of BP and a trace amount of [³H]BP. The Tris-HCl (pH 7.4) of buffer 2 or 3 was replaced by Mes-Tris (pH 6.0) to create a pH gradient across the vesicle membranes. After incubation of vesicles for various times at 25°C as described by Okano et al. (1986), 5 ml of ice-cold buffer 2 or 3 was added, and the mixture was passed through a filter (GF/F; Whatman, Maidstone, England). The filter was washed three times with 5 ml of ice-cold buffer 2 or 3, and the remaining filter-associated radioactivity was determined by scintillation spectroscopy after placing the filter in a vial containing 10 ml of scintillation fluid. In separate experiments, nonspecific adsorption was estimated by the addition of ice-cold buffer 2 or 3 containing various concentrations of BP and a trace amount of [³H]BP to BBM vesicle suspension or BLM vesicle suspension, respectively. The amount of BP incorporated into vesicles was expressed in nanomoles per milligram of protein.

Other Methods. cAMP and protein were measured according to the method described by Longbottom and Heyningen (1989).

Statistical Analysis. All data are given as mean ± S.E. The significance of differences between mean values was determined by the unpaired Student's t test or by one-way ANOVA followed by Fisher's t test. A value of P < .05 was considered statistically significant.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Effects of Epinephrine and Acetylcholine on BP Transport in Rat Small Intestine. The addition of epinephrine at concentrations of 1 µM to 1 mM to the solution perfusing the mucosal side of rat small intestinal lumen markedly increased the amount of [³H]BP transported to the serosal side (Fig. 1). This effect of epinephrine was dose-dependent and reached a maximum (4-fold increase) at a concentration of 0.1 mM, with an EC₅₀ value of 4.22 ± 0.86 µM. The amount of BP transported in the presence of 0.1 mM epinephrine increased as the concentration of BP was increased from 0 to 1.2 mM (Fig. 2), reaching a maximum at 1.0 mM. Kinetic analysis according to the Michaelis-Menten equation of BP transport in the presence and absence of 0.1 mM epinephrine yielded apparent K_m values of 0.28 ± 0.01 and 0.21 ± 0.02 mM and V_max values of 323 ± 27 and 88.6 ± 5.6 nmol/g wet wt./30 min, respectively. Acetylcholine (1.0 mM) did not have a stimulatory effect on the BP transport in rat small intestine (Table 1).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Effect of epinephrine concentration on the transport of BP in rat small intestine. Rat small intestinal lumen that had been equilibrated with Ringer's solution containing 1 mM BP and a trace amount of [3H]BP were perfused for 30 min with the same solution containing the indicated concentrations of epinephrine. The amount of BP transported from the mucosal to the serosal side of the loops was then determined as described in Experimental Procedures. Data are mean ± S.E. for four to eight loops. *P < .05, **P < .01, ***P < .001 versus value for the absence of epinephrine.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Concentration-response relation for BP transport in rat small intestine perfused with or without epinephrine. a, rat small intestinal lumen that had been equilibrated with Ringer's solution containing the indicated concentrations of BP and a trace amount of [3H]BP was perfused for 30 min with the same solution in the presence or absence of 0.1 mM epinephrine. The amount of BP transported from the mucosal to the serosal side of the loops was then determined. Data are mean ± S.E. for four to eight loops. b, Eadie-Hofstee plot (V versus V/S), where V is the velocity of BP transport (nmol/g wet wt./30 min). and S is BP concentration (mM).

Effects of Various Adrenergic Agents on BP Transport in Rat Small Intestine. The effect of various adrenergic agonists and antagonists were investigated to identify the adrenoceptor subtype that mediates the stimulatory action of epinephrine on the mucosal-to-serosal transport of BP in rat small intestine. The addition of phenylephrine, clonidine, dobutamine, or salbutamol to the luminal perfusate did not affect BP transport (Table 1). Isoproterenol stimulated BP transport but to an extent lesser than that observed with epinephrine. However, the addition of phenylephrine or clonidine together with isoproterenol increased BP transport to the same extent as that observed with epinephrine. The addition of phenylephrine in the presence of salbutamol, but not in the presence of dobutamine, also increased of BP transport by the same extent as did epinephrine alone. Tolazolin or propranolol inhibited the epinephrine-induced increase in BP transport (Table 2). These results thus indicate that the stimulatory effect of epinephrine on BP transport is mediated by both ₂-adrenoceptors and either ₁- or ₂-adrenoceptors.

Effects of Dibutyryl cAMP, Forskolin, Staurosporine, and H-8 on BP Transport in Rat Small Intestine. The addition of 10 µM dibutyryl cAMP or 2.5 µM forskolin to the luminal perfusate increased BP transport but to a lesser extent than did epinephrine (Table 3). However, the addition of dibutyryl cAMP or forskolin in the presence of 0.1 mM phenylephrine or 0.1 mM clonidine stimulated BP transport to an extent similar to that observed with epinephrine. These results suggest that the contribution of -adrenoceptors to the effect of epinephrine on BP transport is mediated by an increase in the intracellular concentration of cAMP. Indeed, this suggestion was supported by the observation that perfusion of the loops with epinephrine or forskolin significantly increased the cAMP content of the mucosal cells (Table 4). To evaluate whether the effect of cAMP on BP transport is mediated by cyclic AMP-dependent protein kinase (PKA), we treated intestinal lumen for 15 min with 1 µM staurosporine or 60 µM H-8, both of which are inhibitors of this enzyme, before perfusion with 0.1 mM epinephrine. Both staurosporine and H-8 abolished the epinephrine-induced increase in BP transport in the continued presence of them (Table 3), suggesting that the stimulatory effect of cAMP is mediated by PKA.

Stereospecificity of BP Transport in Rat Small Intestine. To investigate the stereospecificity of the BP transport system in rat small intestine, we investigated the effects of various compounds known to be transported by PepT1. Glycyl-L-alanine, glycyl-L-leucine, glycylsarcosine, and glycyl-glycyl-glycine were used as representative peptides. Cephalexin and cefadroxil were as representative zwitterionic -lactams, and cephalothin was as a representative anionic -lactam. After perfusion of the lumen with Ringer's solution containing BP in the presence of these compounds, epinephrine was added to the perfusate at a concentration of 0.1 mM and perfusion was continued for an additional 30 min. Glycyl-L-alanine, glycyl-L-leucine, glycylsarcosine, and glycyl-glycyl-glycine reduced the extent of epinephrine-induced BP transport by 80 to 90% (Table 5). Cephalexin, cefadoroxil, and cephalothin reduced epinephrine-induced BP transport by 50%, which is consistent with previous data (Poschet et al., 1996). These results suggested that like peptides and zwitterionic or anionic cephalosporins, BP is transported by PepT1 in rat small intestine.

Effects of Epinephrine and a Membrane H⁺ Gradient on Uptake of BP into BBM and BLM Vesicles. BBM vesicles prepared from rat small intestine that had been perfused in the absence or presence of 0.1 mM epinephrine for 60 min were suspended in buffer 3 and incubated for various times at 25°C in buffer 3 containing 1 mM BP and a trace amount of [³H]BP. The extent of BP uptake into BBM vesicles prepared from epinephrine-perfused intestinal loops was greater than that for vesicles prepared from control loops (Fig. 3). The initial rate of BP uptake into BBM vesicles has previously been shown to be markedly increased in the presence of an H⁺ gradient (Poschet et al., 1996). To generate such a gradient, we suspended BBM vesicles in buffer 3 and then with buffer 3 in which Tris-HCl (pH 7.4) was replaced with Mes-Tris (pH 6.0). The initial rate of BP uptake was markedly increased in the presence of the H⁺ gradient, showing the characteristic overshoot phenomenon (Fig. 3). In the presence or absence of the H⁺ gradient, the extent of BP uptake into BBM vesicles prepared from epinephrine-perfused intestinal loops was greater than that for vesicles prepared from control loops at all incubation times. The effect of the initial H⁺ gradient on the extent of BP uptake was no longer significant at 30 min. Even in the absence of driving gradient, the equilibrium uptake at 30 min was larger in BBM vesicles prepared from epinephrine-stimulated loops than that for vesicles from control loops.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Effect of the presence of a H+ gradient on the time course of BP uptake by BBM vesicles prepared from rat small intestine perfused in the absence or presence of epinephrine. BBM vesicles were prepared from rat small intestinal lumen that had been perfused for 60 min in the absence or presence of 0.1 mM epinephrine, as indicated. Vesicles were suspended in buffer 3 and then incubated for the indicated times at 25°C with 1 mM BP and a trace amount of [3H]BP either in buffer 3 or in buffer 3 in which Tris-HCl (pH 7.4) was replaced with Mes-Tris (pH 6.0). Uptake of BP was then terminated and assayed as described in Experimental Procedures. Data are mean ± S.E. for vesicles prepared from three loops.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Concentration-response relation for BP uptake into BBM vesicles from rat small intestine perfused in the absence or presence of epinephrine. a, BBM vesicles were prepared from rat small intestinal lumen that had been perfused for 60 min in the absence or presence of 0.1 mM epinephrine, as indicated. Vesicles were suspended in buffer 3 and then incubated for 30 min at 25°C with a trace amount of [3H]BP in buffer 3. Uptake of BP was then terminated and assayed as described in Experimental Procedures. Data are mean ± S.E. for vesicles prepared from three loops. b, Eadie-Hofstee plot (V versus V/S), where V is the velocity of BP transport (nmol/g wet wt./h). and S is BP concentration (mM).

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Effect of the presence of a H+ gradient on the time course of BP uptake by BLM vesicles prepared from rat small intestine perfused in the absence or presence of epinephrine. BLM vesicles were prepared from rat small intestinal loop that had been perfused for 60 min in the absence or presence of 0.1 mM epinephrine, as indicated. Vesicles were suspended in buffer 2 and then incubated for the indicated times at 25°C with 1 mM BP and a trace amount of [3H]BP either in buffer 2 or in buffer 2 in which Tris-HCl (pH 7.4) was replaced with Mes-Tris (pH 6.0). Uptake of BP was then terminated and assayed as described in Experimental Procedures. Data are mean ± S.E. for vesicles prepared from three loops.

Effect of a Membrane H⁺ Gradient on BP Transport in Intestinal Loops. To evaluate the role of an H⁺ gradient in epinephrine-induced BP transport in intestinal loops, we investigated the effect of FCCP, amiloride, and ouabain. Perfusion with FCCP inhibited epinephrine-induced BP transport by 60% at the concentration that collapses H⁺ gradients (Table 6). The addition of 1 mM amiloride, an inhibitor of the Na⁺/H⁺ exchanger, to the perfusate abolished the BP transport by 90%. Similarly, ouabain, at the concentration that inhibits Na⁺,K⁺-ATPase activity, reduced epinephrine-induced BP transport by 90%.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The intestinal mucosa is extensively innervated by adrenergic fibers, and high concentrations of norepinephrine can potentially be achieved in the vicinity of the epithelial cells (Valet et al., 1993). ₂-Adrenoceptors have been detected in the intestinal mucosa, and the interaction of adrenergic receptors and their agonists augmented the net absorption of Na⁺ and Cl ions from the intestinal lumen (Paris et al., 1990; Hirdebrand and Brown, 1992). -Adrenoceptors are also present in the intestine, and their quantity was markedly increased in BLM of mucosal cells of intestinal loops perfused with epinephrine (Ishikawa et al., 1997). It appears that - and -adrenergic agonists or antagonists added into the luminal perfusate of intestinal loops migrate along transcellular or paracellular pathways and then either interact directly with the respective adrenoceptors located in the BLM of epithelial cells or affect these cells indirectly via the enteric nervous system.

Peptides are transported by PepT1 in a manner that is dependent on an H⁺ gradient generated by the Na⁺/H⁺ exchanger and Na⁺,K⁺-ATPase (Ganapathy and Leibach, 1985). Zwitterionic -lactams are also transported by PepT1 in the intestine (Fei et al., 1994). Although Daniel (1996) pointed out the controversy as to whether anionic -lactams such as BP are transported by the same transporter as that responsible for the transport of zwitterionic -lactams, it was recently demonstrated that zwitterionic and anionic -lactams do indeed share the same transporter (Poschet et al., 1996). In this study, the epinephrine-induced increase in BP transport in rat small intestine was reduced in the presence of peptides or of zwitterionic or anionic -lactams, suggesting that these compounds are all transported by a common transporter.

The uptake of BP across the BBM into enterocytes of the small intestine was previously shown to occur in a saturable manner and to be stimulated by an inwardly directed H⁺ gradient (Kramer et al., 1990). We have now shown that the treatment of rat small intestine with FCCP at the concentration that sufficient to collapse the H⁺ gradient inhibited epinephrine-induced BP transport by 60%, suggesting that the membrane H⁺ gradient in this loop is also important for this effect of epinephrine. The Na⁺/H⁺ exchanger located in the BBM, together with Na⁺,K⁺-ATPase in the BLM, has been proposed to be responsible for the generation and maintenance of such an H⁺ gradient (Ganapathy and Leibach, 1985). The Na⁺/H⁺ exchanger has also been detected in the BLM in rabbit ileum (Sundaram et al., 1991). In this study, treatment of rat small intestine with amiloride, at a concentration that inhibits the Na⁺/H⁺ exchanger, or with ouabain, at a concentration that inhibits Na⁺,K⁺-ATPase, also inhibited epinephrine-induced BP transport by 90%. These observations demonstrate the importance of the Na⁺/H⁺ exchanger and Na⁺,K⁺-ATPase in epinephrine-induced BP transport in rat small intestine. The stimulation of ₁- and ₂-adrenoceptors induces activation of the Na⁺/H⁺ exchanger rat proximal nephrons (Liu et al., 1997), although ₂-adrenoceptor stimulation inhibits the Na⁺/H⁺ exchanger activity in renal BBM as a result of binding of the Na⁺/H⁺ exchanger regulatory factor to the COOH-terminal tail of these receptors (Hall et al., 1998). Na⁺,K⁺-ATPase is also activated by either ₁- or ₂-adrenergic agonists via an activation of a calcium calmodulin-dependent protein phosphatase in renal tubule cells (Aperia et al., 1992). It is therefore likely that epinephrine induces the generation and maintenance of an H⁺ gradient across the BBM by interacting with either ₁- or ₂-adrenoceptors and thereby activating the Na⁺/H⁺ exchanger and Na⁺,K⁺-ATPase.

The activity of PepT1 in human intestinal Caco-2 cells is increased by incubation of cells with the dipeptide glycyl-glycine for 21 days; the amounts of PepT1 protein and mRNA are also increased by this treatment (Walker et al., 1998). However, in our experiments, the epinephrine-induced increase in the extent of BP transport is not likely to be mediated by an increase in the abundance of PepT1 mRNA because of the short incubation time (30 min). Several types of transporters and channels traffic between the plasma membrane and intracellular membranes in response to hormones or other agents. Such trafficking is thought to be mediated by the exocytotic insertion of vesicles containing transporters or channels into the plasma membrane and by the endocytotic retrieval of these proteins (Bradbury and Bridges, 1994). Stimulation of -adrenoceptors or vasopressin receptors with respective agonists induces the activation of PKA and either the translocation of SGLT1 to the BBM in rat intestine (Ishikawa et al., 1997) or the translocation of aquaporin-2 to the apical membrane in rat kidney (Nielsen et al., 1993), respectively. Recently, PepT1 was shown to be translocated to the apical membrane from a cytoplasmic pool in human intestinal cell monolayer by insulin (Thamotharan et al., 1999). In this study, BP uptake into BBM vesicles prepared from epinephrine-perfused intestinal loops was greater than that into those prepared from control loops, in either the presence or absence of an H⁺ gradient. Even in the absence of an H⁺ gradient, the V_max value for the uptake of BP into BBM vesicles prepared from intestinal loops perfused with epinephrine was twice that for BP uptake into control vesicles. These observations suggest that epinephrine induces the activation of PKA in rat small intestine and thereby promotes the trafficking of PepT1 to the BBM.

In conclusion, our data suggest that epinephrine acts at either ₁- or ₂-adrenoceptors on rat intestinal epithelial cells to generate and maintain an H⁺ gradient across the BBM and simultaneously acts at ₂-adrenoceptors to increase the amount of PepT1 in the BBM. As a result of these interactions with its receptors, epinephrine increases BP transport in the rat small intestine.