Introduction

Top Abstract Introduction Methods Results Discussion References

Cefixime is a dianionic carboxymethoxyimino-cephalosporin. Despite its poor lipophilicity and ionization at physiological pH, cefixime is significantly absorbed unchanged after oral administration (Roche, 1988). Human kinetic studies (Faulkner et al., 1988) and in vitro experiments (Tsuji et al., 1987a, 1987b) have shown that cefixime absorption is saturable. Because cephalosporin antibiotics have a chemical structure similar to that of dipeptides, transport studies have suggested common transport systems (Dantzig et al., 1992; Kimura et al., 1983; Tsuji et al., 1987b). However, simple diffusion into enterocytes (Dantzig et al., 1994) and a paracellular pathway (Inui et al., 1992) may also be involved for cefixime. Studies in healthy volunteers have shown that nifedipine, a calcium antagonist, increases both the absorption rate and the bioavailability of cefixime without modifying its distribution or elimination (Duverne et al., 1992). Similar observations have been reported by Westphal et al. (1990) with another -lactam antibiotic, amoxicillin, which is also absorbed by the dipeptide transport system (Nakashima et al., 1984). Because cefixime and amoxicillin are not metabolized in the liver, the beneficial effect of nifedipine could be restricted to their absorption. Several mechanisms might be involved in the absorption-promoting effect of nifedipine, including a direct action on the intestinal epithelial transport of cefixime, a reduction in intestinal motility leading to an increased contact time between cefixime and the transporter and a local hemodynamic effect that could promote passive absorption. Because a direct interaction was not demonstrated on Caco-2 cells,¹ an indirect mechanism involving a neurohormonal regulation was proposed.

The objectives of this study were to (1) analyze the absorption characteristics of cefixime, (2) reproduce the cefixime absorption-promoting effect of nifedipine and (3) investigate the possible involvement of hemodynamic or indirect neurohormonal effects of nifedipine. For these purposes, we used the well-validated in situ intestinal perfusion model in the rat (Schanker et al., 1958). To discriminate hydroelectrolytic vs. neurohormonal effects, selected drugs and chemicals with defined pharmacological actions were coperfused with cefixime and nifedipine. PYY, a gut hormone that is produced and released from mucosal endocrine cells of the colon and distal small bowel, was chosen to study the inhibition of intestinal and colonic motility resulting from the local inhibition of hydroelectrolytic secretions (Sheikh, 1991). HM, a cholinergic and nicotinic blocker, and TTX, a nervous conduction blocker that inhibits the release of secondary neurotransmitters (Bulbring and Tomita, 1967), were used to characterize further neurohormonal regulation.

	Methods

Top Abstract Introduction Methods Results Discussion References

Chemicals and Solutions

Cefixime was obtained from Pharmuka (Neuilly s/Seine, France). [7-¹⁴C]Salicylic acid (2.2 GBq/mmol) was from DuPont-NEN Research Products (Les Ulis, France). The nifedipine perfusate was obtained through 2-fold dilution of Adalate (0.2 mg/0.2 ml; Bayer pharma, Puteaux, France) in distilled water. The final nifedipine solution contained 0.1 mg of nifedipine, 150 mg of 95% ethanol, 150 mg of PEG 400 and 2 ml. of distilled water. HM and PYY were from Neosystem (Strasbourg, France), and TTX was from Sigma Chemical Co. (St. Louis, MO). All other chemicals were of analytical grade. For perfusion experiments, Krebs-Ringer-Tris buffer (pH 5.5) was prepared with 118 mM NaCl, 2.5 mM Tris, 4.7 mM KCl, 2.5 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄ and 9.2 mM citric acid. Isotonic Sorensen buffer (pH 7) contained 26.1 mM KH₂PO₄, 40.5 mM Na₂HPO₄ and 122 mM glucose.

Animals

Male Sprague-Dawley rats (Iffa Credo, L'Abresle, France) weighing 280 to 320 g were fasted for 20 hr before the experiment, with free access to water. The rats were anesthetized with urethane (1.5 g/kg i.p.) 30 min before surgery.

Absorption Studies

Intestinal absorption was studied with an in situ perfusion technique (Schanker et al., 1958). Briefly, the small intestine was exposed through a midline abdominal incision. The proximal end of the selected intestinal segment was cannulated with an Insyte W22g cannula (Becton Dickinson, Meylan, France), and the distal end was cannulated with a polyethylene tube. It was then washed with saline for 30 min. In all experiments, the biliary duct was ligated to avoid increased cefixime absorption through enterohepatic circulation (Yamaoka et al., 1990). Rectal temperature was maintained at 37 ± 1°C with a warming blanket (Harvard Apparatus, Boston, MA). All solutions were maintained at 37°C except for the nifedipine solution, which was kept at room temperature. A peristaltic pump (Gilson Minipuls 2, Villiers Le Bel, France) was used at a flow rate of 0.6 ml/min. Perfusion rates were 0.3 mg/min with 0.5 mg/ml cefixime and 0.6 mg/min with 1 mg/ml cefixime.

Single-perfusion technique. To localize the cefixime absorption site, the antibiotic was perfused throughout the duodenojejunum (20 cm) or jejunoileum segment (20 cm). In subsequent experiments, 40-cm duodenojejunum segments were chosen. After an initial washing period, cefixime was diluted in Krebs-Ringer/Tris buffer (pH 5.5) or isotonic Sorensen buffer (pH 7.0) and perfused throughout the intestine for 1 hr.

Recirculation technique. After an initial washing period, 0.05 mg/ml nifedipine or a drug-free solution was perfused for 15 min throughout the duodenojejunum segment (20 cm). Then, a 0.4 mg/ml salicylic acid solution containing 1.85 KBq/ml [7-¹⁴C]salicylic acid in isotonic buffer (Schanker et al., 1958) was perfused, first in single-pass mode (10 min) and then in recirculating mode (50 min). During the recirculation period, aliquots (0.1 ml) were sampled every 10 min.

Study of neural regulation and ionic movements. For intravenous infusion, a catheter (Microflex 0.4 mm/G27 Vygon, Ecouen, France) was connected to an electric syringe. The following solutions were administered: (1) an intravenous bolus of HM (6.7 mg/kg) followed by intravenous perfusion (6.7 mg/kg/hr), started 30 min before and stopped at the end of the cefixime intestinal perfusion; (2) an intravenous bolus of TTX (5 mg/kg), followed by an intravenous saline infusion, started 30 min before the beginning of the cefixime intestinal perfusion and stopped at the end of the cefixime intestinal perfusion; (3) an intravenous PYY infusion of 240 pmol/kg/hr, started 15 min before and stopped at the end of cefixime intestinal perfusion or (4) an intravenous infusion of a control saline solution, started 30 min before the beginning of the cefixime intestinal perfusion and stopped at the end of the cefixime intestinal perfusion. The same experiments were conducted with or without a previous nifedipine (0.05 mg/ml) intestinal perfusion for 15 min started before the cefixime perfusion. Seven rats were used for each experiment.

Regional Blood Flows

In the cefixime and cefixime-plus-nifedipine groups, regional blood flows were determined in 6 and 7 rats, respectively, using a microsphere technique (Hadengue et al., 1988). ¹¹³Sn- or ¹⁴¹Ce-labeled microspheres (15 ± 3 µm; specific activity, 10 mCi/g; New England Nuclear, Boston, MA) were suspended in 0.9% NaCl containing one drop of 0.05% Tween 80 and sonicated for 10 min. A precounted aliquot of 80 µl (60,000 microspheres) was aspirated into the syringe and then flushed over 45 sec into the left ventricle with 700 µl of saline. Cardiac output was calculated according to the reference blood sample method.

Simultaneous with the microsphere injection, a reference blood sample was drawn from the femoral artery at 0.8 ml/min for 75 sec. The arterial catheter was connected to a multichannel recorder for arterial pressure and heart rate monitoring. Hemodynamic parameters were measured at the beginning of (T0) and 1 hr after (T60) cefixime perfusion throughout the small intestine. The animals were killed after 60 min, and individual organs were dissected. The radioactivity of each organ and of the reference blood sample was determined using a scintillation counter (Gamma 4000, Intertechnique, Plaisir, France).

Blood samples (600 µl) were collected 15, 30, 45 and 60 min after the beginning of the cefixime perfusion. Arterial blood samples were collected via a septum-equipped catheter inserted in the carotid artery (Insyte W22g, Becton Dickinson). Portal blood samples were collected in two ways: either (1) via a polyethylene catheter (0.3 mm external diameter) inserted directly into the portal vein and fixed with cyanoacrylate glue or (2) according to the method of Hadengue et al. (1988) (i.e., via a polyethylene catheter inserted at the junction of two small ileal veins and advanced by visual observation up to the confluence of the superior mesenteric and splenic veins). Each sample was replaced with 300 µl of 0.9% NaCl. Serum was separated by centrifugation and stored at 20°C.

Analytical Procedures

Cefixime concentrations were determined by reverse-phase high performance liquid chromatography with UV detection. After a deproteinization step using trichloracetic acid, the supernatant was injected into the chromatography system. The column was an Ultrabase C18 (5-µm, 4.6 × 150 mm). The mobile phase was a mixture of acetonitrile and water (13:100 v/v) containing 1% perchloric acid and 1% triethylamine. The flow rate was 1.7 ml/min. The detection wavelength was set at 280 nm (Shimatzu SPD6A, Touzart et Matignon, France).

[7-¹⁴C]Salicylic acid concentrations were determined by measuring radioactivity: 100 µl of sample was transferred into vials containing 10 ml of scintillation fluid (Ultima Gold, Packard, Rungis, France) and directly counted by scintillation (beta counter, Intertechnique, Montigny le Bretonneux, France).

Statistical Analysis

Results are presented as mean ± S.E.M. Student's t test for unpaired data was used to compare treatment groups. For the intestinal absorption site study, the Mann-Whitney U test was used. Differences were considered significant at a level of P < .05.

	Results

Top Abstract Introduction Methods Results Discussion References

Cefixime Uptake

Mean cefixime concentrations in arterial blood after a 0.5 mg/ml perfusion into the duodenojejunum or jejunoileum segments are shown in table 1. After 30 min, cefixime absorption was significantly better in the duodenojejunum segment than in the jejunoileum segment, promoting the use of the former in subsequent experiments.

Cefixime portal concentrations increased linearly with time, regardless of the perfusate concentration (0.5 or 1 mg/ml) (table 2). However, achieved cefixime concentrations did not differ significantly between the groups receiving the two concentrations: at 60 min, mean cefixime concentrations were 13.6 mg/liter in the 0.5 mg/ml perfusion group compared with 14.4 mg/liter in the 1.0 mg/ml group.

Cefixime absorption was also compared at perfusate pH 5.5 and 7.0. As shown in figure 1, at 30 min, mean cefixime concentrations in arterial blood were significantly higher when cefixime was perfused at pH 5.5 compared with pH 7.0 (P < .05). At 60 min, cefixime concentrations in the pH 5.5 group were twice those in the pH 7.0 group (9.4 ± 1.2 vs. 4.2 ± 0.7 mg/liter, P < .01).

View larger version (28K): [in this window] [in a new window]   Fig. 1.   Mean ± S.E.M. (n = 5) arterial blood cefixime concentrations (mg/liter) during a 1-hr single-pass mode intestinal perfusion of cefixime (0.5 mg/ml) at pH 5.5 and 7.0. Cefixime was diluted in Krebs-Ringer/Tris buffer at pH 5.5 or in isotonic Sorensen buffer, at pH 7.0, containing glucose. *P < .05.

Interaction Studies

Effects of PYY on cefixime absorption. PYY was used to study the influence of ionic secretion and motility decrease on cefixime absorption in the small intestine.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effect of a PYY continuous intravenous infusion (240 pmol/kg/hr) on mean ± S.E.M. (n = 7) cefixime arterial blood concentrations during a 1-hr single-pass mode intestinal perfusion of cefixime (0.5 mg/ml) at pH 5.5. PYY intravenous perfusion was started during the intestine washing period, 15 min before start of the cefixime perfusion.

Effect of nifedipine on salicylic acid absorption. Salicylic acid was used to study the effect of nifedipine on the passive diffusion. Over 50 min, residual percentages of salicylate fell from 85.1 ± 5.6% to 57.1 ± 2.8% with a previous nifedipine perfusion compared with a decrease from 87.4 ± 1.4% to 52.8 ± 1.6% without nifedipine (table 3).

Effect of nifedipine on cefixime absorption. Cefixime concentrations in arterial (table 4) or portal blood (fig. 3) were compared between two groups who received 1 hr of 0.5 mg/ml cefixime perfusion with or without previous perfusion of 0.05 mg/ml nifedipine for 15 min. Cefixime absorption was markedly enhanced by the 15-min nifedipine perfusion. At 45 min, cefixime concentrations increased by 60% in arterial blood and by 76% in portal blood (P < .05). Cefixime concentrations in arterial blood were always lower than those measured in portal blood. Because arterial concentrations are dependent on distribution processes and portal concentrations are not, the latter more closely reflects the absorption phase. The hyperosmolarity of the nifedipine solution did not influence cefixime absorption, as preliminary experiments showed a similar time course of cefixime concentrations with or without a previous perfusion of an hyperosmolar solution for 15 min (data not shown). Thus, the absorption of cefixime appears to be the specific target of nifedipine.

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Effect of a nifedipine perfusion before 1-hr cefixime intestinal perfusion (0.5 mg/ml, pH 5.5) on mean cefixime portal blood concentration (mean ± S.E.M., n = 12). When applicable, nifedipine was perfused for 15 min before start of the cefixime perfusion. *P < .05.

1

1

View larger version (42K): [in this window] [in a new window]   Fig. 4.   Effect of a previous nifedipine (N) perfusion with or without concomitant intravenous administration of HM or TTX on cefixime intestinal absorption. Saline was perfused in the duodenojejunum segment for 15 min; then, 0.1 mM nifedipine or a drug-free solution (control group) was perfused for 15 min. When applicable, at the beginning of the saline perfusion, HM was administered as an intravenous bolus (6.7 mg/kg) followed by a constant rate infusion (6.7 mg/kg/hr). Similarly, TTX was first injected as an intravenous bolus (5 µg/kg) followed by a saline infusion. Then, a 0.5 mg/ml cefixime intestinal perfusion was started. *P < .05.

	Discussion

Top Abstract Introduction Methods Results Discussion References

Our results clearly indicate that in rats, the absorption of cefixime is better in the duodenojejunum segment, is saturable and is improved at low intraluminal pH. Moreover, cefixime absorption is enhanced by nifedipine via a blood flow-independent mechanism. The action of nifedipine appears to be an indirect one, involving a neural regulation. The lower absorption rate of cefixime from the ileum compared with the duodenum is in agreement with results obtained by Maekawa et al. (1977), who showed that after 1 hr, the disappearance of another oral cephalosporin, cephalexin, was much higher from the proximal small intestine (56%) than from the distal small intestine (15.6%). The reason for such a discrepancy between different portions of the digestive tract is unknown; it could be due to the more basic environment of the ileum, to a lower number of carrier-mediated transport systems or to a reduced number of enterocytes per unit of surface area.

The active transport of cefixime and of other -lactams via the same transporter as that of dipeptides has been widely described (Amidon and Lee, 1994; Dantzig et al., 1994; Tsuji et al., 1987a). However, the existence of multiple transporters has been suggested for the uptake of dipeptides and oral cephalosporins (Kimura et al., 1978; Kramer et al., 1993). Transporter-substrate interactions probably involve histidine residues on intestinal membrane proteins (Kato et al., 1989). In rabbits (Inui et al., 1988) and rats (Tsuji et al., 1987a, 1987b), cefixime uptake follows a mixed-type kinetic pattern, which involves both a nonsaturable process (passive diffusion) and dose-dependent saturable Michaelis-Menten kinetics (active transport) (Sugawara et al., 1990). Indeed, the inhibition kinetics of cefixime uptake by glycyl-L-proline and cyclacillin are consistent with a competitive type of inhibition, as shown by Tsuji et al. (1987a) in everted rat jejunums. Their model keeps the physiological characteristics of the intestinal mucosa; so does our model, which is independent of intestinal blood flow and motility. Our results were confirmed in vitro in cultured Caco-2 cells (Harcouet, 1995), in which cefixime absorption was inhibited, at acid pH, by glycine-L-proline and cefadroxil.

Cefixime transport is influenced by the pH of the medium; we show its improvement at low intraluminal pH 5.5. Inui et al. (1988) suggested that cefixime does not interact with the transport system of -amino -lactam antibiotics at neutral pH but is transported via a different peptide carrier in an acidic pH region.

We found that cefixime was better absorbed at pH 5.5 than at pH 7.0, confirming in vitro observations obtained with intestine brush border membrane vesicles from rabbits (Inui et al., 1988), rats (Sugawara et al., 1991; Tsuji et al., 1987b) and humans (Sugawara et al., 1992). Experiments carried out with an everted gut sac method (Tsuji et al., 1986, 1987a) were also consistent with these results. All these studies concluded that the acidic environment of the mucosal surface of the small intestine (Lucas, 1983; Lucas et al., 1985) maximizes the uptake of cefixime. Low intestinal pH appears to be maintained by a mucus coating (Shiau et al., 1985) or by hydrogen ion secretions (Murer et al., 1976). In the proximal and middle jejunum, hydrogen ions would be controlled by a Na⁺/H⁺ antiporter located on the luminal surface of the enterocytes. The Na⁺/K⁺-ATPase of the basolateral cell membrane would pump the Na⁺ out of the cell (Ganapathy and Leibach, 1985).

Nifedipine could increase cefixime absorption via several mechanisms: (1) by slowing the fluid movement through the gut, thus allowing more time for absorption to occur; (2) by increasing intestinal blood flow, thus favoring passive absorption (provided that transmembrane passage is not a limiting factor); (3) by increasing, directly or undirectly, the rate of uptake by epithelial carriers and (4) by causing intestinal lesions secondary to the hyperosmolarity of its solution. The last hypothesis appears to be very unlikely on the basis of the experiments of Mogard and Nylander (1982), who studied the effect of intraluminal hyperosmolarity on absorption by using a recirculation intestinal perfusion technique in the rat. They showed that passive absorption of iodine and active absorption of selenomethionine were not modified by intraluminal hyperosmolarity (830 mOsmol/kg water).

Calcium antagonists inhibit transmembrane calcium influx via calcium channels expressed in various cell types, including cardiac and vascular smooth muscle cells. They might also have some actions on smooth muscles of the gastrointestinal tract, such as those involved in motility or local microcirculation (De Ponti et al., 1993). Although nifedipine does not alter jejunum motility in humans (Santander et al., 1988), it reduces spike and mechanical activities in dogs (De Ponti et al., 1989). In rats, nifedipine also decreased intestinal motility by spontaneously inhibiting migratory motor complexes (Thollander et al., 1993). In our experimental conditions, the intestinal mucosa was always in contact with a constant concentration of cefixime and the perfusion rate therefore was sufficiently high to rule out the influence of an increased residence time in the intraluminal site. Indeed, we showed that the absorption rate of salicylic acid was not modified by nifedipine.

Furthermore, PYY, which delays the propagation of migratory motor complexes and inhibits spike activity in rat jejunum (Sheikh, 1991), had no effect on cefixime absorption with or without nifedipine. Therefore, cefixime-absorption and nifedipine-promoting effect on the latter appears to occur independent of intestinal motility. In vitro, calcium antagonists relax the mesenteric-portal vein (Sutter, 1990), but in vivo, their hemodynamic effects are more complex. In humans, nifedipine is an effective vasodilator, especially of the splanchnic vascular bed, increasing blood flow in splanchnic territory (Gasic et al., 1987) and hepatic and portal veins (Reiss et al., 1991). Other calcium antagonists do not influence splanchnic hemodynamics (Agner et al., 1984). Nifedipine has no effect on the suprahepatic pressure gradient during portal and hypertension, in animals or humans (Combis and Vinel, 1991). We also found that nifedipine did not significantly alter small bowel blood flow. From the beginning of the cefixime perfusion, our values of organ and tributary vessel blood flow were lower than those previously reported by Hadengue et al. (1988). The use of different experimental conditions, such as anesthesia with urethane or the fact that the abdomen was kept open during the washing period, may account for these differences. At the end of the experiment, when cefixime blood concentrations were markedly different between treatment groups, blood flow was similar in the group that received a previous nifedipine perfusion and in the control group. Therefore, a hemodynamic effect of nifedipine was not responsible for its absorption-promoting effects. Nifedipine did not increase the cefixime concentration gradient between intravascular and extravascular compartments, which might have resulted in an increased passive diffusion of the antibiotic (Winne, 1980). This is also supported by the lack of any promoting effect of nifedipine on salicylic acid absorption, which has been shown to occur via passive diffusion, in an analogous in vivo model and at similar concentrations to those used in our experiments (Schanker et al., 1958).

Wetsphal et al. (1990) showed that in humans, the absorption kinetics of amoxicillin were enhanced by nifedipine, a result of its effect on specific electrical movements. Similarly, the effect of nifedipine on cefixime transport could be explained in terms of electrical movements. By decreasing calcium uptake by the enterocytes (Hurt et al., 1993), calcium channel blockers might Westphal (Westphal et al., 1995) stimulate the apical Na⁺/H⁺ antiport (Donowitz, 1983), thereby setting up a proton gradient favorable to cefixime absorption by the intestinal H⁺/dipeptide cotransport system. Globally, we did not observe any effect of PYY, which is known not only to inhibit intestinal motility but also to inhibit proximal small intestinal hydroelectrolytic secretions in rats. Therefore, cefixime absorption is not modified by a decrease in hydroelectric secretions.

Additional experiments are, however, necessary to specifically address the role of electrical movements in the nifedipine-cefixime interaction, particularly the coadministration of nifedipine with amiloride, which is known to block the effect of the Na⁺/H⁺ antiporter. Calcium antagonists, and particularly verapamil and nifedipine, inhibit the active efflux of many drugs, with a direct effect on P-glycoprotein within the cell apical membrane. P-glycoprotein expression is responsible for the multidrug resistance phenotype of tumoral cells and has been identified in human enterocytes. It works like an ATP-dependent pump, expelling drugs from the cells, probably as a detoxification mechanism or as a way to intestinal absorption. A verapamil-sensitive (P-glycoprotein) exists on the apical membrane of Caco-2 cells and is responsible for the efflux of vinblastin, vincristine and ciprofloxacin (Griffith et al., 1993; Hunter et al., 1993) This P-glycoprotein, which is pHi and pHe independent (Altenberg et al., 1993) probably is not involved in the mechanism of cefixime absorption enhancement in the presence of nifedipine because in Caco-2 cells, nifedipine, verapamil and diltiazem did not modify cefixime epithelial transport at neutral and acid pH levels (Harcouët, 1995).

The mechanism of the interaction appears quite different from that suggested by Westphal (Westphal et al., 1990). Rather, it seems characterized by an indirect neural effect on the intestinal epithelium of the entire system. Indeed, we showed that the interaction was abolished by a neural blockage with either HM or TTX. Intrinsic and extrinsic innervation play an important role in the regulation of electrolyte transport throughout the small intestine (Cooke and Reddix, 1994).

Numerous factors of luminal, neural, endocrine or circulatory origin may modulate the level of expression and the function of the transport within epithelial cells; the details of their complex interactions are still poorly understood. Calcium is a key regulatory element of many cellular functions, such as contraction, differentiation and secretion. Calcium antagonistssensitive voltage-dependent channelswere identified in many cell types. Neural circuits also show a great dependence on calcium. In neurons, calcium fluxes play an important role in the regulation of neuronal excitability and the release of neurotransmitters. Many receptors of neurotransmitters and neuromodulators have been identified on intestinal epithelial cells, and a neural regulation of the dipeptide transport system has recently been suggested by Brandsch et al. (1994). These authors have shown that PKC inhibits the transport of the dipeptide glycylsarcosine across the intestine. Furthermore, PKC, which is stimulated by several gut hormones, also inhibits the Na⁺/H⁺ antiport. Therefore, the effect of nifedipine would not be directly mediated via calcium channels of enterocyte membranes but would result from indirect mechanisms involving, at least in part, a neural regulation. One hypothesis would be that nifedipine modulates the activity of peripheric or central autonomic neurons (intraparietal myenteric and submucuous plexus). The nervous stimulus to the enterocytes would then be modified, and one could envision that one or more neurotransmitters would be delivered in different amounts to the enterocytes. A modulation of PKC could be occurring via such a mechanism. One would have to show that in vitro PKC modulation can modulate cefixime absorption.

In conclusion, nifedipine significantly increased cefixime intestinal absorption without modifying intestinal blood flow or intestinal motility. Nifedipine could affect cefixime absorption, directly or indirectly, via a neural mechanism by increasing the activity of carrier-mediated transport systems within the intestinal epithelium. Further study of dihydropyridine--lactam interactions may help to better understand some of the regulatory processes involved in the control of the intestinal absorption of antibiotics.