Mathematical modelling of in situ and in vitro efflux of ciprofloxacin and grepafloxacin.

The efflux process due to p-glycoprotein-like mechanisms of ciprofloxacin (CIP) and grepafloxacin (GRX) has been studied "in situ" in rats and "in vitro" in Caco-2 cells. The results were modelled by a curve fitting procedure which allowed the characterization of the passive (Pd) and carrier mediated parameters (Vm and Km) from the raw data without initial velocities estimation. CIP absorption in rat was characterized as a passive diffusion at the assayed concentrations. Although the involvement of an efflux transporter cannot be ruled out, its relevance in the transport of the fluoroquinolone is negligible. In GRX absorption, an efflux process is implicated and it is detected in both absorption models. GRX permeability depends on the intestinal segment, reflecting the previously reported different expression level of the efflux transporters along the gut in rat. A first attempt to correlate the "in vitro" and the "in situ" data has been done. The mathematical model has been constructed using very simplistic assumptions and it will require further refinement but, nevertheless, the results are promising and demonstrate that a good modelling approach helps to identify the system critical parameters and how the system behaviour change when the parameters are modified as it happens when we move from the "in vitro" to the "in situ" level. Predicted versus experimental permeability values show a good correlation, demonstrating that the relevance of the secretion process "in situ" in rat can be predicted from the "in vitro" cell results.

Correlation between epithelial cell permeability of cephalexin and expression of intestinal oligopeptide transporter.

The proton-coupled oligopeptide transporter (PEPT1) has been shown to mediate mucosal cell transport of di- and tripeptide, and some peptidomimetic drugs. In this study, we determined the correlation between PEPT1 protein expression and the permeability of cephalexin, a substrate of PEPT1, in human PEPT1 (hPEPT1)-overexpressed Caco-2 cells (Caco-2/hPEPT1 cells) and rat jejunum. Caco-2/hPEPT1 cells with various levels of hPEPT1 expression were established by an adenoviral transfection system. The effective intestinal permeability (P(eff)) in rat jejunum was evaluated using a single pass in situ perfusion method. The level of PEPT1 in Caco-2/hPEPT1 cells and rat intestinal mucosal samples was quantitated by densitometry after immunoblotting and enhanced chemiluminescence detection. In Caco-2/hPEPT1 cells, an excellent correlation was observed between cephalexin uptake and hPEPT1 expression (R2 = 0.96, P < 0.005). This demonstrates that cephalexin uptake is directly proportional to hPEPT1 expression. In the rat perfusion study, the mean P(eff) +/- S.D. (n = 15) of cephalexin was 3.89 +/- 1.63 x 10(-5) cm/s. A very significant correlation between PEPT1 expression and cephalexin permeability with an R2 = 0.63 (P < 0.001) was observed. This indicates that the variation in PEPT1 expression is one of the major factors accounting for variable intestinal cephalexin absorption. To our knowledge, this is the most direct evidence that variation of PEPT1 expression is correlated with absorption permeability variation of peptide-like compounds in vitro and in vivo.

Intrinsic absolute bioavailability prediction in rats based on in situ absorption rate constants and/or in vitro partition coefficients: 6-fluoroquinolones.

A preliminary study attempting to predict the intrinsic absolute bioavailability of a group of antibacterial 6-fluoroquinolones-including true and imperfect homologues as well as heterologues-was carried out. The intrinsic absolute bioavailability of the test compounds, F, was assessed on permanently cannulated conscious rats by comparing the trapezoidal normalized areas under the plasma concentration-time curves obtained by intravenous and oral routes (n = 8-12). The high-performance liquid chromatography analytical methods used for plasma samples are described. Prediction of the absolute bioavailability of the compounds was based on their intrinsic rat gut in situ absorption rate constant, k(a). The working equation was: where T represents the mean absorbing time. A T value of 0.93 (+/-0.06) h provides the best correlation between predicted and experimentally obtained bioavailabilities (F' and F, respectively) when k(a) values are used (slope a = 1.10; intercept b = -0.05; r = 0.991). The k(a) values can also be expressed in function of the in vitro partition coefficients, P, between n-octanol and a phosphate buffer. In this case, theoretical k(a) values can be determined with the parameters of a standard k(a)/P correlation previously established for a group of model compounds. When P values are taken instead of k(a) values, reliable bioavailability predictions can also be made. These and other relevant features of the method are discussed.

Pharmacokinetics, bioavailability and absorption of flumequine in the rat.

The study demonstrates that the oral extent of bioavailability of flumequine in the rat, relative to the intravenous injection, is complete (0.94 +/- 0.04) and not significantly different from that found by the intraduodenal route (0.95 +/- 0.04). The rate of oral bioavailability, however, is slow (ka = 1.20 +/- 0.07 h-1; Tmax = 2.0 h), but enough to maintain plasma levels above the minimal inhibitory concentration of the most common pathogens for an extended period of time (about 10 h). The reason for the oral absorption slowness could be a slow gastric emptying, an adsorption to the gastric mucosae, a precipitation in the gastric medium or any other feature concerning the stomach as the intraduodenal administration is very quick (kid = 38.1 +/- 4.7 h-1; Tmax = 0.05 h). A possible precipitation of flumequine cannot be discarded as the solubility of flumequine is very low in the pH range of 3 to 6 (mean pH values for rat stomach and rat intestine, respectively; T.T. Kararli, Biopharm. Drug Dispos. 16 (1995) 351-380). Flumequine was shown to be not substantially excreted in bile (2-3% of the dose). Surprisingly, plasma levels and AUC values found for animals with interrupted bile flow always surpass those found for animals with enterohepatic circulation. This could be due to experimental model features, which might bias plasmatic flumequine concentrations if the homeostatic equilibrium of the animal is not completely restored due to the volume reduction induced by biliary extraction.