pH-dependent bidirectional transport of weakly basic drugs across Caco-2 monolayers: implications for drug-drug interactions

FIRST-PASS HYDROLYSIS OF A PROPRANOLOL ESTER DERIVATIVE IN RAT SMALL INTESTINE

To evaluate the first-pass hydrolysis of O-isovaleryl-propranolol (isovaleryl-PL), which was used as a model ester-compound, rat intestinal jejunum and blood vessels were perfused simultaneously. The membrane permeability of isovaleryl-PL was greater than that of PL because it was more lipophilic. Isovaleryl-PL was almost completely hydrolyzed to PL and isovaleric acid (IVA) in epithelial cells at a rate limited by its uptake. Based on pH partitioning, PL and IVA were transported into both vascular (pH 7.4) and luminal sides (pH 6.5). Therefore, when isovaleryl-PL was perfused into the jejunal lumen, more than 90% permeated into the blood vessel as PL. In addition, PL appeared in the lumen at a rate 6-fold greater than that in blood vessels. When isovaleryl-PL was perfused, its disappearance (50.5+/-1.95 nmol/min) was the sum of the absorption and secretion rates of PL. In contrast, IVA was transported into blood vessels rather than the jejunal lumen. In addition, the calculated degradation clearance from in vitro hydrolysis (Km 13.7+/-1.71 microM, Vmax 29.1+/-3.81 nmol/min/mg protein) was 3.42 ml/min/10 cm jejunum, which was 24-fold greater than the observed degradation clearance (CLdeg) (0.14+/-0.02 ml/min/10 cm jejunum). These findings indicate that in addition to the liver, the intestine markedly contributes to first-pass hydrolysis.

Dose dependency in the oral bioavailability of an organic cation model, tributylmethyl ammonium (TBuMA), in rats: Association with the saturation of efflux by the P-gp system on the apical membrane of the intestinal epithelium

The oral bioavailability of tributylmethyl ammonium (TBuMA), an organic cation (OC), exhibited a dose-dependency (i.e., 17, 27, and 35% at doses of 0.4, 4, or 12 micromol/kg, respectively) in the rat. Relevant mechanisms were investigated in the present study by estimating the mucosal to serosal (m-s) and serosal to mucosal (s-m) transport of TBuMA across the rat ileum in an Ussing chamber experiment. The m-s permeability rapidly increased with TBuMA concentration in the mucosal side, and then becoming constant at high TBuMA concentrations. Various studies, including temperature- and potential-dependency and inhibition experiments, revealed that carrier-mediated transport mechanisms (most likely OCT1, OCT3, and P-gp) are involved in the s-m transport of TBuMA, and the saturation of the transport at higher concentrations is responsible for the concentration-dependency in the m-s permeability or dose-dependency of the bioavailability of TBuMA. A nonlinear regression of the m-s transport, based on the assumption of a mixed process of linear diffusion and saturable efflux, exhibited a clearance (CLlinear) of 0.343 microL/min/cm2 for the passive diffusion, and an apparent Km of 241 microM for the saturable process. The Km value is consistent with the concentration range in the intestine which is expected to be achieved after the oral dosing of TBuMA at a dose of 0.4 micromol/kg (i.e., 68 approximately 185 microM). Interestingly, the m-s transport of TBuMA was increased by the presence of P-gp substrates or inhibitors in the mucosal side, but not by the mucosal presence of OCT substrates or inhibitors, suggesting that only efflux transport systems on the apical membrane (e.g., P-gp), but not those on the serosal membrane (e.g., OCT1 and OCT3), of the intestinal epithelial cells, are involved in the dose-dependency or concentration dependency. A similar relationship seems likely for drugs that are substrates of efflux transporters on the apical membrane of the intestinal epithelium.

pH-dependent functional activity of P-glycoprotein in limiting intestinal absorption of protic drugs: Kinetic analysis of quinidine efflux in situ

The purpose of this investigation is to evaluate the quantitative contribution of pH-dependent passive permeability on the functional activity of P-glycoprotein (P-gp) in limiting intestinal absorption of weakly basic drugs, in order to include this effect in prediction models. pH-dependent octanol/buffer partition coefficient, artificial membrane permeability and in situ rat intestinal permeability of quinidine were determined in the physiological pH range of gastrointestinal tract. In situ permeability, as a function of luminal pH, was also determined in the presence of P-gp inhibitor, verapamil (500 microM). Octanol/buffer partition coefficient, transport across artificial membrane, and rat in situ permeability showed high pH-dependency. Absorption quotient (AQ), calculated from in situ permeability to express the functional activity of P-gp, declined with increase in luminal pH or increase in luminal quinidine concentration because of the increased passive permeability or saturation of P-gp. AQ was 0.57 +/- 0.02 and 0.41 +/- 0.05, while passive permeability was 0.32 +/- 0.01 x 10(-4) cm/sec and 0.43 +/- 0.02 x 10(-4) cm/sec, in jejunum and ileum, respectively, at pH 7.4. Further, apparent Michaelis-Menten constants (K(M), J(P-gp,max)) for the quinidine efflux in jejunum indicated that efflux activity was more at luminal pH 4.5 over pH 7.4. K(M) values for jejunum quinidine efflux at pH 4.5 and pH 7.4 were determined to be 77.63 +/- 10.90 and 22.86 +/- 5.22 microM, with J(P-gp,max) values of 1.47 +/- 0.08 and 0.62 +/- 0.04 nM/cm2/sec, respectively. AQ vs passive permeability showed significant relationship indicating dependency of P-gp-mediated efflux on pH-dependent passive permeability, which is dictated by ionization status for a protic or ampholytic drug. In conclusion, an orally administered drug is absorbed from various segments of intestine, which inherit difference in luminal pH, transcellular permeability and P-gp expression. In situ data suggests that pH-dependency and regional variability in passive permeability of protic substrates significantly influence their P-gp-mediated efflux and may have implications on predictions of the in vivo drug absorption.

Absorption properties and P-glycoprotein activity of modified Caco-2 cell lines

Caco-2 cell line is extensively used as an in vitro model in studying small intestinal absorption but it lacks proper expression of efflux pumps and cytochrome P450 enzymes that are involved in absorption and first pass metabolism of drugs. We created two novel Caco-2 cell lines expressing orphan nuclear receptors pregnane X receptor and constitutive androstane receptor that regulate many genes involved in xenobiotic metabolism. We conducted a systematic study on expression of some metabolic genes, P-glycoprotein activity and absorption properties of several drugs with these new cell lines and previously described modified Caco-2 cell lines (MDR1 transfection, vincristine treatment and 1alpha,25-dihydroxyvitamin D3 treatment). A short culture time medium was also included in the study. Most modified cell lines formed tight differentiated monolayers. MDR1, CYP2C9 and CYP3A4 genes were upregulated in some cell lines. Elevated P-glycoprotein activities were observed by calcein-AM uptake experiments but this did not affect significantly the permeability of selected P-glycoprotein substrates. Some cell lines had similar passive and active permeability properties to Caco/WT cells while in few cell lines these were altered. Passive transcellular permeability was modestly elevated in all modified cell lines. In addition, several compounds showed pH-dependent permeability properties.

Regional intestinal absorption and biliary excretion of fluvastatin in the rat: possible involvement of mrp2

The first purpose of this study was to investigate the in vivo absorption, biliary secretion, and first-pass effect of fluvastatin following regional intestinal dosing in the rat. We also examined the membrane transport mechanisms and made in silico predictions of the relative importance of various intestinal regions to the human absorption of fluvastatin. Fluvastatin was administered intravenously (2, 10, and 20 micromol/kg) and into the duodenum (1.46, 2.92, 7.32, and 14.6 micromol/kg), jejunum (14.6 micromol/kg), ileum (1.46 and 14.6 mciromol/kg), and colon (1.46 and 14.6 micromol/kg) as a solution to conscious rats. In a separate group of rats, bile was collected after an i.v. dose of fluvastatin (2 micromol/kg). In the Caco-2 model the bidirectional transport of fluvastatin (16 microM) was investigated with and without various efflux inhibitors (verapamil, vinblastine, probenecid, and indomethacin, 160 microM). The human in vivo absorption of fluvastatin from an oral immediate release tablet and that from an oral extended release tablet (both 40 mg) were simulated in GastroPlus. Neither the dose nor the intestinal region influenced the bioavailability of fluvastatin significantly. The rate of absorption was, however, affected by both the dose and the site of administration; duodenum = jejunum > colon > ileum, and higher following the high dose. Increasing the i.v. dose from 2 to 20 micromol/kg decreased the clearance (26 +/- 3 to 12 +/- 1 mL/min/kg), the hepatic extraction (66 +/- 8 to 30 +/- 2%), and the volume of distribution (7.3 +/- 0.3 to 2.1 +/- 0.7 L/kg) for fluvastatin (p < 0.05). Neither bile cannulation nor bile sampling affected the pharmacokinetics. Fluvastatin was secreted into the bile, probably by active transport. The in vitro permeability for fluvastatin was high (>10 x 10(-6) cm/s). Indomethacin, but not the other inhibitors, affected the transport in both directions suggesting mrp2 to be involved. In silico, 93% of the dose was absorbed from the small intestine and 6% from the colon when given as an immediate release formulation. The corresponding values for an extended release formulation were 21% and 74%, respectively. In conclusion, fluvastatin exhibits dose-dependent pharmacokinetics in the rat. The rate of absorption (Cmax, Tmax, and Cmax/AUC(lqc)) from the intestinal tract is both region and dose-dependent in the rat. This may be due to the involvement of mrp2 in the intestine and/or in the liver. These absorption properties have to be considered in the development of an extended release formulation of fluvastatin.

Cell culture-based models for intestinal permeability: a critique

The model systems that are currently used to determine the intestinal permeability characteristics of discovery compounds often represent a combination of high-throughout, but less predictive, in silico and in vitro models and low-throughput, but more predictive, in vivo models. Cell-based permeability models have been integrated into the discovery paradigm for some time and represent the "method of choice" across the industry. Here, in addition to an objective analysis of the utility of cell culture models for permeability screening, anticipated future trends in the field of cell culture models are discussed.

pH-Dependent passive and active transport of acidic drugs across Caco-2 cell monolayers

The aim of this study was to investigate pH-dependent passive and active transport of acidic drugs across Caco-2 cells. Therefore, the bidirectional pH-dependent transport of two acidic drugs, indomethacin and salicylic acid, across Caco-2 cells was studied in the physiological pH range of the gastrointestinal tract. The transport of both drugs decreased with increased pH, as expected from the pH-partition hypothesis. Net absorption occurred when the basolateral pH exceeded the apical pH. Concentration dependence and transporter inhibition studies indicated passive transport for indomethacin and a mixture of pH-dependent passive and active influx for salicylic acid. Unexpectedly, active and passive drug transport results were indistinguishable in temperature dependency studies. The transport of salicylic acid (apical pH 5.0; basolateral pH 7.4) was partly blocked by inhibitors of the proton-dependent transporters MCT1 (SLC16A1) and OATP-B (SLC21A9, SLCO2B1). This study shows that the asymmetry in bidirectional transport of acidic drugs is affected by both passive and active components in the presence of pH gradients across Caco-2 cells. Thus, combined studies of concentration-dependency and transport-inhibition are preferred when acidic drug transport is studied in a pH gradient. The findings of this in vitro study can be extrapolated to in vivo situations involving an acidic microclimate.

pH-dependent functional activity of P-glycoprotein in limiting intestinal absorption of protic drugs 1. Simultaneous determination of quinidine and permeability markers in rat in situ perfusion samples

A simple, specific and sensitive reversed-phase high performance liquid chromatographic (RP-HPLC) method with UV absorbance detection was developed and validated for simultaneous determination of quinidine, verapamil and passive permeability markers, in samples obtained from rat intestinal in situ single-pass perfusion studies. Chromatography was carried out on C18 column with mobile phase comprising of acetate buffer (pH 5.0) and methanol in the ratio of 40:60 (v/v) pumped at a flow rate of 0.6 ml/min and UV detection was employed at 230 and 275 nm. The average retention times for hydrochlorthiazide, frusemide, quinidine, propranolol, and verapamil were 4.9, 5.8, 6.9, 8.9 and 11.3 min, respectively. The calibration curves were linear (R(2)>0.9995) in the selected range for each analyte. The method is specific and sensitive with limit of quantification as 25 ng/ml for quinidine and verapamil. The intra- and inter-day accuracy and precision were found to be good for all the five analytes. The method was found to be reliable in permeability determination and to estimate pH-dependent P-glycoprotein (P-gp)-mediated efflux transport of quinidine. Weak bases quinidine, propranolol and verapamil showed pH-dependent permeability, where quinidine permeability increased by 3.6-fold when the luminal pH was changed from pH 4.5-7.4. Inhibition of P-gp by verapamil (200 microM) indicated that about 68% and only 35% of passive transport of quinidine was attenuated by P-gp-mediated efflux at pH 4.5 and 7.4, respectively. In conclusion, low passive transport rates of weakly basic P-gp substrates at lower pH, may lead to more accessibility of these molecules to P-gp within enterocytes thus resulting in pH-dependent functional activity of P-gp as protic drugs moves along the gastrointestinal tract.

Concentration and pH dependent aggregation of hydrophobic drug molecules and relevance to oral bioavailability

We have examined selected physicochemical properties of compounds from the diaryltriazine/diarylpyrimidine (DATA/DAPY) classes of non-nucleoside reverse transcriptase inhibitors (NNRTIs) and explored possible correlations with their bioavailability. In simple aqueous solutions designed to mimic the gastrointestinal (GI) environment of a fasting individual, all NNRTIs demonstrated formation of aggregates as detected by dynamic light scattering and electron microscopy. Under various conditions mimicking physiological transitions in the GI environment, aggregate size distributions were shown to depend on compound concentration and pH. NNRTIs with good absorption were capable of forming aggregates with hydrodynamic radii of </=100 nm at higher concentrations and over wide ranges of pH, while poorly absorbed inhibitors form aggregates with radii of >/=250 nm at concentrations above 0.01 mM, probably representing precipitate. We propose a model in which the uptake rate into systemic circulation depends on having hydrophobic drug aggregates of appropriate size available for absorption at different locations within the GI tract.

Contribution of the paracellular route to the pH-dependent epithelial permeability to cationic drugs

The aim of this work was to investigate the contribution of the paracellular route to the pH-dependent permeability to cationic drugs in three models expressing different drug permeabilities: hexadecane membranes (HDMs), Caco-2, and 2/4/A1 cell monolayers. The high- and low-permeability drugs alfentanil and cimetidine were used as model drugs. The paracellular permeability was calculated: 1. from the assumption that the ionized form (P(mi)) permeates a cell monolayer only by the paracellular route, and 2. on basis of the pore-restricted diffusion. For both drugs, sigmoidal relationships between membrane permeability and pH were observed in all models. The P(mi) was in excellent agreement with the paracellular permeability of cimetidine in the two cell models, whereas no significant P(mi) of the drugs could be observed in HDM. The results showed that the paracellular route has a significant role in the permeability of small basic hydrophilic drugs, such as cimetidine in leaky, small intestinal-like epithelia such as 2/4/A1. By contrast, in tighter epithelia such as Caco-2 and in artificial membranes such as HDM, the permeability of the ionized forms of the drugs and the paracellular permeability are lower or insignificant, respectively. These findings will have implications in the experimental design and data interpretation of pH-dependent drug transport experiments in cell culture models as well as in artificial membrane models such as HDM and parallel artificial membrane permeability assay (PAMPA).

Caco-2 permeability of weakly basic drugs predicted with the Double-Sink PAMPA method

The aim of this study was to analyze pH-dependent permeability of cationic drugs in Caco-2 cell monolayers using the pK(a)(flux) method and to correlate the results with those obtained in PAMPA (parallel artificial membrane permeability assay). The pH-dependent permeability of verapamil and propranolol was studied in Caco-2 cell monolayers. The data were subsequently processed using software developed for the PAMPA pK(a)(flux) method. Literature values for an additional nine cationic drugs were also analyzed. Double-Sink PAMPA data were also obtained for the same cationic drugs, to compare with the Caco-2 data. The Algorithm Builder program was then used to develop a predictive model of Caco-2 permeability based on PAMPA permeability and calculated Abraham molecular descriptors. From the relationship between permeability and pH it was shown that in PAMPA only the uncharged form of the drugs permeated across the membrane barrier, while charged and ionized forms of the drugs were significantly permeable in Caco-2. The charged-form permeability, P(i), was therefore determined and subsequently subtracted from all permeability coefficients in Caco-2 prior to the comparison with PAMPA. The resulting intrinsic permeability coefficients (P(o)) obtained in Caco-2 were successfully related to those derived from the PAMPA model. In this study we have shown that permeability coefficients obtained in PAMPA can predict the passive transcellular permeability in Caco-2.

Caco-2 replace or refine?

Multiple screening techniques have been developed to gain simplicity and rapidness in prediction of human intestinal permeability. The most extensively used method for years has been the Caco-2 cell monolayers. Are the less time- and resource-consuming artificial membranes and computational-based predictions on their way to replace the Caco-2 cells? The importance of mechanistic approaches and correct interpretation of the data using the techniques available is compared and discussed in this article.:

Global and Local Computational Models for Aqueous Solubility Prediction of Drug-Like Molecules

The aim of this study was to develop in silico protocols for the prediction of aqueous drug solubility. For this purpose, high quality solubility data of 85 drug-like compounds covering the total drug-like space as identified with the ChemGPS methodology were used. Two-dimensional molecular descriptors describing electron distribution, lipophilicity, flexibility, and size were calculated by Molconn-Z and Selma. Global minimum energy conformers were obtained by Monte Carlo simulations in MacroModel and three-dimensional descriptors of molecular surface area properties were calculated by Marea. PLS models were obtained by use of training and test sets. Both a global drug solubility model (R(2) = 0.80, RMSE(te) = 0.83) and subset specific models (after dividing the 85 compounds into acids, bases, ampholytes, and nonproteolytes) were generated. Furthermore, the final models were successful in predicting the solubility values of external test sets taken from the literature. The results showed that homologous series and subsets can be predicted with high accuracy from easily comprehensible models, whereas consensus modeling might be needed to predict the aqueous drug solubility of datasets with large structural diversity.