Search

In this study, we determined the pH and buffer capacity of human gastrointestinal (GI) fluids (aspirated from the stomach, duodenum, proximal jejunum, and mid/distal jejunum) as a function of time, from 37 healthy subjects after oral administration of an 800 mg immediate-release tablet of ibuprofen (reference listed drug; RLD) under typical prescribed bioequivalence (BE) study protocol conditions in both fasted and fed states (simulated by ingestion of a liquid meal). Simultaneously, motility was continuously monitored using water-perfused manometry. The time to appearance of phase III contractions (i.e., housekeeper wave) was monitored following administration of the ibuprofen tablet. Our results clearly demonstrated the dynamic change in pH as a function of time and, most significantly, the extremely low buffer capacity along the GI tract. The buffer capacity on average was 2.26 μmol/mL/ΔpH in fasted state (range: 0.26 and 6.32 μmol/mL/ΔpH) and 2.66 μmol/mL/ΔpH in fed state (range: 0.78 and 5.98 μmol/mL/ΔpH) throughout the entire upper GI tract (stomach, duodenum, and proximal and mid/distal jejunum). The implication of this very low buffer capacity of the human GI tract is profound for the oral delivery of both acidic and basic active pharmaceutical ingredients (APIs). An in vivo predictive dissolution method would require not only a bicarbonate buffer but also, more significantly, a low buffer capacity of dissolution media to reflect in vivo dissolution conditions.

The purpose of this human intestinal perfusion study (in vivo) was twofold. Firstly, we aimed to determine the effective in vivo jejunal permeability (P(eff)) of amoxicillin and to classify it according to the Biopharmaceutics Classification System (BCS). Secondly, we investigated the acute effect of amiloride on the jejunal P(eff) of amoxicillin. Amoxicillin, a beta-lactam antibiotic, has been reported to be absorbed across the intestinal mucosa by both passive diffusion and active transport. A regional single-pass perfusion of the jejunum was performed using a Loc-I-Gut perfusion tube in 14 healthy volunteers. Each perfusion lasted for 200 min and was divided into two periods of 100 min each. The concentration of amoxicillin entering the jejunal segment was 300 mg/l in both periods, and amiloride, an inhibitor of the Na+/H+ exchanger, was added at 25 mg/l in period 2. The concentrations of amoxicillin and amiloride in the outlet jejunal perfusate were measured with two different HPLC-methods. Antipyrine and [14C]PEG 4000 were added as internal standards to the perfusion solution. Amiloride had no significant effect on the jejunal P(eff) of amoxicillin. The human in vivo jejunal P(eff) for amoxicillin was 0.34+/-0.11 x 10(-4) and 0.46+/-0.12 x 10(-4) cm/s in periods 1 and 2, respectively. The high jejunal P(eff) for amiloride was 1.63+/-0.51 x 10(-4) cm/s which predicts an intestinal absorption of more than 90%. Following the BCS amoxicillin was classified as a low P(eff) drug, and amiloride had no acute effect on the in vivo jejunal P(eff) of amoxicillin.

The plasma concentration-time profiles following oral administration of drugs are often irregular and cannot be interpreted easily with conventional models based on first- or zero-order absorption kinetics and lag time. Six new models were developed using a time-dependent absorption rate coefficient, ka(t), wherein the time dependency was varied to account for the dynamic processes such as changes in fluid absorption or secretion, in absorption surface area, and in motility with time, in the gastrointestinal tract. In the present study, the plasma concentration profiles of propranolol obtained in human subjects following oral dosing were analyzed using the newly derived models based on mass balance and compared with the conventional models. Nonlinear regression analysis indicated that the conventional compartment model including lag time (CLAG model) could not predict the rapid initial increase in plasma concentration after dosing and the predicted Cmax values were much lower than that observed. On the other hand, all models with the time-dependent absorption rate coefficient, ka(t), were superior to the CLAG model in predicting plasma concentration profiles. Based on Akaike's Information Criterion (AIC), the fluid absorption model without lag time (FA model) exhibited the best overall fit to the data. The two-phase model including lag time, TPLAG model was also found to be a good model judging from the values of sum of squares. This model also described the irregular profiles of plasma concentration with time and frequently predicted Cmax values satisfactorily. A comparison of the absorption rate profiles also suggested that the TPLAG model is better at prediction of irregular absorption kinetics than the FA model. In conclusion, the incorporation of a time-dependent absorption rate coefficient ka(t) allows the prediction of nonlinear absorption characteristics in a more reliable manner.

Purpose: The objective of this study was a systematic characterization and evaluation of cell culture models based on mixtures of Caco-2/HT29-MTX co-cultures for their use in screening for drug absorption and intestinal permeability in comparison to the properties of the respective mono-cultures., Methods: Co-cultures of Caco-2 cells (absorptive-type) and HT29-MTX cells (goblet-type) were set up. Three different co-cultures (initial seeding ratios Caco-2/HT29-MTX: 90/10, 70/30, and 50/50) were grown on permeable filter supports, and monolayers were used for permeability studies with model compounds for paracellular absorption (atenolol, furosemide, H334/75, mannitol, terbutaline), transcellular absorption (antipyrine, ketoprofen, metoprolol, piroxicam), carrier-mediated absorption (D-glucose, Gly-Pro, and L-phenylalanine) as well as substrates for carrier-mediated secretion via P-glycoprotein (cimetidine and talinolol). Electrophysiological and microscopic controls were performed to characterize the cell cultures., Results: For compounds undergoing passive intestinal absorption permeabilities were generally higher in co-cultures than in Caco-2 monolayers, yielding highest values in pure HT29-MTX monolayers. This difference was most obvious for compounds transported via the paracellular pathway, where HT29-MTX cells may be up to 30 times more permeable than Caco-2 cells, whereas for lipophilic and highly permeable compounds, the difference in permeability values was less obvious. For drugs undergoing intestinal secretion mediated by P-glycoprotein, co-cultivation of Caco-2 cells with HT29-MTX cells led to increased apical to basolateral permeability which was decreased in the opposite direction, consistent with the fact that HT29-MTX cells do not express P-glycoprotein. When a carrier-mediated absorption mechanism is involved, the permeabilities observed were lower than the values reported for human small intestine and co-cultivation of HT29-MTX cells with Caco-2 cells resulted in even lower values as compared to the plain Caco-2 cultures., Conclusions: Co-cultures of HT29-MTX and Caco-2 cells offer the opportunity of modifying the permeability barrier of the cell monolayers both with respect to paracellular resistance and secretory transport via P-gp. Thus, in special cases, they allow more flexibility in adapting the in vitro system to the in vivo situation as compared to the monocultures. Another advantage is the obvious robustness of the method with respect to the reproducibility of the results. A problem remaining, however, is the quantitative expression of carriers involved in intestinal uptake of many nutrients and drugs., (Copyright 2000 Wiley-Liss, Inc. and the American Pharmaceutical Association J Pharm Sci 89: 63-75, 2000)

Purpose: The objective was to investigate the ability of a glycosteroid (TC002) to increase the oral bioavailability of gentamicin., Methods: Admixtures of gentamicin and TC002 were administered to the rat ileum by injection and to dogs by ileal or jejunal externalized ports, or PO. Bioavailability of gentamicin was determined by HPLC. 3H-TC002 was injected via externalized cannulas into rat ileum or jejunum, or PO and its distribution and elimination was determined. The metabolism of TC002 in rats was evaluated by solid phase extraction and HPLC analysis of plasma, urine and feces following oral or intestinal administration., Results: The bioavailability of gentamicin was substantially increased in the presence of TC002 in both rats and dogs. The level of absorption was dependent on the concentration of TC002 and site of administration. Greatest absorption occurred following ileal orjejunal administration. TC002 was significantly more efficacious than sodium taurocholate, but similar in cytotoxicity. TC002 remained primarily in the GI tract following oral or intestinal administration and cleared rapidly from the body. It was only partly metabolized in the GI tract, but was rapidly and completely converted to its metabolite in plasma and urine., Conclusions: TC002 shows promise as a new drug transport agent for promoting intestinal absorption of polar molecules such as gentamicin.

Endothelins are potent vasoconstrictor peptides which have a wide range of tissue distribution and three receptor subtypes (ET(A), ET(B) and ET(C)). Among the linear hexapeptide ET(A)/ET(B) receptor antagonists, PD 145065 (Ac-D-Bhg-L-Leu-L-Asp-L-Ile-L-Ile-L-Trp, Bhg = (10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5-yl)-Gly) and PD 156252 (Ac-D-Bhg-L-Leu-L-Asp-L-Ile-(N-methyl)-L-Ile-L-Trp) were selected to evaluate the metabolic stability and intestinal absorption in the absence and/or in the presence of protease inhibitors. In vitro stability of both compounds was investigated in fresh plasma, lumenal perfusate, intestinal and liver homogenates. PD 156252 was more stable than PD 145065 in intestinal tissue homogenate (63.4% vs. 20.5% remaining) and liver homogenate (74.4% vs. 35.5% remaining), while both compounds showed relatively good stability in the fresh plasma (94.5% vs. 86.7% remaining) and lumenal perfusate (85.8% vs. 72.3% remaining). The effect of protease inhibitors on the degradation of PD 145065 and PD 156252 was also investigated. Amastatin, thiorphan, chymostatin and the mixture of these three inhibitors were effective in reducing the degradation of both compounds. The pharmacokinetic parameters of PD 156252, calculated by using a non-compartmental model, were 6.95 min (terminal half-life), 191 mL (Vss), and 25.5 mL/min (Cl(tot)) after intravenous administration in rats. The intestinal absorption of PD 156252 in rats was evaluated in the absence and/or in the presence of protease inhibitors. The results indicate that the major elimination pathway of PD 156252 appears to be the biliary excretion and protease inhibitors increase the intestinal absorption of PD 156252 through increasing metabolic stability.

Hydrophilic drugs are often poorly absorbed when administered orally. There has been considerable interest in the possibility of using absorption enhancers to promote absorption of polar molecules across membrane surfaces. The bile acids are one of the most widely investigated classes of absorption enhancers, but there is disagreement about what features of bile acid enhancers are responsible for their efficacy. We have designed a class of glycosylated bile acid derivatives to evaluate how increasing the hydrophilicity of the steroid nucleus affects the ability to transport polar molecules across membranes. Some of the glycosylated molecules are significantly more effective than taurocholate in promoting the intestinal absorption of a range of drugs, showing that hydrophobicity is not a critical parameter in transport efficacy, as previously suggested. Furthermore, the most effective glycosylated compound is also far less damaging to membranes than the best bile acid absorption promoters, presumably because it is more hydrophilic. The results reported here show that it is possible to decouple absorption-promoting activity from membrane damage, a finding that should spark interest in the design of new compounds to facilitate the delivery of polar drugs.

The goal of this study was to determine the most appropriate hydrodynamic model for the Loc-I-Gut in-vivo perfusion system. The general mixing-tank-in-series model, which can approximate single mixing tank and laminar and plug-flow hydrodynamics, was fitted to the observed experimental residence-time distribution curves for the non-absorbable marker [14C]PEG 4000. The residence-time distribution analysis shows that the hydrodynamics of the perfusion solution within the jejunal segment in man is well approximately by a model containing on average between 1-2 mixing tanks in series. The solution is well mixed when using perfusion rates of 2.0, 3.0 and 6.0 mL min-1. The average mean residence time estimates from the fitted residence-time distribution were 12 +/- 7.6, 15 +/- 4.2 and 7.7 +/- 4.6 min, respectively, at these three perfusion rates. The mean volumes of the segment (Vs) were 25 +/- 15, 45 +/- 12 and 46 +/- 27 mL, respectively. There were no statistical differences between 2.0, 3.0 and 6.0 mL min-1 in respect of the number of mixing tanks (n) and mean residence times. This residence-time distribution analysis indicates that the luminal fluid in the Loc-I-Gut perfusion system is well-mixed, and that permeability calculations based on the well-mixed assumption most closely approximate the actual local (average) membrane permeability within the perfused segment.

Purpose: To investigate the effect of microparticle size on gastrointestinal tissue uptake., Methods: Biodegradable microparticles of various sizes using polylactic polyglycolic acid (50:50) co-polymer (100 nm, 500 nm, 1 micron, and 10 microns) and bovine serum albumin as a model protein were formulated by water-in-oil-in-water emulsion solvent evaporation technique. The uptake of microparticles was studied in rat in situ intestinal loop model and quantitatively analyzed for efficiency of uptake., Results: In general, the efficiency of uptake of 100 nm size particles by the intestinal tissue was 15-250 fold higher compared to larger size microparticles. The efficiency of uptake was dependent on the type of tissue, such as Peyer's patch and non patch as well as on the location of the tissue collected i.e. duodenum or ileum. Depending on the size of microparticles, the Peyer's patch tissue had 2-200 fold higher uptake of particles than the non-patch tissue collected from the same region of the intestine. Histological evaluation of the tissue sections demonstrated that 100 nm particles were diffused throughout the submucosal layers while the larger size nano/microparticles were predominantly localized in the epithelial lining of the tissue., Conclusions: There is a microparticle size dependent exclusion phenomena in the gastrointestinal mucosal tissue with 100 nm size particles showing significantly greater tissue uptake. This has important implications in designing of nanoparticle-based oral drug delivery systems, such as an oral vaccine system.

For almost a half century scientists have striven to develop a theoretical model capable of predicting oral drug absorption in humans. From the pH-partition hypothesis to the compartmental absorption and transit (CAT) model, various qualitative/quantitative approaches have been proposed, revised and extended. In this review, these models are classified into three categories; quasi-equilibrium models, steady-state models and dynamic models. The quasi-equilibrium models include the pH-partition hypothesis and the absorption potential concept, the steady-state models include the film model and the mass balance approaches, and the dynamic models include the dispersion, mixing tank and CAT models. The quasi-equilibrium models generally provide a basic guideline for understanding drug absorption trends. The steady-state models can be used to estimate the fraction of dose absorbed. The dynamic models predict both the fraction of dose absorbed and the rate of drug absorption and can be related to pharmacokinetic models to evaluate plasma concentration profiles.

The influence of various fasting-state gastrointestinal parameters on variability in absorption of cimetidine was studied using simulation and cimetidine administration as a duodenal infusion and as an oral tablet in fistulated mongrel dogs. In the simulation studies, the frequency of double-peak occurrence in plasma profiles was estimated employing average gastric emptying rates as well as interdigestive-migrating-motor-complex (IMMC) phase lengths that were systematically altered. Emptying rates and phase lengths were modeled as periodic step functions. Simulations indicated that double peaks occur when gastric emptying of the drug begins in early phase I or late phase II/III, which represent the periods of very low, medium, and high gastrointestinal motility, respectively. The incidence is increased for longer phase-I duration and higher elimination rate constants. For a compound with a 2 h elimination or disposition half-life, two concentration maxima (double peaks) were found in 12% of the simulated concentration-time profiles. The double peak frequency in simulated curves was considerably higher for t1/2 < 30 min. When cimetidine was administered to dogs as a duodenal infusion in the active and quiescent motility phases, discontinuous profiles were observed, although the variability of the various parameters was reduced when compared. Pharmacokinetic models were set up that were characterized by multi-segmental input (one- to 3-lag-time models were used) for the profiles resulting from oral and duodenal administration. The lag time for the first process characterized the onset of absorption. A significant difference between phases was detected for infusions at pH 8, where the initial lag time was longer in the quiescent phase. The mean input time (MIT) was calculated as the integral input parameter. There was a tendency for the MIT to be higher for pH 6 infusions than for pH 4 and pH 8. Bioavailability analysis indicated that cimetidine was more rapidly and completely absorbed at pH 8 than at pH 6. Bioavailability was also slightly higher at pH 4 than at pH 6. We concluded that gastric emptying increased the variability of the cimetidine concentration-against-time profiles and that it plays a role with respect to double-peak occurrence, although it is only one of several causative factors.

A microscopic mass balance approach has been developed to predict the fraction dose absorbed of suspensions of poorly soluble compounds. The mathematical model includes four fundamental dimensionless parameters to estimate the fraction dose absorbed: initial saturation (Is), absorption number (An), dose number (Do), and dissolution number (Dn). The fraction dose absorbed (F) increases with increasing Is, An, and Dn and with decreasing Do. At higher Dn and lower Do, the fraction dose absorbed reaches the maximal F, which depends only on An. The dissolution number limit on F can appear at both lower Do and lower Dn. Likewise, at higher Do and Dn, the fraction dose absorbed reaches a Do limit. Initial saturation makes a significant difference in F at lower Do and Dn. It is shown that the extent of drug absorption is expected to be highly variable when Dn and Do are approximately one. Furthermore, by calculating these dimensionless groups for a given compound, a formulation scientist can estimate not only the extent of drug absorption but also the effect, if any, of particle size reduction on the extent of drug absorption.

A theoretical analysis for estimating the extent of intestinal peptide and peptide analogue absorption was developed on the basis of a mass balance approach that incorporates convection, permeability, and reaction. The macroscopic mass balance analysis (MMBA) was extended to include chemical and enzymatic degradation. A microscopic mass balance analysis, a numerical approach, was also developed and the results compared to the MMBA. The mass balance equations for the fraction of a drug absorbed and reacted in the tube were derived from the general steady state mass balance in a tube: [formula: see text] where M is mass, z is the length of the tube, R is the tube radius, Pw is the intestinal wall permeability, kr is the reaction rate constant, C is the concentration of drug in the volume element over which the mass balance is taken, VL is the volume of the tube, and vz is the axial velocity of drug. The theory was first applied to the oral absorption of two tripeptide analogues, cefaclor (CCL) and cefatrizine (CZN), which degrade and dimerize in the intestine. Simulations using the mass balance equations, the experimental absorption parameters, and the literature stability rate constants yielded a mean estimated extent of CCL (250-mg dose) and CZN (1000-mg dose) absorption of 89 and 51%, respectively, which was similar to the mean extent of absorption reported in humans (90 and 50%). It was proposed previously that 15% of the CCL dose spontaneously degraded systematically; however, our simulations suggest that significant CCL degradation occurs (8 to 17%) presystemically in the intestinal lumen.(ABSTRACT TRUNCATED AT 250 WORDS)

The absorption mechanism of several penicillins was characterized using in situ single-pass intestinal perfusion in the rat. The intrinsic membrane parameters were determined using a modified boundary layer model (fitted value +/- S.E.): Jmax* = 11.78 +/- 1.88 mM, Km = 15.80 +/- 2.92 mM, Pm* = 0, Pc* = 0.75 +/- 0.04 for ampicillin; Jmax* = 0.044 +/- 0.018 mM, Km = 0.058 +/- 0.026 mM, Pm* = 0.558 +/- 0.051, Pc* = 0.757 +/- 0.088 for amoxicillin; and Jmax* = 16.30 +/- 3.40 mM, Km = 14.00 +/- 3.30 mM, Pm* = 0, Pc* = 1.14 +/- 0.05 for cyclacillin. All of the aminopenicillins studied demonstrated saturable absorption kinetics as indicated by their concentration-dependent wall permeabilities. Inhibition studies were performed to confirm the existence of a nonpassive absorption mechanism. The intrinsic wall permeability (Pw*) of 0.01 mM ampicillin was significantly lowered by 1 mM amoxicillin and the Pw* of 0.01 mM amoxicillin was reduced by 2 mM cephradine consistent with competitive inhibition.

The feasibility of targeting prolidase as a peptide prodrug-converting enzyme has been examined. The enzymatic hydrolysis by prolidase of substrates for the peptide transporter L-alpha-methyldopa-pro and several dipeptide analogues without an N-terminal alpha-amino group (phenylpropionylproline, phenylacetylproline, N-benzoylproline, and N-acetylproline) was investigated. The Michaelis-Menten parameters Km and Vmax for L-alpha-methyldopa-pro are 0.09 +/- 0.02 mM and 3.98 +/- 0.25 mumol/min/mg protein, respectively. However, no hydrolysis of the dipeptide analogues without an N-terminal alpha-amino group is observed, suggesting that an N-terminal alpha-amino group is required for prolidase activity. These results demonstrate that prolidase may serve as a prodrug-converting enzyme for the dipeptide-type prodrugs, utilizing the peptide carrier for transport of prodrugs into the mucosal cells and prolidase, a cytosolic enzyme, to release the drug. However, a free alpha-amino group appears to be necessary for prolidase hydrolysis.

The effects of microgravity on the physiologic response of the human body, the physical properties of gastrointestinal contents, and the influence these responses have on drug absorption are becoming more and more critical as the duration of humans in the hostile space environment dramatically increases. In this environment, some conventional oral dosage forms may be severely limited as an effective drug regimen. To understand the effects of microgravity, one must first understand the basic forces acting on a particle moving through a walled-tube such as the small intestine: gravity (FG), buoyancy (FB), and drag (FD). These forces can be combined and rearranged into a dimensionless ratio of gravitational forces to viscous forces. This is the most important dimensionless group influencing the motion of a particle relative to the fluid. Gastric emptying is highly influenced by several factors: volume, calories, exercise, size, density, temperature, viscosity, osmolality as well as those factors associated with physiologic responses: splanchnic blood flow, body position, and electrolyte balance. This array of factors can lead to variability in drug plasma levels. In the absence of gravity, the factors of size and density would appear to be most directly altered due to their dependence on the force of gravity. Intestinal transit rate in a gravity environment is highly dependent on the motility state of the GI tract either fasted or fed partly due to the higher viscosities of chyme in the fed state. In space, the absence of gravity may tend to increase the transit rate along the small intestine by decreasing the dimensionless ratio of gravitational forces to viscous forces. In zero gravity, therefore, these alterations in GI emptying and intestinal transit rate could lead to erratic plasma levels and inefficient absorption.

Polyethylene glycols (PEGs) 600, 1000, and 2000 were used to study the molecular weight permeability dependence in the rat nasal and gastrointestinal mucosa. Absorption of the PEGs was measured by following their urinary excretion over a 6-hr collection period. HPLC methods were used to separate and quantitate the individual oligomeric species present in the PEG samples. The permeabilities of both the gastrointestinal and the nasal mucosae exhibited similar molecular weight dependencies. The steepest absorption dependence for both mucosae occurs with the oligomers of PEG 600, where the extent of absorption decreases from approximately 60% to near 30% over a molecular weight range of less than 300 daltons. Differences in the absorption characteristics between the two sites appear in the molecular weight range spanned by PEG 1000. For these oligomers, the mean absorption from the nasal cavity is approximately 14%, while that from the gastrointestinal tract is only 9%. For PEG 2000, mean absorption decreases to 4% following intranasal application and below 2% following gastrointestinal administration. Within the PEG 1000 and 2000 samples, however, very little molecular weight dependency is seen among the oligomers. In the range studied, a distinct molecular weight cutoff was not apparent at either site.

The intestinal absorption of alpha-methyldopa from perfused segments of rat intestine was determined. Jejunal and ileal segments were studied at pH 4.5, 6.0 and 7.4 at an alpha-methyldopa concentration of 0.01 mM, and at pH 6.0 at concentrations of 0.01, 0.10 and 10 mM. Intestinal absorption was found to depend on both concentration and pH but not on segment. A weak positive correlation (r = 0.5) was observed between net water absorption and alpha-methyldopa absorption, similar to that observed for amino acids. The low intestinal wall permeability of alpha-methyldopa under normal intestinal pH conditions (pH 7.4) is consistent with the incomplete oral bioavailability of this drug.

The intestinal absorption of L-lysine-p-nitroanilide, L-alanine-p-nitroanilide, and glycine-p-nitroanilide was studied in perfused rat intestine in the presence of a variety of potential competitive inhibitors. The results indicate that the hydrolysis site(s) show side-chain specificity, and that inhibitors require a free amino group in the alpha-position and must be in the L-configuration to be effective. Glycyl-L-proline, a peptide transport inhibitor, had no effect on the absorption rate.

A physiological flow model is presented to account for plasma level double peaks based on cyclical gastric emptying and intestinal motility in the fasted state. Central to the model is the assumption that gastric emptying and intestinal transit rates will vary directly with the strength of the contractile activity characteristic of the fasted state motility cycle. Simulated curves clearly indicate that variable gastric emptying rates can result in variable absorption rates from the gastrointestinal tract and double peaks in the plasma level curves of cimetidine. Vital to the occurrence of double peaks are (i) dosing time relative to phasic activity, (ii) variability in flow out of the stomach, and (iii) a small emptying rate constant Qs/Vs, for a period of time within the first hour after administration. Variability in intestinal flow rates alone does not cause a double peak in the plasma level curve. Results of the simulations, as well as experimental results, can be categorized according to the shapes of the plasma level curves into four types: type A, Cpmax (1) less than Cpmax (2); type B, single peak; type C, Cpmax (1) greater than Cpmax (2); type D, Cpmax(1) = Cpmax(2). Assuming that the experimental results were obtained from fasted subjects, with the time of dose administration being a random variable, the frequency of the experimental curves having shape A, B, C, or D correlates extremely well with theoretical predictions. It is concluded that variable gastric emptying rates due to the motility cycle can account for plasma level double peaks. Furthermore, variable gastric emptying rates combined with the short plasma elimination half-life and poor gastric absorption of cimetidine can be the cause of the frequently observed plasma level double peaks.

Based on a simple tube model for drug absorption, the key parameters controlling drug absorption are shown to be the dimensionless effective permeability, P*eff, and the Graetz number, Gz, when metabolism or solubility/dissolution is not rate controlling. Estimating the Graetz number in humans and assuming that P*aq is not rate controlling give the following equation for fraction dose absorbed: F = 1 - e-2P*w. The correlation between fraction dose absorbed in humans and P*w determined from steadystate perfused rat intestinal segments gives an excellent correlation. It is of particular significance that the correlation includes drugs that are absorbed by passive and carrier-mediated processes. This indicates that P*w is one of the key variables controlling oral drug absorption and that the correlation may be useful for estimating oral drug absorption in humans regardless of the mechanism of absorption.

Intestinal permeabilities of five dipeptidyl derivatives of L-alpha-methyldopa (I) were studied by an in situ intestinal perfusion method. The dipeptides displayed a significant increase in their permeabilities compared to L-alpha-methyldopa. The increases ranged from 4 to 20 times. These results suggest that the peptide transport system is less structurally specific than the amino acid transport systems and can be used as an absorption pathway for peptide analogues. The kinetic advantage demonstrated by the dipeptide, L-alpha-methyldopa-L-phenylalanine, over the amino acid analogue, L-alpha-methyldopa, suggests that the peptide carrier would be a possible route for improving the intestinal absorption of pharmacologically active amino acid analogues. Furthermore, the preliminary results of in vitro hydrolysis studies of selected dipeptidyl derivatives indicate that the peptide carrier system could be used as a base for a prodrug strategy.

The intestinal absorption mechanism of two ACE inhibitor prodrugs, enalapril and fosinopril, was investigated in rats using a single-pass perfusion method. A modified boundary layer solution was applied to determine the apparent intestinal wall permeability. The prodrug enalapril is well absorbed from rat jejunum, whereas the parent drug, enalaprilat, is poorly absorbed. The permeability of enalapril is concentration dependent and is decreased by the dipeptide Tyr-Gly and by cephradine but not by the amino acids L-leucine or L-phenylalanine, indicating a nonpassive absorption mechanism via the small peptide carrier-mediated transport system. In contrast, fosinopril is readily absorbed by a concentration-independent mechanism without the involvement of the peptide carrier.

A strategy for improving the intestinal absorption of water-insoluble drugs was developed and tested. The strategy is based on making a soluble derivative of an insoluble compound which, in turn, is a substrate for enzymes in the surface coat of the brush border region of the microvillous membrane. Consequently, just prior to reaching the membrane, the physical properties of the diffusing species are changed from polar to nonpolar. The experimental test used two drug-drug derivative pairs, estrone-lysine estrone ester and p-nitroaniline-lysine p-nitroanilide. Wall permeabilities were determined using an external perfusion technique in the rat intestine and a laminar flow convective diffusion model for transport in the lumen. Analysis of the permeability results indicates that the derivatives have higher wall permeabilities than the parent compounds and that the microvillous surface coat may be a significant contributor to the intestinal wall resistance. Comparison of the absorption rates for estrone and the lysine estrone ester indicates that the absorption rate of the derivatives could be up to five orders of magnitude greater than that for the parent compound.

The intestinal absorption mechanism of captopril was investigated in fasted rats using a single-pass perfusion method. Captopril and captopril disulfide were analyzed by HPLC. A modified boundary-layer solution was applied to determine the apparent intestinal wall permeabilities (Pw*). The results indicated that captopril is very permeable in the small intestine but not in the colon, and the permeability in the small intestine is both pH and concentration dependent. The estimated parameters for the carrier are: Km*, 6 mM; Jmax*, 12 mM; and Pc*, 2. There is also a significant passive component to captopril absorption in the small intestine. Furthermore, the intestinal permeability of captopril was significantly decreased by the withdrawal of sodium from the perfusate (3 times), and the addition of 85 mM of gly-gly (2.5 times), 15 mM of gly-pro (4 times), dipeptide mixture (4.5 times), 0.5 mM of 2,4-dinitrophenol (4 times), and 10 mM of cephradine (6 times). This is the first demonstration that an angiotensin converting enzyme (ACE) inhibitor is at least in part transported by a carrier-mediated process in the intestine via the peptide carrier system. In addition, the results showed that the peptide carrier system can transport a substrate without a "N" terminal nitrogen atom.

A two-tank perfect-mixing tank model was used to stimulate GI absorption. The effect of drug parameters (pK alpha, solubility, and intrinsic wall permeability) and system parameters (pH profile, volume of intestinal contents, and intestinal flow rate) on drug absorption were studied by numerical data stimulation. When the dose did not exceed the solubility of the drug in the intestinal lumen, the fraction absorbed depended on the transit rate relative to the absorption rate and the pK alpha relative to the pH profile, but was independent of drug dose. Saturation of one or both tanks led to dose-dependent absorption. The model was used to simulate absorption of chlorothiazide. Good agreement between simulated and experimental data led to the conclusion that the physical characteristics of chlorothiazide, rather than a saturable transport mechanism at the intestinal wall, may be responsible for the dose-dependent absorption observed for this drug. The model was also used to simulate hydrochlorothiazide absorption. By applying the same system parameters used for chlorothiazide, the model simulation correctly predicted the dose proportionality of hydrochlorothiazide absorption. The lack of dose dependency in this case may be attributed to the higher solubility and pK alpha of hydrochlorothiazide compared with chlorothiazide.

The intestinal absorption mechanism of two nonsulfhydril lysyl-proline angiotensin converting enzyme (ACE) inhibitors, lisinopril (1) and SQ 29,852 (2; [(S)-1-[6-amino-2-[[hydroxy (4-phenylbutyl)-phosphinyl]oxy[-1-oxohexyl]-L-proline) were investigated in rats using a single-pass perfusion method. Compound 2 is well absorbed from rat jejunum, whereas lisinopril absorption is relatively low. The permeability of both ACE inhibitors is concentration dependent and is decreased by the dipeptide Tyr-Gly and by cephradine, indicating a nonpassive absorption mechanism via the peptide carrier-mediated transport system. Compound 2 is well absorbed by a nonpassive mechanism, in parallel with a small passive component. The estimated dimensionless carrier parameters for 2 are J*max = 0.16, Km = 0.08 mM, P*c = 2.0, and P*m = 0.25; for lisinopril, passive absorption is not significant and its absorption is nonpassive: J*max = 0.032, Km = 0.082 mM, and P*c = 0.39 (where J*max is the maximal flux, Km is the Michaelis constant, P*c is the carrier permeability, and P*m is the passive permeability). These results offer a mechanistic explanation for the prolonged ACE inhibition and the low oral bioavailability of lisinopril, and for the nonlinear pharmacokinetics of 2.

Gastrointestinal pH as a function of time was recorded for 4 beagle dogs and 10 human subjects using radiotelemetric pH measuring equipment. Results indicated that in the quiescent phase, gastric pH in the dogs (mean = 1.8 +/- 0.07 SEM) was significantly (p less than 0.05) higher than in humans (1.1 +/- 0.15). No significant difference in the time for the pH monitoring device to empty from the stomach was noted for the two species (99.8 +/- 27.2 min for dogs, 59.7 +/- 14.8 min for humans, p greater than 0.05). The fasting intestinal pH in dogs was consistently higher than in humans, with an average canine intestinal pH of 7.3 +/- 0.09 versus 6.0 +/- 0.14 for humans. The implication of these observations for extrapolation of drug absorption data from dogs to humans are discussed.