Kinetics of peritoneal drug transport in rats: an application of the pore theory of transcapillary exchange.

In order to quantitatively describe the peritoneal transport of drugs, this paper proposes a kinetic model that is based on the hydrodynamic pore theory of transcapillary exchange, and incorporates an explicit description of volume and osmolality changes in the dialysate. Sulfisoxazole (SIX) and benzoic acid (BA) were used as model compounds. Following intraperitoneal administration of dialysate in rats, the osmolality, volume, and drug concentration in the dialysate were measured with respect to time. The obtained data were analyzed to give hydrodynamic parameters for solvent and a solute (including drug) by a computer-aided curve-fitting procedure according to the differential equations derived from the model. The present method, requiring no approximation of the changes in dialysate volume, made it possible to predict the concentration profiles of BA under different initial conditions of dialysate (i.e., different osmolality and volume). Solvent drag effect contributed little to the peritoneal transport of SIX and slightly to that of BA. It was also found that the peritoneal transport of BA is blood-flow limited while that of SIX is diffusion limited.

Interindividual changes in volume of distribution of cefazolin in newborn infants and its prediction based on physiological pharmacokinetic concepts.

The purpose of this study was to investigate factors affecting the volume of distribution of cefazolin (a beta-lactam antibiotic) in newborn infants with bacterial infections, and to propose a method for predicting the volume of distribution at steady state per body weight (Vdss/BW). Cefazolin and tobramycin (an aminoglycoside) were simultaneously given to newborn infants (aged 2 to 28 d), and plasma concentration-time data were analyzed on the basis of model-independent moment analysis. The Vdss/BW values ranged from 0.212 to 0.373 L/kg for cefazolin and from 0.384 to 0.541 L/kg for tobramycin. The unbound fraction of cefazolin in plasma (fp) fluctuated widely, from 0.22 to 0.83, among patients. The Vdss/BW value for cefazolin was characterized by both large extracellular water volume and a remarkable change in fp, and could be predicted as a function of fp using physiological pharmacokinetic concepts. Moreover, interindividual changes in the unconjugated bilirubin:albumin molar ratio were predominantly responsible for the individual variation in the fp values of cefazolin in newborn infants.

Effects of fasting on biperiden pharmacokinetics in the rat.

The effects of fasting on the pharmacokinetics of biperiden in rats were examined. Total clearance of biperiden was greater than 90% ascribable to hepatic clearance and was essentially blood-flow dependent. The number of compartments in the preferred pharmacokinetic model of biperiden changed from three (for normal rats) to two (for fasted rats). The smaller mean residence time (MRT) values found for fasted rats were attributable to decreases in distribution volume. Biperiden showed much higher lipophilicity than haloperidol, thiopental, and hexobarbital, and its tissue-to-plasma partition coefficient in adipose tissue was 20-fold higher than that in muscle. The influence of changes in volumes of adipose tissue and muscle on distribution volume (Vdss/BW) was evaluated from tissue-to-plasma partition coefficients. The value of Vdss/BW was predicted to decrease with decrease of adipose tissue, and to increase with decrease of muscle tissue. These results suggest that the observed decrease of Vdss/BW in fasted rats reflects reduced capacity to trap biperiden in the body, especially in adipose tissue. Possible clinical implications of these results are discussed.

Reversal of drug sensitivity in MDR subline of P388 leukemia by gene-targeted antisense oligonucleotide.

We attempted to reverse multidrug resistance (MDR) by treatment with 25-mer antisense phosphorothioate oligonucleotide. The phosphorothioate analogs, the sequences of which are sense or antisense to the initiation codon of mouse mdr1 mRNA, were tested against murine leukemic P388/S and adriamycin-resistant P388/ADR cell lines. A weak inhibitory effect on the growth of P388/S and P388/ADR cells was observed at a sense and antisense oligonucleotide concentration of 30 microM. Using the monoclonal antibody to P-glycoprotein and a flow cytometry technique, we showed that the level of expression of P-glycoprotein in P388/ADR cells treated with antisense oligonucleotide was lower than when treated with sense oligonucleotide. The antisense oligonucleotide potentiated the growth-inhibitory effect of vinblastine on P388/ADR cells, whereas sense oligonucleotide did not. This was accompanied by an increase in vinblastine retention in the cells. The reversal of the resistance by antisense oligonucleotide was increased by the combination with 1 microM verapamil. These results suggest that the antisense oligonucleotide and low dose verapamil may be useful in circumventing the resistance to anticancer drugs of MDR tumors.

Peritoneal transport of beta-lactam antibiotics: effects of plasma protein binding and the interspecies relationship.

In order to examine quantitatively the effect of plasma protein binding on the peritoneal transport of beta-lactam antibiotics, we employed a kinetic model based on the pore theory of transcapillary exchange. This model incorporates the changes in the volume, osmolality, and antibiotic concentration in the dialysate, so that the apparent capillary membrane permeability (Pd) and the reflection coefficient (sigma d) of an antibiotic could be assessed. Six cephalosporins (cefatrizine, cefazolin, cefpiramide, ceftazidime, ceftriaxone, cephaloridine) were used as model compounds. While the unbound fractions of these antibiotics ranged widely from 0.08 to 0.57, including linear and nonlinear protein binding, the concentration-time profiles in plasma and the peritoneal dialysate after intravenous administration in rats could be interpreted well by our model, assuming that only the unbound antibiotic is available for the peritoneal transport. The estimated Pd values were almost the same among the drugs examined. Moreover, the Pd values of cefazolin in mice, rats, and rabbits exhibited a 0.83-power dependency on the animal body weight, indicating that Pd is significantly related to the peritoneal surface area. On the other hand, the sigma d values of cefazolin were found to be almost the same among the animal species examined. Finally, the concentration-time profile of cefazolin in the dialysate after intravenous administration in a patient with end-stage renal failure was successfully predicted using the Pd value extrapolated from those of the experimental animals.

Kinetic phenotypic diagnosis of N-acetylation polymorphism in patients based on ratio of urinary metabolites of salicylazosulfapyridine.

We found that N-acetylation polymorphism can be evaluated from the disposition kinetics of sulfapyridine (SP) and 5-aminosalicylic acid (5-ASA) and their acetylated metabolites generated by N-acetyltransferase (NAT2) after oral administration of salicylazosulfapyridine (SASP). In 126 Japanese subjects, the homozygote of NAT2*4 was the most frequent (40%), followed by heterozygotes of NAT2*4 and mutant genes (28% NAT2*4/*6A, 15% NAT2*4/*7B, and 2% NAT2*4/*5B). Combinations of mutant genes accounted for 16%. When the relationship between the molar ratio of N-acetyl-SP (Ac-SP)/SP or N-acetyl-5-ASA(Ac-5-ASA)/5-ASA in serum and five genotypes of polymorphic NAT2* was examined in patients who received multiple doses of SASP, the molar ratios of Ac-SP/SP, rather than Ac-5-ASA/5-ASA tended to decrease according to the classification of genotype. We calculated the pharmacokinetic parameters in healthy subjects with various genotypes of polymorphic NAT2* after a single p.o. administration of SASP, according to a model of the SP metabolic pathways. The molar ratios of Ac-SP/SP in serum and urine were simulated using these parameters, and the molar ratio of Ac-SP/SP in urine at 4 days after the first administration could be categorized into ranges that were specific to various NAT2* genotypes. Thus, we were able to predict the N-acetylation polymorphic genotypes of patients by measuring the molar ratio of Ac-SP/SP in urine, after administration of SASP.

Kinetic evidence for facilitated peritoneal transport of benzoic acid in rats.

To evaluate the dose dependency in apparent peritoneal permeability (Pd) of benzoic acid as a model compound for a monocarboxylic acid transport system, a kinetic model, which involves changes in the volume and osmolality of the dialysate as well as the diffusion and convection of drugs across the peritoneum, was applied. We compared the Pd value of benzoic acid to that of phenobarbital which is a more lipophilic drug than benzoic acid. The concentration-time courses of phenobarbital in both peritoneal cavity and serum after the intraperitoneal administration with various doses were parallel according to dose, whereas those of benzoic acid varied in a dose-dependent manner. Using the values of unbound fraction (Fu), the value of Pd for unbound drugs was estimated. The Pd values of benzoic acid at 20 micrograms mL-1 was three times the value determined at 1000 micrograms mL-1. We suggest that certain facilitated transport systems constitute the mechanism of enhanced peritoneal membrane permeability of benzoic acid.

Prediction of changes in the clinical pharmacokinetics of basic drugs on the basis of octanol-water partition coefficients.

A physiologically based pharmacokinetic model for basic drugs has been established on the basis of octanol-water partition coefficients of the non-ionized, unbound drugs (Poct). The parameters for the physiological model in man were estimated from a regression equation obtained for the relationships between the Poct and the tissue-plasma partition coefficient, the hepatic intrinsic clearance (CLint,h) and the blood-to-plasma concentration ratio in rabbits. The plasma concentrations observed after intravenous administration of ten basic drugs (3.2 mg kg-1) to rabbits agreed with the levels predicted using the physiological model (r = 0.710-0.980). In man, the predicted plasma concentrations of basic drugs were in good agreement with reported values (r = 0.729-0.973), except for diazepam and pentazocine. Variations in plasma and brain-concentration profiles of clomipramine and nitrazepam in various disease states were simulated using the model. We assumed that the changes in unbound fraction of drug in serum (fp), CLint,h and the hepatic blood flow rate were from 0.25- to 4-fold that of the control and that fat volume changed by 0.2- to 5-fold. With regard to changes in fp, we predicted that the brain-plasma concentration ratio of clomipramine was 1.5- to 25-fold that of the control 24 h after intravenous administration, although the variations in the plasma concentration-time profiles were less marked. Plasma concentrations predicted for several basic drugs were in good agreement with reported values and this physiological model could be useful for predicting drug-disposition kinetics in man.

Relationships between the hepatic intrinsic clearance or blood cell-plasma partition coefficient in the rabbit and the lipophilicity of basic drugs.

The relationships between drug lipophilicity and hepatic intrinsic clearance (CLint,h) or red blood cell-plasma partition coefficients (D) have been elucidated for ten highly lipophilic basic drugs with apparent octanol-water partition coefficients at pH 7.4 (Papp,oct) or 150 or above. The true octanol-water partition coefficients of the non-ionized drugs (Poct) were used to determine CLint,h and D for the unbound drugs (CLint,h,f and Df, respectively), and CLint,h,f and Df for the non-ionized and unbound drugs (CLint,h fu and Dfu, respectively). The total clearance values were determined at steady state by infusion studies of individual drugs in rabbits. There was better correlation between log Poct and log CLint,h,fu (r = 0.974) than between log Poct and log CLint,h,f (r = 0.864). The D values were calculated from the blood-plasma concentration ratio. There was a better correlation between log Poct and log Dfu (r = 0.944) than between log Poct and log Df (r = 0.612). The regression equations obtained were CLint,h,fu = 0.0875 x Poct1.338 and Dfu = 0.0108 x Poct0.970, respectively. These results show that the CLint,h and D of highly lipophilic basic drugs can be predicted from Poct by taking fu into consideration. By applying these parameters to a physiologically based pharmacokinetic model it might be possible to predict the pharmacokinetics of unknown basic drugs.

Influence of ammonium chloride on the tissue distribution of anticholinergic drugs in rats.

Ammonium chloride (NH4Cl) increases lysosomal pH and thereby abolishes intralysosomal accumulation of drugs. Its effect on the tissue distribution of biperiden and trihexyphenidyl in rats has been investigated. The tissue-plasma concentration ratios (Kp) of these drugs in various tissues were determined by infusion studies at steady-state in the presence or absence of NH4Cl. Treatment with NH4Cl reduced the Kp values for both drugs, causing the largest reduction in Kp in the lung (52.1 to 11.8 for biperiden and 59.5 to 18.9 for trihexyphenidyl; ratios of decrease 0.77 and 0.68, respectively), followed by the heart and kidneys, with relatively small reductions in the brain, gut, muscle and fat. Subcellular fractionation studies in the lung indicated that the subcellular fraction-plasma concentration ratio of each drug at the steady state (K(p,sf)) was reduced by treatment with NH4Cl, with the largest decrease in the post-nuclear fraction (ratio of decrease 0.82 for biperiden and 0.74 for trihexyphenidyl), followed by the nucleus, microsomes and supernatant, in that order. A strong correlation was found between the ratio of decrease in K(p,sf) after NH4Cl treatment and the specific activity of acid phosphatases, a marker of lysosomes, in each fraction (biperiden, r = 0.948; trihexyphenidyl, r = 0.945). These results suggest that acidic organelles contribute significantly to the distribution kinetics of anticholinergic drugs.

Effect of sequence of administration on the pharmacokinetic interaction between the anticholinergic drug biperiden and [3H]quinuclidinyl benzylate or [3H]N-methylscopolamine in rats.

In rats the pharmacokinetic interactions between the anticholinergic drug biperiden and [3H]quinuclidinyl benzylate ([3H]QNB) or [3H]N-methylscopolamine ([3H]NMS) is affected by the sequence in which the drugs are administered. Drug concentrations in various tissues were determined after intravenous administration of [3H]QNB or [3H]NMS (325 ng kg(-1)). Biperiden (6.4 mg kg(-1)) was administered either 5 min before, concomitantly with or 20 min after injection of [3H]QNB or [3H]NMS. When biperiden was administered concomitantly with or before [3H]QNB, distribution of [3H]QNB among the regions of the brain and other tissues was reduced; at 4 h the ratio of the distribution of [3H]QNB for experimental animals to that for control animals ranged from 0.15 to 0.9. When biperiden was administered after [3H]QNB, the distribution of [3H]QNB in the brain and other tissues was significantly higher than for the other two treatments (P < 0.01). However, for [3H]NMS the sequence of administration had no effect on the distribution of the drug in the brain and other tissues except for the kidney. In-vitro, in crude synaptosomal membranes, the amount of [3H]QNB at 2 h relative to the control concentration at equilibrium was 87% when biperiden was added before [3H]QNB and 56% when biperiden was added after [3H]QNB. In both instances the concentration of [3H]NMS reached equilibrium within 30 min. These findings suggest that the difference between the rate constant of association and dissociation at the possible site of action gives rise to the effect of the sequence of administration on the pharmacokinetic interaction.

[Studies on the mechanism of subcellular distribution of basic drugs based on their lipophilicity].

This paper described the studies on the mechanism of subcellular distribution of lipophilic weak bases. Although the tissue distribution of basic drugs appeared to decrease with time simply in parallel with their plasma concentration, their subcellular distribution in various tissues exhibited a variety of patterns. Basic drugs were distributed widely in various tissues, but were concentrated in lung granule fraction, where their accumulation was dependent on their lipophilicity and lysosomal uptake. As the plasma concentration of drugs decreased after maximum level, the contribution of lysosomes to their subcellular distribution increased. The uptake of the basic drugs into lysosomes depended both on their intralysosomal pH and on the drug lipophilicity. As the lipophilicity of the basic drugs increased, they accumulated more than the values predicted from the pH-partition theory and raised the intralysosomal pH more potently, probably owing to their binding with lysosomal membranes with or without additional intralysosomal aggregation. These phenomena should be considered as a basis of drug interaction in clinical treatments.

Effect of fat tissue volume on the distribution kinetics of biperiden as a function of age in rats.

The relationship between the volume of fat tissue and variations in the time course of plasma biperiden concentration in rats has been examined in three different groups (4-, 10-, and 50-week-old rats). The plasma concentrations at 24 hr after iv injection of 3.2 mg/kg varied between 0.8 ng/ml (4-week-old rats) and 5.0 ng/ml (50-week-old rats). The rank order of the steady state distribution volume of biperiden was: 50-week-old rats greater than 10-week-old rats greater than 4-week-old rats. The fat volume of the whole body, extracted from the dried carcass with ether, varied between 42 g/kg (4-week-old rats) and 167 g/kg (50-week-old rats). There was a good correlation between the steady state distribution volume of biperiden per lean mass body weight and the fat volume per lean mass body weight (r = 0.987). The fat/plasma concentration ratios at 8 hr after the iv injection varied between 600 (4-week-old rats) and 200 (50-week-old rats), whereas the brain/plasma concentration ratios were identical to those at steady state among the three groups. The time courses of biperiden concentration in plasma, brain, and fat were simulated using a physiological pharmacokinetic model. There was reasonable agreement between the model predictions and the observed data, suggesting that the change in the fat volume is a dominant determinant of the distribution volume of biperiden in rats. Age-related changes in tissue and plasma concentrations are discussed in relation to the clinical usefulness of the blood level monitoring.

Fundamental pharmacokinetic properties of biperiden: tissue distribution and elimination in rabbits.

The pharmacokinetics of biperiden in rabbits were examined at three doses (0.2, 0.8, and 3.2 mg/kg i.v.). The data were interpreted in terms of a three-compartment open model with a linear excretion rate. The serum unbound fraction and the blood-to-plasma concentration ratio were determined as 0.39 and 1.2, respectively, over a wide concentration range (25-10000 ng/ml). Rapid and complete absorption from the injection site in muscle to the systemic circulation was observed. The bioavailability of muscular injection was unity. The hepatic extraction ratio was 0.94, and the high plasma clearance could be explained in terms of hepatic blood flow rate-limited elimination. The major tissues in which biperiden was distributed were fat and muscle. The highest tissue-to-plasma partition coefficient in the steady-state was obtained for the lung. These three tissues comprised 56% of the total distribution volume.

Uptake of imipramine in rat liver lysosomes in vitro and its inhibition by basic drugs.

We investigated the uptake of imipramine (IMP) in highly purified lysosomes from rat liver and its inhibition by a variety of basic drugs in vitro. The uptake of [(3)H]IMP into lysosomes peaked in less than 20 s, showing little temperature dependency or countertransport phenomena. It was accelerated by increase of extralysosomal pH, stimulated by Mg(2+)-ATP in KCl buffer, and suppressed by acidic ionophores. However, the uptake of [(3)H]IMP in lysosomes was approximately 140-fold higher than the value expected from the pH-partition theory. IMP and other weak lipophilic bases like chlorpromazine and propranolol raised the intralysosomal pH, and their potency was stronger than that of NH(4)Cl, a typical pH-perturbing weak base. A variety of basic drugs inhibited the uptakes of [(3)H]IMP and [(14)C]methylamine into lysosomes, their 50% inhibitory concentrations (IC(50)) being almost the same for [(3)H]IMP and [(14)C]methylamine uptake (r = 0.842). A high correlation (r = 0.946) was observed between the IC(50) values (for the inhibition of [(3)H]IMP uptake) and the lipophilicity (P(oct) values). These results suggest that the accumulation of lipophilic basic drugs is driven primarily by the transmembrane pH difference (pH-partition theory) but with the involvement of some additional mechanism(s) related to drug lipophilicity, possibly binding (partition or adsorption) to lipophilic substance(s) and/or aggregation within lysosomes. Based on this idea, we have established a model that described and successfully simulated the weak base-induced pH increase, the accumulation of a lipophilic weak base (IMP), and the inhibition of accumulation of IMP by lipophilic basic drugs.