Fetal hepatic drug elimination.

The majority of studies of fetal hepatic elimination have concentrated on the expression and activity of the metabolizing enzymes, but the unique physiologic milieu of the fetal liver should also be considered. The basic structure of the liver is formed by the end of the first trimester. The fetal hepatic circulation differs substantially from that of the adult in that there is an extra input vessel, the umbilical vein, and there is shunting of 30-70% of hepatic blood flow via the ductus venosus. The left and right lobes of the fetal liver seem to function independently with respect to a variety of biochemical parameters, due at least in part to the lower oxygen supply to the right lobe. The zonation of drug-metabolizing enzymes along the hepatic acinus, which is prominent in the adult liver, is absent in the fetal liver. Unlike rodent species, the human fetal liver has a significant capacity for drug metabolism. Of the oxidative enzymes, CYP3A7 accounts for up to 50% of total fetal hepatic cytochrome P450 content. Expression of this enzyme decreases dramatically after birth. CYP1A1 and CYP2D6 have also been detected in human fetal liver, but whether CYP2E1 is expressed remains controversial. Several other cytochrome P450s have been identified and await characterization. Fetal hepatic drug conjugation may prolong fetal exposure to the metabolites produced, which, being more water soluble, do not readily cross the placenta back to the mother and, if excreted in fetal urine, can be recycled in the fetus via amniotic fluid and fetal swallowing. Limited activity of glucuronidation enzymes has been demonstrated in human fetal liver in contrast to the activity of sulfation enzymes, which is significant. Limited in vivo studies in fetal sheep have demonstrated significant fetal hepatic drug elimination, and this has been confirmed in studies of the isolated perfused fetal sheep liver. Our understanding of fetal hepatic elimination processes has advanced steadily over the years. Future developments, however, should consider more fully the influence of the unique physiological milieu of the fetal liver, in addition to the expression and activity of drug metabolizing enzymes.

Role of cytochrome P450 2D6 (CYP2D6) in the stereospecific metabolism of E- and Z-doxepin.

The tricyclic antidepressant, doxepin, is formulated as an irrational mixture of E (trans) and Z (cis) stereoisomers (85%: 15%). We examined the stereoselective metabolism of doxepin in vitro, with the use of human liver microsomes, recombinant CYP2D6 and gas chromatography-mass spectrometry. In human liver microsomes over the concentration range 5-1500 microM, the rate of Z-doxepin N-demethylation exceeded that of E-doxepin above 100 microM in two of three livers. Eadie-Hofstee plots were curvilinear indicating the involvement of several enzymes in N-demethylation. Coincubation of doxepin with 7,8-naphthoflavone and ketoconazole reduced the rates of N-demethylation of E- and Z-doxepin by 30-50% and 40-60%, respectively, suggesting the involvement of CYP1A and CYP3A4, whilst quinidine had little effect on N-demethylation. In contrast, doxepin hydroxylation was exclusively stereo-specific; E-doxepin and E-N-desmethyldoxepin were hydroxylated with high affinity in liver microsomes and by recombinant CYP2D6 (Km in the range of 5-8 microM), but there was no evidence of Z-doxepin hydroxylation. In 'metabolic consumption' experiments with liver microsomes (having measurable CYP2D6 activity) and initial substrate concentration of 1 microM, the consumption of E-doxepin was greater (P < 0.05, n = 5) than that of Z-doxepin. Quinidine inhibited the consumption of E-doxepin but did not affect the consumption of Z-doxepin. With N-desmethyldoxepin, quinidine inhibited the consumption of E-N-desmethyl-doxepin whereas Z-N-desmethyldoxepin appeared to be a terminal oxidative metabolite. In summary, CYP2D6 is a major oxidative enzyme in doxepin metabolism; predominantly catalysing hydroxylation with an exclusive preference for the E-isomers. The relatively more rapid metabolism of E-isomeric forms, and the limited metabolic pathways for the Z-isomers may explain the apparent enrichment of Z-N-desmethyldoxepin that is observed in vivo.

Metabolism of dexfenfluramine in human liver microsomes and by recombinant enzymes: role of CYP2D6 and 1A2.

Dexfenfluramine has been widely used as an appetite suppressant in the treatment of obesity. It was recently shown that the apparent non-renal clearance of dexfenfluramine was significantly lower in poor metabolizers than in extensive metabolisers of debrisoquine which suggested the involvement of the polymorphically expressed enzyme, CYP2D6, in dexfenfluramine metabolism. In this study, human liver microsomes and yeast-expressed recombinant enzymes were used to examine dexfenfluramine metabolism in vitro. In human liver microsomes, the major product of dexfenfluramine was nordexfenfluramine with lesser amounts of a novel metabolite, N-hydroxynordexfenfluramine, and ketone and alcohol derivatives being formed. Eadie-Hofstee plots (v against v/[s]) of nordexfenfluramine formation between 1 and 1000 microM substrate concentration were biphasic in three of four liver microsome samples examined, with mean Km values of 3 and 569 microM for the high and low affinity enzymes, respectively. At a substrate concentration (0.5 microM) around the known therapeutic plasma concentration, there was negligible inhibition of microsomal dexfenfluramine N-dealkylation by sulphaphenazole and ketoconazole, but between 33 and 100% inhibition by quinidine, and 0-58% inhibition by 7,8-naphthoflavone in seven liver samples. In human liver microsomes, there was also a significant correlation (rs= 0.79, n = 10, P < 0.01) between dextromethorphan O-demethylation and dexfenfluramine (at 1 microM) N-dealkylation activities. Dexfenfluramine was a specific inhibitor (IC50 46 microM) of CYP2D6-mediated dextromethorphan O-demethylation in human liver microsomes but did not appreciably inhibit six other cytochrome P450 isoform-selective activities for CYP1A2, 2A6, 2C9, 2C19, 2E1 and 3A activities in human liver microsomes. Yeast-expressed recombinant human CYP2D6 metabolized dexfenfluramine with high affinity (Km 1.6 microM, Vmax 0.18 nmol min(-1) nmol P450(-1)) to nordexfenfluramine which was the sole product observed. Recombinant CYP1A2 was a lower affinity enzyme (Km 301 microM, Vmax 1.12 nmol min(-1) nmol P450(-1)) and produced nordexfenfluramine with small amounts of N-hydroxynordexfenfluramine. This is the first detailed study to examine the in-vitro metabolism of dexfenfluramine in human liver microsomes and by recombinant human P450s. We were able to identify CYP2D6 (high affinity) and CYP1A2 (low affinity) as the major enzymes catalysing the N-dealkylation of dexfenfluramine in human liver microsomes.

Elimination of amphotericin B in impaired renal function.

The influence of impaired renal function on the steady-state plasma clearance of amphotericin B was determined in seven patients with creatinine clearances ranging from zero to normal. Contrary to previous reports, steady-state plasma concentrations of total drug were lower in uremic patients than in patients with normal renal function. Total plasma clearance of amphotericin B ranged from 16.7 to 39.9 ml/min, correlated directly with the plasma creatinine concentration, and correlated inversely with the creatinine clearance. Urinary excretion of unchanged drug accounted for less than 10% of the dose. In 10 healthy subjects, mean percent of amphotericin B unbound in plasma was 3.55 +/- 0.32 (SD). Binding was determined in a further group of 10 uremic patients. Mean unbound percent (4.15 +/- 0.73, SD) was higher than in the healthy subjects, and the binding ratio (molar concentration of bound to unbound drug) correlated weakly with the creatinine clearance. This suggests that plasma clearance of unbound amphotericin B and, therefore, steady-state plasma concentrations of unbound drug are not affected by renal impairment, and that dosage requirements will be overestimated if based on measurements of total drug plasma concentration.

Catalytic activities of human debrisoquine 4-hydroxylase cytochrome P450 (CYP2D6) expressed in yeast.

A 1.57kb BamH1 fragment containing a full-length human debrisoquine 4-hydroxylase cytochrome P450 (CYP2D6) cDNA was inserted into the BglII site of the yeast expression plasmid pMA91 and the resulting recombinant plasmid, PELT1, introduced into Saccharomyces cerevisiae strain AH22. Microsomes prepared from AH22/pELT1 cells gave an absorption maximum at 448 nm and a P450 content of 67 +/- 31 pmol/mg of microsomal protein. No P450 was detectable in microsomes prepared from AH22/pMA91 control cells. A western blot of microsomes prepared from yeast transformed with pELT1 were probed with a monoclonal antibody to CYP2D6 and revealed a strong band with a molecular mass consistent with that of CYP2D6 from human liver microsomes. No corresponding band was observed with microsomes from control yeast transformed with pMA91 alone. Microsomes from AH22/pELT cells showed catalytic activity towards metoprolol (alpha-hydroxylation and O-demethylation, 0.17 and 0.78 nmol/mg protein/h, respectively); and towards sparteine (2- and 5-dehydrogenation, 1.82 and 0.59 nmol/mg protein/h, respectively). The inhibition of metoprolol metabolism by quinidine (Qd) was 200 times more potent than that of quinine (Qn), both for alpha-hydroxylation (Qd IC50 = 0.05 microM; Qn IC50 = 4 microM) and O-demethylation (Qd IC50 = 0.05 microM; Qn IC50 = 4 microM). Negligible metabolism of tolbutamide and S-mephenytoin, substrates of the 2C sub-family, and of p-nitrophenol, a substrate of CYP2E1, was detected, although a trace of the N-deethylated metabolite of lignocaine, thought to be metabolised by CYP3A4, was detected with microsomes from CYP2D6-expressing yeast cells. The results indicate that yeast cells containing human CYP2D6 cDNA express a functionally active form of the enzyme, the immunochemical and catalytic properties of which are consistent with those of human liver.

Potent inhibition of yeast-expressed CYP2D6 by dihydroquinidine, quinidine, and its metabolites.

The inhibitory effects of dihydroquinidine, quinidine and several quinidine metabolites on cytochrome P450 2D6 (CYP2D6) activity were examined. CYP2D6 heterologously expressed in yeast cells O-demethylated dextromethorphan with a mean Km of 5.4 microM and a Vmax of 0.47 nmol/min/nmol. Quinidine and dihydroquinidine both potently inhibited CYP2D6 metabolic activity (mean Ki = 0.027 and 0.013 microM, respectively) in yeast microsomes and in human liver microsomes. The metabolites, 3-hydroxyquinidine, O-desmethylquinidine and quinidine N-oxide also inhibited CYP2D6, but their Ki values (0.43 to 2.3 microM) were one to two orders of magnitude weaker than the values for quinidine and dihydroquinidine. There was a trend towards an inverse relationship between Ki and lipophilicity (r = -0.90, N = 5, P = 0.07), as determined by the retention-time parameter k' using reverse-phase HPLC. Thus, although the metabolites of quinidine have the capacity to inhibit CYP2D6 activity, quinidine and the impurity dihydroquinidine are the important inhibitors of CYP2D6.

Hepatic uptake and excretion of [14C]sodium taurocholate by the isolated perfused fetal sheep liver.

We have developed an in situ isolated perfused fetal sheep liver preparation to study fetal hepatic function free from the confounding influences of the mother and other fetal organs, and we have used the preparation to study the fetal hepatic clearance and biliary excretion of sodium taurocholate (TC). The viability and stability of this model were established by monitoring perfusion pressure, oxygen consumption, perfusate enzymes and electrolytes, the perfusate concentration ratio of lactate to pyruvate, bile flow, and liver histology. Perfusate delivery was 300 mL/min with a mean value of 3.94 mL/min/g liver (range: 2.46-6.72 mL/min/g liver). Gadolinium radiolabeled 15 microns microspheres were used to quantify the ductus venosus shunt through the liver and to determine relative flow rates between right and left hepatic lobes. TC was added to the reservoir either as a [14C]TC tracer bolus dose (2 microCi, N = 5) followed by a constant infusion of unlabeled TC, or as an initial bolus of [14]TC (54 mumol) followed by a [14C]TC constant infusion (30 mumol/hr, specific activity 30 microCi/mmol; N = 3). Perfusate samples were taken from the reservoir every 15 min and bile was collected in 30 min aliquots. Perfusion pressure (7.9 +/ 0.30 mmHg), perfusate potassium and oxygen consumption (0.9 +/- 0.07 mumol/min/g liver) were constant throughout, and the perfusate lactate/pyruvate concentration ratio was low (< 20). Liver histology showed no hypoxic changes. Bile flow fell slightly over the 150 min experiment time from 0.6 to 0.5 muL/min/g liver. These data indicate preparation viability and stability. The extent of the ductus venosus shunt was 16-66% (mean 35 +/- 6%) of umbilical vein flow, which correlated inversely with fetal gestational age (r = 0.94, P < 0.001). Relative flow to right and left lobes of liver was 1:1.4. In bolus dose experiments, TC t1/2 was 81.6 +/- 26 min, clearance (Cl) was 35.0 +/- 22.6 mL/min, shunt corrected extraction (E*) was 0.29 +/- 0.17 and biliary clearance (ClB) was 35.5 +/- 19.5 mL/min. In constant infusion experiments the corresponding results were Cl: 34.7 +/- 18.2, E*: 0.23 +/- 0.16, and ClB 32.7 +/- 17.7. The cumulative biliary excretion of [14C]TC in bolus dose experiments was 86.5 +/- 8.7% of the dose, and in constant infusion experiments, concentration of TC in bile was on average over 800 times that in plasma.(ABSTRACT TRUNCATED AT 400 WORDS)

The expression of human cytochrome P450IA1 in the yeast Saccharomyces cerevisiae.

Data from animal studies suggest that cytochrome P450IA1 catalyses the metabolic activation of several procarcinogenic compounds. In the present study, we have expressed human cytochrome P450IA1 in yeast cells. A 1.70 kb BclI/BamHI fragment containing a full-length human cytochrome P450IA1 cDNA was inserted into the BglII expression site of the yeast expression plasmid pMA91 thereby allowing the ATG initiation codon to be located adjacent to the PGK (phosphoglycerate kinase) promoter. The resulting recombinant plasmid, pCK-1, was introduced into Saccharomyces cerevisiae strains ATCC 44773 and AH22. Microsomes prepared from yeast transformatants of strain ATCC 44773 contained undetectable levels of cytochrome P450. In contrast, microsomes from strain AH22 contained cytochrome P450 with a specific content of 33.3 +/- 10.8 pmol/mg of microsomal protein and showed a reduced carbon monoxide difference spectrum with a peak at 448 nm. Control yeast cells transformed with pMA91 showed no cytochrome P450. Western blots were carried out using an antibody that reacts against rat cytochrome P450IA1 and an antibody that reacts against a synthetic peptide representing a short sequence of human cytochrome P450IA1. A band with a molecular weight of 54 kD was observed in microsomes of yeast transformed with pCK-1, but not with pMA91. When microsomes from yeast transformed with pCK-1 were incubated with benzo(a)pyrene (10 min, 10-160 microM), an estimated Km value of 7 microM was obtained. The availability of yeast cells with functionally active human cytochrome P450IA1 will facilitate molecular structure-activity studies of procarcinogen and drug metabolism by this enzyme in man.

Quinidine single dose pharmacokinetics and pharmacodynamics are unaltered by omeprazole.

Omeprazole has been shown in previous studies to inhibit the hepatic metabolism of selected drugs. Quinidine is an antiarrhythmic and antimalarial agent with a low therapeutic index. We therefore examined the effect of 40 mg omeprazole daily for one week or placebo on the pharmacokinetics and pharmacodynamics of a single 400 mg dose of quinidine in 8 healthy volunteers in a double-blind crossover study. During placebo and omeprazole treatment, there was no significant difference in area under the time-plasma quinidine concentration curve, (17.0 +/- 4.83 micrograms.h/ml, 18.6 +/- 4.43 micrograms.h/ml, respectively; P greater than 0.2) or renal clearance of quinidine (56.2 +/- 26.0 ml/min, 55.6 +/- 12.7 ml/min, respectively; P greater than 0.5). Quinidine unbound fraction in plasma (0.170 +/- 0.041 vs. 0.166 +/- 0.041 in the presence of omeprazole; P greater than 0.5) was not altered by omeprazole. Peak plasma quinidine concentration and the time this occurred did not differ. Omeprazole also had no effect on these parameters for the metabolite 3-hydroxyquinidine. There was no significant difference in the change in the corrected Q-T interval on the electrocardiogram due to quinidine (mean area under the time versus delta Q-Tc curve = 351 +/- 192 ms.h, placebo; 414 +/- 303 ms.h, omeprazole) showing that quinidine pharmacodynamics were unaltered by omeprazole. We conclude that omeprazole does not affect the pharmacokinetics of quinidine.

Influence of maternal plasma protein binding on fetal unbound plasma concentration of propranolol in the pregnant ewe.

According to theory, for a drug of nonrestrictive, flow-limited clearance, a change during pregnancy of the unbound fraction (fu) of drug in maternal plasma should cause a change in steady-state unbound plasma drug concentration (Cu) in maternal plasma, which should also cause a change in fetal Cu. This theory was examined in 14 chronically cannulated, unanesthetized pregnant ewes in which 28 separate experiments were performed during the latter part of gestation. An initial bolus dose and 3-h constant rate infusion of propranolol were administered via the maternal jugular vein and steady-state maternal and fetal carotid arterial plasma total and unbound propranolol concentrations were measured. Fetal Cu (32 +/- 21 ng/mL) was significantly less than maternal Cu (78 +/- 52 ng/mL), due to previously demonstrated fetal hepatic extraction of propranolol. Notwithstanding fetal elimination, there was a significant correlation between fetal Cu and maternal Cu (r = 0.41, p less than 0.025). There was also a strong correlation between fetal Cu and the maternal unbound fraction of drug (fu; r = 0.75, p less than 0.001). We conclude that for propranolol, a drug of nonrestrictive, flow-limited clearance, changes in maternal fu can have a significant influence on fetal Cu, and therefore would be expected to influence the pharmacological effect of the drug in the fetus.

Disposition of the diastereoisomers quinine and quinidine in the ovine fetus.

The disposition of the diastereoisomers quinine and quinidine was investigated in the near-term pregnant ewe. Five sheep were administered quinine and quinidine separately in random order by a combination of bolus and 30-h iv infusion. On a subsequent occasion, four of the five sheep were also administered the two drugs simultaneously. After separate dosage, systemic clearance of quinine tended to be greater than that of quinidine (714 +/- 299 versus 422 +/- 146 mL/min, p = 0.08). Maternal renal clearance exhibited no stereoselectivity and represented less than 2% of total clearance. Simultaneous administration did not alter the disposition of either drug in the mother. After separate dosage, fetal total concentrations (Cf) of quinine and quinidine were substantially lower than maternal total concentrations, as reflected in Cf:Cm ratios of 0.15 +/- 0.06 versus 0.10 +/- 0.08, respectively. Similarly, fetal unbound concentrations (Cfu) were substantially lower than maternal unbound concentrations (Cmu; Cfu/Cmu = 0.46 +/- 0.09 for quinine and 0.23 +/- 0.09 for quinidine). This indicates the presence of fetal elimination of both isomers. Fetal renal clearances of quinine and quinidine were similar (0.34 +/- 0.24 mL/min versus 0.38 +/- 0.24 mL/min) and less than that of endogenous creatinine, indicating the absence of net renal tubular secretion. After simultaneous dosage of quinine and quinidine, Cf:Cm (0.48 +/- 0.24 and 0.31 +/- 0.19, respectively) and Cfu:Cmu (0.73 +/- 0.14 and 0.52 +/- 0.20, respectively) were greater than for separate dosages. Fetal renal clearance of both drugs was unchanged, suggesting that the higher Cfu:Cmu ratios after simultaneous dosage were due to mutual inhibition of the fetal metabolism of these drugs.

Neonatal hepatic propranolol elimination: studies in the isolated perfused neonatal sheep liver.

Using the isolated perfused neonatal sheep liver model, we examined the disposition of propranolol (n = 8, age 0.25-10 days) and compared our findings with our previous study from the perfused near-term fetal sheep liver (Ring JA, et al. 1995. Drug Metab Dispos 23:190-196). Within 45 min of dosage, perfusate propranolol levels had fallen by three orders of magnitude to be less than the limit of detection. Perfusate disappearance curves were monoexponential in six experiments and biexponential in two experiments. The mean shunt-corrected hepatic extraction ratio was 0.92 +/- 0.09, much greater than that seen in the fetal sheep liver (0.26 +/- 0.13, P < 0.0001) but still less than values in the adult sheep (0.97). At the conclusion of the perfusion, 4-hydroxypropranolol was the major metabolite present and 5-hydroxypropranolol and N-desisopropylpropranolol were minor metabolites. We conclude that the isolated perfused neonatal sheep liver is a useful model with which to study the maturation of neonatal hepatic drug oxidation. Our study shows that propranolol is rapidly eliminated by the neonatal liver to form several metabolites at rates far greater than in the fetal liver, but rates of elimination have not yet reached that reported in the adult sheep liver.

Conjugation of para-nitrophenol by the isolated perfused neonatal sheep liver.

We examined the metabolism of para-nitrophenol (PNP) in the isolated perfused neonatal sheep liver (n = 8, 0.25-11 days) and compared the findings with our previous data from the perfused near-term fetal sheep liver (Ring, J. A., et al. Drug Metab Dispos 1996, 24, 1378). A three-step dosage regimen was used (72, 144, and 288 micromol of PNP). At the end of each dosage phase, PNP had fallen below detectable levels, and 101 +/- 16% of the dose was accounted for as PNP conjugates. Elimination of PNP from perfusate varied with dose. Elimination was first order with the 72-micromol dose; with the 144-micromol dose, elimination was first order in four livers but Michaelis-Menten kinetics in the remaining four. With all the 288-micromol doses, elimination was Michaelis-Menten and gave the following biochemical parameters: K(m) = 255 +/- 138 microM (fetal = 14.7 microM, P < 0.01), V(max) = 515 +/- 285 nmol/min/g liver (fetal = 34.3 nmol/min/g liver, P < 0.01), and intrinsic hepatic clearance = 2.36 +/- 1.21 mL/min/g liver (fetal = 4.74 mL/min/g liver, P > 0. 05). The mean shunt-corrected hepatic extraction ratio of PNP was 0. 82 (range, 0.40-1.0) and strongly correlated with neonatal age (r = 0.90, P < 0.05). We conclude that PNP is highly extracted by the isolated perfused neonatal sheep liver at much higher efficiency than in the near-term fetal sheep, reflecting a maturation of conjugation that progresses further in the early neonatal period.

Uptake and excretion of sodium taurocholate by the isolated perfused neonatal sheep liver.

We present a model for perfusion of the isolated perfused neonatal sheep liver which allows examination of drug disposition by the intact organ. We studied the disposition of sodium taurocholate (TC) in seven neonatal lambs (ages 2-11 days) and compared the results with earlier data from the perfused fetal sheep liver (Ring, J. A. et al. Biochem. Pharmacol. 1994, 48, 667-674). Measurements of perfusion pressure, oxygen consumption, lactate:pyruvate ratio, bile flow, and liver histology indicated that the preparation was both viable and stable over a 2 h period. [14C]-labeled TC was added to the reservoir by constant infusion (30 micromol/h) and the ductus venosus shunt quantitated by injection of [153Gd]-labeled microspheres. Shunt-corrected hepatic extraction ratio of TC was 0. 56 +/- 0.14 (fetal 0.23 +/- 0.16, p < 0.005) and clearance of TC was 0.92 +/- 0.35 mL/min/g liver (fetal 0.44 +/- 0.23 mL/min/g, p < 0. 01). We conclude that the isolated perfused neonatal sheep liver is a useful experimental model which will facilitate the study of the developmental physiology and pharmacology of the liver. There is considerable maturation of the biliary excretion of TC between the late fetal and early neonatal periods in the lamb.

The inhibitory effects of ranitidine and cimetidine on propranolol elimination by the rat isolated perfused liver.

The effect of low (50 micrograms) and high (1 mg) doses of the histamine H2-receptor antagonists cimetidine and ranitidine on the first pass extraction of propranolol was studied in the rat isolated perfused liver. Both low and high dose cimetidine increased the area under the perfusate propranolol concentration time curve (AUC) 4 to 5-fold. Although low dose ranitidine did not alter propranolol AUC, high dose ranitidine increased it to the same extent as cimetidine. These results indicate that ranitidine has a clear propensity for microsomal inhibition, but one which is unlikely to be manifest at therapeutic dosage.

Models of hepatic elimination: implications from studies of the simultaneous elimination of taurocholate and diazepam by isolated rat liver under varying conditions of binding.

Various kinetic models have been developed to describe the elimination of substances by the liver, but there is no agreement about which model is the most appropriate, as experimental evaluation is incomplete and results conflict. We have shown previously that hepatic elimination of taurocholate is best described by models that embody a high degree of sinusoidal heterogeneity or mixing, whereas another study showed that elimination of diazepam is best described by models that embody a low degree of sinusoidal heterogeneity or mixing. To investigate this discrepancy we examined, in the isolated perfused rat liver, the simultaneous elimination of taurocholate and diazepam. The effect on hepatic availability of varying unbound fraction (fu) in a single pass, steady-state system was studied in six experiments for taurocholate (fu, 0.066-0.966) and diazepam (fu, 0.051-0.675) by varying perfusate albumin concentration (0-60 g l-1). As before, elimination of taurocholate was best described by models that embody a high degree of sinusoidal heterogeneity or mixing (i.e., venous equilibrium model, dispersion model with dispersion number = infinity). Diazepam elimination was best described by models that embody a low degree of sinusoidal heterogeneity or mixing (undistributed sinusoidal model, dispersion model with low dispersion number = 1.04). A third model, a distributed model incorporating heterogeneity of sinusoidal blood flows and intrinsic clearances, fitted the data for both taurocholate and diazepam. The fitted coefficients of variation of both flow and intrinsic clearance among the sinusoids were large for taurocholate (176 and 164, respectively), but for diazepam blood flow was large (200) whereas that for intrinsic clearance was small (0.807).(ABSTRACT TRUNCATED AT 250 WORDS)

Differences in the binding of quinine and quinidine to plasma proteins.

1. Little is known about the comparative plasma protein binding of the antimalarial agents quinine (QN) and its isomer quinidine (QD). We have examined the in vitro binding of QN and QD to albumin, alpha 1-acid glycoprotein, normal human plasma, and maternal and foetal umbilical cord plasma. 2. QN was more avidly bound than QD, and binding of both drugs was substantially higher to alpha 1-acid glycoprotein than to albumin, indicating that alpha 1-acid glycoprotein is the more important binding protein. 3. Protein and drug concentration dependent binding was evident for both QN and QD. The unbound fraction of both drugs fell with increasing albumin (10 to 60 g l-1) and alpha 1-acid glycoprotein (0.5 to 2.0 g l-1) concentration, and there was a marked increase in unbound fraction of QN (6 to 19%) and QD (13 to 36%) in human plasma when drug concentrations were increased over the antimalarial therapeutic range (0.5 to 10 mg l-1). 4. In human volunteer plasma, the unbound fractions of QN and QD were 7.5 +/- 2.2% and 12.3 +/- 2.3% respectively, whilst the unbound fractions for both drugs were significantly higher in maternal plasma (QN = 13.0 +/- 5.4%, QD = 18.3 +/- 2.5%) and significantly higher still in foetal umbilical cord plasma (QN = 25.7 +/- 10%, QD = 35 +/- 5.3%).

Measurement of dexfenfluramine metabolism in rat liver microsomes by gas chromatography-mass spectrometry.

A specific and useful method was developed for the determination of dexfenfluramine metabolism by microsomal systems utilising GC-MS. The synthesis of two metabolites 1-(3-trifluoromethylphenyl)propan-2-ol ('alcohol') and 1-(3-trifluoromethylphenyl)-1,2-propanediol ('diol') via straightforward routes, were confirmed by MS and NMR spectra. The conditions for extraction from alkalinised microsomal mixtures of the metabolites nordexfenfluramine, 1-(3-trifluoromethylphenyl)propan-2-one ('ketone'), alcohol and diol, their conversion to trifluoroacetate derivatives and analysis by GC-MS-SIM are described. Calibration curves were constructed between 48 and 9662 nM and fitted to quadratic equations (r2>0.999). The method precision was good over low (121 nM) medium (2415 nM) and above medium (9662 nM) concentrations for all metabolites; the within- and day-to-day coefficients of variation ranged between 2.5-12.4% and 6.7-17.5%, respectively. The accuracy, measured as bias, was very good both within- and day-to-day (range: -0.4-12.6%, 0.8-18.9%). For most metabolites, the C.V. for the assay and bias increased at 121 nM. Dexfenfluramine metabolism by rat liver microsomes was investigated using the assay method and showed a concentration dependent increase in nordexfenfluramine and ketone metabolites over the substrate range of 5-200 microM.

A carrier-protein receptor is not a prerequisite for avid hepatic elimination of highly bound compounds: a study of propranolol elimination by the isolated perfused rat liver.

The highly efficient hepatic extraction of propranolol by the isolated perfused rat liver does not diminish when albumin binding is increased from 30 to 75%. One possible explanation of this insensitivity of propranolol uptake to changes in albumin binding is the mediation of uptake of bound ligand by an albumin receptor on the hepatocyte as postulated for oleate, taurocholic acid and rose bengal. To test this hypothesis, the hepatic extraction of propranolol was studied in the isolated perfused rat liver using alpha 1-acid glycoprotein, which lacks a hepatocyte receptor, as the carrier protein in the perfusate rather than albumin. Livers were perfused with a medium containing propranolol (4 microM) and varying concentrations of alpha 1-acid glycoprotein (0 to 25 microM). Hepatic extraction of propranolol was very high (0.990 +/- 0.006; mean +/- S.D.) and did not alter significantly despite an increase in bound fraction from 0.2 to 0.8, thus closely paralleling the findings when albumin is the carrier protein. This result indicates that bound propranolol is efficiently cleared by the liver, presumably by a "free intermediate" mechanism, in the absence of a specific carrier-protein receptor on the hepatocyte. This study does not, therefore, support the albumin receptor hypothesis.