Thesis (Ph. D.)--Ohio State University, 2005. / Available online via OhioLINK's ETD Center; full text release delayed at author's request until 2008 Nov 30

There are many benefits of predicting pharmacokinetic properties before administration to man. Recently it has become apparent that the in vivo clearance of drugs from in vitro metabolic data can be predicted with some degree of accuracy. To substantiate and extend this approach, a database of metabolic in vitro and in vivo data in rat was established for a total of 35 drugs. The database was then utilised to validate in vitro intrinsic clearance (CLint) scaling factors (1200M hepatocytes/SRW and 660mg microsomal protein/SRW). Poorly predicted drugs were then identified by applying discrimination criteria for predictive success of 50% under prediction to 100% over prediction. Of those, ethoxycoumarin (EC) and phenacetin (PNC) were chosen for closer examination with their in vivo and in vitro estimates of CLint being determined using the isolated perfused liver preparation (IPL), freshly isolated hepatocytes and hepatic microsomes. The CLints obtained for both EC and PNC from the IPL experiments under predicted the observed CLints but gave reproducible results and demonstrated the ability to reflect the trends of increasing CLint with decreasing dose and induction with beta-Naphthoflavone (betaNF). The IPL system used in the experiments was a simple one in terms of the perfusate. There were no blood cells or protein added nor was monitoring of oxygen and pH carried out. Therefore, if the IPL was to be used as a substitute for an in vivo study to determine hepatic clearance it may be necessary to use a more sophisticated system. EC depletion data from hepatocytes gave a good 84% prediction of in vivo CLint with microsomes giving a poor 38% prediction, due possibly, to end product inhibition. However, on induction with BNF, hepatocytes and microsomes both gave good predictions of 130 and 70% respectively. It was suggested that induction with BNF induces CYP isoforms which are less susceptible to end product inhibition. Predictions based upon 7-hydroxycoumarin formation data were consistent with those carried out previously and expectedly under predicted CLint. This is consistent with the existence of other pathways of metabolism. These studies demonstrated the utility of drug depletion data to predict in vivo CLint when the complete metabolic fate of a drug is unknown. The metabolism of PNC was monitored via the formation of paracetamol. When compared to the in vivo CLint, corrected for the fraction metabolised to paracetamol, hepatocytes gave a prediction of 50% with microsomes giving a prediction of 31%. End product inhibition was investigated and was shown not to be a factor in the microsomal prediction. However, recently published studies demonstrated that paracetamol could be produced from routes other than via direct O-deethylation of PNC. A further correction was made to the in vivo CLint to take into account the role of futile deacetylation which then allowed an acceptable microsomal prediction of 50% to be made. The predictions of EC and PNC CLint arising from the work carried out in this thesis were an improvement on those previously obtained by other workers. If the 50% under prediction to 100% over prediction acceptance criteria are applied, only the microsomal prediction of EC CLint would be considered as an unacceptable prediction. Use of the drug depletion approach, as demonstrated with EC, may become the method of choice to predict in vivo clearance as it overcomes metabolite identification and separation difficulties often encountered during drug metabolism studies.

Charles University Faculty of Pharmacy in Hradec Králové Department of Pharmacology & Toxicology Student: Lucie Ungerová Supervisor: PharmDr. Jana Pourová, Ph.D. Title of diploma thesis: The pharmakokinetics of isoflavonoids The aim of this diploma thesis is to summarize currently available informations about the pharmacokinetics of the polyfenolic compounds from a large group of flavonoids, isoflavonoids. Isoflavonoids are coumpounds found in many plant families, but especially in some members of Fabaceae family. One of the main well-known sources are soy beans containing the most famous isoflavones - genistein and daidzein, their pharmakokinetics is the main theme of this diploma thesis. Since isoflavonoids are mainly contained in the diet, their oral administration is very relevant. Isoflanonoids in the human diet occurs predominantly in the form of glykosides and therefore, after ingestion, they have to be deglycosylated by β- glucosidase enzyme family to aglycones. After that they are able to pass through the intestine into the systemic circulation and they are available for action. This is already happening in the oral cavity and the absorption occurs in the small intestine. The intestinal bacterial microflora composition is also very important for pharmacokinetics, for example I can mention...

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / O perfil farmacocinÃtico de duas diferentes formulaÃÃes de norfloxacino - comprimido revestido de 400 mg - como formulaÃÃo teste versus a formulaÃÃo referÃncia, Floxacin comprimido revestido de 400 mg (Merck Sharp & Dohme) foi avaliado em 32 voluntÃrios saudÃveis de ambos os sexos. A pesquisa consistiu de um estudo aberto, aleatorizado, cruzado, com 2 tratamentos, 2 perÃodos (duas sequÃncias) com intervalo mÃnimo de sete dias entre os internamentos. Amostras de sangue foram coletas em intervalos regulares para construÃÃo das curvas de concentraÃÃo plasmÃtica versus tempo. As concentraÃÃes de norfloxacino foram analisadas por Cromatografia LÃquida de Alta EficiÃncia acoplada à Espectrometria de Massa (LC-MS/MS). A mÃdia geomÃtrica do Norfloxacino teste/Norfloxacino referÃncia e o intervalo de confianÃa (IC) de 90% para Cmax e ASCo-t Ãltimo foram 103,9020% (89,5247-120,5881%) e 99,3030% (87,8186-112,2892%) respectivamente. Considerando que as formulaÃÃes estudadas apresentaram razÃo entre as mÃdias de Cmax e ASC0-tÃltimo entre 80 e 125%, e considerando um intervalo de confianÃa de 90% como proposto pelo Food and Drug Administration e ANVISA, conclui-se que as duas formulaÃÃes de norfloxacino apresentaram biodisponibilidades semelhantes tanto na taxa quanto na extensÃo de absorÃÃo. / The pharmacokinetic profile of two different formulations of Norfloxacin tablets - 400 mg; as test formulation versus FloxacinÂ, Norfloxacin reference formulation (- 400 mg; Merck Sharp & Dohme) was evaluated in 32 healthy volunteers of both sexes. The research was consisted of an open, randomized, crossover study, designed with two treatments, two periods (two sequences) with a minimum interval of seven days (washout) between the admissions. Blood samples were collected at regular intervals to construct the plasma concentration versus time curves. The concentrations of Norfloxacin were analyzed by High Performance Liquid Chromatography coupled to Mass Spectrometry (HPLC-MS/MS). The geometric mean of test Norfloxacin / reference Norfloxacin and confidence interval (CI) of 90% for Cmax and AUC0-t last were 103.9020% (89.5247 to 120.5881%) and 99.3030% (87.8186 - 112.2892%) respectively. Whereas the studied formulations showed the ratio between the mean Cmax and AUC0-t last between 80 and 125%, considering a confidence interval of 90%, as proposed by the Food and Drug Administration and ANVISA, it appears that the two formulations of Norfloxacin showed similar bioavailability in rate and in extent of absorption.

Erythromycin, a macrolide antibiotic isolated from Streptomyces erythreus, was first introduced into clinical medicine in 1952. It is active against most gram-positive bacteria, some gram-negative bacteria and is currently the agent of choice for Legionella pneumophila. Erythromycin is an acid-labile compound rapidly degrading in acidic solutions such as the acid environment of the stomach. As such, erythromycin absorption following oral administration of solid dosage forms is relatively poor. Accordingly there have been various approaches used to protect the drug against gastric inactivation. These precautions include enteric-coating of tablets, capsules or pellets of erythromycin base, the synthesis of acid stable 2' esters of erythromycin (ethylsuccinate and propionate) and salts of these esters (erythromycin estolate), and more recently, the synthesis of a range of new acid-stable, semi-synthetic macrolide antibiotics. The 2' esters are antimicrobially inactive or much less active than the parent compound and must be converted to the free erythromycin base in vivo in order to exhibit antibacterial activity. Intrinsic dissolution rates determined on raw material can provide extremely useful information relating to the gastrointestinal absorption of drugs from solid dosage forms. The large inter- and intrasubject variability associated with erythromycin base has, to date, mainly been attributed to gastric acid inactivation of the drug. However, changes in duodenal pH resulting in altered solubility and intrinsic dissolution rates may account for the observed variability. Thus, the intrinsic dissolution rates of erythromycin base at pH 6.0, 6.5, 7.0, 7.5 and 8.0 were compared in order to investigate the possible effects of pH changes which may occur in the duodenal contents, on the in vivo dissolution and subsequent absorption of this compound. The standard intrinsic dissolution rate test procedure employing a rotating disc of pure erythromycin base powder which only allows for dissolution from a constant surface area, was adapted and the drug quantitatively determined by reversed phase high performance liquid chromatography (HPLC) using ultraviolet detection. Results of intrinsic dissolution studies at both 22°C and 37°C indicate that the solubility, and therefore the rate of dissolution of erythromycin base is pH dependent, being more soluble at pH 6.0 than pH 8.0 (an approximate 800 times and 1000 times reduction in the amount dissolved after 30 minutes, at 22°C and 37°C respectively, when the pH of the medium was increased from 6 to 8). Although the stability of erythromycin and its ester derivatives in aqueous acidic solutions has been well documented, very little has been reported on the compound's stability in organic solvents. Methanol is recommended by official drug compendia (U.S.P. and B.P.) for use in erythromycin identification tests as well as in the sample preparation steps during assay procedures. Thus, the effect of methanol and acetonitrile, organic solvents of similar polarities and densities, on the stability of erythromycin base, erythromycin ethylsuccinate, propionyl erythromycin and erythromycin estolate at room temperature (22°C ± 0.5°C), using HPLC with electrochemical detection, was investigated. Erythromycin base is relatively stable in both methanol and acetonitrile, remaining intact for over 168 hours in acetonitrile and showing less than 5% degradation in methanol over the same period. Erythromycin ethylsuccinate in acetonitrile shows less than 5% degradation over 168 hours whereas in methanol, rapid hydrolysis occurs resulting in almost total conversion to base within 40 hours. Approximately 87% of erythromycin propionyl ester remained intact after 168 hours in acetonitrile whilst methanol caused rapid hydrolysis to erythromycin base (35% remaining after 28 hours). Erythromycin estolate appeared to be unstable in both acetonitrile and methanol. In acetonitrile, only 13% of the estolate remained intact after 168 hours, whereas in methanol, the reaction was much more rapid with 35% of the estolate remaining after 28 hours. The use of methanol as a solvent for erythromycin estolate reference standards is thus contraindicated. A number of conflicting reports on the half- life as well as the body compartment model that best describes erythromycin base serum concentration-time profiles (lBCM generally used to describe orally administered erythromycin, whilst a 2BCM has been used to describe erythromycin administered intravenously), appear in the literature. These differences may be largely attributed to the sampling period (between 6 and 12 hours) used in the repective studies. The objective of this study was to determine the body compartment model that best describes erythromycin base serum concentration-time curves by increasing the sampling time to 24 hours. In addition, the effect of chronic dosing of erythromycin on erythromycin pharmacokinetics, in the same group of subjects, was investigated. The single and multiple oral dose pharmacokinetics of erythromycin enteric coated base pellets within a gelatin capsule (250mg), were studied in 6 healthy, normal volunteers (19.5 ± 0.76 years, 71.5 ± 8.18 kg, 180.33 ± 5.99 cm). Furthermore, steady state concentrations were predicted using the pharmacokinetic parameters obtained from the single dose study, and compared with those obtained in the multiple dose study. Plasma concentrations were determined using a sensitive high-performance liquid chromatographic method with electrochemical detection. For the single dose study, after a tlag of 2.5 ± 0.71 hr, Cmax (1.12 ± 0.47 μ/ml) was reached at a tmax of 4.08 ± 0.93 hr post dose, with serum concentrations ranging from 0.31 - 1.62 μ/ml. The half-life was found to be 5.42 ± 1.31 hr. On multiple dosing (250mg six hourly), serum concentrations for the fifth, ninth and thirteenth dosing intervals ranged from 0.67 - 2.92 μ/ml, 1.69 - 3.65 μ/ml and 0.61 - 3.01 μ/ml, occurring at 3.75 ± 0.69 hr, 3.17 ± 1.03 hr and 3.17 ± 1.03 hr post dose with a Cmax of 1.89 ± 0.68 μ/ml, 2.35 ± 0.70 μ/ml and 1.94 ± 0.74 μ/ml, respectively. The area under the serum concentration- time curve for the single dose study (AUC₀₋∞) was 4.67 ± 0.88 hr.μ/ml, whilst the AUC₀₋τ. for the fifth, ninth and thirteenth dosing intervals of the multiple dose study were 5.77 ± 1.76 hr.μ/ml, 6.46 ± 1.33 hr.μ/ml and 5.97 ± 2.36 hr.μ/ml respectively, indicating an approximately 33% increase in AUC on chronic dosing of erythromycin. The observed increase in AUC may be a result of increased bioavailability or a decrease in clearance on chronic dosing.

Class of 2005 Abstract / Objectives: To determine the pharmacokinetic parameters of a patient population from drug concentration measurements records created by the pharmacokinetic service at a regional hospital in South Carolina, to predict peak and trough concentrations from three large dose-extended interval (LDEI) protocols to determine which method produce the highest percentage of concentrations that fall in the desired ranges, and to compare pharmacokinetic parameters of overweight and normal weight patients. Methods: This was a descriptive, retrospective study that used clinical data from 121 of 208 patient data forms. The collected data included patient age, gender, weight, height, serum creatinine (Scr), measured serum peak and trough concentrations, time of dosing, dose and dosing interval. These were used to determine individual pharmacokinetic parameters and predict peak and trough concentrations from three LDEI dosing protocols. Results: Method II produced the highest percentage of patients with peaks and troughs falling into the target range (95.9%). The Hartford method produced the highest percentage (79.3%) of patients achieving peak concentrations >20mg/L. All three methods achieved low troughs of <2mg/L, <1mg/L, <0.5mg/L, and <0.1mg/L at least 95%, 80%, 70%, and 50% of the time, respectively. No statistical significance was found between the group having actual body weight/ideal body weight ratio (ABW/IBW) greater than 1.2 and another group having ABW/IBW <1.2 for ABW, volume of distribution (V), elimination half-life (T1/2) and aminoglycoside clearance (Clag). Also, when overweight patients were excluded, a higher correlation between elimination rate constant (k) and creatinine clearance (CrCl) was found than when all patients were combined. In other words, as k increases, CrCl increases. Implications: Even though Method II produced the greatest percentage of peak and trough concentrations within its stated target range, the Hartford method may be the best dosing protocol to use since it achieves high peak concentrations (>20mg/L) while maintaining low trough concentrations. In addition, based on our data, we can assume that overweight people affect the predicted k value. There was no statistical significance between actual and predicted pharmacokinetic characteristics in overweight patients.

The investigations detailed herein have been conducted to address various aspects of the biopharmaceutics and pharmacokinetics of josamycin which to-date, have received little or no attention in the literature. Areas of investigation have included the selective determination of josamycin in serum and urine samples, the stability of josamycin in stored biological samples, intrinsic dissolution rates, solubility, acid and alkali stability and bioavailability and pharmacokinetics after dosing with a solution, powder and tablets. High performance liquid chromatography (HPLC) was used as the main analytical tool throughout these studies and proved to be highly versatile for the determination of josamycin in a number of different media. HPLC analysis afforded simple yet accurate determination of josamycin in samples from dissolution, solubility, tablet content and stability studies. Furthermore, the specificity afforded by HPLC was particularly useful for the separation of josamycin from degradation products formed in acid and alkali media. Since metabolites of josamycin are microbiologically active, microbiological assays do not determine the concentration solely of josamycin. An analytical method capable of the selective determination of josamycin in serum and urine samples is therefore required for the procurement of reliable bioavailability and pharmacokinetic data. HPLC affords this selectivity and a method for the selective determination of josamycin in serum and urine was successfully developed. The assay was simple yet precise, accurate and sensitive. Furthermore, it was well suited to the determination of josamycin in a large number of biological samples. Its success was largely due to the use of a solid phase extraction step using C₁₈ extraction columns, with a highly specific wash sequence followed by a phase separation step after elution from the extraction column. Chromatography was performed on a C₁₈ reversed-phase analytical column with UV detection of josamycin and internal standard at 231 nm and at 204 nm respectively using a programmable multi-wavelength detector. Only slight modification of the assay described should enable the selective determination of the metabolites of josamycin. This assay, therefore, lays the groundwork for future investigations into the pharmacokinetics of these metabolites. The re-usability of extraction columns was assessed in an attempt to reduce the cost of sample analysis. It was found that extraction columns could be used twice for the extraction of serum samples and up to four times for the extraction of urine samples. The difference between the re-usability of extraction columns for serum and urine samples was ascribed to various differences in the composition of the sample matrix. The stability of josamycin in stored serum and urine samples was also assessed.

Mexiletine [(2',6'-dimethylphenoxy)-2-amino propane] is a class 1 antiarrhythmic agent with a similar chemical structure and electrophysiological effects to those of lidocaine. It is a chiral drug which is used clinically in the racemic form (i.e. 50:50 ratio of two enantiomers). This thesis describes the stereoselective HPLC analysis, pharmacokinetics, tissue distribution and pharmacodynamics of mexiletine enantiomers. The development of a highly sensitive and stereoselective HPLC assay for mexiletine enantiomers, using 2-anthroyl chloride as a derivatization reagent, was attempted. The synthesis and characterization of the acid chloride was successfully carried out. The 2-anthroyl derivatives of the enantiomers were resolved on a Pirkle[formula omitted] ionic (phenyl glycine) chiral column using a mobile phase of ethyl acetate/2-propanol/Hexane (4:6:90). Detection was accomplished by fluorescense (ex = 270 nm, em = 400 nm) with a lower limit of 0.5 ng/ml. However, there was an interfering peak coeluting with S(+)-mexiletine which could not be resolved. This precluded the use of the assay for the proposed pharmacokinetic and pharmacodynamic studies. A previously developed stereoselective HPLC method, with 2-naphthoyl chloride as a derivatization reagent, was subsequently used. The in vitro protein binding of mexiletine enantiomers was examined with human serum, lipoprotein deficient serum, albumin and α[formula omitted]-acid glycoprotein. The binding of the enantiomers to human serum was moderate (45 to 50%) within the therapeutic range of mexiletine. This binding was due, mainly, to albumin and α[formula omitted]-acid glycoprotein. The free fractions of the enantiomers decreased significantly (P<0.05) as pH was increased from 7.0 to 8.0. Stereoselective binding was apparent at pH 8.0 such that the free fraction of S(+)-mexiletine was significantly (p<0.05) greater than that of the R(-)-enantiomer. However, stereoselective binding was not observed at physiological pH (≈ 7.4). These results indicated that the serum binding of mexiletine enantiomers is pH-dependent. Binding was not concentration-dependent, nor was there any competitive binding interaction between the enantiomers, within the therapeutic range. Scatchard analysis of the binding data obtained with serum and albumin both showed the presence of 2 classes of binding sites. A high affinity, low capacity site and a low affinity, high capacity site. In contrast, α[formula omitted]-acid glycoprotein showed only 1 class of binding sites and this was a high affinity, low capacity site. Pharmacokinetic and tissue distribution studies in rats following the administration of racemic mexiletine (10 mg/kg) indicated extensive tissue uptake and rapid elimination of the enantiomers. R(-)-Mexiletine showed a 32% greater systemic clearance (161.8 ml/min/kg vs 122.9 ml/min/kg) than the S(+)-enantiomer. The steady state-volume of distribution was also greater for the R(-)-enantiomer (9.0 L/kg vs 7.4 L/kg), while the elimination half-lives of the enantiomers (1.4 and 1.3 h for R(-)- and S(+)-mexiletine, respectively) were not different. Maximum tissue concentrations were observed at 5 min in all the tissues studied (heart, brain, lung, kidney, liver and fat). These concentrations were not significantly different, except for the liver tissue where a 2.4-fold greater concentration of the S(+)-enantiomer was found. High tissue/serum ratios (>20) were observed for each enantiomer in the brain, lungs and kidneys. The brain accumulated 3-fold the heart concentrations of the enantiomers. Pharmacodynamic studies on the relative antiarrhythmic effects of racemic mexiletine and its enantiomers were carried out using electrical and ischaemia-induced arrhythmias in rats. Racemic mexiletine and its enantiomers significantly (P<0.05) increased VFT and ERP. However, the differences between the effects of the 3 drugs on these variables were not statistically significant. R,S-, S(+)- and R(-)-mexiletine caused significant bradycardia and PR prolongation in both pentobarbitone anaesthetized and conscious rats. These effects of the drugs were also not significantly different from each other. In the ischaemic conscious rats, the 3 drugs did not significantly reduce the incidence of VT and VF, the number of PVCs nor the "arrhythmia score" when compared to saline (control). Racemic mexiletine and its enantiomers produced comparable CNS toxicity in the conscious rats. / Pharmaceutical Sciences, Faculty of / Graduate

Mexiletine (MexitilR) is an orally effective antiarrhythmic drug used clinically as a racemate of the R(-)- and S(+)-enantiomers. A stereoselective high-performance liquid chromatographic assay was developed for the determination of mexiletine enantiomers in serum, saliva, red blood cells and urine. The mexiletine enantiomers were resolved as their N-anthroyl derivatives on a PirkleR phenylglycine ionic chiral HPLC column. The present study examined the serum free (unbound) and total drug kinetics for the mexiletine enantiomers in twelve healthy volunteers following oral administration of 200 mg of racemic mexiletine hydrochloride. To further characterize serum free mexiletine in the body, the concentrations of mexiletine enantiomers in saliva and in red blood cells were examined. Since mexiletine was largely eliminated by metabolic processes, p-hydroxy-mexiletine and hydroxymethyl-mexiletine metabolites were examined in the urine of four healthy subjects. Following oral drug administration, the disposition of mexiletine enantiomers was described by one or two-compartment open models. The mean peak serum total mexiletine concentration of 217 ± 68 ng/ml for R(-)-mexiletine was found to be significantly greater (p<0.01) than a mean value of 196 ± 56 for S(+)-mexiletine. The mean serum total R(-)-mexiletine concentrations were also found to be significantly greater than those for S(+)-mexiletine during the first six hours. The absorption, rapid and terminal disposition kinetic parameters between the two enantiomers were not significantly different. From urinary data, the mean percentages of mexiletine enantiomers recovered from the urine over 72 hours were found to be 3.5 ± 3.4% and 3.7 ± 3.9% for R(-)- and S(+)-mexiletine, respectively. The mean terminal elimination half-lives were found to be 5.8 ± 1.5 h and 5.6 ± 1.4 h for R(-)- and S(+)-mexiletine, respectively. Both the urinary recoveries and the half-1ives for the enantiomers were not significantly different. Comparative in vitro studies on the serum protein binding of mexiletine enantiomers by ultrafiltration and by equilibrium dialysis indicated a serum pH-dependent stereoselective protein binding of mexiletine enantiomers. A serum pH range from 6.3 to 9.4 was correlated with the serum protein binding of mexiletine enantiomers from ≈30% to ≈80%. Within this pH range, the serum free drug R(-)/S(+) ratio was found to decrease from 1.0 to 0.7. At serum pH 7.4, the serum protein binding of mexiletine enantiomers was similar, and was not dependent on the therapeutic concentration range of 0.25 to 3.0 µg/ml. The in vivo serum protein binding of mexiletine enantiomers was found to be non-stereoselective. The mean serum free fractions of 0.57 ± 0.07 and 0.56 ± 0.06 for R(-)- and S(+)-mexiletine, respectively, were not significantly different. The overall mean serum free R(-)/S(+) mexiletine ratio of 1.09 was also indicative of a non-stereoselective binding of mexiletine enantiomers. Following the collection of unstimulated saliva, the overall mean saliva / serum free mexiletine area under the concentration-time curve ratios of 6.10 ± 2.82 and 7.49 + 3.48 for R(-)- and S(+)-mexiletine, respectively, were found to be significantly different (p<0.01). The overall mean saliva R(-)/S(+) ratio of 0.89 ± 0.02 (mean ± S.E.) over 48 hours suggested that the disposition of mexiletine enantiomers in saliva was stereoselective. In addition, saliva mexiletine concentrations were found to correlate poorly with serum free mexiletine concentrations. In vitro studies on the distribution of mexiletine enantiomers into red blood cells indicated a distribution equilibrium of ≈40 minutes. The mean red blood cell mexiletine area under the concentration-time curve of 2.3 ±1.5 µg/ml/h and 2.8 ± 2.1 µg/ml/h for R(-)- and S(+)-mexiletine, respectively, were not significantly different. The overall mean red blood cell mexiletine R(-)/S(+) ratio of 0.91 ± 0.13 suggested a similar distribution of the enantiomers into the red blood cells. A stereoselective HPLC assay was developed for the simultaneous determination of mexiletine, p-hydroxy-mexiletine, and hydroxymethyl-mexiletine enantiomers in urine. Mexiletine and hydroxymethyl -mexiletine enantiomers were resolved on a PirkleR isoleucine covalent HPLC column as their N-anthroyl derivatives, while p-hydroxy-mexiletine enantiomers were resolved as their 0-ethyl-N-anthroyl derivatives. For mexiletine and p-hydroxy-mexiletine, chromatographic retention was found to favour the R(-)-enantiomer, thus leading to the initial elution of the S(+)-enantiomer. For hydroxymethyl-mexiletine, the order of elution was found to be reversed. The mean cumulative amounts of p-hydroxy-mexiletine enantiomers recovered from the urine over 72 hours were found to be 1.31 ± 0.33 mg and 1.27 + 0.39 mg for the R(-) and S(+)-enantiomers, respectively. The two values were not significantly different. The mean cumulative amounts of R(-)-hydroxymethyl-mexiletine (2.94 ± 1.70 mg) recovered from the urine over 72 hours were found to be significantly (p<0.01) greater than a value of 1.17 ± 0.60 mg for the S(+)-enantiomer, suggesting the presence of a stereoselective metabolic pathway for hydroxymethyl-mexiletine. / Pharmaceutical Sciences, Faculty of / Graduate

Valproic acid (VPA), an antiepileptic drug, possesses a delay in maximum pharmacological response upon initial drug administration, and a prolonged duration of activity following discontinuation of the drug. Metabolites of VPA are thought to be involved as evidence from previous studies in mice demonstrated that (E)-2-ene VPA and (E,E)-2,3'-diene VPA, major products of VPA metabolism in serum, exerted some degree of anticonvulsant activity against pentylenetetrazole (PTZ)-induced seizures. Also associated with VPA therapy is a fatal idiosyncratic hepatotoxicity possibly involving two metabolites, 4-ene VPA and (E)-2,4-diene VPA. Preliminary tissue distribution studies had suggested that (E)-2-ene VPA may not be as hepatotoxic as VPA based solely on (E)-2-ene VPA concentrations in liver. The main objectives of this study were to investigate the kinetic and metabolic profiles, disposition, and anticonvulsant activity of VPA, (E)-2-ene VPA, and (E,E)-2,3'-diene VPA in rats. Results of these experiments were intended to provide insight into the possible contributions of these metabolites towards VPA activity or toxicity. Synthesis of (E)-2-ene VPA and (E,E)-2,3'-diene VPA was accomplished by the regiospecific addition of propionaldehyde to an ester enolate, followed by nucleophilic elimination of the mesylate ester with l,8-diazabicyclo[5.4.0]undec-7-ene or potassium hydride. The synthesis provided good yields and was stereoselective. The isomeric purity of the synthesized compounds was found to be 95 - 97% based on nuclear magnetic resonance and gas chromatographyc-mass spectrometric data. The assay of VPA and its metabolites in rat plasma and tissue homogenate extracts was achieved by negative ion chemical ionization gas chromatography-mass spectrometry. This method proved to be selective, sensitive, reproducible, and amenable to automation. In order to compare the disposition and pharmacokinetics of VPA and its analogues, VPA was administered intraperitoneally to rats and the kinetic profiles in plasma, liver, heart, lungs, and nine brain regions were determined. Selective binding of VPA to liver was observed with the liver/plasma ratio at 10 hours after dosing being 4.6. VPA did not persist in brain and the distribution in brain tissue appeared uniform. Metabolites of VPA also were not retained in brain. A most interesting observation was the absence of (E,E)-2,3'-diene VPA in brain while a minor plasma metabolite, (E,Z)-2,3'-diene VPA, was the only detectable diene. A stereoselective active transport mechanism could account for this unusual result. Present in plasma but not detected in liver was (E)-2,4-diene VPA, the hepatotoxic metabolite of VPA. It was proposed that the diene may be covalently bound to liver tissue. Following single dose administration to rats, (E)-2-ene VPA appeared to persist in all tissues assayed following an initial decline phase. The prolonged terminal elimination phase may be attributed to the extensive plasma protein binding of (E)-2-ene VPA (>99%). No selective binding of (E)-2-ene VPA in brain was observed. Brain/plasma ratios at 10 hours after dosing did not exceed 0.03. Metabolites of (E)-2-ene VPA were mainly products of β-oxidation and reduction. Both hepatotoxic metabolites were observed in plasma with concentrations of 4-ene VPA in liver higher than normally seen following VPA administration. Questions arise regarding the potential hepatotoxicity of (E)-2-ene VPA. After single dose administration of (E,E)-2,3'-diene VPA to rats, clearance of the diene was rapid compared to that of VPA or (E)-2-ene VPA. Selective binding of the diene was observed in the superior and inferior colliculus and substantia nigra but the concentrations were too low to be considered clinically significant. Reduction of (E,E)-2,3'-diene VPA appeared to be the main route of metabolism. 4-Ene VPA and (E)-2,4-diene VPA were not detected in plasma or tissues suggesting (E,E)-2,3'-diene VPA may have a lower potential for liver toxicity. The anticonvulsant activities of VPA, (E)-2-ene VPA, and (E,E)-2,3'-diene VPA were compared in rats by the PTZ-induced seizure test. Based on ED50 values, the anticonvulsant potencies of VPA and (E)-2-ene VPA were comparable and significantly greater than (E,E)-2,3'-diene VPA. The detection of (E,Z)-2,3'-diene VPA in brain following VPA administration led to the testing of this diene isomer. The potency of the (E,Z)-isomer was found to be equivalent to VPA and (E)-2-ene VPA. Sedation was a severe side effect of (E)-2-ene VPA and the (E,E)-2,3'-diene VPA was stereoselectively unique in causing skeletal muscle rigidity. Sedation was minimal and muscle rigidity was not a property of the (E,Z)-isomer over the dose range studied. Based on the results of these studies, it can be concluded that neither (E)-2-ene VPA nor (E,E)-2,3'-diene VPA is responsible for the pharmacodynamic effects of VPA. From the metabolism of (E,E)-2,3'-diene VPA and the results of anticonvulsant testing, it was proposed that (E,Z)-2,3'-diene VPA may have potential as a relatively safe and useful anticonvulsant drug. / Pharmaceutical Sciences, Faculty of / Graduate