An assay has been developed for the simultaneous quantitative analysis of VPA and twelve of its metabolites namely 2-ene VPA, 3-ene VPA, 4-ene VPA, (E)-2,4-diene VPA, (E,E)-2,3'-diene VPA, 3-OH VPA, 4-OH VPA, 5-OH VPA, 3-keto VPA, 4-keto VPA, 2-propylsuccinic acid and 2-propylglutaric acid. The assay is accomplished in a single run using the Hewlett-Packard 5987A GC-MS. VPA and its metabolites were measured using selected ion monitoring of the characteristic (M-57)⁺ ions of the tertbutyldimethylsilyl (tBDMS) derivatives. The 4-OH VPA was measured as the underivatized ϒ-lactone. [²H₆]-VPA, [²H₃]-2-ene VPA and 3-octanone served as internal standards. In a typical assay 1 mL of urine or serum is adjusted to pH 13 to hydrolyze the conjugates, then acidified to pH 2 and extracted with ethyl acetate. The solvent is then dried, concentrated to 200 μL and derivatized with tBDMS reagent. The chromatographic run time with a 25 m x 0.32 mm I.D. bonded phase OV 1701 capillary column was 22 minutes. All metabolites were well resolved and extracts of control serum or urine showed no interferences. The tBDMS and TMS derivatives of VPA metabolites were compared. Mixed tBDMS-TMS derivatives were also investigated. This assay has been applied to the analysis of urine and serum samples from pediatric patients on VPA therapy. The effects of acetyl salicylic acid (ASA) on the metabolism of VPA were investigated. Urine samples of six pediatric patients under valproic acid therapy and a normal adult subject taking valproic acid were obtained before and after administration of ASA. The analyses showed that the excretion of conjugated and unchanged VPA was enhanced in all subjects following administration of ASA. The 3-keto VPA, the second major urinary metabolite after VPA-glucuronide, was markedly reduced. The average excretion of 2-ene VPA and 3-OH VPA was also lowered. These results suggest that the β-oxidation pathway which includes 2-ene VPA, 3-OH VPA and 3-keto VPA was inhibited by salicylate at some stage prior to the formation of 2-ene VPA. Since serum samples of the normal subject were available the kinetics of VPA metabolites before and after administration of ASA were determined. It was found that the formation clearance, the fraction metabolized and the elimination clearance of the β-oxidation metabolites were decreased following ASA administration. These results confirmed the urinary data obtained with the seven subjects. Salicylate was found to inhibit the β-oxidation of VPA. Other metabolic pathways of VPA were not significantly affected by ASA administration. The mechanism by which salicylate affects the β-oxidation of valproic acid is discussed. / Pharmaceutical Sciences, Faculty of / Graduate

Valproic acid (VPA) is a major anticonvulsant drug widely used in the treatment of absence seizures. VPA is extensively metabolized in humans. Several VPA metabolites possess anticonvulsant activity and other metabolites are implicated in rare but fatal cases of hepatotoxicity. A highly sensitive and more specific analytical method was required to analyze the large number of VPA metabolites, some of which are present at trace levels. The objective of this study was to develop such a method and to make a preliminary application of the method to the determination of trace VPA levels and to search for new VPA metabolites. The suitability of analyzing halogenated derivatives of VPA and its metabolites by negative ion chemical ionization (NICI) GCMS was evaluated for the desired sensitivity and specificity. An assay was thus developed for VPA in serum and saliva based on NICI-GCMS of the pentafluorobenzyl (PFB) derivative. The NICI spectrum of the PFB ester of VPA was dominated by a single fragment ion, the m/z 143 ([M-181]⁻) ion. When the m/z 143 ion was monitored the lower limit of detection was 2 ng/mL of VPA in serum or saliva. Using [²H₆]-VPA as the internal standard, the intra- and inter-assay variations were less than 10 % at serum VPA concentrations of 10 to 800 ng/mL. Linearity was observed over the concentration range of 10 ng/mL to 25 µg/mL. The NICI assay was employed to quantitate VPA in serum (total and free) and saliva in five healthy volunteers who took part in a drug interaction study between VPA and carbamazepine (CBZ). A total of 63 paired saliva and serum samples were analyzed by NICI-GCMS; 33 before the administration of CBZ and 30 after CBZ. The % decrease in the average VPA concentration after CBZ was 27.91 ± 3.48, 36.85 ± 13.64, and 48.13 ± 7.70, for serum total, serum free and saliva VPA, respectively. There was a significant reduction (p<0.025) in the average VPA concentration in all three biological fluids. The average saliva to serum free VPA ratio was 18.92% ± 6.25 before CBZ and 16.37% ± 2.82 following CBZ. The average saliva to serum total VPA ratio was 2.43% ± 0.86 before CBZ and 1.67% ± 0.50 following CBZ, indicating that the saliva to serum total VPA ratio was concentration dependent. A strong correlation was found between saliva and both serum free (r = 0.9035 ± 0.0784) and serum total VPA (0.9058 ± 0.0450) (after CBZ). The free fraction of VPA did not increase after CBZ administration suggesting that the decrease in VPA concentration after CBZ was not related to changes in the free fraction of VPA. PFB derivative formation of VPA metabolites was facile and resulted in uniform derivatization of all metabolites studied. In the NICI mass spectra most of the ion current was carried by the [M-181]⁻ fragment ion, the only exception being that of 3-keto VPA. The base peak in the NICI spectrum of PFB derivatized 3-keto VPA was [M-181-C0₂]⁻. Isolated metabolites were identified with the help of twin ions (deuterated and undeuterated) in the mass spectra and by comparison of mass spectra and retention times with synthetic reference compounds. Urine or serum metabolites were analyzed in one chromatographic run and SIM chromatograms obtained. Serum and urine controls showed no interfering peaks and the analytical method appears suitable for a sensitive assay of VPA metabolites. The NICI method employing PFB derivatives was sensitive enough to detect VPA metabolites in saliva. Seven metabolites were detected. The ratio of Z to E isomers of 2-ene VPA was much greater in saliva than in serum (3.82 vs. 0.458), suggesting differences in the transport or plasma protein binding properties of these two isomers. A new VPA metabolite, assigned the structure 4¹-keto-2-ene VPA was detected in urine. The mass spectrum and retention time of this new metabolite matched that of one compound which was present in a synthetic mixture containing 4¹-keto-2-ene VPA. Another new metabolite which appears to be 2-(2¹-propenyl)-glutaric acid was also detected in urine. The synthesis of 4¹-keto-2-ene VPA was attempted using two different synthetic methods. The first method which involved the dehydrogenation of the 0-TMS dialkyl ketene acetal of ethyl 2-propyl-4-oxopentanoate apparently resulted in the formation of the positional isomer, 4-keto-2-ene VPA. The second synthetic route was based on the dehydration of 4-carboethoxy-2-ethylenethioketal-5-hydroxyheptane and produced 4¹-keto-2-ene VPA. However, it was not possible to isolate sufficient product for NMR characterization. / Pharmaceutical Sciences, Faculty of / Graduate

Valproic acid (VPA) is a broad spectrum antiepileptic agent used widely in the treatment of absence and tonic-clonic seizures. VPA is extensively metabolized and forms 17 metabolites in man. A monounsaturated metabolite, (E)-2-ene VPA, is at least as potent as the parent drug VPA in several animal models of epilepsy. Moreover, (E)-2-ene VPA appears to be free of two serious side effects of VPA, namely hepatotoxicity and teratogenicity. Another monounsaturated metabolite of VPA, 4-ene VPA, has been incriminated in the pathogenesis of fatal hepatic failure in children on VPA therapy. This thesis describes the synthesis of (E)-2-ene VPA and 4-ene VPA and the development of a simple and sensitive capillary gas chromatographic-mass spectrometric (GCMS) assay method for the estimation of (E)-2-ene VPA and 4-ene VPA in the biological fluids of the rat. This thesis also describes the pharmacokinetics of (E)-2-ene VPA and 4-ene VPA at two dose levels of 20 and 100 mg/kg in normal and bile exteriorized rats. A simple capillary GCMS assay method was developed that involves a single extraction of 80 µL of plasma, urine or bile with ethyl acetate followed by derivatization with MTBSTFA (N-tertiarybutyldimethylsilyl-N-methyl-trifluoroacetamide). For an 80 µL biological sample employed for extraction, the lowest detection limit for (E)-2-ene VPA was 60 ng/mL and for 4-ene VPA, 100 ng/mL. The calibration curves for (E)-2-ene VPA were linear over a fairly wide concentration range of 0.4-35 /µg/mL in plasma and 2-200 µg/ml in urine of the rat. Standard curves for 4-ene VPA were prepared in concentration ranges of 0.5-45 µg/mL in plasma and 2-80 µg/ml in urine. The assay method is reliable, reproducible, and is able to separate the diene metabolites of (E)-2-ene VPA. For pharmacokinetic studies, a single intravenous (IV) bolus dose of either (E)-2-ene VPA or 4-ene VPA was administered to normal or bile-exteriorized rats. On increasing the dose from 20 to 100 mg/kg in normal rats, the apparent plasma clearance of (E)-2-ene VPA changed from 4.9 ± 1.7 (SD) to 3.0 ± 0.3 mL/min.kg, and of 4-ene VPA decreased from 8.7 ± 0.6 to 5.9 + 0.5 mL/min.kg. A total (conjugates and unconjugates) of 32 + 6% of the low dose and 50 ± 11% of the high dose of (E)-2-ene VPA was recovered in the urine of the rat. The second metabolite, 4-ene VPA, was eliminated in the urine to a relatively smaller extent (22 ± 3% of the low dose and 28 ± 6% of the high dose). In bile-duct cannulated rats, the apparent plasma clearance of (E)-2-ene VPA was 7.7 ± 1.8 mL/min.kg at the low dose and 6.0 ± 1.1 mL/min.kg at the high dose. The corresponding values for 4-ene VPA were 11 ± 1.8 mL/min.kg and 7.4 ± 1.1 mL/min.kg, respectively. The apparent elimination half-life of (E)-2-ene VPA remained unchanged at 20-21 min at the two dose levels, compared to a 1.5 fold increase in the t½ °f 4-ene VPA from 13 ± 2 to 19 ± 3 min. The fraction of the low dose (29 ± 5%) eliminated in bile was significantly larger than at the high dose (21 ± 4%), when calculated as the sum of conjugated and unconjugated 4-ene VPA. The biliary elimination of (E)-2-ene VPA showed a non-significant change from 38 ± 10 to 31 ± 9% on increasing the dose. Like the parent drug VPA, (E)-2-ene VPA and 4-ene VPA showed enterohepatic recirculation in the rat which produced secondary plasma peaks in normal animals. Moreover, both (E)-2-ene VPA and 4-ene VPA showed a rapid but transient choleretic effect in the rat. The plasma protein binding of 4-ene VPA was apparently low (14-25%), in the concentration range of 20-350 µg/mL. The results indicate that 4-ene VPA is cleared much faster from the plasma than (E)-2-ene VPA in the rat. The plasma levels of 4-ene VPA required to show a non-linear decline (>200 µg/mL) in the rat are two orders of magnitude higher than 4-ene VPA levels (<1 µg/ml) seen in patients on VPA therapy. It is, therefore, unlikely that 4-ene VPA is eliminated more slowly than VPA in man. On the other hand, the plasma elimination t½ of (E)-2-ene VPA in bile-exteriorized rats is longer than that reported for VPA, indicating that (E)-2-ene VPA may have a longer lasting pharmacologic effect than VPA. / Pharmaceutical Sciences, Faculty of / Graduate

Modifications to the GCMS assay for valproic acid and 12 metabolites were attempted with respect to internal standards and derivatizing reagents. Four new internal standards, octanoic acid and 2-methylglutaric acid for analysis of VPA and metabolites and hexanoic acid and di-ռ-butylacetic acid for the analysis of hexadeuterated VPA and metabolites were used. Two new derivatizing reagents, MSTFA and MTBSTFA, were tested as alternatives to the reagent previously used. TMS (MSTFA) and tBDMS derivatives were compared with respect to sensitivity, stability, and chromatographic time. The derivatives formed from MTBSTFA were extremely stable a major drawback was the formation of a diderivative of 3-keto VPA upon increased heating time and storage. Preliminary data on the metabolism of D₆-VPA was obtained in one volunteer. The substitution of six deuterium atoms for six hydrogen atoms resulted in an isotope effect with decreased serum trough concentrations of 4-ene VPA and 2,4-diene VPA. Valproic acid and carbamazepine are frequently coadministered in efforts to optimize seizure control. VPA is extensively metabolized while CBZ is known to induce the hepatic microsomal enzyme system, and thus, this is a potentially toxic interaction. Pharmacokinetic parameters for VPA were obtained before and after CBZ administration in five, healthy male volunteers. Increased plasma clearance of VPA accompanied by decreased plasma concentrations, serum half-life, and AUC values were observed after CBZ comedication. This was consistent with the ability of CBZ to induce the hepatic microsomal enzyme systems in a manner similar to phenobarbital. Serum trough and steady state concentrations and AUC values for 12 metabolites were determined before and after CBZ administration. The AUC values for the monounsaturated metabolites decreased after CBZ administration while the AUC values of the polar metabolites increased. The amount of 4-ene VPA, a potential hepatotoxin, was not increased in the serum after administration of CBZ. The amounts of the two diunsaturated metabolites, 2,3'-diene VPA and 2,4-diene VPA, were increased in the serum of the volunteers after CBZ administration. The amount of 2-ene trans VPA in the serum was significantly decreased after CBZ administration, while the amount of 3-keto VPA did not increase. Urinary metabolic profiles were determined individually and grouped in pathways for the five volunteers before and after CBZ administration. Increased recoveries of 4-ene VPA, 4-keto VPA, and 2-PSA after CBZ administration were consistent with enhanced ω-1 oxidation. Formation clearance, metabolic clearance, and fraction metabolized were determined for the metabolic pathways and for the individual metabolites. CBZ adminstration resulted in increased formation clearances for all pathways. The results obtained from this study indicate that CBZ caused a general induction of VPA metabolism and did not specifically affect a particular pathway. The effect of CBZ on the beta-oxidation pathway is not clearly understood. CBZ may cause a metabolic shift away from beta-oxidation, or actually inhibit beta-oxidation to some extent. As well, peroxisomal beta-oxidation may be involved. / Pharmaceutical Sciences, Faculty of / Graduate