Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Valproic acid (VPA, Fig. 1)¹ , a simple, branched-chain fatty acid with a broad spectrum of anticonvulsant activity, is widely used for treatment of various forms of epilepsy (Rimmer and Richens, 1985). A major metabolic pathway of VPA is glucuronidation of the carboxylic acid. The other minor pathways are -oxidation and -hydroxylation at aliphatic hydrocarbon side chains (Eadie et al., 1988). Valproate glucuronide (VPA-Glu) metabolite is known to be excreted into urine of rat, dog, monkey, and human (Vree and Van der Kleijn, 1977; Dickinson et al., 1979).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Chemical structures of PAPM and VPA.

The chemical structure of VPA differs very much from other antiepileptics, thus, VPA is often applied in combination with other anticonvulsant agents (Cloyd et al., 1985). Numerous types of drug-drug interactions between VPA and concomitantly administered drugs have been reported, e.g., reduction in the plasma concentration of VPA by phenytoin or phenobarbital due to induction of hepatic drug-metabolizing enzymes (Richens et al., 1975); enhanced VPA hepatotoxicity by phenytoin or carbamazepine due to the formation of 4-ene-VPA, a minor but toxic metabolite of VPA (Levy et al., 1990); and inhibition of plasma protein binding of VPA by salicylic acid (Fleiman et al., 1980; Yu et al., 1989) or by endogenous free fatty acid (Bowdle et al., 1982). A clinically interesting drug interaction of VPA with Carbenin [panipenem (PAPM)/betamipron, Sankyo Co., Ltd., Tokyo, Japan], a carbapenem antibiotic, has been reported recently in three patients, who showed a reduction of trough plasma level of VPA during concomitant treatment with Carbenin (Nagai et al., 1997). A similar interaction was also found during concomitant therapy with VPA and meropenem, another carbapenem antibiotic. Carbapenem antibiotics have a broad spectrum of antibacterial activity that includes moderate activity against Gram-positive bacteria, excellent activity against Gram-negative aerobics and anaerobics (Shimada and Kawahara, 1994; Fish and Singletary, 1997), and are used frequently in treating various infections. These interactions resulted in the recurrence of epileptic seizures in patients and, therefore, prohibition of the concomitant use of carbapenem antibiotics with VPA was newly added to Information on Adverse Reactions to Drugs (Ministry of Health and Welfare, Japan, 1996).

More recently, it was established that PAPM (Fig. 1), a pharmacologically active constituent of Carbenin, affects the pharmacokinetics of VPA also in cynomolgus monkey (Ministry of Health and Welfare, Japan, 1996). We previously reported that the VPA-PAPM interaction also occurred in Beagle dog and was related to the glucuronidation of VPA (Yamamura et al., 1998). The details of the interaction mechanism, however, have not yet been clarified.

In this report we examined the effect of PAPM on the hepato-biliary excretion of VPA-Glu in rat. Additionally, we investigated whether PAPM alters VPA uptake by rat hepatocytes and plasma-protein binding of VPA.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Male, Sprague-Dawley (SD) rats (Charles River, Ibaragi, Japan) were housed in a well ventilated room maintained at 24°C and were allowed free access to food and water. The body weight of the rats used for the experiments ranged from 200 to 350 g. Rats were fasted for 12 h before use.

In Vivo Study.

Plasma concentration-time profile of VPA in normal SD rats, nephrectomized rats, and hepatectomized rats The normal rats were lightly anesthetized with ether during the experiments and the femoral vein was cannulated with polyethylene tubing (PE-50, Becton Dickinson & Co., Sparks, MD). A loading dose of 10 mg/kg PAPM followed by a constant infusion of 1.43 mg/min/kg PAPM was administered through the femoral vein cannula using a constant rate infusion pump (pump 11, Harvard Apparatus, Inc., South Natick, MA). The maintenance dose of PAPM was expected to produce a steady-state plasma level of about 50 µg/ml. Animals injected with VPA alone were not cannulated. Thirty minutes after starting the PAPM treatment, a bolus dose of 30 mg/kg VPA was injected through the jugular vein. Blood samples of about 0.3 ml each were collected from the jugular vein at scheduled intervals for a period of up to 120 min and immediately centrifuged at 12,000 rpm for 3 min to separate plasma samples. Plasma concentrations of VPA were measured by fluorescence polarization immunoassay (TDx/TDxFLx system, Abbott Laboratories, Abbott Park, IL). Nephrectomized rats were prepared by removing both kidneys after ligating both the renal vein and artery under light ether anesthesia. Functionally hepatectomized rats were prepared by ligating the hepatic portal vein and hepatic artery (Takenaka et al., 1995). Both nephrectomized and hepatectomized rats were given VPA and PAPM as described above and blood samples were collected for up to 120 min and 60 min, respectively.

Cumulative biliary excretion of metabolites after bolus injection of ¹⁴C-VPA. The normal rats were lightly anesthetized with ether and the femoral vein and bile duct were cannulated with polyethylene tubing. Thirty minutes after starting the PAPM infusion as described above, a bolus dose of 30 mg/kg ¹⁴C-VPA (0.05 mCi/kg) was injected through the jugular vein. Blood samples were collected from the jugular vein and centrifuged to obtain plasma samples. Bile was collected at 30-min intervals for 120 min. Bile flow was determined gravimetrically assuming a density of 1.0 (g/ml). The total radioactivity of plasma and bile samples was measured in a liquid scintillation spectrophotometer (LSC-3500, Aloka Co., Tokyo, Japan). VPA and its metabolites were extracted from plasma and bile samples with ethanol and subjected to silica gel TLC (n-butanol:acetic acid:distilled water = 4:1:1). Radioactivity was quantified by a Bio-Imaging Analyzer (BAS-2000, Fuji Photo Film Co., Ltd., Tokyo, Japan). VPA-Glu on the chromatogram was identified as a spot that disappeared in the separate samples after -glucuronidase treatment (400 U).

Hepatobiliary excretion of VPA-Glu at steady state. Catheterized rats were injected with a loading dose of 10 mg/kg PAPM, followed by a constant infusion of 1.43 mg/min/kg PAPM to achieve steady state. A loading dose of 16.8 mg/kg followed by a constant infusion of 24.6 mg/h/kg of ¹⁴C-VPA (specific activity: 2 µCi/mg) was administered through the femoral vein cannula. The maintenance dose of ¹⁴C-VPA was expected to produce a steady-state plasma level of about 75 µg/ml. In control animals, PAPM treatment was replaced by saline at 10 µl/min. After a stabilization period of 30 min, bile was collected at 15-min intervals for 45 min. Blood samples were collected simultaneously from the jugular vein before each collection of bile and centrifuged to separate plasma samples. In typical experiments, the plasma concentrations at 15, 30, and 45 min postdose were 86.1, µg/ml, 85.8 µg/ml, and 89.0 µg/ml, demonstrating the steady state has been achieved. After the final collection of bile, the liver was quickly removed and weighed. A small portion of liver was weighed and solubilized with NCS Tissue Solubilizer (DuPont-NEN Research Products, Boston, MA) to determine total radioactivity. Another portion of excised liver was homogenized with ethanol containing 1% acetic acid to extract VPA, VPA-Glu, and other metabolites. After centrifugation at 900g, the supernatant was evaporated to dryness. The extract was analyzed by TLC to determine VPA and VPA-Glu in the liver using a Bio-Image Analyzer as described above.

In Vitro Study.

Initial uptake of VPA into isolated rat hepatocytes Rat hepatocytes were isolated with a slight modification of the method described by Moldeus et al. (1978). The viability of cells was assessed by the trypan blue exclusion method and cells with 90% or higher viability were used for the experiments. The initial rate of VPA uptake was measured by the method described by Schwenk (1980). The hepatocytes were suspended in Krebs-Henseleit buffer at a final cell density of 2 × 10⁶ cells/ml; 0.95 ml of this solution was preincubated at 37°C for 3 min before addition of ³H-VPA (0.05 ml). At 0.5 min after addition of the substrate, 0.2-ml aliquots were removed and centrifuged immediately through a silicone oil layer into 3 M KOH to separate the cells from the incubation medium. The radioactivity of the solubilized cells in the alkaline layer was determined in a liquid scintillation spectrophotometer. VPA uptake was calculated as an amount of VPA in pmol (the radioactivity in the pellet divided by the specific activity of ³H-VPA) incorporated in 10⁶ cells in 30 s (pmol/10⁶ cells/30 s).

Plasma unbound fraction of VPA. Animals were sacrificed by exsanguination from the abdominal aorta under light ether anesthesia. Blood samples collected into plastic tubes containing EDTA were immediately centrifuged to separate plasma samples. To determine the free fraction of VPA, with or without PAPM or salicylate, VPA (final concentration 50, 100, or 150 µg/ml) and PAPM (final concentration 50 µg/ml) or salicylate (final concentration 250 µg/ml) were added to plasma and, after incubation at 37°C for 15 min, centrifugal ultrafiltration of the plasma was carried out through an MPS-3 membrane (Amicon Division, W.R. Grace & Co., Beverly, MA) at 2500 rpm for 15 to 30 min. The concentration of VPA in plasma (Cp) and filtrate (Cu) was measured by fluorescence polarization immunoassay. The adsorption of VPA to MPS-3 membrane was negligible. The plasma unbound fraction was calculated as Cu/Cp.

Hepatic tissue unbound fraction of VPA. Animals were sacrificed by exsanguination from the abdominal aorta under light ether anesthesia. The liver was quickly isolated and homogenized with twofold volumes of 10 mM Tris-HCl (pH 7.4). Cytosol from 30% homogenate was prepared by centrifugation at 9,000g for 20 min at 4°C followed by ultracentrifugation (105,000g for 60 min at 4°C). To determine the free fraction of VPA (final concentration 25 µg/ml) in the presence of PAPM (final concentration 0, 20, or 200 µg/ml), VPA and PAPM were added to the cytosol and the mixture was treated in the same way as above for plasma protein binding.

-Glucuronidase activity in the liver toward VPA-Glu. Three rats were sacrificed by exsanguination under ether anesthesia and the livers were quickly isolated and homogenized in 3-fold volumes of 20 mM potassium bicarbonate buffer (pH 6.8). This homogenate was used as the enzyme source of -glucuronidase. Because VPA-Glu was the major metabolite in the bile after i.v. administration of ¹⁴C-VPA to the rats above in in vivo study, the bile samples collected over a period of 2 h after administration of VPA (30 mg/kg) to three rats were combined and used as VPA-Glu. To measure the concentration of VPA-Glu in the pooled bile sample, the bile was hydrolyzed under alkaline condition as follows. A 0.2-ml aliquot of the bile was added with 0.1 ml of 5 N NaOH and incubated at room temperature for 2 h. The reaction was stopped by the addition of 0.1 ml 5 N HCl and the pH of the sample was adjusted to approximately 7 with 0.2 ml 1 M potassium bicarbonate buffer (pH7.4). Then, the sample was applied to the TDx/TDxFLx system to measure the concentration of VPA.

Pharmacokinetic Analysis. For individual rats, model-independent pharmacokinetic parameters, i.e., area under the plasma concentration-time curve (AUC_inf), biological half-life (T_1/2), and distribution volume (V_d) were calculated using WinNONLIN (Scientific Consulting, Inc., Cary, NC) software. Total body clearance of VPA (CL_tot), the apparent metabolic clearance of VPA by glucuronidation [CL_m(glu)], and the biliary excretion clearance of VPA-Glu [CL_bile(glu)] were calculated as follows:

Statistical Method. All results are shown as means ± S.D. Statistical differences in pharmacokinetic parameters between VPA alone and PAPM-treated groups were tested by Student's t test. The criterion for statistical significance was p < .05.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

In Vivo Study.

Effects of PAPM on plasma concentration-time profiles of VPA in normal SD rats, nephrectomized rats, and hepatectomized rats The time courses of the plasma concentration of VPA, with or without PAPM treatment, in normal SD rats and in nephrectomized rats are shown in Fig. 2 and Fig. 3, respectively. VPA rapidly disappeared from plasma in both groups of rats with PAPM treatment. Significant increases in the CL_tot as well as a significant reduction in the T_1/2 by PAPM treatment were observed in both the normal SD rats and nephrectomized rats (Table 1). In contrast, PAPM did not affect the pharmacokinetic behavior of VPA in hepatectomized rats, which showed markedly prolonged elimination of VPA (Fig. 4 and Table 1).

View larger version (11K): [in this window] [in a new window]   Fig. 2.   The plasma concentration-time course of VPA in control (, n = 6) and PAPM treatment (, n = 4). In the case of PAPM-treated rats, VPA (30 mg/kg) was administered intravenously after 30 min for infusion of PAPM (see detail in text).

View larger version (11K): [in this window] [in a new window]   Fig. 3.   The plasma concentration-time course of VPA in nephrectomized rats with (, n = 3) or without (, n = 3) PAPM treatment. Nephrectomized rats were prepared by removing both kidneys and VPA (30 mg/kg) was administered intravenously (see details in text).

View larger version (10K): [in this window] [in a new window]   Fig. 4.   The plasma concentration-time course of VPA in hepatectomized rats with (, n = 3) or without (, n = 3) PAPM treatment. Hepatectomized rats were prepared by ligating the hepatic artery and hepatic portal vein and VPA (30 mg/kg) was administered intravenously (see details in text).

Effect of PAPM on cumulative biliary excretion of metabolites after bolus injection of ¹⁴C-VPA. The thin-layer chromatogram of bile and the time profile of the plasma concentration of ¹⁴C-VPA in bile-fistula rats are shown in Fig. 5. In bile-fistula rats, PAPM still enhanced the elimination rate of VPA. As shown in Table 2, 90% or more of the radioactivity excreted in the bile was identified as VPA-Glu. The cumulative excretion of the glucuronide in the bile up to 2 h after the i.v. dose of ¹⁴C-VPA in the PAPM-treated group was 1.4-fold higher than that in the control group (VPA alone). On the other hand, no difference in the cumulative amount of the unchanged VPA or other minor metabolites (BM-1 and BM-2), found at amounts not more than 2.4%, was observed between the PAPM-treated group and the control group.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Typical chromatogram of bile after administration of 14C-VPA (A) and the plasma concentration-time course of VPA in rats with a bile duct fistula (B), with (, n = 3) or without (, n = 5) PAPM treatment. Rats were prepared with a bile duct cannulated and collected bile after intravenous administration of 14C-VPA (30 mg/kg). Bile sample with ethanol was subjected to TLC with n-butanol:acetic acid:distilled water (4:1:1; see details in text).

Effect of PAPM on hepatobiliary excretion of VPA-Glu under steady-state condition of VPA. The result under steady-state condition of VPA is shown in Table 3. The PAPM treatment caused a decreasing tendency of Css,plasma(vpa) and an increasing tendency of Css,liver(glu), although the differences in these values were not statistically significant between PAPM-treated and control rats. CL_tot and Vbile(glu) were 1.2- and 1.5-fold higher, respectively, in the PAPM-treated rats than in the control rats. CL_m(glu) was also 1.8-fold higher in the PAPM-treated rats than in the control rats. CL_bile(glu) values were nearly identical in the PAPM-treated and the control rats. In addition, the contribution of the clearance parameters other than CL_m(glu) to CL_tot (CL_tot minus CL_m(glu)), namely the clearance due to other elimination pathways, was not affected by the PAPM treatment. Thus, the increase in CL_m(glu) by the PAPM treatment was concluded to be the major reason for the increased CL_tot in the PAPM-treated rats.

In Vitro Study.

Effects of PAPM on initial uptake of VPA by rat hepatocytes The initial velocity of VPA uptake by rat hepatocytes was linear over a concentration range from 1 to 200 µM and independent of temperature (Fig. 6). Addition of PAPM at 50 or 200 µM did not affect the initial uptake.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   The uptake of VPA into rat hepatocytes, with or without PAPM. The hepatocytes were prepared from three animals and suspended at final cell density of 2 × 106 cells/ml (see details in text).

Effects of PAPM on plasma unbound fraction of VPA. As shown in Fig. 7, there was no significant difference in the plasma-unbound fractions of VPA determined in vitro at 50, 100, or 150 µg/ml between plasma containing PAPM (50 µg/ml) and the control plasma. In contrast, a clear increase in the plasma-unbound fraction of VPA (p < .001) was observed by addition of salicylic acid (final concentration 250 µg/ml), a known inhibitor of plasma protein binding of VPA.

View larger version (54K): [in this window] [in a new window]   Fig. 7.   Plasma-unbound fraction of VPA in the absence and presence of PAPM or salicylic acid in vitro. VPA and PAPM or salicylate were added to plasma and, after incubation at 37°C for 15 min, centrifugal ultrafiltration of the plasma was carried out through an MPS-3 membrane at 2500 rpm for 15 to 30 min.

Effect of PAPM on hepatic tissue-unbound fraction of VPA. As shown in Table 4, addition of PAPM (20 or 200 µg/ml) did not significantly affect the tissue-unbound fractions of VPA in hepatic cytosol determined in vitro at 25 µg/ml.

Effect of PAPM on hepatic -glucuronidase activity toward VPA-Glu. The concentration of VPA-Glu in the bile, which was used as the substrate solution in this experiment, was 0.415 µmol/ml (60 µg/ml as VPA). VPA produced from VPA-Glu by the action of -glucuronidase in 25% liver homogenate proportionally increased as a function of incubation time up to 60 min (result not shown). Therefore, the effect of PAPM on hepatic -glucuronidase activity toward VPA-Glu was investigated after incubation for 60 min as shown in Table 5. Addition of PAPM (50 or 500 µg/ml) caused no effect on the VPA production from VPA-Glu by liver homogenate.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The drug interaction between VPA and PAPM is due to the pharmacokinetic interaction, because the marked reduction of trough plasma level of VPA was recognized in every patient (Nagai et al., 1997). The mechanism of the interaction is noteworthy because the physiochemical and pharmacokinetic characteristics of VPA and PAPM are completely different.

In a series of in vivo studies, it was found that i.v. administration of PAPM significantly reduces T_1/2, but did not alter the maximum plasma concentration (C_max) of VPA after oral administration of VPA to monkey and dog (Ministry of Health and Welfare, Japan, 1996; Nouda and Perez, 1997; Yamamura et al., 1998). In the present study, a similar result was obtained in rat with a significant reduction in AUCinf and a significant increase in CL_tot of VPA by PAPM cotreatment (Fig. 2). Thus, the interaction between VPA and PAPM was reproduced in almost all animal species.

PAPM enhanced the elimination of VPA in the nephrectomized rats (Fig. 3), but not in the hepatectomized rats (Fig. 4). These results demonstrate that the interaction of PAPM with VPA occurs in the liver. The fact that the CLtot of VPA in the hepatectomized rats without PAPM treatment was about one-third of that in the normal rats also indicates that VPA is metabolized mainly in the liver. The CLtot value of VPA was less than the hepatic blood flow (1800-2400 ml/h/kg). Thus, the CLtot of VPA in rats is regarded as the hepatic clearance (CL_H) and can be approximated by the following equation (Gibaldi and Perrier, 1982):
where fp is the plasma-unbound fraction of VPA and CL_int,H is the hepatic intrinsic clearance. Hence, an increase in CL_tot should be accompanied by an increase in fp and/or CL_int,H. Schematic representation of possible routes of VPA disposition in the liver is shown in Fig. 8. Step 1 is the plasma-protein binding step, step 2 is the uptake step by hepatocytes, step 3 is the hepatic tissue-protein binding step, step 4 is the -oxidation or -hydroxylation step, step 5 is the glucuronidation step, step 6 is the deconjugation step, and step 7 is the biliary excretion step.

View larger version (24K): [in this window] [in a new window]   Fig. 8.   Schematic representation of possible routes of VPA disposition in the hepatocyte. Step 1 is the plasma-protein binding step (dynamic equilibrium), step 2 is the uptake step by hepatocytes (passive diffusion), step 3 is the hepatic tissue-protein binding step, step 4 is the -oxidation or -hydroxylation step, step 5 is the glucuronidation step, step 6 is the deconjugation step, and step 7 is the biliary excretion step.

An increase in fp will occur by displacement of plasma protein binding (Fig. 8-1). The VPA binding protein is mainly albumin and the albumin binding of VPA is inhibited by many drugs (Fleiman et al., 1980; Dasgupta and Emerson, 1996; Rambeck et al., 1996). The displacement of the plasma-protein binding of VPA by salicylic acid in vivo and in vitro is known to increase the CL_tot of VPA (Yu et al., 1989). However, PAPM altered the plasma-protein binding of VPA neither in vitro (Fig. 7) nor in vivo (data not shown), demonstrating that the plasma-protein binding step of VPA is not affected by PAPM. Therefore, the increase in CL_tot of VPA in PAPM-treated rats is considered to be mainly due to an increase in CL_int,H.

CL_int,H is a hybrid parameter including uptake by the liver, metabolism, and biliary excretion of VPA and/or its metabolites. As shown in Fig. 8, the increase of CL_int,H can occur at steps 2, 3, 4, 5, 6, and 7. The initial uptake of VPA by hepatocytes occurred by passive diffusion consistent with the previous report (Booth et al., 1996) and was not affected by PAPM (Fig. 6). Thus, the uptake step of VPA was not considered to increase the CL_int,H by PAPM (Fig. 8-2). It is reported that the binding of VPA in the liver is partly ascribed to the intracellular ligandin (Yu and Shen, 1995). However, more than 90% of VPA in the hepatic cytosol was protein unbound and was not affected by the PAPM treatment in vitro (Table 4). Therefore, the displacement of hepatic-protein binding of VPA by PAPM is negligible (Fig. 8-3). The glucuronidation was confirmed to be the major pathway of VPA in rats as shown in Table 2 and Fig. 5. The -oxidation and -hydroxylation of VPA (Fig. 8-4) are the minor pathways and were not considered to be important in the drug interaction with PAPM. We also observed that PAPM did not inhibit the -glucuronidase activity toward VPA-Glu as shown in Table 5 (Fig. 8-6). The fact that the increase in CL_m(glu) by the PAPM treatment was the major reason for the increased CL_tot in the PAPM-treated rats under steady-state condition means a negligible contribution of -oxidation and -hydroxylation pathways in the interaction (Table 3). This CL_m(glu) is a parameter including steps 5 and 7 in Fig. 8. The CL_bile(glu) indicated the excretion process of VPA-Glu (Fig. 8-7) was not affected. It is considered that unaffected CL_bile(glu) but increased Vbile(glu) is due to the increased Css,liver(glu) and thus, it is indicated that PAPM does not enhance the biliary excretion process of VPA-Glu Therefore, the site of action of PAPM to enhance the glucuronidation of VPA is consequently considered after only one step, namely, the glucuronidation step (Fig. 8-5) from this study.

It has been reported (Dickinson et al., 1979) that the biliary excretion of VPA-Glu in the rat is very high (45-55% of the administered dose) but VPA and VPA-Glu are not excreted in the feces, showing that VPA undergoes extensive enterohepatic recirculation as a consequence of biliary excretion of VPA-Glu and after deconjugation and reabsorption. Kojima et al. (1998) reported recently that the decrease in numbers of enteric bacteria that are able to deconjugate VPA-Glu by PAPM results in shutdown of the enterohepatic recirculation of VPA and is regarded as a possible mechanism for the VPA-PAPM interaction. However, the biliary excretion of VPA-Glu shows a species difference, as it has been reported that the biliary excretion of VPA-Glu in monkeys is very limited (3-7% of the administered dose; Dickinson et al., 1980). The interaction between VPA and PAPM occurred in both species. This fact indicates that the abolished deconjugation of VPA-Glu and the decreased reabsorption of VPA are very unlikely as the mechanisms for the VPA-PAPM interaction.

The biochemical mechanism of the interaction between VPA and PAPM has not been elucidated yet, but the following factors are considered as possible reasons for the increased glucuronidation of VPA: 1) enzyme induction of UDP-glucuronosyltransferase (Fig. 8-5) and 2) increased cofactor availability [UDP-glucuronic acid (UDPGA), Fig. 8-5; Rowland and Tozer, 1980].

Repeated administration of anticonvulsant drugs such as carbamazepine is well known to cause the reduction of plasma level of VPA due to enzyme induction (Richens et al., 1976), which generally takes a relatively long time, whereas the reduction of the plasma VPA level due to interaction with PAPM is immediately observed after coadministration (Nagai et al., 1997). Therefore, it is unlikely that the induction of UDP-glucuronosyltransferase explains this drug interaction.

The cofactor availability is considered to be one of the most possible mechanisms for the drug interaction between VPA and PAPM (Fig. 8-5). Because the glucuronidation is a bimolecular reaction between the substrate and UDPGA, UDPGA is an essential determinant for the conjugation reaction. As an example of the availability of UDPGA being rate-limiting in glucuronidation, Braun et al. (1997) have demonstrated that pretreatment by GSH-depleting agents causes the enhancement of glycogenolysis, the increase of UDP-glucose, from which UDPGA is produced and the enhancement of p-nitrophenol glucuronidation in isolated murine hepatocytes.

In conclusion, the increase in CL_tot of VPA by PAPM is mainly due to the increased hepatic intrinsic clearance, especially the enhanced glucuronidation of VPA.