Species-Associated Differences in the Inhibition of Propofol Glucuronidation by Magnolol

Abstract

Magnolol, a major active constituent in herbal medicine, potently inhibits propofol glucuronidation in human liver microsomes, with inhibition constants in the nanomolar range. This study was conducted to investigate magnolol-induced inhibition of propofol glucuronidation in liver microsomes from Swiss–Hauschka mice, Sprague–Dawley rats, Chinese Bama pigs, and cynomolgus macaques. Results indicated that magnolol (10 μM) inhibited propofol glucuronidation in liver microsomes from Bama pigs and cynomolgus macaques but not in those from mice or rats. Data from liver microsomes from Bama pigs indicated a competitive inhibition mechanism, with a K_i of 1.7 μM. In contrast to that of pig liver microsomes, the inhibition of microsomes from cynomolgus macaques followed a noncompetitive mechanism, with a K_i of 3.4 μM. In summary, this study indicates that magnolol-induced inhibition of propofol glucuronidation varies substantially among species, and the K_i values determined by using liver microsomes from various experimental animal species far exceed that for human liver microsomes. The inhibition of propofol glucuronidation by magnolol in liver microsomes from all animal species tested was significantly lower than the inhibition previously demonstrated in human liver microsomes. Hepatic microsomes from Swiss–Hauschka mice, Sprague–Dawley rats, Chinese Bama pigs, and cynomolgus macaques are not effective models of the inhibition of glucuronidation induced by magnolol in humans.

UDP glucuronosyltransferases (UGT) conjugate many clinical drugs with glucuronosyl groups, making them more water-soluble and readily excreted by the kidney.³ Inhibition of UGT activity slows the clearance of drugs dependent on this pathway, increasing the exposure level and the risk of toxicity.¹⁵

Propofol, a widely used anesthetic drug, can undergo rapid glucuronidation,¹³ which mainly is mediated by UGT1A9.¹⁴ Previous studies have demonstrated that the inhibition of propofol glucuronidation can prolong the anesthesia time.⁶ In addition, because propofol glucuronidation represents a marker reaction of UGT1A9, the inhibition on this metabolic process can be extrapolated to predict the potential influences on other drugs’ glucuronidation in which UGT1A9 is involved.

Magnolol is a chief active constituent of Magnolia officinalis and has been used in traditional medicine to treat several common diseases in Asian countries for thousands of years.⁵ In addition, because this compound displays potent antimicrobial activities, magnolia bark supercritical carbon dioxide extract (containing at least 93% magnolol) is used as a food supplement in mints and gums to freshen breath.¹⁰

Our previous studies have demonstrated that magnolol acts as an atypical substrate of UGT and potently inhibits propofol glucuronidation in humans, thus prompting concerns regarding the safety of this compound.^18,19 To identify a potential model for studying magnolol's toxic effects, we evaluated the inhibition of propofol glucuronidation by magnolol by using liver microsomes from mice, rats, pigs, and monkeys.

Materials and Methods

Chemical reagents.

Magnolol (greater than 98% pure) was purchased from the Victory Company (Chengdu, China). Propofol glucuronide was obtained from Toronto Research Chemicals (North York, Ontario, Canada). Propofol (99%), 4-methylumbelliferone glucuronide (98%), uridine-5-diphosphoglucuronic acid (trisodium salt), and alamethicin were purchased from Sigma–Aldrich (St Louis, MO). All other reagents were HPLC grade.

Enzymes sources.

Pooled liver microsomes from Sprague–Dawley rats (RLM, n = 20, male), Swiss–Hauschka (ICR) mice (MLM, n = 50, male), cynomolgus monkeys (CyLM, n = 2, male), and colony-bred Chinese Bama minipigs (PLM, n = 3, male) were all purchased from the Research Institute for Liver Diseases (Shanghai, China).

Inhibition of magnolol on propofol glucuronidation.

Propofol (100 μM) was incubated with RLM, MLM, PLM, and CyLM in the absence or presence of magnolol (0, 0.1, 1, 5, or 10 μM). Incubations were performed by using 200-μL reaction mixtures of 50 mM Tris-HCl buffer (PH 7.4) containing 5 mM MgCl₂ and 4 mM uridine-5-diphosphoglucuronic acid. Propofol and magnolol were diluted in methanol. In the 200-μL incubation system, the volume of methanol was 2 μL (1 μL propofol and magnolol stock solution, respectively). According to our previous study,⁸ the incubation time was set at 20 min, and microsomal protein concentrations were set at 0.1 mg/mL. The microsomes were fully activated by the addition of alamethicin, with a final concentration of 50 μg/mg microsomal protein.

Determination of IC₅₀ values for the inhibition of PLM and CyLM.

IC₅₀ values (the inhibitor concentration that inhibits 50% of the control activity) were determined for the inhibition of magnolol on propofol glucuronidation (100 μM) in PLM and CyLM. To obtain the IC₅₀ values, various concentrations of magnolol (0 to 10 μM) were used. IC₅₀ values were calculated by using Origin 7.5 (Origin Lab, Northampton, MA) as described previously.¹⁶

Kinetic assays for the inhibition of PLM and CyLM.

Kinetic assays were performed for inhibition in PLM and CyLM. To obtain the inhibition constants (K_i), various concentrations of magnolol (0 to 10 μM) and propofol (31.25 to 500 μM) were used.

Analytical methods.

In all incubations, the consumption of magnolol and propofol were both less than 10%. Propofol glucuronidation (200-μL reaction mixture) was terminated by adding 100 μL methanol containing 15 μM 4-methylumbelliferone glucuronide as the internal standard. The vessel then was vortexed for 20 s and subsequently put in an ice bath for 30 min. Propofol glucuronidation samples were centrifuged at 20,000 × g for 10 min to remove protein, and the supernatants (10 μL) were analyzed on a Prominence (Shimadzu, Kyoto, Japan) ultrafast liquid chromatography system as described previously.^8,19

Data analysis.

Inhibition constants (K_i) were derived by fitting data to either a competitive inhibition model (equation 1), noncompetitive inhibition model (equation 2), or mixed inhibition model (equation 3). The type of inhibition and the K_i values were decided by using Origin 7.5.

where v is the velocity of the reaction; [S] and [I] are the substrate and inhibitor concentrations, respectively; K_i is the constant describing the affinity of the inhibitor to the enzyme; αK_i describes the affinity of the inhibitor to the complex of enzyme and substrate; and K_s is the constant describing the affinity of the substrate to the enzyme.

Results

Inhibition of propofol glucuronidation in RLM and MLM by magnolol.

The inhibition of propofol glucuronidation in RLM and MLM by magnolol is displayed in Figure 1. Magnolol (0 to 10 μM) slightly decreased the activities of RLM and MLM. In the presence of 10 μM magnolol, the propofol glucuronidation activity of RLM was decreased to 78% of the control. Similarly, 10 μM magnolol decreased the activity of MLM to 70% of the control. In the presence of 5 μM magnolol or less, the activities of RLM and MLM both exceeded 80% of control values.

Figure 1.

Inhibitory effects of magnolol (0, 0.1, 1, 5 and 10 μM) on propofol glucuronidation activities of (A) RLM and (B) MLM. For each incubation, the propofol concentration was set at 100 μM, reaction time was set at 20 min, and microsomal content was set at 0.1 mg/mL. Each column and error bar represents the computer-calculated mean ± 1 SD of triplicate determinations.

Inhibition of propofol glucuronidation in PLM by magnolol.

The inhibition of propofol glucuronidation in PLM by magnolol is shown in Figure 2. The inhibition of the marker reaction was concentration-dependent manner (Figure 2 A). In the presence of 10 μM magnolol, the remaining propofol glucuronidation activity of PLM was about 26% of the control. The IC₅₀ value was calculated to be about 2.4 ± 0.2 μM (R² = 0.997). Additional kinetic assays demonstrated that the inhibition proceeded according to a competitive inhibition mechanism (Figure 2 B), with a K_i value of 1.7 ± 0.5 μM (R² = 0.977).

Figure 2.

(A) IC₅₀ and (B) Lineweaver–Burk plots of inhibition of propofol glucuronidation by magnolol in PLM. Each point represents the mean of triplicate determinations, and lines are from model fitting.

Inhibition of propofol glucuronidation in CyLM by magnolol.

The inhibition of propofol glucuronidation in CyLM by magnolol is shown in Figure 3. Similar to that in assays with PLM, magnolol markedly inhibited the activity of CyLM. The presence of 10 μM magnolol decreased glucuronidation to 25% of the control value; the IC₅₀ was determined to be 2.1 ± 0.2 μM (R² = 0.998). Different from the inhibition of PLM, the inhibition against CyLM followed a noncompetitive inhibition mechanism (Figure 3 B), with a K_i value of 3.4 ± 0.2 μM (R² = 0.951).

Figure 3.

(A) IC₅₀ and (B) Lineweaver–Burk plots of inhibition of propofol glucuronidation by magnolol in CyLM. Each point represents the mean of triplicate determinations, and lines are from model fitting.

Discussion

Due to potential extensive exposure of humans to magnolol, assessing the safety of this chemical is of vital importance. Our previous studies demonstrated that magnolol potently inhibits the activity of UGT1A7 and 1A9.¹⁸ As are described in other studies, UGT are important protective agents against many carcinogens and drugs with side or even toxic effects.^2,11,17 Therefore, the inhibition of UGT may serve as the potential toxic mechanism for magnolol and warrants additional investigation, particularly given the potential extensive exposure of adolescents to this compound due to their excess consumption of gums and mints, which may contain magnolol.

This in vitro study was designed to investigate the inhibition by magnolol of propofol glucuronidation in MLM, RLM, PLM and CyLM and sought to identify a suitable experimental animal in which the inhibition mimics that in HLM. Results indicate that magnolol inhibits the propofol glucuronidation activity of PLM and CyLM with K_i values of 1.7 and 3.4 μM, respectively, which are higher than that in HLM.¹⁸ No significant inhibition of propofol glucuronidation in MLM and RLM was noted.

Rats and mice are 2 common animal models in drug metabolism and safety research. However the UGT expression in these species is significantly different from that in human. UGT1A9, a major isoform responsible for glucuronidation of propofol in humans, is not detected in rats.¹² Propofol glucuronidation in RLM is catalyzed by other isoforms, perhaps explaining why magnolol exerted little inhibition of the reaction in RLM. Unlike rats, mice have prominent levels of UGT1A9.¹ However, because other UGT in mouse liver may contribute significantly to propofol glucuronidation, magnolol failed to show efficient inhibition in MLM as well.

Nearly all whole-animal pharmacologic research of magnolol has been conducted in rats and mouse.^4,7 The pharmacokinetic study in rats indicates that after oral administration, magnolol levels in liver can reach about 10 μM.⁹ At such a high magnolol concentration, propofol glucuronidation is almost completely abolished in human, but only limited inhibition is observed in rats. This difference may explain, at least in part, why in rats magnolol displays many desired pharmacologic activities, rather than toxic effects, even though magnolol can potently inhibit UGT in humans.

In comparison with rats and mice, pigs and monkeys more closely resemble humans pharmacologically. Magnolol inhibits propofol glucuronidation in PLM and CyLM at K_i values of 1.7 and 3.4 μM, respectively. Compared with that in HLM, the inhibition against PLM and CyLM is much weaker, with K_i values that are about 10 times higher. However, to our knowledge, UGT expression in pigs and monkeys has not been reported previously. The current study may provide useful information indicating that the UGT with propofol glucuronidation activity in the liver of pigs and monkeys are much less sensitive toward magnolol than is UGT1A9 in humans.¹⁸ Therefore, these 2 animal species are unsuitable for studies investigating the toxic or side effects of magnolol due to the inhibition of UGT.

Although glucuronidation metabolism and the inhibition has drawn increased attentions, systemic evaluation of the function of UGT in different experimental animals remains unavailable. In particular, the species-associated differences in the inhibition of UGTs by clinical drugs and frequently administered food chemicals have been neglected. When no significant inhibition of UGT is observed in experimental animals, the potential inhibition and corresponding toxic effects in humans are dismissed. As demonstrated in the current study, the inhibition by magnolol of propofol glucuronidation differs dramatically between different animal species. The inhibitory effects of magnolol on the activity of liver microsomes from the several animal species investigated in this study were all weaker than that in HLM.

The difference in the magnolol-induced inhibition of UGT between humans and animals shown in this study suggests that adequately modeling clinical drug interactions involving UGT inhibition in any of these animal species may be difficult. How to model the inhibition of UGT and to validate the corresponding potential toxic effects is a continuing story.