An In Vitro Study for Evaluating Permeability and Metabolism of Kurarinone

Kurarinone is a major component found in the dried roots of Sophora flavescens Ait. that participates in vital pharmacological activities. Recombinant CYP450 supersomes and liver microsomes were used to study the metabolic profiles of kurarinone and its inhibitory actions against cytochrome P450 (CYP) and UDP-glucuronosyltransferase (UGT) enzymes. 100 μM of kurarinone strongly inhibited more than 90% of UGT1A1, UGT1A6, CYP1A2, and CYP2C9. CYP1A2 and CYP2D6 played important roles in catalyzing the biotransformation of kurarinone. Moreover, metabolism of kurarinone considerably differs among species, and metabolic characteristics were similar between monkey and human. Kurarinone demonstrated moderate permeability at values of pH 4.0 and 7.4. Our findings offer a clearer idea to understand the pharmacological and toxicological mechanisms of kurarinone.

Cytochrome P450 (CYP) and UDP-glucuronosyltransferase (UGT) are important phase-I and -? drug-metabolizing enzymes that actively participate in the metabolism of more than 90% of currently available drugs [19]. Inhibition or induction of CYPs or UGTs might cause potential drug-drug interactions (DDIs). S. flavescens induces CYP3A expression by activating the pregnane X receptor (PXR) [20]. S. flavescens extract has been found to reduce blood theophylline concentration in rats by inducing hepatic CYPs including CYP1A2, CYP2B, CYP2E1, and CYP3A [21]. Several other studies reported that S. flavescens extract dose-dependently inhibited human hepatic CYP2C8, CYP2C9, CYP1A2, CYP2C19, CYP2B6, and CYP3A4 [2,22,23]. Such varying experimental outcomes may be due to the complexity of S. flavescens extracts. Although kurarinone is a marker compound of S. flavescens, the interactions between former and CYPs or UGTs are yet to be investigated earlier.
Currently, the parallel artificial membrane permeation assay (PAMPA) is used as a rapid in vitro assay of passive biomembrane permeation in the drug discovery stage [24]. is technique was developed by Kansy et al. [25] and had been originally established to rapidly predict passive permeability through the gastrointestinal tract [26]. Nevertheless, the permeability of kurarinone has not yet been reported.
In this study, the liver microsomes and recombinant human supersomes were used to explore the interaction of kurarinone and UGTs or CYPs, which would contribute to its safe clinical usage and minimize the occurrence of DDIs.

In Vitro Incubation Systems in Liver Microsomes or
Recombinant CYP Supersomes. Samples were prepared in a total volume of 200 μL, containing 0.3 mg/mL protein in human liver microsomes or 15 nM protein in recombinant human supersomes, probe substrates or 50 μM kurarinone, 10 mM MgCl 2 , 10 mM glucose-6-phosphate, and 1 unit/mL of glucose-6-phosphate dehydrogenase. e reactions were activated by adding NADPH following preincubation at 37°C for 5 min. e reactions were maintained at 37°C for 1 h and terminated by adding 200 μL ice-cold acetonitrile. e supernatant (10 μL) collected from the samples that were centrifuged (20,000 g, 20 min, 4°C) was injected into the HPLC system for analysis. Control incubations devoid of NADPH or kurarinone were performed parallelly to ensure that the produced metabolites were NADPH-or kurarinonedependent.

Inhibitory Effect of CYP Activities by Kurarinone.
RLMs were used to assess the inhibitory effects of kurarinone (100 μM) on six different human isoforms (CYP1A2, CYP2C9, CYP2D6, CYP2E1, CYP3A4, and CYP2C19). e probe substrates used in this experiment were 40 μM phenacetin for CYP1A2, 10 μM diclofenac for CYP2C9, 25 μM dextromethorphan for CYP2D6, 120 μM chlorzoxazone for CYP2E1, 35 μM testosterone for CYP3A4, and 20 μM (S)-mephenytoin for CYP2C9 as described earlier [27]. e reaction system has been described in Section 2.3. Kinetic parameters including IC 50 and K i were determined and analyzed for CYP isoforms whose activities were strongly inhibited by more than 90%. Dixon and Lineweaver-Burk plots were formed to confirm the reversible inhibition type, and a second plot of slopes from the Lineweaver-Burk plot over the concentrations of kurarinone was used to estimate the K i value. All incubations were performed in triplicates, and mean values were used for analysis.  were initiated by adding UDPGA, and incubations were carried out at 37°C in a shaking waterbath for 60 min. e reactions were terminated by addition of 200 μL of ice-cold acetonitrile containing 7-hydroxycoumarin (100 μM) as the internal standard. e incubation mixture was then centrifuged at 20,000 ×g for 20 min to obtain the supernatant, and an aliquot of supernatant (20 μL) was transferred to an autoinjector vial for HPLC analysis.

Chemical Inhibition Study.
Chemical inhibition studies were carried out by adding various human CYP inhibitors to the incubation mixture of kurarinone (20 μM) prior to the addition of the NADPH-generating system. HLMs were utilized as metabolic enzymes. e period of incubation and conditions are detailed in Section 2.3. (1-200 μM) were incubated with RLMs, MLMs, PLMs, RAMs, CYP1A2, and CYP2D6 to estimate kinetic parameters. e rest of the other conditions remained the same as described in Section 2.3. Preliminary experiments evidenced that the formations of metabolites occurred in a linear range of reaction time and concentrations of microsomes. e apparent K m and V max values were calculated from nonlinear regression analysis of experimental data according to the Michaelis-Menten equation, and the results were represented graphically by Eadie-Hofstee plots. All incubations were carried out as triplicate independent experiments, and kinetic constants were recorded as mean ± SD.

Prediction of In Vivo Hepatic Clearance.
e following equations were utilized to predict the in vivo hepatic clearance of kurarinone in humans and rats [28].
where the SF (scaling factor) represents the milligrams of microsomal protein per gram of the liver multiplied by the grams of liver weight; CL int is the intrinsic metabolic clearance; CL H is hepatic clearance; f u is the free fraction in the blood (no data were available for kurarinone; thus, f u was arbitrarily proposed to be 1); and Q H is the hepatic blood flow. e CL H of kurarinone was calculated using equations (1)-(3). e physiological parameters for calculating the intrinsic clearance in rats and humans are described as follows: the amounts of microsomal protein were 44.8 mg and 48.8 mg of protein/g of the liver; the liver weight per kilogram of body weight values were 40 g and 25.7 g; and the liver blood flow values were 55.2 mL/min/kg and 20.7 mL/ min/kg, respectively [29].
2.10. Molecular Docking. Molecular docking analysis was performed to further assess the molecular mechanism of interaction between kurarinone and CYPs. e X-ray structures of CYP 1A2 (PDB code 2HI4), 2C9 (PDB code 3QM4), and 2D6 (PDB code 4WNU) were obtained from RCSB Protein Databank (http://rcsb.org/). Molecular docking evaluations were performed with AutoDock 4.02. Schrodinger Maestro software was utilized for graphic display.
e protein structures were prepared through ProteinPrep wizard within Schrodinger package, and the energy minimization was completed through the external Tripos forcefield. e cluster analysis with AutoDock results was performed to determine best poses of kurarinone within investigated CYP sites.

PAMPA Permeability Study.
e PAMPA method was used to measure the membrane permeability values, as described by Singh et al. in an earlier study [30]. Drug samples were dissolved in DMSO and then diluted to 10 mM concentrations as a stock solution. 300 μL diluted drugs solution was filled in the PAMPA plate as the "donor" wells, and the 96-well filter plate with a synthetic phospholipid membrane was then placed on the donor wells. e "acceptor" wells were filled with 200 μl of buffer solution, and the PAMPA instruments were incubated at room temperature for 5 h. Kurarinone permeability was assessed in quadruplicate at pH 4.0 and 7.4. Naproxen was taken as a high permeability marker and furosemide as a low permeability marker. At the end of the incubation, kurarinone and the marker compounds in the donor and acceptor wells were determined by HPLC, and permeability was calculated.

Metabolic Profiling of Kurarinone in Different Liver
Microsomes. After kurarinone (20 μM) was incubated for one hour with different liver microsomes such as HLMs, RLMs, RAMs, PLMs, MOMs, MLMs, and DLMs, HPLC was utilized to confirm the biotransformation of kurarinone. As showed in Figure 2, a new peak (P 1 ) was detected at 6.77 min compared with that of the control group, which was inferred as the main metabolite of kurarinone owing to its high peak area for all species. e formation of metabolite was time-, NADPH-, and microsome-dependent.

Inhibitory Effects of Kurarinone against UGT and CYP
Activities. 100 μM kurarinone inhibited the activities of UGT1A1, UGT1A6, CYP1A2, and CYP2C9 by more than 90%. Inhibition kinetic parameters of UGT1A1 and CYP1A2 were established. Kurarinone displayed concentration-dependent inhibitions against UGT1A1 and CYP1A2 with IC50 values of 13.64 μM and 10.02 μM, respectively (Figures 3(a) and 4(a)). Dixon (Figures 3(b), and 4(b)) and Lineweaver-Burk plots (Figures 3(c), and 4(c)) demonstrated that UGT1A1 and CYP1A2 inhibition by kurarinone was fitted well by noncompetitive and competitive inhibition, respectively. e K i values were calculated to be 13.04 μM and 8.13 μM for UGT1A1 and CYP1A2 by a second plot of the slopes from the Lineweaver-Burk plots versus kurarinone concentrations (Figures 3(d), and 4(d)). Kurarinone exhibited competitive inhibition against CYP1A2 and noncompetitive inhibition against UGT1A1.

CYP Isoforms Involved in Catalyzing the Formation of
Metabolite. After kurarinone was incubated with different liver microsomes, its main metabolite (P 1 ) was determined. Kurarinone was incubated with 13 cDNA-expressed human CYP isoforms for confirming the CYP isoform involved in metabolizing kurarinone. Figure 5 indicates that CYP1A2 and CYP2D6 played a significant role in catalyzing the formation of P 1 , whereas the other enzymes hardly contributed in metabolizing kurarinone.
e results were confirmed through the chemical inhibition study. Kurarinone was incubated with selective chemical inhibitors of the eight CYP isoforms in HLMs, and the results showed that furafylline (CYP1A2 inhibitor) and quinidine (CYP2D6 inhibitor) remarkably inhibited the formation of P 1 compared with other CYP isoform inhibitors ( Figure 6). Figure 7, the metabolic profiles of kurarinone (1-200 μM) in RLMs, MLMs, PLMs, HLMs, CYP1A2, and CYP2D6 exhibited typical monophasic Michaelis-Menten kinetics, which were further confirmed by Eadie-Hofstee plots. e kinetic parameters (K m , V max , and CL int ) were calculated using P 1 data and are summarized in Table 1.

Prediction of In Vivo Hepatic Clearance in Human and
Rat. CL H was calculated using the kinetic parameters of P 1 generated from nonlinear regression in HLMs and RLMs, and the results were 16.82 mL/min/kg and 46.01 mL/min/kg body weight for humans and rats. e percentages of CL H versus hepatic blood flow (Q H ) values for human and rat were 81.27% and 83.35%, respectively.

PAMPA Permeability of Kurarinone.
Drugs are primarily known to be absorbed in the small intestines, wherein the pH may vary from acidic to neutral and slightly basic. In this study, the PAMPA assay was carried out at pH 4.0 and 7.4, respectively. e PAMPA permeability results of kurarinone and permeability markers are showed in Table 2. Kurarinone exhibited moderate permeability at pH 4.0 and 7.4.

Discussion
Owing to its various pharmacological activities, kurarinone will be coadministered with other drugs. Hence, assessment of potential DDIs is an essential process in the clinical development of kurarinone. Both Food and Drug Administration (FDA) and European Medicines Agency (EMA) affirmed that C max /K i ratio was calculated as an indicator of in vivo DDI potential. e interaction likely occurred if the C max /K i value of the inhibitor is greater than 1, possibly occurred if the ratio is between 0.1 and 1, and unlikely occurred if below 0.1 [31]. Kurarinone potently inhibited CYP1A2 with the Ki value of 8.13 μM in RLMs. e C max  Evidence-Based Complementary and Alternative Medicine value of kurarinone in rats was recorded to be 1668.01 ng/mL [32], and C max /K i value was calculated to be 0.46, indicating that the in vivo adverse effects possibly occur due to the C max /K i value of kurarinone against CYP1A2. us, potential metabolism-based DDIs might occur during coadministration of kurarinone with other CYP1A2 substrate such as tizanidine [33]. Nevertheless, it is noteworthy that S. flavescens extracts could induce CYP1A, CYP2B1/2, CYP2C11, and CYP3A in rats and mice [22]. e complexity of S. flavescens extracts may be the reason for varying experimental results, and while the alkaloids matrine and oxymatrine participated in inducting CYP isoforms [3]. Hence, clinicians should be aware of the in vivo concentration alterations of CYP1A2-metabolized substrates when their patients are administered S. flavescens with a high content of kurarinone.
Reaction phenotyping assays and chemical inhibition study evidenced that CYP1A2 and CYP2D6 play a significant role in catalyzing the formation of P 1 , whereas other enzymes exhibited limited ability to metabolize kurarinone. If kurarinone was coadministered with a CYP1A2 or CYP2D6 inhibitor, the CYP inhibition might likely amplify the blood concentration of kurarinone and lead to adverse effects.
Animals are often used as a model to predict kinetics and toxicity in humans. As we know, CYPs are classified into Evidence-Based Complementary and Alternative Medicine distinct subfamilies on the basis of amino acid sequence identity. A high degree of sequence identity does not necessarily indicate similar catalytic specificity, and a single amino acid substitution may even result in a change in substrate specificity. erefore, the substantial differences in CYPs among various species may cause varying drug metabolism across species [34]. In the current experiments, enzyme kinetic studies were performed to further assess species difference. Differences in K m value have shown significance to comprehend the basis of species-related differences in P450-catalyzed drug oxidation reactions [35].
Similar K m values among various species reflect equivalent binding affinities toward the metabolic site and indicate species similarities [36]. e comparable K m values for the formation of P 1 in MLMs and HLMs in this study evidenced similar CYPs affinity in these two mammals. Monkeys and humans have a common ancestor since approximately 25 million years ago. Monkeys are genetically and physiologically similar to humans and are therefore the most extensively used nonhuman primate in basic and applied biomedical research [37]. Owing to higher V max values, the CL int values in MLMs, RLMs, and PLMs were found to be  Evidence-Based Complementary and Alternative Medicine much higher than those in HLMs, while all the differences were statistically significant, evidencing that these liver preparations transformed kurarinone more efficiently than that in HLMs. e K m , V max , and CL int values for the oxidative metabolism of kurarinone in CYP2D6 and CYP1A2 were similar. e K m values were found to be much higher for kurarinone metabolism in liver microsomes than those in CYP2D6 and CYP1A2, which may be due to the nonspecific binding of kurarinone to microsomal material. e kinetic constant values of single CYP in cDNA expression systems and multiple CYPs in liver microsomal systems are incomparable [38]. Compounds can be classified into high-clearance (>70% liver blood flow), low-clearance (<30% liver blood flow), and intermediate-clearance drugs on the base of CL H value projected from the in vitro data. Here, kurarinone was categorized as a high-clearance drug in humans and pigs. After intravenous administration of 10 mg/kg kurarinone in six rats, the t 1/2 was calculated, 1.81 h [32], proving that kurarinone could be cleared quickly in vivo.
Molecular docking was performed in this study to investigate the molecular mechanism of interactions between kurarinone and CYPs. e results exhibited that kurarinone interacted with CYP1A2 (Asn312 and Asp320), CYP2C9 (Asp301), and CYP2D6 (Gln244 and Ser304) through formation of H-bonds. Asn312 and Asp320 are significant residues in the substrate and inhibitor recognition regions of CYP1A2 [39,40]. Gln244 and Ser304 in CYP2D6 are active site residues involved in hydrogen bond formation with substrates [41].
More than 80% of orally administered drugs are absorbed into the bloodstream through passive diffusion in the small intestines. us, to detect the intestinal absorption and bioavailability, the passive diffusion of a drug must be comprehended. PAMPA is one of the most widely used assays to predict transcellular passive absorption in vitro   [42]. In the present study, permeability of kurarinone was evaluated at pH 4.0 and 7.4, which corresponded to conditions in the stomach, small intestine, and plasma. e permeability of kurarinone was found to be moderate, indicating that kurarinone was absorbed in the stomach and small intestine.
In this study, Kurarinone was thus evidenced to strongly inhibit UGT1A1, UGT1A6, CYP1A2, and CYP2C9, which could lead to DDIs. Kurarinone metabolism showed significant differences among species. CYP1A2 and CYP2D6 played important roles to metabolize kurarinone. e PAMPA permeability of kurarinone was moderate at both pH 4.0 and 7.4.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.