Corydaline Inhibits Multiple Cytochrome P450 and UDP-Glucuronosyltransferase Enzyme Activities in Human Liver Microsomes

Corydaline is a bioactive alkaloid with various antiacetylcholinesterase, antiallergic, and antinociceptive activities found in the medicinal herb Corydalis Tubers. The inhibitory potential of corydaline on the activities of seven major human cytochrome P450 and four UDP-glucuronosyltransferase enzymes in human liver microsomes was investigated using LC-tandem MS. Corydaline was found to inhibit CYP2C19-catalyzed S-mephenytoin-4’-hydroxylatoin and CYP2C9-catalyzed diclofenac 4-hydroxylation, with Ki values of 1.7 and 7.0 μM, respectively. Corydaline also demonstrated moderate inhibition of UGT1A1-mediated 17β-estradiol 3-glucuronidation and UGT1A9-mediated propofol glucuronidation with Ki values of 57.6 and 37.3 μM, respectively. In the presence of corydaline, CYP3A-mediated midazolam hydroxylation showed a decrease with increasing preincubation time in a dose-dependent manner with Ki values of 30.0 μM. These in vitro results suggest that corydaline should be evaluated for potential pharmacokinetic drug interactions in vivo due to potent inhibition of CYP2C19 and CYP2C9.

Modulation of drug-metabolizing enzymes, such as CYP and UGT enzymes, is one of the important mechanisms of drug interaction. Among the numerous CYP and UGT enzymes identified to date, seven human hepatic CYP enzymes (CYPs 1A2, 2A6, 2C8, 2C9, 2C19, 2D6, and 3A) and four UGT enzymes (UGTs 1A1, 1A4, 1A9, and 2B7) have been shown to a play important role in drug metabolism [16,17]. Studies have shown that bergamottin, a major furanocoumarin found in grapefruit juice, is reported to increase the blood concentration of drugs through inhibition of hepatic CYP3A activity, thereby enhancing the toxicity of drugs such as atorvastatin, felodipine, quinidine, sildenafil, and verapamil [18].
To the best of our knowledge, no previous study has reported on the effect of corydaline, a pharmacologically active isoquinoline alkaloid isolated from Corydalis tubers, on human CYP and UGT enzymes. In this study, the effects of corydaline on activities of seven major human CYP and four major human UGTs (1A1, 1A4, 1A9, and 2B7), were examined using pooled human liver microsomes in order to evaluate the possibility of corydaline-drug interactions.

Results and Discussion
The inhibitory effects of corydaline toward seven major human CYP isoforms and four major human UGT isoforms were evaluated using cocktail CYP probe substrates and each UGT probe substrate in human liver microsomes, respectively. Corydaline induced moderate inhibition of CYP2C19-mediated S-mephenytoin 4-hydroxylation and CYP2C9-mediated diclofenac 4-hydroxylation with IC 50 values of 11.7 and 26.2 μM, respectively. Corydaline also induced weak inhibition of CYP2D6-catalyzed bufuralol 1'-hydroxylation with an IC 50 value of 64.5 μM. This is not an unusual finding, given that berberine and palmatine, isoquinoline alkaloids like corydaline, also showed an inhibitory effect on CYP2D6 activity (with IC 50 values of 49.4 and 92.6 μM, respectively) [19,20]. Corydaline at 200 μM showed negligible inhibition of CYP1A2-mediated phenacetin O-deethylation, CYP2A6-mediated coumarin 7-hydroxylation, CYP2C8-mediated amodiaquine N-deethylation, and CYP3A-mediated midazolam 1'-hydroxylation. The inhibitory potencies of corydaline were not significantly affected after a 30-min period of pre-incubation with human liver microsomes in the presence of NADPH, excluding CYP3A activity (Table 1), indicating that corydaline is a mechanism based inhibitor of CYP3A. To determine whether midazolam metabolism is inhibited by corydaline, we performed time-and concentration-dependent inactivation assays in human liver microsomes. In the presence of corydaline, CYP3A-mediated midazolam hydroxylation was decreased with increasing preincubation time in a dose-dependent manner (Figure 2a). Values for k obs were plotted against corydaline concentration and used for estimation of maximal rates (k inact ) and K i values in human liver microsomes ( Figure 2b). Estimated maximal rates (k inact ) and apparent K i values were 0.064 min −1 and 30.0 μM in human liver microsomes, respectively.  In an inhibition study, the apparent K i value is a better parameter for defining the interaction of the inhibitor with a particular enzyme. K i values and inhibition types (competitive, noncompetitive, uncompetitive, or mixed) for corydaline were determined using Lineweaver plots, Dixon plots, and secondary reciprocal plots; the results are summarized in Table 1. Corydaline showed noncompetitive inhibition for CYP2C9-catalyzed diclofenac 4-hydroxylation (K i , 7.1 μM) and competitive inhibition for CYP2C19-catalyzed S-mephenytoin 4'-hydroxylation (K i , 1.7 μM) ( Figure 3 and Table 1). Corydaline competitively inhibited CYP2D6-catalyzed bufuralol 1'-hydroxylation with a K i value of 27.3 μM.
Corydaline was found to be a potent competitive inhibitor of CYP2C19 with K i values of 1.7 μM, indicating that corydaline may be used carefully with drugs metabolized by CYP2C19, such as proton pump inhibitors (lansoprazole and omeprazole), antiepileptics (diazepam and phenytoin), and antidepressants (amitriptyline and imipramine) in order to avoid drug interactions. Some natural compounds, including anthocyanidin [23], beauvericin [24], eupatilin [25], and ursolic acid [26], have demonstrated strong inhibition of CYP2C19. Corydaline was also shown to be an inhibitor of CYP2C9 with K i values of 7.1 μM, suggesting that corydaline should be used carefully with drugs metabolized by CYP2C9, such as celecoxib, diclofenac, glyburide, losartan, tolbutamide, torasemide, and Swarfarin in order to avoid drug interactions [27]. However, corydaline was not found to be a potent inhibitor of CYP2C9, as compared with the inhibition produced by sulfaphenazole (IC 50 = 0.8 μM), a selective inhibitor of CYP2C9 [28]. Herbal preparations containing corydaline may also affect CYP2C19 and CYP2C9 activity. These results suggest the possibility that high amounts of corydaline might alter the bioavailability of drugs that are metabolized by CYP2C19 or CYP2C9, although in vitro inhibition of CYP2C19 and CYP2C9 activities don't necessarily translate into in vivo drug interactions. Up to now, there was no available data on the pharmacokinetics of corydaline in human which is indispensable for the prediction of drug-drug interaction potential of corydaline. According to the recent report of Zhang et al. [29], corydaline contents in various batches of Yuanhu Zhitong tablets, a traditional Chinese medicine used for the treatment of nerve pain and stomach pain, didn't exceed 1 mg/g of original preparations, which may cause very low blood concentrations (<0.5 μM), even when an entire oral dose reaches the systemic circulation. Therefore, the probability of clinically important drug interactions by corydaline might be negligible based on available data.

Inhibitory Effects of Corydaline on Seven Major CYP Activities in Human Liver Microsomes
Inhibitory potencies (IC 50 values) of corydaline were determined using CYP assays in the presence and absence of corydaline (final concentrations of 1-200 μM with an acetonitrile concentration less than 0.5% v/v) using pooled human liver microsomes. Phenacetin O-deethylase, coumarin 7-hydroxylase, amodiaquine N-deethylase, diclofenac 4-hydroxylase, S-mephenytoin 4-hydroxylase, bufuralol hydroxylase, and midazolam 1'-hydroxylase activities were determined as probe activities for CYP1A2, CYP2A6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, and CYP3A4, respectively, using cocktail incubation and liquid chromatography-tandem mass spectrometry (LC/MS/MS) [25]. The incubation mixtures were prepared in a total volume of 200 μL, as follows: pooled human liver microsomes (0.25 mg/mL), 1.3 mM NADPH, 3.3 mM MgCl 2 , 50 mM potassium phosphate buffer (pH 7.4), and a cocktail of probe substrates and various concentrations of corydaline (1−200 μM). The substrates were used at concentrations approximately equal to their respective K m values: 50 μM phenacetin, 2.5 μM coumarin, 2.5 μM amodiaquine, 10 μM diclofenac, 100 μM S-mephenytoin, 5 μM bufuralol, and 2.5 μM midazolam. After a 3-min pre-incubation period at 37 °C, the reactions were initiated by addition of an NADP generating system and incubated for 20 min at 37 °C in a shaking water bath. After incubation, the reaction was stopped by placement of the tubes on ice and addition of 100 μL of ice-cold methanol containing internal standards ( 13 C 2 , 15 N-acetaminophen for acetaminophen and N-desethylamodiaquine, d 5 -7-hydroxycoumarin for 7-hydroxycoumarin, 13 C 6 -4-hydroxydiclofenac for 4-hydroxydiclofenac, d 3 -4-hydroxy-mephenytoin for 4-hydroxymephenytoin and hydroxymidazolam, d 9 -1-hydroxybufuralol for 1-hydroxybufuralol). The incubation mixtures were then centrifuged at 13,000 × g for 5 min. All incubations were performed in triplicate and the mean values were used. For evaluation of NADPH-dependent mechanism-based inhibition of CYP activities, various concentrations of corydaline (1−200 μM) were pre-incubated for 30 min with human liver microsomes in the presence of NADPH. The reaction was started by addition of a cocktail of CYP probe substrates.
All 7 metabolites produced from the cocktail of CYP isoform-specific substrates were simultaneously determined by LC/MS/MS [25]. The system consisted of a tandem quadrupole mass spectrometer (TSQ Quantum Access, ThermoFisher Scientific, San Jose, CA, USA) coupled with a Nanospace SI-2 LC system. Separation was performed on an Atlantis dC 18 column (5 μm, 2.1 mm i.d. ×100 mm, Waters, Milford, MA, USA) using the gradient elution of a mixture of 5% methanol in 0.1% formic acid (mobile phase A) and 95% methanol in 0.1% formic acid (mobile phase B) at a flow rate of 0.25 mL/min: 10% mobile phase B for 1 min, 10% to 95% mobile phase B for 1 min, 95% mobile phase B for 5 min. The column and autosampler temperatures were 50 and 6 °C, respectively. After 1.5 min, the LC eluent was diverted from waste to the mass spectrometer fitted with the electrospray ionization (ESI) source and operated in positive ion mode. ESI source settings for ionization of the metabolites were as follows: electrospray voltage, 5.0 kV; vaporizer temperature, 420 °C; capillary temperature 360 °C; sheath gas pressure, 35 psi; auxiliary gas pressure, 10 psi. Quantification was performed by selected reaction monitoring (SRM) of the [M+H] + ion and the related product ion for each metabolite. SRM transitions for the metabolites and internal standard are summarized in Table 3. Analytical data were processed using Xcalibur ® software (Thermo Fisher Scientific).
Glucuronides produced from UGT isoform-specific substrates were respectively determined by LC/MS/MS. Separation was performed on the Atlantis dC 18 column using the gradient elution of a mixture of 5% methanol in 0.1% formic acid (mobile phase A) and 95% methanol in 0.1% formic acid (mobile phase B) at a flow rate of 0.25 mL/min: 10% mobile phase B for 2 min, 10% to 95% mobile phase B for 4 min. The column and autosampler temperatures were 50 and 6 °C, respectively. After 3.0 min, the LC eluent was diverted from waste to the mass spectrometer fitted with an ESI source. ESI source settings for ionization of trifluoperazine glucuronide and azidothymidine glucuronide in positive ion mode were as follows: Electrospray voltage, 5.0 kV; vaporizer temperature, 420 °C; capillary temperature 360 °C; sheath gas pressure, 35 psi; auxiliary gas pressure, 10 psi. ESI source settings for ionization of 17β-estradiol 3-glucuronide and propofol glucuronide in negative ion mode were as follows: Electrospray voltage, −4.0 kV; vaporizer temperature, 420 °C; capillary temperature 360 °C; sheath gas pressure, 35 psi; auxiliary gas pressure, 10 psi. Quantification was performed by SRM of the [M+H] + ion for trifluoperazine glucuronide and azidothymidine glucuronide or [M−H] − for 17β-estradiol 3-glucuronide and propofol glucuronide and the related product ion for each metabolite. SRM transitions for the metabolites and internal standard are summarized in Table 3.

Data Analysis
IC 50 values (concentration of inhibitor causing 50% inhibition of the original enzyme activity) were calculated using WinNonlin software, a non-linear regression analysis program (Pharsight, Mountain View, CA, USA). Apparent kinetic parameters for inhibitory potential (K i values) of CYPs were estimated from the fitted curves using Enzyme Kinetics Ver. 1.3 program (Systat Software Inc., San Jose, CA, USA). For the enzyme kinetic analysis of UGT1A1 and UGT1A9 inhibition, the following mixed competitive-noncompetitive equation was fitted to the data points by nonlinear regression [30]:

Conclusions
In conclusion, an evaluation of the effects of corydaline on seven CYPs and four UGTs was conducted across a wide range of substrate and corydaline concentrations using in vitro human liver microsomes. CYP2C19 and CYP2C9 activity were potently inhibited by corydaline during incubation with NADPH in microsomes. Corydaline also inhibited CYP2D6-mediated bufuralol hydroxylation, CYP3A-mediated midazolam hydroxylation, UGT1A1-mediated 17β-estradiol 3-glucuronidation, and UGT1A9-mediated propofol glucuronidation in a dose-dependent manner. These results suggest that high uptake of corydaline from herbal medications may cause an interaction with drugs metabolized by CYP2C19 and CYP2C9 in some individuals. It is important to note, however, that the inhibition of CYP activities in vitro does not necessarily translate into drug interactions in clinical situations. Clinical trials to evaluate the inhibitory effects of corydaline on CYP2C19 and CYP2C9 remain to be conducted.