Abstract
The role of human UDP glucuronosyltransferase (UGT) 2B10 in the N-glucuronidation of a number of tricyclic antidepressants was investigated and compared with that of UGT1A4 in both the Sf9 expressed system and human liver microsomes. The apparent Km (S50) values for the formation of quaternary N-glucuronides of amitriptyline, imipramine, clomipramine, and trimipramine were 2.60, 16.8, 14.4, and 11.2 μM in UGT2B10 and 448, 262, 112, and 258 μM in UGT1A4, respectively. The kinetics of amitriptyline and imipramine glucuronidation in human liver microsomes exhibited a biphasic character, where the high- and low-affinity components were in good agreement with our results in expressed UGT2B10 and UGT1A4, respectively. The kinetics of clomipramine and trimipramine glucuronidation in human liver microsomes were sigmoidal in nature, and the S50 values were similar to those found for expressed UGT1A4. The in vitro clearances (CLint or CLmax) were comparable between UGT2B10 and UGT1A4 for glucuronidation of imipramine, clomipramine, and trimipramine, whereas CLint of amitriptyline glucuronidation by UGT2B10 was more than 10-fold higher than that by UGT1A4. Nicotine was found to selectively inhibit UGT2B10 but not UGT1A4 activity. At a low tricyclic antidepressant concentration, nicotine inhibited their glucuronidation by 33 to 50% in human liver microsomes. Our results suggest that human UGT2B10 is a high-affinity enzyme for tricyclic antidepressant glucuronidation and is likely to be a major UGT isoform responsible for the glucuronidation of these drugs at therapeutic concentrations in vivo.
Glucuronidation, catalyzed by the uridine diphosphate glucuronosyltransferases (UGTs), is an important metabolic pathway for the detoxification and elimination of many endobiotics (bilirubin and bile acids) and xenobiotics (zidovudine and acetaminophen) (Miners et al., 2006). UGT enzymes transfer the glucuronic acid moiety from a cofactor, uridine 5′-diphosphoglucuronic acid (UDPGA), to hydroxyl, carboxyl, or amine groups of aglycone substrates (Tukey and Strassburg, 2000). The resulting β-glucuronide metabolites are usually water-soluble, less active, or toxic and can be readily excreted from the body via bile and urine. At least 17 human UGT isoforms have been identified to date and can be categorized into two subfamilies, UGT1 and UGT2, based on similarities of their amino acid sequences and gene organization (Mackenzie et al., 2005). UGT enzymes are localized primarily in the endoplasmic reticulum and are expressed mainly in the liver, although UGT1A7, UGT1A8, UGT1A10, and UGT2A1 are expressed only in extrahepatic tissues (Miners et al., 2006). UGT enzymes have generally been shown to display very broad and overlapping substrate selectivity (Kiang et al., 2005). Only a limited number of isoform-selective UGT substrates and inhibitors have been identified to date, whereas no isoform-selective UGT inhibitory antibody is available (Miners et al., 2006).
UGT-catalyzed glucuronidations are responsible for approximately one-third of all drugs metabolized by phase II enzymes (Evans and Relling, 1999). Aliphatic tertiary amine or aromatic amine functional groups are quite common in drugs, e.g., in tricyclic antidepressants (TCAs) and anticonvulsants. UGT conjugations of these drugs can result in the formation of quaternary ammonium glucuronide metabolites. The quaternary ammonium glucuronides of a number of TCAs, including amitriptyline, imipramine, clomipramine, and trimipramine (Fig. 1), have been identified as major phase II metabolites in human urine (Luo et al., 1995). The urinary excretion of amitriptyline N-glucuronide metabolite was 2.5 to 21% of the oral dose administered (Breyer-Pfaff et al., 1997). Human UGT1A3 and UGT1A4 were shown to contribute to the N-glucuronidation of imipramine and amitriptyline (Green et al., 1995, 1998). However, the contribution of UGT1A3 to drug conjugation may be insignificant to overall metabolism as it exhibited very high apparent Km values compared with those for UGT1A4 (Green and Tephly, 1998; Kubota et al., 2007). Eadie-Hofstee plots of imipramine and amitriptyline N-glucuronide metabolites indicated that more than one enzyme or active site participated in the N-glucuronidation in human liver microsomes (HLM). UGT1A4 was suggested as the low-affinity enzyme for conjugations of both drugs, whereas the high-affinity enzyme was not identified in these studies (Breyer-Pfaff et al., 1997, 2000; Nakajima et al., 2002).
It was well accepted that only UGT1A3 and UGT1A4 could catalyze quaternary N-glucuronidation reactions until recent reports demonstrated that UGT2B10 catalyzed direct quaternary N-conjugations at the aromatic nitrogen of nicotine, cotinine, and nitrosamines with higher affinities than UGT1A4 (Kaivosaari et al., 2007; Chen et al., 2008). UGT2B10 has been one of the least characterized UGT isoforms because early research indicated no activity toward a large variety of compounds (Jin et al., 1993). However, identification of new UGT2B10 substrates suggests that this enzyme could be more relevant than previously thought, especially in N-glucuronidation reactions. In the present study, we expressed human UGT2B10 enzyme in Sf9 insect cells and demonstrated that UGT2B10 catalyzes the quaternary N-glucuronidation of TCAs with higher affinity than UGT1A4. UGT2B10 is likely to be a major contributor to the glucuronidation of amitriptyline, imipramine, clomipramine, and trimipramine at therapeutic concentrations in vivo.
Materials and Methods
Materials.
Human liver QUICK-Clone cDNA was purchased from Clontech (Mountain View, CA). PfuUltra High-Fidelity DNA Polymerase was obtained from Agilent Technologies (Santa Clara, CA). To enhance specificity, amplifications included AmpliWax PCR Gems obtained from Applied Biosystems (Foster City, CA); thermal cycling was performed on an Applied Biosystems GeneAmp PCR System 9700. A Zero Blunt PCR Cloning Kit, calf intestine alkaline phosphatase, One-Shot TOP10 chemically competent Escherichia coli, Bac-to-Bac Baculovirus Expression System, Cellfectin II Reagent, Sf-900 II SFM insect cell media and Sf9 insect cells (Spodoptera frugiperda) were obtained from Invitrogen (Carlsbad, CA). DNA sequences were determined by using BigDye Terminator v3.1 Cycle Sequencing Kits (Applied Biosystems). Viability of Sf9 cells was determined by using a Vi-CELL Series Cell Viability Analyzer (Beckman Coulter, Fullerton, CA). A BaculoELISA titer kit was purchased from Clontech. Pooled HLM and recombinant UGT1A3 and UGT1A4 expressed in Sf9 insect cells were purchased from BD Gentest (Woburn, MA).
NuPAGE 4 to 12% bis-Tris precast gels, MOPS SDS running buffer, antioxidant, LDS Sample Buffer, Sample Reducing Agent, Transfer Buffer, polyvinylidene difluoride membranes, and SeeBlue Plus2 prestained molecular weight standard were obtained from Invitrogen. UGT2B goat polyclonal antibody, donkey anti-goat IgG-horseradish peroxidase, Blotto, nonfat dry milk, Tris-buffered saline/Tween 20 and Tris-buffered saline wash solutions, and Luminol reagent were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Amersham Hyperfilm ECL was from GE Healthcare (Little Chalfont, Buckinghamshire, UK).
Amitriptyline, imipramine, clomipramine, trimipramine, nicotine, and UDPGA were obtained from Sigma-Aldrich (St. Louis, MO). Hecogenin was purchased from Voigt Global Distribution Inc. (Lawrence, KS).
Expression of UGT2B10 in Sf9 Insect Cells.
Full-length cDNA for human UGT2B10 was PCR-amplified from human liver QUICK-Clone cDNA with PfuUltra High-Fidelity DNA Polymerase and AmpliWax PCR Gems following the manufacturers' protocols. Forward and reverse primers for amplification of UGT2B10 cDNA were 5′-ATGGCTCTGAAATGGACTACAGTT and 5′-CCAGCTTCAAATCTCAGATATAAC, respectively. The PCR product was ligated into pCR-Blunt and subcloned into restriction enzyme-digested and dephosphorylated pFastBac1. Clones and subclones were propagated in TOP10 E. coli cells. The cDNA sequence was confirmed by sequencing, and the GenBank accession number for UGT2B10 cDNA sequence is NM_001075.4.
The Bac-to-Bac Baculovirus Expression System was used to generate recombinant baculovirus according to the manufacturer's protocol. In brief, pFastBac1 plasmid containing UGT2B10 was transfected into E. coli DH10Bac competent cells harboring Bacmid vector DNA and a helper plasmid designed to transpose the pFastBac1 insert into Bacmid in situ. Recombinant Bacmid DNA was propagated, isolated, and then used to transfect Sf9 insect cells via Cellfectin II-mediated gene transfer to generate recombinant baculovirus. Recombinant baculovirus-containing culture supernatants were harvested (passage 1 viral stock), amplified, and titered by using BaculoELISA kits.
High-titer passage 3 stock was used for UGT2B10 protein expression. Sf9 insect cell liquid cultures were grown in Sf-900 II SFM medium at 27°C with orbital shaking at 120 rpm. Sf9 cells were infected at ∼1 to 1.5 × 106 viable cells/ml. Cells were infected at a multiplicity of infection of 1.0 and harvested by centrifugation at 48 h postinfection. Microsomes were prepared by homogenization and two-speed centrifugation (10,000 and 105,000g) and were reconstituted in phosphate-buffered saline (pH 7.4). Microsomal protein concentrations were measured by using a Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA). Microsomes were stored at −70°C until use.
Western Blotting.
UGT2B10 and uninfected control Sf9 cell microsomal fractions (20 μg each) were adjusted to contain 1× NuPAGE LDS Sample Buffer and 1× NuPAGE Sample Reducing Agent and then heated at 70°C for 10 min. Samples were resolved on a NuPAGE Novex bis-Tris 4 to 12% gel in NuPAGE MOPS SDS running buffer. SeeBlue Plus2 prestained standards were used as molecular weight markers. Cathode gel running buffer and electroblot transfer buffer contained NuPAGE Antioxidant. Polyvinylidene difluoride membranes were then probed with goat polyclonal UGT2B antibody (1:200 dilution), followed by horseradish peroxidase-conjugated donkey anti-goat IgG (1:3500 dilution). UGT2B10 protein was visualized using the Luminol chemiluminescent reagent and Hyperfilm ECL.
Activity Assays.
The enzyme reactions were conducted under linear conditions with respect to incubation time (up to 90 min) and protein concentration (up to 0.25 mg/ml). In all cases, 200 μl (or 400 μl for low substrate concentration incubations) of incubation mixtures containing 0.25 mg/ml microsomal protein, 50 μg/mg alamethicin, 50 mM (pH 7.4) potassium phosphate buffer, and 5 mM MgCl2 were kept on ice for 15 min before incubation to activate UGT enzymes. After a 3-min preincubation, reactions were initiated with the addition of 2 mM UDPGA. Reactions were allowed to proceed for 90 min at 37°C and were terminated by addition of 3 volumes of acetonitrile-methanol (1:1). After centrifugation, the supernatant was transferred to a 96-well plate. The samples were evaporated to dryness and reconstituted, and the glucuronide metabolites were quantified. Amitriptyline, imipramine, clomipramine, and trimipramine were dissolved in methanol, and the final incubation contained 1% methanol (v/v). Control incubations contained the same concentration of organic solvent.
To assess the contribution of each isoform to the formation of N-glucuronides, TCA compounds at 5 and 50 μM were first incubated with recombinant human UGT1A3, UGT1A4, or UGT2B10 in triplicate. The assays to study enzyme reaction kinetics were then performed in triplicate at 9 concentrations of TCA (0.15–100 μM) in recombinant UGT2B10 and 13 concentrations (0.15–500 μM) in recombinant UGT1A4 and pooled HLM.
Selective Inhibition.
The selective inhibitory effects of hecogenin and nicotine on TCA glucuronidation were evaluated in recombinant UGT1A4 and UGT2B10. The incubation conditions were the same as described above. Amitriptyline, imipramine, clomipramine, and trimipramine were incubated at 5 μM in duplicate with UGT1A4 or UGT2B10 in the presence of 10 or 100 μM hecogenin or 10, 100, or 500 μM nicotine. The formation of glucuronide metabolite in the presence of nicotine or hecogenin was compared with the formation in their absence (vehicle control).
The contribution of UGT1A4 and UGT2B10 to glucuronidation of TCA was further assessed in pooled HLM. Amitriptyline, imipramine, clomipramine, and trimipramine were incubated at 5 and 200 μM with HLM in triplicate in the presence of 100 μM hecogenin or 500 μM nicotine or both. The formation of glucuronide metabolite in the presence of nicotine or hecogenin was compared with the formation in their absence (vehicle control).
Quantification of Quaternary N-Glucuronide Metabolites.
TCA glucuronides were determined with the method described for the quantification of trifluoperazine N-glucuronide (Uchaipichat et al., 2006) with a few modifications, as the UV absorption characteristics of aliphatic N-glucuronide metabolite resemble those of the aglycone substrate (Hawes, 1998). The N-glucuronides were generated in HLM incubations at 100 μM concentrations of each TCA compound. After evaporation to dryness and reconstitution in 20% methanol, N-glucuronides were quantified using high-performance liquid chromatography-UV spectroscopy. The UV spectra were recorded from 256 to 264 nm (encompassing the UV maxima), and calibration curves were prepared using corresponding TCA over a concentration range of 0.5 to 20 μM in incubation matrix. The quantified N-glucuronides were then diluted in matrix containing 200 nM hydroxytriazolam (internal standard) sequentially to serve as calibration standards in high-performance liquid chromatography-mass spectrometry analysis. The calibration curve of each metabolite acquired from accurate mass extracted ion chromatograms was used for quantification of N-glucuronides in sample incubations. The lower limits of quantification for the glucuronides of amitriptyline, imipramine, clomipramine, and trimipramine in prepared incubation mixtures were 0.016, 0.013, 0.027, and 0.024 μM, respectively.
Chromatographic separations of glucuronide metabolites from corresponding aglycones were performed on a Luna 100 × 2.0 mm C18 (2) column (particle size 5 μm; Phenomenex, Torrance, CA) connected to an Acquity ultraperformance liquid chromatography system equipped with photodiode array detector (Waters, Milford, MA). The components in the incubation were separated under a 10-min linear gradient condition at a flow rate of 0.3 ml/min using mobile phases of ammonium formate in water (20 mM, pH 5.0) (A) and acetonitrile (B). Starting conditions consisted of 20% B and were maintained for 0.5 min. The gradient was increased to 50% B over 5 min with a subsequent 1 min wash at 60% B. The mobile phase was returned to initial conditions to equilibrate for a further 4 min. Mass spectrometric analysis were carried out on an LTQ Orbitrap mass spectrometer (Thermo Fisher Scientific) using electrospray ionization in positive mode, at sheath and auxiliary gas settings of 65 and 20, respectively, and a capillary temperature of 300°C. Full-scan accurate mass analyses were performed, and the resolution of the Orbitrap was set at 15,000 for two scan events at the mass ranges of m/z 358 to 361 and 450 to 495. Accurate mass extracted ion chromatograms were obtained by measuring hydroxytriazolam (internal standard), and the N-glucuronides of amitriptyline, clomipramine, imipramine, and trimipramine at m/z 359.0466, 454.2223, 491.1924, 457.2333, and 471.249, respectively.
Data Analysis.
Kinetic constants for TCA glucuronidation in recombinant UGT1A4, UGT2B10, and HLM were obtained by fitting the following kinetic equations to the experimental data using nonlinear regression (Prism 4; GraphPad Software Inc., San Diego, CA).
The Michaelis-Menten equation (eq. 1) for one-enzyme hyperbolic kinetics is and the Michaelis-Menten equation (eq. 2) for two-enzyme hyperbolic kinetics is where Vmax is apparent maximal velocity and Km is the concentration of substrate at which half-maximal velocity is achieved. Vmax2 and Km2 are constants for the second enzyme or binding site. The Hill equation (eq. 3), which describes sigmoidal kinetics, is where S50 is the substrate concentration resulting in 50% of Vmax (analogous to the Km in previous equations) and n is the Hill coefficient. The substrate inhibition equation (eq. 4) is where Ksi is the inhibition constant describing the reduction in rate. The biphasic kinetics equation (eq. 5) is The biphasic kinetic profile described by eq. 5 is different from that described in eq. 2. Biphasic kinetics has two distinct phases but does not follow saturation kinetics. At a low substrate concentration, the kinetic profile is curved as hyperbolic kinetics; however, at a high substrate concentration, the velocity of the reaction continues to increase, developing a linear, upward slope. CLint2 represents the slope of the linear portion.
For reactions exhibiting Michaelis-Menten kinetics, substrate inhibition, and biphasic kinetics, intrinsic clearance (CLint) was calculated as Vmax/Km. For reactions exhibiting sigmoidal kinetics, maximum clearance (CLmax) was calculated using eq. 6 (Houston and Kenworthy, 2000):
Goodness of fit to kinetic models was assessed from S.E., 95% confidence intervals, and r2. Kinetic curves were also analyzed using Eadie-Hofstee plots when fitted with different kinetic models. Kinetic constants were reported as the mean ± S.E. of the parameter estimated.
Results
Protein Expression.
Full-length cDNA of human UGT2B10 was cloned, and protein was expressed in Sf9 insect cells. Expression of UGT2B10 was confirmed by immunoblotting analysis using a commercially available anti-UGT2B antibody. According to the manufacturer, the anti-UGT2B antibody was raised against a peptide mapping near the C terminus of UGT2B of human origin and therefore could be used for detection of a broad range of UGT2B family members of mouse, rat, and human origin by Western blotting. It also has been successfully used to detect human UGTs 2B4, 2B7, 2B10, and 2B11 by Western blot (Nishiyama et al., 2006). UGT2B10 protein exhibited an apparent molecular mass of ∼51.8 kDa (Fig. 2) that was consistent with another report (Uchaipichat et al., 2004). Microsomes prepared from uninfected Sf9 cells exhibited no cross-reactivity to anti-UGT2B antibody.
Quaternary N-Glucuronidation in Recombinant UGTs.
Two TCA substrate concentrations, 5 and 50 μM, were used in the activity assays in recombinant UGT1A3, UGT1A4, and UGT2B10. At a TCA concentration of 5 μM, UGT1A3 did not exhibit detectable N-glucuronidation activity toward imipramine or trimipramine. Low levels of amitriptyline and clomipramine N-glucuronidation, 3.2 and 0.27 pmol/min/mg, respectively, were detected in UGT1A3. However, glucuronide formation in UGT1A3 was at least 30-fold lower than that observed in UGT1A4 or UGT2B10. UGT2B10-catalyzed imipramine N-glucuronidation was approximately 2-fold more efficient than that catalyzed by UGT1A4. UGT2B10 and UGT1A4 exhibited comparable activity in metabolizing amitriptyline, clomipramine, and trimipramine at a 5 μM substrate concentration. (Fig. 3A).
At 50 μM TCA concentration, UGT1A4 appeared to be the most active human UGT for TCA glucuronidation, whereas UGT1A3 exhibited the least activity (Fig. 3B). Results with UGT1A3 are consistent with previous findings, indicating that UGT1A3 is probably insignificant to the overall metabolism of these drugs (Green and Tephly, 1998). UGT2B10-catalyzed imipramine N-glucuronidation was comparable with that catalyzed by UGT1A4, whereas UGT1A4 was approximately 5-fold more efficient than UGT2B10 in the glucuronidation of amitriptyline, clomipramine, and trimipramine (Fig. 3B).
Kinetics of Quaternary N-Glucuronidation.
The kinetics of TCA glucuronidation was characterized in HLM and recombinant UGT2B10 and UGT1A4 (Fig. 4). Kinetic parameters of glucuronidation for each substrate are shown in Table 1. Amitriptyline N-glucuronidation by HLM was best described with two-enzyme Michaelis-Menten kinetics (eq. 2) (Fig. 4), with apparent Km values of 1.75 and 343 μM for the high- and low-affinity component, respectively. These are consistent with the previous reported Km values of 1.4 and 311 μM in HLM (Breyer-Pfaff et al., 1997). Substrate inhibition (eq. 4) and one-enzyme Michaelis-Menten kinetics (eq. 1) were observed for amitriptyline glucuronidation by UGT2B10 and UGT1A4, respectively (Fig. 4). The apparent Km values of amitriptyline N-glucuronidation were 2.60 and 448 μM in UGT2B10 and UGT1A4, in good agreement with Km values observed in HLM. The Ksi value for UGT2B10 (353 μM) was approximately 136-fold higher than the Km value. The intrinsic clearance for amitriptyline N-glucuronidation by UGT2B10 was more than 10-fold higher than that determined for UGT1A4.
Biphasic kinetics was observed for imipramine N-glucuronidation by HLM (eq. 5) (Fig. 4). The high-affinity component exhibited apparent Km and Vmax values of 9.92 μM and 55.7 pmol/min/mg, whereas the low-affinity component was not saturable at 500 μM substrate concentration and showed an apparent CLint of 0.98 μl/min/mg. Imipramine glucuronidation by UGT2B10 exhibited one-enzyme Michaelis-Mention kinetics (eq. 1) (Fig. 4). The apparent Km and Vmax values of imipramine glucuronidation by UGT2B10 were 16.8 μM and 59.6 pmol/min/mg, in good agreement with the high-affinity component in HLM. Sigmoidal kinetics was observed for imipramine glucuronidation by UGT1A4 (eq. 3), with an apparent S50 value of 262 μM (Table 1). The in vitro clearances for imipramine glucuronidation by UGT2B10 and UGT1A4 were comparable, although it should be noted that CLint and CLmax are not equivalent parameters.
Clomipramine N-glucuronidation by HLM and recombinant UGT1A4 exhibited sigmoidal kinetics (eq. 3) (Fig. 4). The kinetic constants were in good agreement between HLM and UGT1A4, exhibiting apparent S50 values of 108 and 112 μM, Vmax values of 1750 and 1430 pmol/min/mg, and Hill coefficients of 1.9 and 1.8, respectively. Clomipramine glucuronidation by UGT2B10 fit best to the one-enzyme Michaelis-Menten equation (eq. 1), with an apparent Km value of 14.4 μM, approximately 7.5-fold less than the apparent S50 that was obtained from HLM or UGT1A4, although caution should be exercised in comparing Km and S50 values.
In contrast to the sigmoidal kinetics observed with HLM, trimipramine glucuronidation by UGT2B10 and UGT1A4 exhibited one-enzyme Michaelis-Menten kinetics (eq. 1) (Fig. 4). The apparent Km values of trimipramine glucuronidation were 11.2 and 258 μM by UGT2B10 and UGT1A4. HLM exhibited an apparent S50 value of 277 μM, similar to the apparent Km value by UGT1A4. The intrinsic clearances for trimipramine glucuronidation by UGT2B10 and UGT1A4 were similar (0.9 and 1.2 μl/min/mg for UGT2B10 and UGT1A4, respectively).
Selective Inhibition of UGT Activities.
The inhibitory effects of hecogenin and nicotine on UGT1A4 and UGT2B10 activities were investigated using TCA as substrate at a 5 μM concentration. As shown in Fig. 5, A and B, nicotine and hecogenin demonstrated highly selective inhibition of UGT2B10 and UGT1A4, respectively. Nicotine at 500 μM inhibited 76, 89, 88, and 90% of the glucuronidation of amitriptyline, imipramine, clomipramine, and trimipramine by UGT2B10, while exhibiting no inhibition of UGT1A4 activities. On the contrary, hecogenin at 100 μM inhibited 83, 81, 75, and 83% of the glucuronidation of amitriptyline, imipramine, clomipramine, and trimipramine by UGT1A4, while exhibiting less than 10% inhibition of UGT2B10 activities.
The contribution of UGT1A4 and UGT2B10 to glucuronidation of TCAs was assessed in pooled HLM using hecogenin and nicotine as selective inhibitors. As shown in Fig. 6A, at 5 μM TCA concentration, 500 μM nicotine significantly inhibited 33, 50, 36, and 40% of the glucuronidation of amitriptyline, imipramine, clomipramine, and trimipramine by HLM. Hecogenin at 100 μM exhibited no inhibitory effect on amitriptyline glucuronidation and inhibited 15, 13, and 32% of glucuronidation of imipramine, clomipramine, and trimipramine at 5 μM substrate concentration. At 200 μM TCA concentration (Fig. 6B), hecogenin at 100 μM significantly inhibited 45, 75, 72, and 49% of the glucuronidation of amitriptyline, imipramine, clomipramine, and trimipramine by HLM, respectively. On the other hand, 500 μM nicotine did not inhibit TCA glucuronidation by HLM (Fig. 6B). Addition of hecogenin increased the inhibitory effect of nicotine at 5 μM TCA concentration (Fig. 6A), whereas addition of nicotine did not increase the inhibitory effect of hecogenin at 200 μM TCA concentration (Fig. 6B).
Discussion
Human UGT2B10 has been largely considered as an “orphan” UGT isoform until recent findings demonstrated that UGT2B10 was primarily responsible for the quaternary N-glucuronidation of nicotine, cotinine, and nitrosamines (Kaivosaari et al., 2007; Chen et al., 2008). The present study expanded the UGT2B10 substrate list to include a number of aliphatic tertiary amines, amitriptyline, imipramine, clomipramine, and trimipramine. These observations may prompt other investigators to consider the importance of UGT2B10 in the metabolism and clearance of xenobiotics for drugs that undergo quaternary N-glucuronidation.
The quaternary N-glucuronides of TCAs were major phase II metabolites identified in human urine (Luo et al., 1995). The steady-state plasma concentrations of amitriptyline, imipramine, clomipramine, and trimipramine are approximately 0.2, 0.6, 0.2, and 0.2 μM, respectively (Bailey and Jatlow, 1976; Musa, 1989; Reis et al., 2009). In the present study, the glucuronidation kinetics studied was assessed at a wide concentration range (0.15–500 μM) in HLM. In this way, in vitro observations would not be skewed without consideration of therapeutic concentrations of these drugs. Both amitriptyline and imipramine N-glucuronidation by HLM demonstrated biphasic kinetics, consistent with previous reports (Breyer-Pfaff et al., 1997; Nakajima et al., 2002). The investigators in previous studies suggested UGT1A4 as the low-affinity enzyme in HLM but were unable to identify the high-affinity component. In the present study, the Km values of amitriptyline and imipramine glucuronidation by recombinant UGT2B10 were in good agreement with those observed for the high-affinity component in HLM. This suggested that UGT2B10 is the high-affinity enzyme in the glucuronidation of amitriptyline and imipramine. For clomipramine and trimipramine glucuronidation, the apparent Km (S50) values by recombinant UGT1A4 were similar to the apparent S50 values by HLM fitted with the sigmoidal model. The apparent Km values of clomipramine and trimipramine glucuronidation by UGT2B10 were approximately 7.5- and 23-fold lower than Km (S50) by UGT1A4 or HLM. These results clearly demonstrated that UGT2B10 also exhibited higher affinity toward these two drugs than UGT1A4. However, the high-affinity component associated with UGT2B10 was probably hidden by the sigmoidal kinetic character observed in HLM. Although the glucuronidation capacity (Vmax) by UGT2B10 was lower than that by UGT1A4 for these TCAs, the in vitro clearance was comparable between UGT2B10 and UGT1A4 because of the very high affinity of UGT2B10 to TCAs. The intrinsic clearance of amitriptyline by UGT2B10 was more than 10-fold higher than that by UGT1A4. These results suggested that UGT2B10 was likely to be a high-affinity but low-capacity enzyme, whereas UGT1A4 was a low-affinity but high-capacity enzyme in the glucuronidation of amitriptyline, imipramine, clomipramine, and trimipramine. These TCA compounds are hydrophobic bases and have been reported to bind nonspecifically to HLM and insect cell microsomes (McLure et al., 2000; Venkatakrishnan et al., 2000; Austin et al., 2002). It would be important to determine nonspecific binding of these drugs in the in vitro incubation matrices to improve any extrapolation of the in vitro findings to predict in vivo observations (Obach, 1997).
In present study, we demonstrated that hecogenin selectively inhibited UGT1A4 glucuronidation activity toward TCAs at concentrations up to 100 μM. Nicotine demonstrated highly selective inhibition toward UGT2B10 and no inhibition toward UGT1A4 activities at concentrations up to 500 μM in the in vitro conditions. Although the inhibitory effects of nicotine toward other UGT isoform activities requires further exploration, nicotine could be used as a selective inhibitor in vitro to differentiate UGT1A4 and UGT2B10 activities.
At low TCA concentrations, nicotine significantly inhibited their glucuronidation activities in HLM, whereas hecogenin exhibited no inhibition of amitriptyline glucuronidation and much less inhibition of imipramine, clomipramine, and trimipramine glucuronidation. The inhibition profiles at high TCA concentration were opposite to that observed at low TCA concentration; however, considering low therapeutic concentration of these drugs, observations at high TCA concentration might not be relevant in vivo. Similar inhibition profiles were also reported for amitriptyline (Dehal et al., 2001), for which the authors showed that hecogenin significantly inhibited amitriptyline glucuronidation at a high substrate concentration (200 μM) but not at a low concentration (10 μM). The selective inhibition and enzyme kinetic results observed here for recombinant UGT1A4, UGT2B10, and HLM clearly demonstrated that UGT2B10 would be the major UGT isoform responsible for amitriptyline glucuronidation at therapeutic concentrations in vivo. UGT2B10 and UGT1A4 probably contribute equally in the glucuronidation of imipramine, clomipramine, and trimipramine in vivo. Nicotine glucuronidation exhibited a high Km value of 290 μM in recombinant UGT2B10 (Kaivosaari et al., 2007), which was much higher than the Km values observed in this study for TCA glucuronidation by UGT2B10. This may be one of the reasons that 500 μM nicotine did not fully inhibit TCA glucuronidation in HLM.
Both UGT1A4 and UGT2B10 are primarily found in human liver, yet the relative protein abundance of each UGT isoform in human tissues needs further evaluation. Several recent publications applied the real-time reverse transcriptase-PCR method to determine mRNA copy numbers of each UGT isoform in human liver and other tissues (Izukawa et al., 2009; Ohno and Nakajin, 2009). The reverse transcriptase-PCR method can provide useful information on the mRNA levels of each UGT isoform but does not directly measure the protein level. These studies suggest that UGT2B10 is one of the most abundant isoforms in liver and in the same range as UGT1A4, indicating that contributions of UGT2B10 to the N-glucuronidation of drugs could be of great importance in vivo. In addition, drugs that modulate UGT2B10 activity in patients might affect the clearance of TCAs in vivo.
In conclusion, this study demonstrated for the first time that UGT2B10 was able to catalyze aliphatic tertiary amines to form quaternary N-glucuronide. This work also revealed that UGT2B10 was the high-affinity component for TCA glucuronidation in HLM. UGT2B10 would be a major UGT isoform responsible for the glucuronidation of amitriptyline, imipramine, clomipramine, and trimipramine at therapeutic concentrations in vivo. Nicotine could potentially be used as selective UGT2B10 inhibitor in enzyme identification of xenobiotic N-glucuronidation pathways in humans.
Footnotes
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
doi:10.1124/dmd.109.030981.
-
- UGT
- uridine diphosphate glucuronosyltransferase
- UDPGA
- uridine 5′-diphosphoglucuronic acid
- TCA
- tricyclic antidepressants
- HLM
- human liver microsome(s)
- PCR
- polymerase chain reaction
- bis-Tris
- 2-[bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-1,3-diol
- MOPS
- 4-morpholinepropanesulfonic acid.
- Received October 30, 2009.
- Accepted February 4, 2010.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics