Mertansine Inhibits mRNA Expression and Enzyme Activities of Cytochrome P450s and Uridine 5′-Diphospho-Glucuronosyltransferases in Human Hepatocytes and Liver Microsomes

Mertansine, a tubulin inhibitor, is used as the cytotoxic component of antibody–drug conjugates (ADCs) for cancer therapy. The effects of mertansine on uridine 5′-diphospho-glucuronosyltransferase (UGT) activities in human liver microsomes and its effects on the mRNA expression of cytochrome P450s (CYPs) and UGTs in human hepatocytes were evaluated to assess the potential for drug–drug interactions (DDIs). Mertansine potently inhibited UGT1A1-catalyzed SN-38 glucuronidation, UGT1A3-catalyzed chenodeoxycholic acid 24-acyl-β-glucuronidation, and UGT1A4-catalyzed trifluoperazine N-β-d-glucuronidation, with Ki values of 13.5 µM, 4.3 µM, and 21.2 µM, respectively, but no inhibition of UGT1A6, UGT1A9, and UGT2B7 enzyme activities was observed in human liver microsomes. A 48 h treatment of mertansine (1.25–2500 nM) in human hepatocytes resulted in the dose-dependent suppression of mRNA levels of CYP1A2, CYP2B6, CYP3A4, CYP2C8, CYP2C9, CYP2C19, UGT1A1, and UGT1A9, with IC50 values of 93.7 ± 109.1, 36.8 ± 18.3, 160.6 ± 167.4, 32.1 ± 14.9, 578.4 ± 452.0, 539.5 ± 233.4, 856.7 ± 781.9, and 54.1 ± 29.1 nM, respectively, and decreased the activities of CYP1A2-mediated phenacetin O-deethylase, CYP2B6-mediated bupropion hydroxylase, and CYP3A4-mediated midazolam 1′-hydroxylase. These in vitro DDI potentials of mertansine with CYP1A2, CYP2B6, CYP2C8/9/19, CYP3A4, UGT1A1, and UGT1A9 substrates suggest that it is necessary to carefully characterize the DDI potentials of ADC candidates with mertansine as a payload in the clinic.

After an intravenous injection of [ 3 H]-mertansine at 0.2 mg/kg in rats, the radioactivity of mertansine was rapidly cleared from the blood and extensively distributed to highly perfused organs such as liver, kidney, spleen, lungs, heart, adrenal, and the gastrointestinal tract with high tissue-toblood radioactivity ratios (ca. 1~11) for 24 h, declining to minimal levels by 120 h [28]. The majority of dosed mertansine radioactivity was recovered in feces over 120 h, with biliary excretion as the major route (~46% of dosed radioactivity over 72 h), but 5% of dosed radioactivity was recovered in urine over 120 h [28,29]. Mertansine was extensively metabolized to 11 metabolites via S-oxidation, hydrolysis, S-methylation, and glutathione conjugation [28,30,31]. It competitively inhibited CYP2C8-mediated paclitaxel 6α-hydroxylation and CYP2D6-mediated dextromethorphan Odemethylation with Ki values of 11 and 14 µ M, respectively, in human liver microsomes; mertansine also inactivated midazolam 1′-hydroxylation in recombinant human CYP3A4 with a Ki of 3.4 µ M and a kinact of 0.058 min −1 , but it exhibited no induction potential up to 1 µ M [31,32].
Other tubulin inhibitors, such as colchicine and monomethyl auristatin E (MMAE), have been reported to downregulate CYP mRNA expression through the disruption of the microtubulin cellular skeletal structure that is necessary for the proper functioning of nuclear receptor signaling cascades [33][34][35]. However, to our knowledge, no studies have investigated the inhibitory potential of mertansine on UGTs, the second major group of enzymes responsible for drug metabolism [27], in human liver microsomes and the suppression potential of mertansine on mRNA expression or activities of major CYPs and UGTs in human hepatocytes. Cytochrome P450s (CYPs) and uridine-5 -diphospho-glucuronosyltransferases (UGTs) are critical drug-metabolizing enzymes and are often involved in drug-drug interactions (DDIs) [21][22][23][24][25][26][27]. The in vitro inhibitory and induction potentials of drugs on CYPs and UGTs in human liver microsomes and hepatocytes have been evaluated to help identify clinical DDIs [26].
After an intravenous injection of [ 3 H]-mertansine at 0.2 mg/kg in rats, the radioactivity of mertansine was rapidly cleared from the blood and extensively distributed to highly perfused organs such as liver, kidney, spleen, lungs, heart, adrenal, and the gastrointestinal tract with high tissue-to-blood radioactivity ratios (ca. 1~11) for 24 h, declining to minimal levels by 120 h [28]. The majority of dosed mertansine radioactivity was recovered in feces over 120 h, with biliary excretion as the major route (~46% of dosed radioactivity over 72 h), but 5% of dosed radioactivity was recovered in urine over 120 h [28,29]. Mertansine was extensively metabolized to 11 metabolites via S-oxidation, hydrolysis, S-methylation, and glutathione conjugation [28,30,31]. It competitively inhibited CYP2C8-mediated paclitaxel 6α-hydroxylation and CYP2D6-mediated dextromethorphan O-demethylation with K i values of 11 and 14 µM, respectively, in human liver microsomes; mertansine also inactivated midazolam 1 -hydroxylation in recombinant human CYP3A4 with a K i of 3.4 µM and a k inact of 0.058 min −1 , but it exhibited no induction potential up to 1 µM [31,32].
Other tubulin inhibitors, such as colchicine and monomethyl auristatin E (MMAE), have been reported to downregulate CYP mRNA expression through the disruption of the microtubulin cellular skeletal structure that is necessary for the proper functioning of nuclear receptor signaling cascades [33][34][35]. However, to our knowledge, no studies have investigated the inhibitory potential of mertansine on UGTs, the second major group of enzymes responsible for drug metabolism [27], in human liver microsomes and the suppression potential of mertansine on mRNA expression or activities of major CYPs and UGTs in human hepatocytes.  The purpose of this study was to investigate the in vitro inhibitory potentials of mertansine on  human UGT activities including UGT1A1, UGT1A3, UGT1A4, UGT1A6, UGT1A9, and UGT2B7 in  ultrapooled human liver microsomes and to evaluate the effect of mertansine on the mRNA levels of  human CYP1A2, CYP2B6, CYP3A4, CYP2C8, CYP2C9, CYP2C19, UGT1A1, UGT1A4, and UGT1A9 in human hepatocytes to assess the potential for mertansine-induced drug interactions.

Inhibitory Potential of Mertansine on Human Major UGTs in Human Liver Microsomes
The inhibitory potential of mertansine on UGT1A1, UGT1A3, UGT1A4, UGT1A6, UGT1A9, and UGT2B7 activities was evaluated using liquid chromatography-tandem mass spectrometry (LC-MS/MS) with a cocktail of UGT substrates and ultrapooled human liver microsomes [36]. Each incubation mixture was prepared to a final volume of 100 µL as follows: ultrapooled human liver microsomes (0.2 mg/mL), 5 mM UDPGA, 10 mM magnesium chloride, alamethicin (25 µg/mL), 50 mM Tris buffer (pH 7.4), various concentrations of mertansine in methanol (final concentrations of 0.01-50 µM), and the cocktail sets of UGT enzyme-specific substrates. Two cocktail sets were used: set A contained 0.5 µM SN-38 for UGT1A1, 2 µM chenodeoxycholic acid for UGT1A3, and 0.5 µM trifluoperazine for UGT1A4; and set B contained 1 µM N-acetylserotonin for UGT1A6, 0.2 µM mycophenolic acid for UGT1A9, and 1 µM naloxone for UGT2B7. The reactions were initiated by adding UDPGA and incubated in a shaking water bath for 60 min at 37 • C. Reactions were terminated by adding 50 µL of ice-cold acetonitrile containing internal standards (IS, propofol glucuronide for set A and meloxicam for set B). Incubation mixtures were centrifuged at 13,000 g for 8 min at 4 • C. Next, 50 µL of each supernatant of sets A and B was mixed, and aliquots (5 µL) were analyzed using LC-MS/MS. All assays were performed in triplicate and average values were used in the analysis. The LC-MS/MS system was comprised of an Agilent 6495 triple quadrupole mass spectrometer coupled with an Agilent 1290 Infinity system (Agilent Technologies, Wilmington, DE, USA). The column and autosampler temperatures were set to 40 • C and 4 • C, respectively. Six glucuronide metabolites and two ISs were simultaneously separated using an Atlantis dC 18 system (3 µm, 2.1 mm i.d. ×100 mm, Waters Technologies, Milford, MA, USA) with a gradient elution of 5% acetonitrile in 0.1% formic acid (MP A) and 95% acetonitrile in 0.1% formic acid (MP B) at a flow rate of 0.3 mL/min. Separation was achieved using the following sequence: 10% MP B for 1 min, 10% to 60% MP B for 1 min, 60% to 95% MP B for 1 min, 95% MP B for 2 min, 95% to 10% MP B for 0.1 min, and 10% MP B for 2.9 min. The electrospray ionization (ESI) source settings in both positive and negative ion modes were as follows: gas temperature, 200 • C; gas flow, 14 L/min; nebulizer, 40 psi; sheath gas temperature, 380 • C; sheath gas flow, 11 L/min; capillary voltage, 4500 V; and nozzle voltage, 500 V. Each metabolite was quantified via selected reaction monitoring in the negative ion mode (chenodeoxycholic acid 24-acyl-β-glucuronide, m/z 567.1 to 391.

Induction of Mertansine on Human Major CYPs and UGTs in Human Hepatocytes
Plateable cryopreserved human hepatocytes (lots 319, 321, and 361) were thawed in cryopreserved hepatocyte recovery medium according to the manufacturer's protocol.

Cytotoxicity of Mertansine in Human Hepatocytes
To estimate the cytotoxicity of mertansine, viable hepatocytes (lot 319) cells were seeded in a collagen type 1 precoated 96-well plate in 100 µL of hepatocyte plating medium (6 × 10 4 cells/well) and incubated for 4 h at 37 • C in 5% CO 2 . Next, the plating medium was removed, and a matrigel medium containing 0.25 mg/mL of Matrigel™ matrix was applied to each cell prior to incubation for 24 h at 37 • C in 5% CO 2 . The hepatocytes were incubated with 0.0125, 0.0625, 0.125, 0.250, 0.625, 1.25, 2.5, and 6.25 µM mertansine in triplicate for 48 h at 37 • C in 5% CO 2 . The medium was exchanged with fresh medium containing mertansine every 24 h. Then, 20 µL of MTS solution was added to each well and the plate was incubated for 1 h at 37 • C in 5% CO 2 . The absorbance of the reaction mixture was measured at 492 nm.

Treatment of Mertansine in Human Hepatocytes
To evaluate the induction effect of mertansine on drug-metabolizing enzymes, three different cryopreserved human hepatocytes (lots 319, 321, and 361) were thawed in cryopreserved hepatocyte recovery medium, and viable cells were seeded in collagen type 1 precoated 48-well plates in 250 µL of hepatocyte plating medium (6 × 10 5 cells/well) and incubated for 4 h at 37 • C in 5% CO 2 . Next, the plating medium was removed and replaced with matrigel medium containing 0.25 mg/mL of Matrigel™ matrix prior to incubation at 37 • C for 24 h. The hepatocytes were incubated with 1.25, 12.5, 125, 625, 1250, and 2500 nM mertansine, vehicle (0.1% DMSO in hepatocyte culture media), and prototypical inducers including 50 µM omeprazole, 10 nM CITCO, and 10 µM rifampicin in triplicate. Samples were incubated for 48 h at 37 • C in 5% CO 2 , and the medium was exchanged with 250 µL of fresh medium containing drugs or the vehicle every 24 h.

CYP1A2, CYP2B6, and CYP3A4 Activity Measurement
The effects of mertansine on CYP1A2, CYP2B6, and CYP3A4 activities were evaluated. Plates were prepared with a vehicle, omeprazole, CITCO, rifampin, and mertansine, and incubated for 48 h. Next, 150 µL of a CYP cocktail solution containing 40 µM phenacetin (CYP1A2 substrate), 20 µM bupropion (CYP2B6 substrate), and 20 µM midazolam (CYP3A4 substrate) in William's E buffer was added to each well and incubated for 30 min, and then 100 µL aliquots of the incubate from each well were stored at −80 • C until LC-MS/MS analysis. 13 C 2 , 15 N-acetaminophen (0.1 µg/mL, IS for acetaminophen), and d 9 -1 -hydroxybufuralol (0.01 µg/mL, IS for hydroxybupropion and 1 -hydroxymidazolam) in methanol were added to 50 µL of the medium obtained from each well. Mixtures were vortexed for 2 min and then centrifuged at 13,000 g for 4 min at 4 • C. The supernatant (40 µL) was diluted with 60 µL of deionized water and then mixed for 2 min by vortexing. An aliquot (5 µL) was analyzed using LC-MS/MS [37], and CYP1A2, CYP2B6, and CYP3A4 enzyme activities were expressed as formation rates (pmol/million cells/min).

RNA Purification and RT-PCR Analysis
At the end point of the experiment, total RNA was immediately isolated using an RNeasy Micro Kit, and RNA concentration and purity were determined using an absorbance test at 260 nm/280 nm using a NanoVue Plus spectrophotometer (GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA). Samples were stored at −80 • C until RT-PCR analysis.

Data Analysis
The percentage changes in enzymatic activities were calculated as (CYP activity with test compound treatment/CYP activity with vehicle control treatment) × 100. The IC 50 (the concentration of the inhibitor needed for half-maximal inhibition) values were calculated using SigmaPlot ver. 12.5 (Systat Software, Inc.; San Jose, CA, USA). K i (the inhibition constant) and the inhibition mode of UGT1A1, UGT1A3, and UGT1A4 activities were determined using Enzyme Kinetics ver. 1.1 (Systat Software, Inc.).

Effects of Mertansine on CYP and UGT mRNA Levels in Human Hepatocytes
In the MTS colorimetric assay, mertansine did not cause toxicity in human hepatocytes (lot 319), as the viability of hepatocytes following 48 h mertansine treatment (1.25-6250 nM) was over 96.2%.

Effects of Mertansine on CYP and UGT mRNA Levels in Human Hepatocytes
In the MTS colorimetric assay, mertansine did not cause toxicity in human hepatocytes (lot 319) as the viability of hepatocytes following 48 h mertansine treatment (1.25-6250 nM) was over 96.2%.

Effects of Mertansine on CYP and UGT mRNA Levels in Human Hepatocytes
In the MTS colorimetric assay, mertansine did not cause toxicity in human hepatocytes (lot 319), as the viability of hepatocytes following 48 h mertansine treatment (1.25-6250 nM) was over 96.2%.

Effects of Mertansine on CYP and UGT mRNA Levels in Human Hepatocytes
In the MTS colorimetric assay, mertansine did not cause toxicity in human hepatocytes (lot 319), as the viability of hepatocytes following 48 h mertansine treatment (1.25-6250 nM) was over 96.2%.

Effects of Mertansine on CYP and UGT mRNA Levels in Human Hepatocytes
In the MTS colorimetric assay, mertansine did not cause toxicity in human hepatocytes (lot 319), as the viability of hepatocytes following 48 h mertansine treatment (1.25-6250 nM) was over 96.2%.
The functionality of the hepatocyte was confirmed by the increase of mRNA levels and enzyme activities of CYPs following 48 h treatment with prototypical inducers using RT-PCR and LC-MS/MS, respectively, compared to the vehicle ( Table 2). Fifty micromoles of omeprazole, a representative aromatic hydrocarbon receptor inducer (AHR), increased the CYP1A2 mRNA levels by enhancing the AHR binding to the promoter region of CYP1A2 [38] and CYP1A2-mediated phenacetin Odeethylase activity by 58.7-299.3 and 11.7-61.8 fold, respectively ( Table 2)

Effects of Mertansine on CYP and UGT mRNA Levels in Human Hepatocytes
In the MTS colorimetric assay, mertansine did not cause toxicity in human hepatocytes (lot 319), as the viability of hepatocytes following 48 h mertansine treatment (1.25-6250 nM) was over 96.2%.
The functionality of the hepatocyte was confirmed by the increase of mRNA levels and enzyme activities of CYPs following 48 h treatment with prototypical inducers using RT-PCR and LC-MS/MS, respectively, compared to the vehicle (Table 2). Fifty micromoles of omeprazole, a representative aromatic hydrocarbon receptor inducer (AHR), increased the CYP1A2 mRNA levels by enhancing the AHR binding to the promoter region of CYP1A2 [38] and CYP1A2-mediated phenacetin Odeethylase activity by 58.7-299.3 and 11.7-61.8 fold, respectively (Table 2). 10 μM rifampin, a potent pregnane X receptor (PXR) inducer, increased mRNA levels of CYP3A4 by enhancing the PXR binding to the promoter region of CYP3A4 [39] and CYP3A4-mediated midazolam 1-hydroxylase by 74.0-146.7 and 3.6-9.8 fold, respectively (Table 2). Additionally, 10 nM CITCO increased CYP2B6 mRNA levels and CYP2B6-mediated bupropion hydroxylase activity by 5.6-8.7 and 3.8-15.7 fold, respectively (Table 2)

Discussion
In this study, the effects of mertansine on the inhibition of UGT activities in human liver microsomes and its effects on mRNA expression of CYPs and UGTs in human hepatocytes were evaluated to assess the potential for mertansine-induced drug interactions.
Mertansine was a noncompetitive inhibitor of UGT1A1-catalyzed SN-38 glucuronidation with a K i value of 13.5 µM, and a competitive inhibitor of UGT1A3-catalyzed 24-acyl-β-glucuronidation and UGT1A4-catalyzed trifluoperazine N-β-d-glucuronidation with K i values of 4.3 and 21.2 µM, respectively, in human liver microsomes (Figure 3). These findings suggest the potential for DDIs between mertansine and UGT1A1, UGT1A3, or UGT1A4 substrates when used concomitantly. However, the maximum plasma concentrations of mertansine were 7.2 ± 2.7 nM, with the highest level of 30 nM after the intravenous infusion of 3.6 mg/kg T-DM1 every 3 weeks in HER2-positve breast cancer patients [16][17][18][19]. Therefore, the ratio of maximal unbound plasma concentrations of mertansine to K i values (0.00004-0.0002) was much lower than the ratio indicating the likelihood of drug interaction (0.1), suggesting that mertansine-induced drug interactions via the inhibition of UGT activity are unlikely during T-DM1 therapies.
In addition, although mertansine is a competitive inhibitor of CYP2C8 and CYP2D6 activities with K i values of 11 and 14 µM, respectively, and it also irreversibly inhibits CYP3A4 activity with K i of 3.4 µM and k inact of 0.058 min −1 , mertansine would not cause serious CYP-mediated DDI during the T-DM1 therapies considering the plasma concentrations [32].
Several ADCs with mertansine as a payload have been under clinical trials since the approval of T-DM1 [7,[10][11][12][13][14]. Liver is the major organ for the distribution and metabolism of antibody maytansinoid conjugates and its catabolites, and the hepatic concentrations of mertansine or ravtansine therefore depend on the catabolism of ADC within the liver [7,18,29,42]. The extensive tissue distribution of mertansine after the administration of mertansine itself in rats led to higher hepatic levels of mertansine compared to plasma levels [28]. Although the maximal plasma concentration of the catabolite mertansine is low (≤7.2 ± 2.7 nM) in T-DM1 treated cancer patients [16][17][18][19], a clinical evaluation of DDIs regarding the reduced mRNA levels by repeated treatment of T-DM1 and the CYP1A2, CYP2B6, CYP2C8/9/19, CYP3A4, UGT1A1, and UGT1A9 substrates may be necessary on the basis of these in vitro findings.

Conclusions
Mertansine inhibited UGT1A1, UGT1A3, and UGT1A4 enzyme activities in human liver microsomes and dose-dependently suppressed the mRNA levels of CYP1A2, CYP2B6, CYP3A4, CYP2C8, CYP2C9, CYP2C19, UGT1A1, UGT1A4, and UGT1A9 after 48 h treatment of 1.25-2500 nM mertansine in three human hepatocytes. Additionally, mertansine treatment resulted in the decrease of CYP1A2, CYP2B6, and CYP3A4 enzyme activities. These in vitro DDI potentials of mertansine with substrate drugs for major CYPs and UGTs enzymes indicate that the evaluation of the DDI potentials of ADC candidates with mertansine as a payload is necessary.