Differential Effects of 1α,25-Dihydroxyvitamin D3 on the Expressions and Functions of Hepatic CYP and UGT Enzymes and Its Pharmacokinetic Consequences In Vivo

The compound 1α,25-Dihydroxyvitamin D3 (1,25(OH)2D3) is the active form of vitamin D3 and a representative ligand of the vitamin D receptor (VDR). Previous studies have described the impacts of 1,25(OH)2D3 on a small number of cytochrome P450 (CYP) and uridine diphosphate-glucuronyltransferase (UGT) enzymes, but comparatively little is known about interactions between several important CYP and UGT isoforms and 1,25(OH)2D3 in vitro and/or in vivo. Thus, we investigated the effects of 1,25(OH)2D3 on the gene and protein expressions and functional activities of selected CYPs and UGTs and their impacts on drug pharmacokinetics in rats. The mRNA/protein expressions of Cyp2b1 and Cyp2c11 were downregulated in rat liver by 1,25(OH)2D3. Consistently, the in vitro metabolic kinetics (Vmax and CLint) of BUP (bupropion; a Cyp2b1 substrate) and TOL (tolbutamide; a Cyp2c11 substrate) were significantly changed by 1,25(OH)2D3 treatment in liver microsomes, but the kinetics of acetaminophen (an Ugt1a6/1a7/1a8 substrate) remained unaffected, consistent with Western blotting data for Ugt1a6. In rat pharmacokinetic studies, the total body clearance (CL) and nonrenal clearance (CLNR) of BUP were significantly reduced by 1,25(OH)2D3, but unexpectedly, the total area under the plasma concentration versus time curve from time zero to infinity (AUC) of hydroxybupropion (HBUP) was increased probably due to a marked reduction in the renal clearance (CLR) of HBUP. Additionally, the AUC, CL, and CLNR for TOL and the AUC for 4-hydroxytolbutamide (HTOL) were unaffected by 1,25(OH)2D3 in vivo. Discrepancies between observed in vitro metabolic activity and in vivo pharmacokinetics of TOL were possibly due to a greater apparent distribution volume at the steady-state (Vss) and lower plasma protein binding in 1,25(OH)2D3-treated rats. Our results suggest possible drug-drug and drug-nutrient interactions and provide additional information concerning safe drug combinations and dosing regimens for patients taking VDR ligand drugs including 1,25(OH)2D3.


1,25(OH) 2 D 3 Treatment in Rats
Male Sprague-Dawley (SD) rats (8 weeks, 260-280 g) were purchased from Nara Biotech Co. (Seoul, Korea). Animals were maintained in cages under controlled conditions under a 12-h dark/light cycle with free access to food and tap water, acclimated for at least five days in the laboratory before experiments, and allocated to a control group or 1,25(OH) 2 D 3 -treated group. The 1,25(OH) 2 D 3 dosing solution was prepared by diluting 1,25(OH) 2 D 3 in ethanol with 5 mL of corn oil (2.56 nmol/mL of final concentration). The dosing solution for the control group was prepared but the 1,25(OH) 2 D 3 was omitted. The dosing solutions were administered intraperitoneally to rats at a dose of 1 mL/kg/day (2.56 nmol/kg/day as 1,25(OH) 2 D 3 ) over four consecutive days. On the fifth day, rats were used for pharmacokinetic study [2,11,12,16,27].

Liver Histology
A segment of liver was removed from control and 1,25(OH) 2 D 3 -treated rats, washed with PBS. After fixing the liver segment with 4% polyoxymethylene for 1 day, the liver segments were cut vertically into thin slices, followed by staining with hematoxylin and eosin (H&E) (Maxdiagnostics, Seoul, South Korea). The H&E-stained samples were examined under a light microscope (200×) (Olympus JP/IX70, Olympus Optical, Tokyo, Japan).

Serum Chemistry
Rat serum was obtained from whole blood on the fifth day after the 1,25(OH) 2 D 3 treatment. The assay of total protein, albumin, serum glutamic oxaloacetic transaminase (sGOT), and serum glutamic pyruvic transaminase (sGPT) was performed by Green Cross Reference Laboratory (Seoul, Korea).

Real-Time Quantitative Polymerase Chain Reaction
Collected livers were frozen immediately with liquid nitrogen and stored at −80 • C. Trizol reagent in RNAiso Plus (Takara Bio, Inc., Shiga, Japan) was used to extract RNA from 100-mg tissue homogenates according to the manufacturer's protocol. After purities and total RNA concentrations extracted from liver samples were determined at a wavelength of 260/280 nm with a Nanodrop 2000c spectrophotometer (Thermo Scientific, Waltham, MA, USA), synthesis of cDNA was processed from 1 µg of total RNA using the following conditions: incubation at 50 • C for 1 h, inactivation at 95 • C for 5 min, and cooling to 4 • C. The synthesized cDNA was then subjected to qPCR assays using SYBR ® Premix Ex TagTM on Stratagene Mx3005P (Agilent Technologies, Boblingen, Germany) using the following conditions: 95 • C for 10 min and 40 cycles of 95 • C for 15 s and 55 • C for 30 s. A "Comparative Quantitation" mode was selected, and fold expressions were calculated using the delta-delta method 2 −( Ct) . GAPDH was used as the internal reference gene for normalization. The forward and reverse primers used for qPCR analysis are listed in Supplementary Table S1 [2,25,28,29].

In Vitro Metabolic Study Using Rat Liver Microsomes (RLMs)
RLMs were prepared according to a previously described method [30]. Rat livers were homogenized in ice-cold microsomal buffer (0.154 M KCl and 1 mM EDTA in 50 mM Tris-HCl (pH 7.4)). Resulting homogenates were centrifuged at 10,000 × g for 30 min and supernatants were further centrifuged at 100,000× g for 90 min. RLMs were obtained by suspending microsomal pellets in microsomal buffer. Protein contents of RLMs were determined using Lowry reagent (Sigma Aldrich Co., St. Louis, MO, USA). RLMs were obtained independently from three different rats for each experimental group. For CYP activity assays, a mixture of RLMs (protein concentration 1 mg/mL), 1.2 mM NADPH, and 100 mM phosphate buffer (pH 7.4) was pre-incubated in a thermomixer for 5 min at 37 • C and at 200 opm. The metabolic reaction was initiated by adding a 2.5 µL aliquot of drug solution in methanol with DMSO (final concentration 1% MEOH with 0.1% DMSO) to the preincubated microsomal reaction mixture (total volume of 250 µL). A system control study using buspirone (1 µM) was first confirmed (data not shown). The concentration ranges used for model substrates (20-1000 µM for BUP; 20-2000 µM for TOL) were obtained from previous studies [31][32][33]. A preliminary study at a specific drug concentration was first performed to assess the linearity of metabolite formation rate and determine a suitable incubation time for each substrate (i.e., 30 min). For UGT activity assays, a microsomal incubation mixture comprised of RLMs (protein concentration 1 mg/mL), MgCl 2 (1 mM), ACET solution in DMSO and 100 mM phosphate buffer solution (pH 7.4) was pre-incubated in thermal mixer for 5 min at 37 • C and at 200 opm, and then 100 mM of UDPGA in phosphate buffer solution was added to initiate the reaction. A substrate concentration range from 0.1 to 30 mM was selected based on previous studies [29,34], and an incubation time of 90 min was chosen after a preliminary study with 1 mM ACET. Sampling was conducted at 0 min and at the end of incubation by transferring 50 µL aliquot of microsomal reaction mixture to 1.5 mL microcentrifuge tubes containing 100 µL of ice-cold internal standard (IS) solution in methanol. Resultant mixtures were immediately vortex-mixed to terminate the enzymatic reaction and then centrifuged at 15,000× g for 15 min at 4 • C. Supernatants were analyzed by ultra-performance liquid chromatography method with diode array detection (UPLC-DAD) or liquid chromatography-tandem mass spectrometry (LC-MS/MS) to determine metabolite levels.

In Vitro Plasma Protein Binding Study
A rat plasma protein binding study was performed using the rapid equilibrium dialysis kit (RED, 8 KDa molecular weight cut-off, Thermo Fisher Scientific, Waltham, MA, USA) according to a manufacture's protocol. The plasma containing drug (5 µg/mL) was placed into the sample chamber, and then dialysis buffer was added to the buffer chamber. Upon sealing the cover of the unit, incubation was applied for 4 h at 37 • C on a shaking water-bath. To determine percent unbound (f up , %) in plasma, the drug concentrations in the plasma and buffer samples were analyzed by LC-MS/MS.

In Vivo Pharmacokinetic Study in Rats
In vivo pharmacokinetic animal experiments were accomplished according to the Guide for the Care and Use of Laboratory Animals issued by the National Institute of Health, as described previously [2,10,11]. Before starting the animal studies, all experimental protocols were reviewed and approved by the Animal Care and Use Committee of Gachon University (Approval no. GIACUC-R2019020; approved on July 1st, 2019). SD male rats (8-9 weeks, 250-300 g, Nara Biotech Co., Seoul, South Korea) had free access to food and water, acclimatized and maintained in the room under a 12-h light/dark cycle for a week before the study. After pretreatment with 1,25(OH) 2 D 3 for 4 continuous days, the rats in the control and 1,25(OH) 2 D 3 groups were anesthetized with Zoletil ® (Vibrac, TX, USA) (10 mg/kg i.m.) and then the femoral vein and artery were cannulated with polyethylene tubing (Clay Adams, NJ, USA) for drug administration and blood sampling, respectively, as described previously [2,11,12]. The pharmacokinetic experiment was initiated by drug administration after the rats were recovered from anesthesia. In the pharmacokinetic study of BUP, BUP in saline was administered intravenously at 5 mg/kg to control and 1,25(OH) 2 D 3 -treated rats. Approximately 100 µL of blood was collected via the femoral artery at 0, 1, 5, 15, 30, 60, 120, 180, 240, 360, 480, and 600 min after the drug administration. In pharmacokinetic study of TOL, TOL in vehicle (DMSO:PEG400:saline = 5:40:55, v/v/v) was administered intravenously at 2 mg/kg. Blood samples were collected at 0, 1, 5, 15, 30, 60, 90, 120, 180, 240, 360, and 480 min later and immediately centrifuged at 14,000× g for 15 min at 4 • C. For compensation of fluid loss, a same volume of saline was intravenously provided after each sample collection. Plasma was then separated from whole blood cells and stored at −20 • C for further analysis. Urine samples were collected 0-4, 4-8, and 8-24 h after drug administration. The collected urine samples were diluted with distilled deionized water (DDW) 20-fold prior to LC-MS/MS analysis [11,12]. For the tissue distribution study of TOL, we sacrificed rats at 8 h after intravenous injection of the drug, and several organs including liver, and kidney, brain, spleen, and heart were taken, as described previously [11]. The weighted tissues were homogenized on ice by an electric homogenizer following adding 2-fold volume of PBS. The homogenates were stored before LC-MS/MS analysis.

Sample Preparation
Calibration standards for plasma, diluted urine samples and tissue homogenate samples of parent drugs and metabolites were prepared by mixing 10 µL of standard working solution with 90 µL of blank rat plasma, blank diluted urine, or blank tissue homogenate. Then, 200-µL IS solution was added to 100-µL biological samples and vortex-mixed for 1 min. Following centrifugation at 14,000× g for 15 min at 4 • C, supernatants were analyzed by UPLC-DAD or LC-MS/MS.

UPLC-DAD Analysis
UPLC-DAD was conducted using an Agilent Technologies 1290 Infinity II UHPLC system (Agilent Technologies, Santa Clara, CA, USA) with an autosampler (G7167B), a flexible pump (G7104A), Pharmaceutics 2020, 12, 1129 6 of 17 a multicolumn thermostat (MCT-G7116B) a DAD detector (G7117A), and a Luna Omega Polar C18 column (100 × 2.1 mm, 1.6 µm; Phenomenex, Torrance, CA, USA). The mobile phase was a mixture of 0.2% acetic acid (pH 3.8, solvent A) and acetonitrile (ACN) (solvent B). For ACET, the mobile phase was eluted using the following gradient program: 10 v/v % solvent B for 3.5 min, 10 to 30 v/v % solvent B over 0.5 min, 30 v/v % solvent B for 6 min, and 10 v/v % solvent B for 5 min. For the measurement of HTOL in microsomal study, the mobile phase was eluted using the following gradient program: 25 v/v % solvent B for 5.5 min; 25 to 35 v/v % solvent B over 1 min; 35 v/v % solvent B for 8.5 min; and 25 v/v % solvent B for 5 min. For AG, the mobile phase consisted of 96 v/v % solvent A and 4 v/v % solvent B was elution isocratically for 19 min. ACET, AG, and HTOL were detected at 245, 245, and 230 nm, respectively. Rutin, CARB, and THEO were used as ISs for the analyses of ACET, HTOL, and AG, respectively. The flow rate used was 1 mL/min, and the sample injection volume was 5 µL for all the analytes except AG (10 µL).

Pharmacokinetic Analysis
In vitro K m and V max for metabolic reactions in RLMs were calculated based on the Michaelis-Menten equation using Sigma Plot software (Jandel Scientific, San Rafael, CA, USA). CL int was calculated by dividing the V max by the K m . Percentage (%) of free form was calculated by the concentration of buffer chamber by dividing the concentration of plasma chamber in the protein binding study. The following pharmacokinetic parameters were determined by non-compartmental analysis using WinNonlin ® 8.3 (Pharsight Co., Mountain View, CA, USA): total area under the plasma concentration versus time curve from time zero to infinity (AUC); total body clearance (CL), elimination half-life (t 1/2 ); apparent distribution volume at the steady-state (V ss ); the first moment of AUC (AUMC); and the mean residence time (MRT). CL R was obtained by dividing the accumulated drug amount excreted in urine over 24 h by AUC [11], assuming the urinary recovery of the drug was completed 24 h after drug administration. Moreover, CL NR was calculated to subtract CL R from CL.

Statistical Analysis
p-values of <0.05 were considered to be statistically significant as determined by the two-tailed Student t-test between unpaired means for control and treatment groups. Results are presented as means ± standard deviation (SD)s, except T max values, which are expressed as medians (ranges).

Effects of 1,25(OH)2D3 on the Functional Activities of CYPs and UGTs in RLMs
Using RLMs, we continued to investigate whether significant expressional changes of Cyp2b1, Cyp2c11, and Ugt1a6 by 1,25(OH)2D3 affected metabolic activity in vitro or not. Concentration dependencies of the metabolic activities (i.e., metabolite formation rates) of BUP, TOL, and ACET using RLMs from control and 1,25(OH)2D3 -treated rats were observed to determine the formed The asterisks indicate statistically significant differences compared to the control group (* p < 0.05).

Effects of 1,25(OH) 2 D 3 on the Functional Activities of CYPs and UGTs in RLMs
Using RLMs, we continued to investigate whether significant expressional changes of Cyp2b1, Cyp2c11, and Ugt1a6 by 1,25(OH) 2 D 3 affected metabolic activity in vitro or not. Concentration dependencies of the metabolic activities (i.e., metabolite formation rates) of BUP, TOL, and ACET using RLMs from control and 1,25(OH) 2 D 3 -treated rats were observed to determine the formed metabolites, HBUP, HTOL, and AG, as shown in Figures 4 and 5.   Tables 2 and 3, respectively. Notably, Vmax values for the metabolism of both BUP and TOL were significantly lower in 1,25(OH)2D3-treated rats than in controls (p = 0.0025 for BUP and 0.0012 for TOL), consistent with quantitative real-time PCR data, but no significant intergroup difference was observed for Km values. Consequently, calculated CLint values were   Tables 2 and 3, respectively. Notably, Vmax values for the metabolism of both BUP and TOL were significantly lower in 1,25(OH)2D3-treated rats than in controls (p = 0.0025 for BUP and 0.0012 for TOL), consistent with quantitative real-time PCR data, but no significant intergroup difference was observed for Km values. Consequently, calculated CLint values were  Tables 2 and 3, respectively. Notably, V max values for the metabolism of both BUP and TOL were significantly lower in 1,25(OH) 2 D 3 -treated rats than in controls (p = 0.0025 for BUP and 0.0012 for TOL), consistent with quantitative real-time PCR data, but no significant intergroup difference was observed for K m values. Consequently, calculated CL int values were significantly lower in 1,25(OH) 2 D 3 -treated rats than in controls (by 82%, p = 0.0075 for BUP and by 70%, 0.0051 for TOL) ( Table 2). In contrast, no significant intergroup difference was observed for K m , V max , and CL int values for the metabolism of ACET (Table 3).

Effects of 1,25(OH) 2 D 3 on the Pharmacokinetics of BUP and TOL
Since calculated CL int values for BUP and TOL were found to be significantly changed in vitro, we further investigated the effects of 1,25(OH) 2 D 3 on the pharmacokinetics of BUP (Cyp2b1 substrate) and TOL (Cyp2c11 substrate) in vivo. Plasma concentration-time curves of BUP and HBUP (formed metabolite) after intravenous administration of 5 mg/kg BUP in control and 1,25(OH) 2 D 3 treated rats are presented in Figure 6. Plasma concentration levels of both BUP and HBUP were increased in the 1,25(OH) 2 D 3 treated group, compared to control group. The pharmacokinetic parameters of BUP and its formed metabolite, HBUP, are provided in Table 4. Several obvious alterations were found in the pharmacokinetics of BUP in rats treated with 1,25(OH) 2 D 3 . In particular, the AUC of BUP was significantly higher by 60.7% (p < 0.001), as expected. CL and CL NR were significantly lower by 34% and 34.8% (p = 0.0004 for CL and p = 0.0003 for CL NR ), respectively, in 1,25(OH) 2 D 3 -treated rats. However, 1,25(OH) 2 D 3 -treated rats did not exhibit any significant change in MRT or terminal half-life versus controls. Notably, the AUC of HBUP and the AUC ratio between HBUP and BUP were significantly higher (p = 0.001 for AUC HBUP and p = 0.015 for AUC HBUP /AUC BUP ) and the CL R of HBUP was significantly lower (p = 0.009) in 1,25(OH) 2 D 3 -treated rats (Table 4).  In addition, the Figure 7 shows the plasma concentration-time profiles of TOL (Cyp2c11 substrate, Figure 7A) and HTOL (formed metabolite, Figure 7B) after intravenous administration of 2 mg/kg TOL to control and 1,25(OH)2D3-treated rats, respectively, and the calculated pharmacokinetic parameters are summarized in Table 5. No significant intergroup difference was found between the AUC values of TOL and HTOL (p = 0.563 for TOL and p = 0.0871 for HTOL). However, the CLR values of TOL and  In addition, the Figure 7 shows the plasma concentration-time profiles of TOL (Cyp2c11 substrate, Figure 7A) and HTOL (formed metabolite, Figure 7B) after intravenous administration of 2 mg/kg TOL to control and 1,25(OH) 2 D 3 -treated rats, respectively, and the calculated pharmacokinetic parameters are summarized in Table 5. No significant intergroup difference was found between the AUC values of TOL and HTOL (p = 0.563 for TOL and p = 0.0871 for HTOL). However, the CL R values of TOL and HTOL were significantly lower in 1,25(OH) 2 D 3 -treated rats (p = 0.0405 for TOL and p = 0.00304 for HTOL). In addition, the V ss of TOL in 1,25(OH) 2 D 3 -treated rats was significantly greater than that in control rats (p = 0.000807, Table 5).
Pharmaceutics 2020, 12, x FOR PEER REVIEW 12 of 17 HTOL were significantly lower in 1,25(OH)2D3-treated rats (p = 0.0405 for TOL and p = 0.00304 for HTOL). In addition, the Vss of TOL in 1,25(OH)2D3-treated rats was significantly greater than that in control rats (p = 0.000807, Table 5).  Then, we further investigated tissue distribution of TOL and HTOL at the terminal phase (i.e., 8 hr) after intravenous administration of TOL. Table 6 summarizes tissue to plasma concentration ratios (Kp) of TOL and HTOL for liver, kidney, spleen, heart, and brain in both groups. To be  Table 5. Pharmacokinetic parameters of TOL and HTOL after intravenous administration of 2 mg/kg TOL in control and 1,25(OH) 2 D 3 -treated rats (n = 9-10). Then, we further investigated tissue distribution of TOL and HTOL at the terminal phase (i.e., 8 h) after intravenous administration of TOL. Table 6 summarizes tissue to plasma concentration ratios (K p ) of TOL and HTOL for liver, kidney, spleen, heart, and brain in both groups. To be consistent with the systemic pharmacokinetic result (i.e., increased V ss ), the tissue to plasma concentration ratios of TOL in most of tissues such as liver, brain, spleen, and heart were significantly increased (p < 0.05), except the kidney, suggesting that tissue distribution of TOL was higher in 1,25(OH) 2 D 3 -treated rats. Similarly, the formed metabolite, HTOL, also showed significantly increased tissue to plasma concentration ratios for the kidney, spleen, and heart, compared to control group (p < 0.05). Table 6. Tissue to plasma concentration ratios (K p ) of TOL and HTOL at 8 h after intravenous injection of 2 mg/kg TOL in control and 1,25(OH) 2 D 3 -treated rats (n = 4-6). In addition, when the percentage of free form (f up , %) for TOL was compared between control and 1,25(OH) 2 D 3 -treated group using rat plasma protein binding assay, f up values were found to be 3.51 ± 0.73% and 13.1 ± 0.3% (n = 3 per group, p < 0.05), respectively. This result suggests that 1,25(OH) 2 D 3 -treatment may affect the plasma protein binding of TOL in rats, resulting in the increased V ss in vivo consequently.

Discussion
In the present study, a rat model was chosen to investigate the effect of 1,25(OH) 2 D 3 on hepatic CYPs and UGTs, due to the similar hepatic VDR distributions in human and rats [35,36]. In a recent study, we reported that 1,25(OH) 2 D 3 treatment affects intravenous and oral pharmacokinetics of buspirone, a CYP3A substrate, in rats due to the differential regulation of hepatic and intestinal CYP3A metabolic activities. Likewise, we designed the present study to investigate if the regulating effect of 1,25(OH) 2 D 3 (the active form of vitamin D) on the expressions and activities of hepatic CYPs and UGTs other than CYP3A, and its consequences on pharmacokinetics of the specific substrates in vivo. Among the five Cyps and six Ugts enzymes tested, the mRNA and/or protein expression levels of hepatic Cyp2b1, Cyp2c11, and Ugt1a6 were found to be significantly reduced by 1,25(OH) 2 D 3 , whereas those of other enzymes were unaltered (Figures 2 and 3). The extent of change in protein levels for Cyp2b1 or Cyp2c11 coincided with its mRNA results.
The H&E staining data with serum chemistry data shows that the apparent liver damage by 1,25(OH) 2 D 3 is unlikely (Figure 1 and Table 1). However, other effects on hepatocytes, such as nucleoli and lipid accumulation, by 1,25(OH) 2 D 3 treatment still need to be investigated. For examples, drug-nutrient interactions and provide practical information on effective and safe drug combinations and dosing regimens for patients taking VDR ligand drugs such as 1,25(OH) 2 D 3 .

Conclusions
The current study shows the mRNA/protein expressions of Cyp2b1 and Cyp2c11, are downregulated by 1,25(OH) 2 D 3 in rats. Furthermore, the in vitro metabolic kinetics (V max and CL int ) of BUP (a Cyp2b1 substrate) and TOL (a Cyp2c11 substrate) were significantly changed by treating RLMs with 1,25(OH) 2 D 3 , as indicated by qPCR and Western blotting results, whereas that of acetaminophen (an Ugt1a6/1a7/1a8 substrate) was unaffected, consistent with data from Western blot analysis. Rat pharmacokinetic studies showed the CL and CL NR of BUP were significantly reduced by 1,25(OH) 2 D 3 treatment, but that surprisingly, the AUC of HBUP was increased (probably due to the markedly reduced CL R of HBUP). Nevertheless, the AUC, CL, and CL NR of TOL and the AUC of HTOL remained unchanged in vivo. These discrepancies between in vitro metabolic activities and in vivo pharmacokinetics of TOL might be partially due to a greater V ss and lower plasma protein binding in 1,25(OH) 2 D 3 -treated rats. To the best of our knowledge, the present study is the first to describe the impacts of 1,25(OH) 2 D 3 on metabolic functions and systemic pharmacokinetics of BUP and TOL in rats.