Human Sulfotransferases Enhance the Cytotoxicity of Tolvaptan

Abstract

Tolvaptan, a vasopressin receptor 2 antagonist used to treat hyponatremia, has recently been reported to be associated with liver injury. Sulfotransferases (SULTs) have been implicated as important detoxifying and/or activating enzymes for numerous xenobiotics, drugs, and endogenous compounds. To characterize better the role of SULTs in tolvaptan metabolism, HEK293 cells stably overexpressing 12 human SULTs were generated. Using these cell lines, the extent of tolvaptan sulfate formation was assessed by reversed-phase high-performance liquid chromatography through comparison to a synthetic standard. Of the 12 known human SULTs, no detectable sulfation of tolvaptan was observed with SULT1A1, SULT1A2, SULT1A3, SULT1C2, SULT1C4, SULT4A1, or SULT6B1. The affinity of individual SULT isozymes, as determined by K_m analysis, was SULT1C3 >> SULT2A1 > SULT2B1 ~ SULT1B1 > SULT1E1. The half inhibitory concentration of tolvaptan on cell growth in HEK293/SULT1C3 cells and HEK293/CYP3A4 & SULT1C3 cells was significantly lower than that in the corresponding HEK293/vector cells or HEK293/CYP3A4 & SULT vector cells. Moreover, exposing cells to tolvaptan in the presence of cyclosporine A, an inhibitor of the drug efflux transporters, significantly increased the intracellular levels of tolvaptan sulfate and decreased the cell viability in HEK293/SULT1C3 cells. These data indicate that sulfation increased the cytotoxicity of tolvaptan.

Tolvaptan (N- (4-{[(5 R)-7-chloro-5-hydroxy-2,3,4,5-tetrahydro-1H-1-benzazepin-1-yl]carbonyl}-3-methylphenyl)-2-methylbenzamide; CAS #: 150683-30-0) is an orally active vasopressin V₂-receptor antagonist that promotes the excretion of water without electrolyte loss. Tolvaptan has been approved for the treatment of autosomal dominant polycystic kidney disease in Europe, Canada, and Japan. Other approved disorders, which vary by country, include hypervolumic hyponatremia associated with congestive heart failure, the syndrome of inappropriate antidiuretic hormone, and edema with liver cirrhosis (Friedman and Cirulli, 2013; Sakaida, 2014; Schrier et al., 2006; Watkins et al., 2015).

In a recent large clinical trial, which tested the use of tolvaptan in patients with autosomal dominant polycystic kidney disease, an elevated level of alanine aminotransferase (>3 times the upper limit of normal) was observed in 4.4% of patients receiving tolvaptan and 1.0% of patients in the placebo group, while an elevation (>3 times the upper limit of normal) of aspartate aminotransferases was observed in 3.1% of patients on tolvaptan compared to 0.8% of patients given a placebo (Torres et al., 2012; Watkins et al., 2015). As a consequence, the U.S. Food and Drug Administration recommended that the drug Samsca (tolvaptan) should not be used for longer than 30 days and should not be used in patients with underlying liver disease (http://www.fda.gov/Drugs/DrugSafety/ucm350062.htm). We recently reported that tolvaptan-induced cytotoxicity in human hepatoma HepG2 cells results from delayed cell cycle progression, the induction of DNA damage, and the execution of apoptosis (Wu et al., 2015).

Several studies have demonstrated that cytochrome P450 3A4 (CYP3A4) is the major CYP isozyme that metabolizes tolvaptan (Shoaf et al., 2012a,c). Our previous study confirmed the activity of CYP3A4 against tolvaptan (Wu et al., 2015); however, the overexpression of human CYP3A4 in HepG2/CYP3A4 cells did not significantly affect tolvaptan-induced cytotoxicity compared with HepG2 cells and HepG2/vector cells, which lack appreciable levels of CYP3A4. The lack of a differential response in cytotoxicity between HepG2 cells and HepG2/CYP3A4 cells suggests that the metabolites resulting from CYP3A4 metabolism are not responsible for the tolvaptan-mediated cytotoxicity (Wu et al., 2015). In a single oral dose study in humans administered 60 mg [¹⁴C]tolvaptan, tolvaptan and 7 of its CYP-catalyzed metabolites accounted for 60.4% of the plasma radioactivity (Joseph, 2008), indicating that there were additional unidentified tolvaptan metabolites resulting from other metabolic pathways.

In human tissues, a family of cytosolic sulfotransferases (SULTs) catalyzes the sulfation of a multitude of xenobiotics, many therapeutic drugs, and endogenous compounds including steroids, thyroid hormones, and monoamine neurotransmitters (Gamage et al., 2006; Glatt, 2000). The SULT isoforms involved in the sulfation of tolvaptan have not been described. In this study, we have investigated the ability of 12 expressed human SULT isoforms to conjugate tolvaptan and the effect of human SULT1C3-mediated metabolism of tolvaptan on cell growth. The results indicate that several of human SULTs are involved in the sulfation of tolvaptan, with SULT1C3 being the predominant isoform. Moreover, the sulfation of tolvaptan enhanced the cytotoxicity of tolvaptan.

MATERIALS AND METHODS

Chemicals and reagents.

Tolvaptan, ammonium acetate, anhydrous pyridine, sulfur trioxide pyridine complex, 2,2-bis(hydroxymethyl)-2,2′,2″-nitrilotriethanol (Bis-Tris), cyclosporine A, dimethyl sulfoxide (DMSO), ethylenediaminetetraacetic acid (EDTA), and thiazolyl blue tetrazolium bromide (MTT) were obtained from Sigma-Aldrich (St. Louis, Missouri). Acetonitrile, Dulbecco’s Modified Eagle Medium (DMEM), methanol, penicillin–streptomycin solution, puromycin, sodium pyruvate, nonessential amino acids, and 2.5% trypsin were purchased from Thermo Fisher Scientific, Inc. (Pittsburgh, Pennsylvania). Adenosine 3′-phosphate 5′-phosphosulfate lithium salt hydrate (PAPS; stated purity ≥ 80%) was acquired from Santa Cruz Biotechnology (Santa Cruz, California). High-performance liquid chromatography-ultra violet (HPLC-UV) analysis at 254 nm, conducted in our laboratory, indicated that the purity of PAPS was 79% and adenosine 3′-phosphate 5′-phosphate was not detected.

Fetal bovine serum (FBS) was acquired from Atlanta Biologicals (Lawrenceville, Georgia). [³⁵S]PAPS (specific activity 2.4 Ci/mmol, purity > 99.0%) was purchased from American Radiolabeled Chemicals, Inc. (St. Louis, Missouri). Triclosan [5-chloro-2 -(2,4-dichlorophenoxy)phenol] was obtained from Alfa Aesar (Ward Hill, Massachusetts). [2,4-Dichlorophenyl-¹⁴C(U)]triclosan (radiochemical purity ≥ 97.0% by HPLC and specific activity 80 mCi/mmol) was purchased from Moravek Biochemicals (Brea, California).

Antibodies.

Mouse monoclonal antibodies to human SULT1E1, SULT2A1, SULT4A1, CYP3A4, and β-actin, goat polyclonal antibody to human SULT2B1, and rabbit polyclonal antibody to breast cancer resistance protein (BCRP) were acquired from Santa Cruz Biotechnology. Mouse monoclonal antibody to human SULT1C2 and rabbit polyclonal antibodies to human SULT1C3 and SULT6B1 were purchased from LifeSpan BioSciences, Inc. (Seattle, Washington). Rabbit polyclonal antibodies to human SULT1A1, SULT1A3, SULT1C4, and the multidrug resistance protein 1 (P-glycoprotein, Pgp) were obtained from Thermo Fisher Scientific Inc. Rabbit polyclonal antibody to human SULT2B1 was purchased from Sigma-Aldrich. Rabbit polyclonal antibody to human SULT1A2 was obtained from Abcam Inc. (Cambridge, Massachusetts).

Synthesis and characterization of tolvaptan sulfate standard.

Tolvaptan sulfate was synthesized by reacting 6 mg tolvaptan in 250 μl anhydrous pyridine with 70 mg sulfur trioxide pyridine complex in an Eppendorf Thermomixer R thermoblock (Eppendorf North America, Hauppauge, New York) at 60°C, with shaking at 1000 rpm, for 5 h. The tolvaptan and sulfur trioxide pyridine complex were previously dried in the presence of phosphorus pentoxide under vacuum for 2 days. After the reaction, the mixture was diluted (1:200) with methanol and the resulting solution was subjected to analysis by reversed-phase HPLC using a Waters HPLC system consisting of a 600 Controller, a 996 Photodiode Array detector, and a 717 Plus autosampler (Waters Corporation, Milford, Massachusetts). Samples were injected onto a 4.6 × 250 mm C18 (5 μm particle size) Luna column (Phenomenex, Torrance, California). HPLC separations were performed with acetonitrile (solvent A) and 50 mM ammonium acetate, pH 5.0 (solvent B) as follows: 5 min with 100% solvent B, a 10-min linear gradient to 70% solvent B, then with 70% solvent B for 10 min, and a final 12-min linear gradient to 0% solvent B. The HPLC flow rate was 1 ml/min. The column was washed with 100% solvent A for 15 min and equilibrated for 15 min with 100% solvent B after every run. When [³⁵S]tolvaptan sulfate was being measured, the HPLC system included a β-RAM radioactive flow detector (LabLogic Systems, Inc., Tampa, Florida), with a scintillation fluid flow rate of 2 ml/min.

The peak corresponding to tolvaptan sulfate was tentatively identified by relative retention time. The peak eluting from the HPLC column was collected, purified, and characterized by ultra-performance liquid chromatography tandem mass spectrometry (UPLC-MS/MS). Aliquots of the synthesized standard were then used as a UV standard for identification of tolvaptan sulfate.

Mass spectrometry analyses.

Synthesized tolvaptan sulfate was dissolved in acetonitrile and analyzed using a Waters ACQUITY UPLC system coupled with a Waters Quattro Premier XE quadrupole MS/MS equipped with an electrospray ionization (ESI) source. The sample was eluted using a Waters C18 BEH column (2.1 × 50 mm, 1.7 μm particle size) maintained at 30°C with a 4 min linear gradient of 10–98% of acetonitrile in 15 mM ammonium acetate containing 5% acetonitrile at a flow rate of 0.4 ml/min. The mass spectrometer was operated in the negative electrospray mode and the collision gas was argon at a flow rate of 0.5 ml/min. The collision energies were optimized to provide an optimal fragmentation of the analyte. A Waters ACQUITY diode array detector programmed to scan from 220 to 400 nm was included in the eluate path.

Cell culture.

Human embryonic kidney HEK293 cells were obtained from American Type Culture Collection (ATCC, Manassas, Virginia) and cultured in DMEM supplemented with 10% FBS and penicillin–streptomycin solution. The 293T cell line used for lentivirus packaging was purchased from Biosettia (San Diego, California) and maintained in DMEM supplemented with 10% FBS, 1 mM sodium pyruvate, and non-essential amino acids. Both cell lines were maintained at 37°C in a humidified atmosphere with 5% CO₂.

Generation of HEK293 cells overexpressing human SULT isoforms.

The coding sequence of the individual human SULT enzymes (Table 1) was cloned into the lentiviral expression vector pLV-EF1a-SULT-IRES-Puro (Biosettia). The generated human SULT expression vector or empty vector and viral packaging plasmids (Biosettia) were cotransfected into 293T cells to produce lentivirus stocks. The titrations of the lentivirus stocks were determined with a tittering kit provided by Biosettia. HEK293 cells were then infected with the lentivirus carrying human SULT expression vector or empty vector at a multiplicity of infection of 10. At 48 h posttransfection, cells were passaged and subsequently grown in puromycin (2 μg/ml media) for the selection of puromycin-resistant cells that stably expressed the individual human SULTs (HEK293/SULT1A1, HEK293/SULT1A2, HEK293/SULT1A3, HEK293/SULT1B1, HEK293/SULT1C2, HEK293/SULT1C3, HEK293/SULT1C4, HEK293/SULT1E1, HEK293/SULT2A1, HEK293/SULT2B1, HEK293/SULT4A1, and HEK293/SULT6B1 cells) or the empty vector (HEK293/vector cells).

TABLE 1.

GenBank access numbers for individual human SULT coding sequences

		Coding Sequences
SULT	Access Number	Regions	Length (base)
1A1	NM_177536	390 … 1043	654
1A2	NM_177528	352 … 1239	888
1A3	NM_177552	105 … 992	888
1B1	NM_014465	299 … 1189	891
1C2	NM_176825	454 … 1377	924
1C3	NM_001008743	1 … 915	915
1C4	NM_006588	374 … 1282	909
1E1	NM_005420	114 … 998	885
2A1	NM_003167	141 … 998	858
2B1	NM_004605	180 … 1232	1053
4A1	NM_014351	117 … 971	855
6B1	NM_001032377	22 … 819	798

When cells reached 80%–90% confluence, culture medium was removed, and the cells were trypsinized and washed 3 times in PBS. The cells were lysed in 5 mM Bis-Tris (pH 7.0) and 0.1 mM EDTA for 30 min on ice, sonicating 3 times for 10 s each time, centrifuging at 14 000 × g for 30 min at 4°C. The resulting supernatant fraction (cell lysate) was collected and the protein concentrations of cell lysate were determined using a bicinchoninic acid (BCA) protein assay (Thermo Fisher Scientific, Inc.).

Western blot analysis.

The levels of human SULT proteins in the SULT-overexpressing cell lines were measured by Western blot analysis. Forty micrograms of cell lysate proteins was separated by SDS polyacrylamide gel electrophoresis and transferred onto a polyvinylene difluoride membrane. The membranes were blocked with 5% nonfat milk and incubated with specific primary antibodies against SULT1A1 (1:1000), SULT1A2 (1:500), SULT1A3 (1:1000), SULT1B1 (1:1000), SULT1C2 (1:500), SULT1C3 (1:500), SULT1C4 (1:1000), SULT1E1 (1:1000), SULT2A1 (1:1000), SULT2B1 (1:500), SULT4A1 (1:1000), SULT6B1 (1:500), Pgp (1:1000), BCRP (1:1000), CYP3A4 (1:1000), or β-actin (1:2000), followed by a secondary antibody conjugated to horseradish peroxidase. The blots were then detected by chemiluminescence using Immobilon Western Horseradish Peroxidase Substrate (Millipore Corporation, Billerica, Massachusett), a UVP BioSpectrum AC Imaging System, and VisionWorks LSD Image Acquisition & Analysis Software (UVP LLC, Upland, California).

Analysis of sulfation activity.

The sulfation activity of the individual SULT-overexpressing cell lines against tolvaptan was assayed in a 125 μl final reaction volume by incubating 3.125–250 μM tolvaptan with human SULT-overexpressing cell lysates (0.4–1.8 mg protein), 2 mM EDTA, 0.5 mM PAPS, and 100 mM sodium phosphate (pH 7.2) for 1 h at 37°C with gentle shaking. An initial sulfation activity screening was performed with a radiochemical assay using 0.5 mM [³⁵S]PAPS (diluted with unlabeled PAPS to a specific activity of 19.8 mCi/mmol), with the assay conditions optimized (the concentration of PAPS, the amount of cell lysate proteins, and the incubation time) to maximize the level of detection within a linear range. Kinetic analyses for SULTs exhibiting tolvaptan sulfation activity were performed as described above, using an incubation time of 1 h, the time at which the rate of tolvaptan sulfate formation was linear for each SULT isoform tested. The K_m and V_max for the sulfation of tolvaptan by individual SULT isoforms were calculated using GraphPad Prism 6.0 (GraphPad Software Inc., La Jolla, California). Negative controls for the sulfation reactions included incubations conducted with cell lysates from parent HEK293 cells and HEK293/SULT vector cells and as well as incubations conducted in the absence of tolvaptan or PAPS. Reactions were terminated by the addition of an equal volume of ice-cold methanol. Precipitated material was removed by centrifugation (5 min, 14 000 × g), and the supernatants (200 μl) were analyzed for tolvaptan sulfate using a reversed-phase HPLC assay described as above. The lower limit of quantification for tolvaptan sulfate was 10 pmol, based on radioflow detection and quantification of radioactivity within tolvaptan sulfate HPLC peak as determined using a IN/US β-RAM radioactivity detection program. Intraday and interday precision variation (coefficient of variation) was less than 10%.

The half inhibitory concentration (IC₅₀) of tolvaptan on cell growth.

The IC₅₀ values of tolvaptan in HEK293, HEK293/vector, and HEK293/SULT1C3 cells were assessed using an MTT assay, as previously described (Wu et al., 2014). Briefly, cells were plated into 24-well plates at cell densities of 4.0 × 10⁴, 2.0 × 10⁴, and 0.5 × 10⁴ cells/cm² culture surface for 48, 72, and 120 h incubations, respectively. Cells were cultured for 24 h prior to treatment with various concentrations of tolvaptan (0.78–75 μM) for 24, 48, or 96 h. When conducting incubations with cyclosporine A, a known efflux transporter inhibitor, cells were exposed to various concentrations of cyclosporine A (0.05–50 μM) for 48 h. Control cells were fed with complete culture medium containing 0.1 % (vol/vol) DMSO, which had no effect on cell growth. For the 96 h treatment, fresh medium and tolvaptan were replaced at 48 h. At the end of the treatment, the absorbance of formazan was determined with a BioTek Synergy 2 reader (BioTek Instruments, Inc., Winooski, Vermont) at 540 nm, with 690 nm as the reference. The IC₅₀ values were obtained from the cell growth curves, using GraphPad Prism 6.0. In addition, the effect of tolvaptan on the growth of HEK293, HEK293/vector, and HEK293/SULT1C3 cells was investigated in the presence of cyclosporine A.

Statistical analyses.

Data are presented as the mean ± standard deviation (SD) of 3 independent experiments. Comparisons among concentrations were conducted by 1-way analysis of variance, with pairwise comparisons versus control group being performed by Dunnett’s method. When necessary, the data were log-transformed to maintain an equal variance or normal data distribution. The results were considered significant at P < .05.

RESULTS

Characterization of Tolvaptan Sulfate

The reaction of tolvaptan in anhydrous pyridine with sulfur trioxide pyridine complex led to the formation of a peak at 32.7 min as shown in Figure 1A. The reaction was quantitative and all tolvaptan was converted to tolvaptan sulfate, as determined by HPLC-UV analysis. The peak, which was collected and purified (purity > 99%, Figure 1C), exhibited a UV spectrum comparable to that of tolvaptan (inserts of Figures 1B and C). Structural characterization of the purified peak by UPLC-ESI-MS/MS in full scan negative electrospray mode showed a prominent ion chromatogram peak at m/z 527.0 and m/z 529.0, consistent with the expected negatively ionized ³⁵Cl and ³⁷Cl isotopologues of tolvaptan sulfate. Fragmentation analysis of the most abundant isotopologue at m/z 527.0 by tandem mass spectrometry at a collision energy of 30 eV afforded 2 major fragments at m/z 447.4 and 429.2 corresponding to the loss of -SO₃ and -OSO₃H, respectively (Figure 1D), confirming that the sulfation occurred at the hydroxyl group of tolvaptan.

FIG. 1.

High-performance liquid chromatography (HPLC) analysis of tolvaptan sulfate formed by reacting 6 mg tolvaptan in 250 μl anhydrous pyridine with 70 mg sulfur trioxide pyridine complex at 60°C for 5 h as described in Materials and Methods. Tolvaptan sulfate formation in reaction mixture (A), tolvaptan (B), and purified tolvaptan sulfate (C) were detected by UV at 254 nm. Insert: spectrum over the range 220–400 nm. (D) Mass chromatogram depicting the fragmentation pattern of tolvaptan sulfate at a collision energy of 30 eV. The 2 major fragments at m/z 447.4 and 429.2 correspond to the loss of -SO₃ and -OSO₃H, respectively.

Stability studies conducted with tolvaptan sulfate using reversed-phase HPLC analyses as described above indicated tolvaptan sulfate was stable in 50 mM Tris-HCl (pH 7.0) at room temperature up to 2 months.

Generation of HEK293 Cells Overexpressing Human SULT Isoforms

To determine if tolvaptan is a substrate for human SULTs and which human SULT isoforms can catalyze the sulfation, HEK293 cells stably overexpressing 12 human SULT isoforms were generated. As shown in Figure 2, when assessed by Western blot analysis, the individual SULTs (SULT1A2, SULT1A3, SULT1B1, SULT1C2, SULT1C3, SULT1C4, SULT2A1, SULT2B1, SULT4A1, and SULT6B1) were readily detected in the corresponding HEK293/SULT overexpressing cells. SULT1A1 and SULT2B1 were also detected in HEK293 cells and HEK293/vector cells; however, the levels of SULT1A1 and SULT2B1 in the overexpressing cells (HEK293/SULT1A1 cells and HEK293/SULT2B1 cells, respectively) were much higher.

FIG. 2.

Western blotting of human individual sulfotransferase (SULT) isoforms using cell lysates from SULT-overexpressing HEK293 cells (40 μg of protein). β-Actin was used as a loading control.

Sulfation of Tolvaptan in SULT-Overexpressing Cell Lysates

To elucidate whether any human SULTs exhibited activity against tolvaptan, a comprehensive screening was conducted of known human SULTs. All SULT-overexpressing cell lysates showed triclosan sulfation activity for each of the SULT isoforms examined in this study (Table 2). As summarized in Table 2, cell lines overexpressing SULT1B1, SULT1C3, SULT1E1, SULT2A1, and SULT2B1 exhibited detectable levels of tolvaptan sulfation. No detectable sulfation of tolvaptan was observed for cell lysates of cell lines overexpressing SULT1A1, SULT1A2, SULT1A3, SULT1C2, SULT1C4, SULT4A1, or SULT6B1. The recovery of exogenously added tolvaptan sulfate from the reaction mixtures was 97% ± 4.4% after 1 h of incubation at 37°C. Typical HPLC chromatograms for the analysis of tolvaptan sulfate from HEK293/SULT1C3 and HEK293 cell lysate are shown in Figure 3. Incubation of tolvaptan with HEK293/SULT1C3 cell lysate at 37°C for 1.0 h led to the formation of a peak (Figures 3A and B) exhibiting retention time identical to that obtained for tolvaptan sulfate UV standard (Figure 1C). No such metabolite was detected in reactions using HEK293 cell lysates (Figures 3C and D), HEK293/vector cell lysates, or in incubations lacking tolvaptan or PAPS.

TABLE 2.

Sulfation of Triclosan and Tolvaptan by Individual Human SULTs

SULT	pmol.min−1.mg cell lysate protein−1
	Triclosana	Tolvaptanb
1A1	15.2 ± 1.5	n.d.
1A2	193 ± 4.3	n.d.
1A3	182 ± 9.6	n.d.
1B1	199 ± 9.8	9.0 ± 0.3
1C2	38.6 ± 3.4	n.d.
1C3	22.9 ± 2.8	18.5 ± 0.5
1C4	210 ± 12.8	n.d.
1E1	190 ± 4.0	2.3 ± 0.1
2A1	26.7 ± 4.7	13.8 ± 0.6
2B1	34.1 ± 1.0	2.3 ± 0.2
4A1	20.8 ± 1.6	n.d.
6B1	8.9 ± 0.3	n.d.

^aSulfation assays were conducted in a 125 μl final reaction volume by incubating 0.25 mM [2,4-dichlorophenyl-¹⁴C(U)]triclosan (specific activity: 22.9 mCi/mmol) with 1.2 mg cell lysate proteins, 2 mM EDTA, 0.5 mM PAPS, and 100 mM sodium phosphate, pH 7.2, at 37°C for 2 h. Reactions were terminated by the addition of an equal volume of ice-cold methanol. Precipitated material was removed by centrifugation at 14,000 g at 4°C for 30 min and the supernatants were analyzed for tolvaptan and tolvaptan sulfate by a reversed-phase HPLC with radiochemical detection as previously described (Fang et al., 2014).

^bSulfation assays using 0.2 mM tolvaptan at 37°C for 1 h, as described in Materials and Methods. n.d.: not detected.

FIG. 3.

HPLC analysis of tolvaptan sulfate formation in cell lysates from individual SULT-overexpressing cells. Cell lysates were incubated 37°C with 0.1 mM tolvaptan and 0.5 mM [³⁵S]PAPS as described in Materials and Methods. Shown are tolvaptan and tolvaptan sulfate from incubations using 0.4 mg of cell lysate proteins from HEK293/SULT1C3 cells (A and B) and 1.8 mg of cell lysate proteins from HEK293 cells (C and D). Incubations were performed for 1 h with HEK293/SULT1C3 and for 6 h with HEK293 cell lysates. (A) and (C), tolvaptan sulfate formation detected by UV at 254 nm; (B) and (D), [³⁵S]tolvaptan sulfate formation as assessed by radiochemical detection. Peaks corresponding to [³⁵S]PAPS, tolvaptan, [³⁵S]tolvaptan sulfate, and tolvaptan sulfate are indicated by arrows.

The relative affinities for tolvaptan as reflected by their apparent K_m’s were SULT1C3 >> SULT2A1 > SULT2B1 ~ SULT1B1 > SULT1E1 (Table 3). As determined by the V_max/K_m ratio, SULT1C3 had the highest overall tolvaptan sulfation activity, which was 16-, 46 -, 115 -, and 230-fold higher than for SULT2A1, SULT1B1, SULT2B1, and SULT1E1, respectively (Table 3).

TABLE 3.

Kinetic Analysis of Individual Human SULTs Against Tolvaptan^a

SULT	Kmb μM	Vmax pmol.min−1. cell lysate protein mg−1	Vmax/Km μl.min−1.mg−1
1B1	87.0 ± 15.3	10.9 ± 0.6	0.1 ± 0.02
1C3	4.1 ± 0.4	19.0 ± 0.2	4.6 ± 0.4
1E1	149 ± 24.8	3.5 ± 0.3	0.02 ± 0.004
2A1	56.8 ± 5.8	16.3 ± 0.4	0.3 ± 0.03
2B1	73.7 ± 17.0	3.0 ± 0.2	0.04 ± 0.01

^aAll the reactions were performed using cell lysates for individually overexpressed SULT isoenzyme, and incubations were at 37°C for 1 h, as described in Materials and Methods. Kinetic parameters are reported as mean ± SD for 3 individual experiments.

^bK_m, apparent K_m.

^cV_max values are adjusted per mg of the corresponding cell lysate proteins of individual SULT overexpressing cell line.

Effect of Tolvaptan Sulfation on Its Cytotoxicity in HEK293/SULT1C3 Cells

To determine the effect of human SULTs on the cytotoxicity of tolvaptan, the effect of tolvaptan on the growth of HEK293, HEK293/vector, and HEK293/SULT1C3 cells was conducted using an MTT assay. As shown in Figure 4A, tolvaptan decreased the growth of the 3 cell lines in a concentration- and time-dependent manner. With HEK293/SULT1C3 cells, the IC₅₀ values were 32.5 and 21.4 μM after a 48 and 96 h of exposure, respectively. These values were significantly lower than those at same exposure time with HEK293 cells and HEK293/vector cells, which had similar IC₅₀ values (Figure 4B). These data clearly indicate that sulfation increased the cytotoxicity of tolvaptan.

FIG. 4.

Cell growth curves of HEK293, HEK293/vector, and HEK293/SULT1C3 cells treated with tolvaptan (0.78–75 μM) for 24, 48, or 96 h (A). The results shown are the mean and standard deviation of 3 independent experiments. (B) The IC₅₀ values obtained from the cell growth curves shown in (A) using GraphPad Prism 6.0. Values in the parenthesis are the 95% confidence intervals of the IC₅₀. *: significantly (P < .05) different from HEK293 cells and HEK293/vector cells.

To evaluate whether or not tolvaptan sulfate was formed in cells upon treatment with tolvaptan, tolvaptan sulfate was measured using reversed-phase HPLC. Exposing of 10 μM tolvaptan to HEK293/SULT1C3 cells for 48 h led to the detection of tolvaptan sulfate in both the culture media (Figure 5A) and the HEK293/SULT1C3 cells (Figure 5B), with the level being much higher in the culture media (Figure 5A). Tolvaptan sulfate was not detected in incubations with HEK293/vector cells (Figures 5C and D). These data demonstrated that tolvaptan sulfate was formed in HEK293/SULT1C3 cells treated with tolvaptan and that most of tolvaptan sulfate was transported out of the cells.

FIG. 5.

HPLC analysis of tolvaptan sulfate formation in HEK293/SULT1C3 cells (A and B) and HEK293/vector cells (C and D) incubated with 10 μM tolvaptan for 48 h. (A) and (C), tolvaptan and tolvaptan sulfate (indicated by arrows) in the culture medium. (B) and (D), tolvaptan and tolvaptan sulfate in the cells.

Effect of Cyclosporine A on the Intracellular Tolvaptan Sulfate Level and the Cytotoxicity in HEK293/SULT1C3 Cells

The efflux of the intracellular formed tolvaptan sulfate out of cells could be associated with the expression of the drug efflux transporters Pgp or BCRP in the HEK293 cells (Ahlin et al., 2009). To investigate the effect of drug efflux transporters on the intracellular tolvaptan sulfate level, cyclosporine A, a specific inhibitor of Pgp and BCRP (Qadir et al., 2005), was used. As shown in Figure 6A, when assessed by Western blot analysis, Pgp and BCRP were readily detected in HEK293/SULT1C3 cells and their expression was not affected significantly by the treatment with either cyclosporine A or tolvaptan or coexposure to cyclosporine A and tolvaptan. The IC₅₀ of cyclosporine A in HEK293/SULT1C3 cells was 9.1 μM after a 48 h exposure (Figure 6B). Based on the IC₅₀ value, non-cytotoxic concentrations at 0.5 and 2.0 μM cyclosporine A were used in the subsequent experiments.

FIG. 6.

A, Western blotting of Pgp and BCRP using cell lysates (40 μg of proteins) from HEK293/SULT1C3 cells. β-Actin was used as a loading control. B, The IC₅₀ of cyclosporine A. HEK293/SULT1C3 cells were exposed to various concentrations of cyclosporine A (0.05–50 μM) for 48 h. After the treatment, the total number of viable cells was determined by an MTT assay. The IC₅₀ value was obtained from the cell growth curves using GraphPad Prism 6.0.

When HEK293/SULT1C3 cells were treated with tolvaptan in the presence of cyclosporine A for 48 h, the intracellular level of tolvaptan sulfate was significantly increased compared with cells treated with tolvaptan only (Figure 7). The increased intracellular level of tolvaptan sulfate was accompanied by a significantly decreased cell viability in HEK293/SULT1C3 cells cotreated with cyclosporine A and tolvaptan for 48 h (Figure 8), suggesting that the elevated intracellular tolvaptan sulfate level resulting from cyclosporine A treatment led to greater cytotoxicity in HEK293/SULT1C3 cells. The concentrations of cyclosporine A used in these incubations were not cytotoxic after a 48 h exposure (Figure 8). Moreover, the cyclosporine A treatment did not affect the intracellular levels of tolvaptan either in HEK293/vector cells or in HEK293/SULT1C3 cells. In addition, there was no significant difference in the cell viability of HEK293/vector cells co-treated with cyclosporine A and tolvaptan for 48 h, compared with cells treated with tolvaptan alone (Figure 8). These data demonstrate that sulfation of tolvaptan potentiates the cytotoxicity of tolvaptan.

FIG. 7.

The intracellular levels of tolvaptan sulfate. Cells were incubated with tolvaptan (10 or 30 μM) in the presence or absence of cyclosporine A (0.5 and 2.0 μM) for 48 h. Following treatment, cell homogenates were prepared and mixed with an equal volume of ice-cold methanol. Precipitated material was removed by centrifugation at 14 000 × g at 4°C for 30 min and the supernatants were analyzed for tolvaptan and tolvaptan sulfate by a reversed-phase HPLC. The results shown are the mean and standard deviation of 3 independent experiments. *: significantly (P < .05) different from cells without cyclosporine A treatment.

FIG. 8.

HEK293/SULT1C3 cells and HEK293/vector cells were incubated with tolvaptan (0, 10, or 30 μM) in the presence or absence of cyclosporine A (0.5 and 2.0 μM) for 48 h. Cell viability was assessed by an MTT assay. The results shown are the mean and standard deviation of 3 independent experiments. *: significantly (P < 0.05) different from cells without cyclosporine A treatment.

Effect of Tolvaptan Metabolites by CYP3A4 and Tolvaptan Sulfation on Its Cytotoxicity in HEK293/CYP3A4 Cells and HEK293/CYP3A4 & SULT1C3 Cells

To validate the role of CYP3A4 and SULT1C3 in the cytotoxicity of tolvaptan, HEK293/CYP3A4 cells, which stably express human CYP3A4, were constructed as previously described (Wu et al., 2015). The HEK293/CYP3A4 cells were then transfected with the lentivirus particles carrying the human SULT1C3 expression vector or SULT empty vector as described in Materials and Methods to generate cell lines stably expressing human CYP3A4 and SULT1C3 (HEK293/CYP3A4 & SULT1C3 cells) or human CYP3A4 and the SULT empty vector (HEK293/CYP3A4 & SULT vector cells). As shown in Figure 9, when assessed by Western blot analysis, CYP3A4 was detected in the 3 cell lines overexpressing human CYP3A4 but not in HEK293/CYP vector cells, while SULT1C3 was expressed only in HEK293/CYP3A4 & SULT1C3 cells.

FIG. 9.

Western blotting of human CYP3A4 and SULT1C3 using 40 μg of cell lysate proteins from HEK293/CYP vector, HEK293/CYP3A4, HEK293/CYP3A4 & SULT vector, and HEK293/CYP3A4 & SULT1C3 cells. β-Actin was used as a loading control.

The metabolites of tolvaptan were determined by reversed-phase HPLC in the 4 cell lines after incubation with 30 μM tolvaptan for 48 h. As shown in Figure 10, tolvaptan metabolites by CYP3A4 (indicated by Arabic numbers) were detected in the 3 cell lines overexpressing human CYP3A4 but not in HEK293/CYP vector cells, while tolvaptan sulfate was found only in HEK293/CYP3A4 & SULT1C3 cells. None of the metabolites were detected in HEK293/CYP vector cells. In HEK293/CYP3A4 cells and HEK293/CYP3A4 & SULT vector cells, 20.8% of the tolvaptan was converted into metabolites by CYP3A4. In cells cooverexpressing human CYP3A4 and SULT1C3, 19.6% of the tolvaptan was converted into metabolites by CYP3A4, whereas tolvaptan sulfate accounted for 4.8% of the metabolites. A comparison of the effect of tolvaptan on cell viability showed no difference in IC₅₀ values between HEK293/CYP vector cells and HEK293/CYP3A4 cells after a 48 h of exposure (Figure 11). With HEK293/CYP3A4 & SULT1C3 cells, the IC₅₀ value was 39.8 μM, which was significantly lower than those at same exposure time with HEK293/CYP vector cells and HEK293/CYP3A4 & SULT vector cells. These data clearly indicated that the cytotoxicity of tolvaptan was enhanced in HEK293/CYP3A4 & SULT1C3 cells and that metabolism of tolvaptan by human CYP3A4 did not cause any change in the cytotoxicity of tolvaptan.

FIG. 10.

HPLC analysis of tolvaptan (A), tolvaptan sulfate (B), and tolvaptan metabolites by CYP3A4 (indicated using Arabic numbers) and tolvaptan sulfate in HEK293/CYP vector cells (C), HEK293/CYP3A4 cells (D), HEK293/CYP3A4 & SULT vector cells (E), and HEK293/CYP3A4 & SULT1C3 cells (F) incubated with 30 μM tolvaptan for 48 h. The metabolites of tolvaptan in cells were analyzed using a reversed-phase HPLC with a 40 min linear gradient of 0–20% of acetonitrile in 50 mM ammonium acetate (pH 5.0) at a flow rate of 1 ml/min. Tolvaptan and tolvaptan sulfate are indicated by arrows.

FIG. 11.

Cell growth curves of HEK293/CYP vector, HEK293/CYP3A4, HEK293/CYP3A4 & SULT vector, and HEK293/CYP3A4 & SULT1C3 cells treated with tolvaptan (0.78–75 μM) for 48 h (A). The results shown are the mean and standard deviation of 3 independent experiments. (B) The IC₅₀ values obtained from the cell growth curves shown in (A) using GraphPad Prism 6.0. Values in the parenthesis are the 95% confidence intervals of the IC₅₀. *: significantly (P < .05) different from HEK293/CYP vector cells and HEK293/CYP3A4 & SULT vector cells.

DISCUSSION

This is the first study to characterize the sulfation of tolvaptan by human SULT isoforms. Several human SULTs were shown to exhibit significant levels of activity against tolvaptan. Whereas the SULT with the lowest apparent K_m and therefore the highest affinity for tolvaptan was SULT1C3, other SULTs, including SULT2A1, SULT2B1, SULT1B1, and SULT1E1, also catalyzed the sulfation of tolvaptan.

HEK293 is a cell line originally derived from human embryonic kidney cells. HEK293 cells readily grow in culture and have been widely used as hosts for gene expression transfection studies (Thomas and Smart, 2005). Although very low protein levels of SULT2B1 were detected in the parent HEK293 cells and HEK293/vector cells (Figure 2), sulfation of tolvaptan was not observed in either HEK293 cells (Figures 3C and D) or HEK293/vector cells (Figures 5C and D). Thus, HEK293 cells were used in this study for transfecting human SULTs.

Treatment with tolvaptan caused cell death in HepG2 cells as assayed by MTT and lactate dehydrogenase (LDH) release assays (Wu et al., 2015). Using MTT assay, cell death was also observed in cells treated with tolvaptan in current study. MTT and LDH release assays have been demonstrated to be highly specific for predicting human hepatotoxicity (O’Brien et al., 2006). There is a consistency between our observations that tolvaptan caused cell death in HepG2 cells, HEK293 cells, and overexpressing cell lines generating from HEK 293 cells and that tolvaptan is associated with elevated aminotransferases in humans.

Although sulfation is generally considered as a detoxication pathway by producing more water-soluble and often less toxic metabolites, the sulfation of benzylic hydroxyl groups has been associated with the formation of reactive electrophiles. This has been demonstrated with 7-hydroxymethyl-12-methylbenz[a]-anthracene, 6-hydroxymethybenzo[a]pyrene, 9-hydroxymethyl-10-methylanthracene, 1-hydroxymethylpyrene, and 1′-hydroxysafrole, and that the resulting sulfate metabolites can covalently bind to DNA and protein to produce both cytotoxic and carcinogenic responses (Michejda and Kroeger Koepke, 1994). More recently, sulfation of the benzylic hydroxyl group of 12-hydroxynevirapine has been shown to result in the covalent binding of nevirapine to skin and possible liver proteins and to be responsible for the skin rash associated with this drug (Sharma et al., 2013a,b). In the present study, we demonstrated that tolvaptan, which also possesses a benzylic hydroxy group, inhibits the growth of HEK293 cells and overexpressing cell lines generating from HEK293 cells, with the greatest sensitivity being observed in the HEK293/SULT1C3 cells and HEK293/CYP3A4 & SULT1C3 cells. The greater cytotoxicity observed in the HEK293/SULT1C3 cells was accompanied by higher levels of tolvaptan sulfate, both intracellularly and extracellularly, suggesting that tolvaptan sulfate contributes to the tolvaptan-mediated cytotoxicity.

The higher levels of tolvaptan sulfate found extracellularly in the culture medium compared to intracellularly in the HEK293/SULT1C3 cells indicate a role for drug efflux transporters. Adenosine triphosphate-binding cassette transporters, such as Pgp (encoded by the ABCB1 gene) and BCRP (by the ABCG2 gene), are expressed in various organs and tissues and transport a wide range of structurally diverse substances out of cells thereby exerting an important role in drug disposition and distribution. The expression of both Pgp and BCRP has been demonstrated in HEK293 cells (Ahlin et al., 2009). In this study, noncytotoxic concentrations of cyclosporine A, a known inhibitor of Pgp and BCRP (Qadir et al., 2005), led to a significant increase in the intracellular level of tolvaptan sulfate, which was accompanied by an increase in tolvaptan-mediated cytotoxicity in the HEK293/SULT1C3 cells.

The findings of this study raise concerns about potential adverse effects resulting from drug-drug interactions between tolvaptan and cyclosporine A. The 5 SULTs responsible for tolvaptan sulfation and drug efflux transporters Pgp and BCRP are expressed in normal human tissues, including the kidney, liver, and intestine (Brand et al., 2010; Gamage et al., 2006; Riches et al., 2009). Thus, it is possible that the concomitant use tolvaptan and cyclosporine A may pose a greater risk for liver and kidney injury than the use of tolvaptan by itself.

Pharmacokinetic studies have shown that the mean maximum plasma concentration of tolvaptan is approximately 1 μM in in healthy subjects given oral doses of 15–60 mg daily (Kim et al., 2011; Sakaida, 2014; Shoaf et al., 2007), which is comparable to the apparent K_m of SULT1C3 (4.1 μM) in the present study. Following a single oral dose of 60 mg [¹⁴C]tolvaptan in humans, only 60% of the plasma radioactivity could be identified as tolvaptan and its CYP3A4 metabolites (Joseph, 2008). It is possible that tolvaptan sulfate is one of the unidentified plasma metabolites of tolvaptan.

CYP3A4 plays an important role in the phase I metabolism of tolvaptan (Shoaf et al., 2012b,c). We found that CYP3A4 metabolized approximately 20% of the tolvptan in all 3 cell lines expressing CYP3A4. Unexpectedly, tolvaptan did not cause any significant change in the cytotoxicity between CYP3A4 overexpressing cells (HEK293/CYP3A4 cells and HEK293/CYP3A4 & SULT vector cells) and cells lacking of human CYP3A4 (HEK293/CYP vector cells). In contrast, although only a low percent of tolvatpan sulfate (4.8% of the initial tolvaptan) was detected only in HEK293/CYP3A4 & SULT1C3 cells, tolvaptan was more toxic in HEK293/CYP3A4 & SULT1C3 cells compared to HEK293/CYP3A4 & SULT vector cells, with the IC₅₀ values being approximately 15% lower (Figure 11B). Thus, these data demonstrate that tolvaptan sulfate increases the cytotoxicity of tolvaptan.

In conclusion, several human SULT isozymes catalyzed the sulfation of tolvaptan and tolvaptan sulfate increased the cytotoxicity of tolvaptan.