Transport Characteristics of Tramadol in the Blood–Brain Barrier

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

Transport Characteristics of Tramadol in the Blood–Brain Barrier ATSUSHI KITAMURA, KEI HIGUCHI, TAKASHI OKURA, YOSHIHARU DEGUCHI Laboratory of Drug Disposition & Pharmacokinetics, Faculty of Pharma-Sciences, Teikyo University, Itabashi, Tokyo 173-8605, Japan Received 9 June 2014; revised 22 July 2014; accepted 29 July 2014 Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.24129 ABSTRACT: Tramadol is a centrally acting analgesic whose action is mediated by both agonistic activity at opioid receptors and inhibitory activity on neuronal reuptake of monoamines. The purpose of this study was to characterize the blood–brain barrier (BBB) transport of tramadol by means of microdialysis studies in rat brain and in vitro studies with human immortalized brain capillary endothelial cells (hCMEC/D3). The Kp,uu,brain value of tramadol determined by rat brain microdialysis was greater than unity, indicating that tramadol is actively taken up into the brain across the BBB. Tramadol was transported into hCMEC/D3 cells in a concentration-dependent manner. The uptake was inhibited by type II cations (pyrilamine, verapamil, etc.), but not by substrates of organic cation transporter OCTs or OCTN2. It was also inhibited by a metabolic inhibitor but was independent of extracellular sodium or membrane potential. The uptake was altered by changes of extracellular pH, and by ammonium chloride-induced intracellular acidification, suggesting that transport of tramadol is driven by an oppositely directed proton gradient. Thus, our in vitro and in vivo results suggest that tramadol is actively transported, at least in C 2014 Wiley Periodicals, Inc. and the American part, from blood to the brain across the BBB by proton-coupled organic cation antiporter.  Pharmacists Association J Pharm Sci Keywords: active transport; blood–brain barrier; drug transport; in vitro models; membrane transporter; microdialysis; organic cation transporter; tramadol

INTRODUCTION Tramadol hydrochloride, (1RS,2RS)-2-[(dimethylamino)ethyl]1-(3-methoxyphenyl)-cyclohexanol hydrochloride, is a widely used centrally acting analgesic. The analgesic effect following parenteral administration of tramadol is due to synergistic interaction between its agonist activity toward opioid receptors and its antinociceptive effect mediated by inhibition of neuronal reuptake of monoamines in the brain.1 Thus, the analgesic activity should be dependent on the concentration of unbound tramadol in the vicinity of both :-opioid receptors and monoamine transporters, such as serotonin transporter (SERT) and norepinephrine transporter (NET), in the brain. Since the concentration of unbound tramadol in the brain would be at least partly determined by the blood–brain barrier (BBB) transport characteristics of tramadol, an understanding of the transport characteristics is important for predicting the onset and duration of analgesic activity of tramadol. The BBB dynamically regulates the transfer of endogenous nutrients, waste products, and drugs between blood and brain interstitial fluid (ISF),2 depending upon the functions of various transporters and receptors localized on the brain capillary endothelial cell membrane.2 Opioids, such as morphine and oxycodone,3 and opioid-like analgesic peptides, such as H-Tyr-D-Arg-Phe-beta-Ala-OH (TAPA)4 and ebiratide,5 are transported through the BBB via both identified and unidentified transporters and receptors. Recently, we have shown that oxycodone is actively taken up into rodent brain capillary endothelial cells by proton-coupled organic cation (H+ /OC) antiporter.3 Other organic cationic drugs, such as diphenhydramine,6,7 pyrilamine,3,7 nicotine,8,9 and clonidine,10 are also transported by the H+ /OC antiporter.

Tramadol (pKa 9.41),11 which contains a tertiary amine moiety, is present in cationic form at physiological pH. The tramadol concentration in brain is approximately five times higher than that in plasma12 and the unbound brain to plasma (Kp,uu,brain ) concentration ratio has been indirectly estimated to be greater than unity using the rat brain slice method.12 Thus, tramadol may be actively transported into the brain by the H+ /OC antiporter across the BBB. Functional transporters at the human BBB are important not only for pharmacotherapy to treat cerebral diseases, but also for development of new central nervous system (CNS)acting drugs. Human immortalized brain capillary endothelial cells (hCMEC/D3)13 retain many of the morphological and functional characteristics of the human BBB in terms of expression of tight-junction proteins, as well as various ABC and several SLC transporters.2,14–16 Our laboratory has shown recently that H+ /OC antiporter is functionally expressed in hCMEC/D3,6 and therefore in vitro uptake studies using hCMEC/D3 cells may be useful to predict tramadol concentration in the human brain, and hence its analgesic effect. However, it is first necessary to establish the transport mechanism of tramadol. The aim of this study, therefore, was to examine the transport mechanism of tramadol in vivo and in hCMEC/D3 cells. First, we measured the Kp,uu,brain value of tramadol using the rat brain microdialysis technique in order to obtain evidence for active uptake of tramadol through the BBB in vivo. We then investigated the mechanism of tramadol transport using hCMEC/D3 cells in vitro.

MATERIALS AND METHODS Reagents

Correspondence to: Yoshiharu Deguchi (Telephone: +81-3-3964-8246; Fax: +81-3-3964-8252; E-mail: [email protected]) Journal of Pharmaceutical Sciences  C 2014 Wiley Periodicals, Inc. and the American Pharmacists Association

Tramadol hydrochloride was purchased from Wako Pure Chemical Industries (Osaka, Japan). All other chemicals and reagents were commercial products of reagent grade. Kitamura et al., JOURNAL OF PHARMACEUTICAL SCIENCES

1

2

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

Cell Culture hCMEC/D3 cells had been immortalized by lentiviral transduction of the catalytic subunit of human telomerase and SV40-T antigen.13 The cells were cultivated at 37◦ C in EBM-2 medium (Lonza, Basel, Switzerland) supplemented with 2.5% fetal bovine serum, 0.025% VEGF, 0.025% R3-IGF, 0.025% hEGF, 0.01% hydrocortisone, 5 :g/mL bFGF, 1% penicillin– streptomycin, and 10 mM HEPES on rat collagen type I-coated dishes in an atmosphere of 95% air and 5% CO2 . Animals Adult male Wistar rats purchased from Japan SLC (Shizuoka, Japan) were housed, two or three per cage, with free access to food and water. The room was maintained on a 12-h dark/12-h light cycle with controlled temperature (24 ± 2◦ C) and humidity (55 ± 5%). This study was conducted according to guidelines approved by the Experimental Animal Ethical Committee of Teikyo University. Transport Studies hCMEC/D3 cells used for the experiments were between passages 25 and 35. The cells were seeded on rat collagen I-coated 24-well plates (Becton Dickinson, Franklin Lakes, New Jersey) at a density of 0.2 × 105 cells/cm2 . At 3 or 4 days after seeding, the cells reached confluence. For uptake experiments, they were washed twice with 1 mL of transport buffer (122 mM NaCl, 3 mM KCl, 25 mM NaHCO3 , 1.2 mM MgSO4 , 1.4 mM CaCl2 , 10 mM D-glucose, 10 mM HEPES, pH 7.4) and preincubated with 0.25 mL of transport buffer for 20 min at 37◦ C. After preincubation, 0.25 mL of the transport buffer containing 5 :M tramadol was added to initiate uptake. The cells were incubated at 37◦ C for a designated time, and then washed three times with 1 mL of ice-cold incubation buffer to terminate the uptake. The cells were collected with a scraper in 200 :L of H2 O containing 100 nM propranolol as an internal standard and stored in a freezer set at −30◦ C until analysis. Uptake was expressed as the cell-to-medium ratio (:L/mg protein or :L/(mg protein·min), obtained by dividing the uptake amount by the concentration of substrate in the transport medium. In order to estimate the kinetic parameters, tramadol uptake data (10, 25, 50, 75, 100, 200, 500, and 1000 :M, for 30 s) were analyzed using Michaelis–Menten plots based on the following equation: V=

Vmax × S + Pdiff × S Km + S

(1)

where V is the initial uptake rate of substrate (nmol/(mg protein·min)), Vmax is the maximum uptake rate (nmol/(mg protein·min)), S is the concentration of tramadol in the medium (:M), Km is the Michaelis–Menten constant (:M), and Pdiff is non-saturable uptake clearance (:L/(mg protein·min)), respectively. The uptake data were fitted to the above equation by nonlinear least-squares regression analysis with Prism software (Graphpad, San Diego, California). To delineate the energy requirements of the transport system, uptake of tramadol was carried out in the presence of a metabolic energy inhibitor, sodium azide (NaN3 ). The uptake was measured after pretreatment with 0.1% NaN3 for 20 min. In this experiment, 10 mM D-glucose in the transport medium was replaced with 10 mM 3-O-methylglucose to Kitamura et al., JOURNAL OF PHARMACEUTICAL SCIENCES

reduce metabolic energy. To examine the sodium ion dependency and the effect of reducing the membrane potential, NaCl was replaced with N-methylglucamine+ and KCl, respectively. Uptake was also measured at medium pH values of 6.4, 7.4, and 8.4. The influence of intracellular pH (pHi ) was examined by pretreatment or incubation with 30 mM NH4 Cl to produce intracellular acidosis or alkalization, respectively.17,18 To measure tramadol uptake at acidic pHi , extracellular NH4 Cl was removed after the preincubation with 30 mM NH4 Cl because intracellular NH3 rapidly diffuses out of the cells, resulting in accumulation of protons released from NH4 + during NH3 generation in the cells. In the inhibition study, uptake was measured after incubation with tramadol in the presence of test compounds (tramadol, 1-methyl-4-phenylpyridinium (MPP+ ), tetraethylammonium (TEA), carnitine, morphine, codeine, oxycodone, apomorphine, clonidine, pyrilamine, diphenhydramine, amantadine, memantine, verapamil, and quinidine) at the concentration of 1 mM. Tramadol uptake (10–1000 :M) was measured in the absence and presence of oxycodone (500 :M). The cells were solubilized with an equal volume of 1 M NaOH at 37◦ C for 60 min and three volumes of H2 O were added. The cellular protein content was determined with a BCA protein assay kit (Pierce Chemical Company, Rockford, Illinois). Plasma Protein Binding Tramadol was added to 1 mL of blank rat plasma to give a concentration of 3 :M. Aliquots of spiked plasma were equilibrated for 20 min at 37◦ C, then ultrafiltered (MPS-1; Millipore Corporation, Billerica, Massachusetts) and centrifuged at 35◦ C for 5 min (1000g). The concentrations remaining in spiked plasma (Cp,tot ) and ultrafiltrate (Cp,u ) samples were measured by LC– MS/MS. The unbound fraction in plasma (fu ) was determined by dividing Cp,u into Cp,tot . In Vitro Microdialysis In vitro brain microdialysis studies were carried out according to our previous report.19 A CMA12 microdialysis probe (3 mm; CMA, Stockholm, Sweden) was inserted into a tube containing 1 :g/mL tramadol and antipyrine (reference compound) in Krebs-Ringer phosphate (KRP) buffer (120 mM NaCl, 2.4 mM KCl, 1.2 mM CaCl2 , 1.2 mM MgSO4 , 0.9 mM NaH2 PO4 , 1.4 mM Na2 HPO4 , pH 7.4) solution. KRP buffer was perfused for 240 min through the probe at a constant flow rate of 5 :L/min by means of a syringe infusion pump (Model 22; Harvard Apparatus, South Natick, Massachusetts). The dialysate was collected every 30 min and the concentrations of tramadol and antipyrine in the dialysate and the buffer were measured by LC–MS/MS. Brain Microdialysis In vivo brain microdialysis studies were carried out according to our previous report.19 The rats were anesthetized with pentobarbital and a hole was drilled 2.7 mm lateral and 0.8 mm anterior to the bregma, and 3.8 mm ventral to the surface of the brain. A CMA12 guide cannula (CMA) was implanted into the striatum and fixed to the skull by a screw and dental cement (GC Fuji I, Tokyo, Japan). A CMA12 probe was inserted through the guide cannula 24 h after the surgery. Forty-eight hours after surgery, brain microdialysis was performed. The rats were anesthetized with pentobarbital and SP31 polyethylene tubes (inner diameter; 0.5 mm, outer diameter; 0.8 mm; Natsume Seisakusho Company, Ltd., Tokyo, DOI 10.1002/jps.24129

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

Japan) were inserted into the femoral vein to administer tramadol and into the femoral artery for blood sampling. The cannulas were filled with 100 IU/mL heparin-saline solution to prevent clotting. Tramadol was administered as an intravenous bolus dose of 2 mg/kg, followed by a 4 h constant infusion of 10 :g/(kg·min) into the femoral vein by using a Harvard 22 pump. Antipyrine was administered with tramadol as a reference compound and its bolus and constant infusion dose were 0.3 mg/kg and 0.6 :g/(kg·min), respectively. KRP buffer was perfused for 240 min through the probe at a constant flow rate of 5 :L/min. The dialysate solution and blood was collected every 30 min. Blood samples were centrifuged at 3000g for 10 min at 4◦ C to obtain plasma. The concentrations of tramadol and antipyrine in dialysate and plasma were measured by LC– MS/MS as described below. Extrapolation of Brain ISF Concentration The in vitro permeability rate constants (PAvitro ) of drug and reference compound were determined according to the following equation19–21 : CLvitro =

   F × Cd,vitro −PAvitro = F 1 − exp Cr F

   F × Cd,vivo −PAvivo = F × 1 − exp Cisf F

PAvivo PAvitro

(3)

(4)

Antipyrine was used as an in vivo reference compound. It is well established that the binding of antipyrine to plasma and tissue proteins is negligible, and there is rapidly equilibration between ISF and plasma. Therefore, assuming that the Rd value of tramadol is identical to that of antipyrine, rearranging Eqs. 2–4 gives the unbound tramadol concentration in the ISF (Cisf ) as follows. Cisf = Cd,vivo /{1 − exp(−Rd × PAvitro /F)}

(5)

Accordingly, the Kp,uu,brain value of tramadol can be determined according to the following formula: K p,uu,brain =

Cisf Cp,u

(6)

LC–MS/MS Analysis Cells and plasma were deproteinized with 4 volumes of acetonitrile, kept in a freezer (ca. –30◦ C) for at least 30 min and DOI 10.1002/jps.24129

Statistical Analysis All values are presented as average ± standard error. Statistical analysis of the data was performed by employing Student’s t-test and by one-way analysis of variance followed by Dunnett’s test for single and multiple comparisons, respectively. Differences were considered statistically significant at p < 0.05.

RESULTS

where Cisf is the unbound concentration in the ISF. To estimate the in vivo permeability rate constant of tramadol, the effective dialysis coefficient (Rd ), which is the ratio of the in vivo and in vitro permeability rate constants, was defined as follows9,19,21 : Rd =

filtered (pore size: 0.2 :m). The filtered samples, ultrafiltrate and dialysate were analyzed on an LC–MS/MS system composed of an Accela HPLC system and a TSQ Quantum Ultra (Thermo Fisher Scientific Inc., Waltham, Massachusetts) mass spectrometer with an electrospray ionization interface in positive ion mode. Chromatographic separation was achieved on a Synergi Hydro-RP column (2.0 × 50 mm, 2.5 :m; Phenomenex, Torrance, California) at a flow rate of 0.3 mL/min. The gradient program was composed of solvent A [ammonium acetate buffer (10 mM, pH 4.0)] and solvent B (methanol) as follows: 0% B for 0–0.5 min, 0%–80% B for 0.5–2 min, 80% B for 2–3.5 min, and 0% B for 3.51–5 min. The column temperature was set at 40◦ C. Xcalibur version 2.1.0 software was used to control the instrument and to collect data.

(2)

where CLvitro is the in vitro dialysis clearance, Cd,vitro and Cr denote the dialysate and reservoir concentrations in the in vitro study, respectively, and F is the dialysis flow rate. The in vivo permeability rate constant (PAvivo ) was also determined from the following equation: CLvivo =

3

In Vitro and In Vivo Permeability Rate Constants and In Vivo Microdialysis Study The in vitro permeability rate constants (PAvitro ) of tramadol and antipyrine, determined in KRP buffer at 37◦ C, were 0.435 ± 0.0152 and 0.497 ± 0.0159, respectively. The in vivo permeability rate constant (PAvivo ) of antipyrine was 0.0929 ± 0.0080. The Rd value was therefore calculated to be 0.190 ± 0.0178. The brain ISF concentration (Cisf ) of tramadol was extrapolated using the reference method (Eq. 5) from the dialysate concentration (Cd,vivo ), PAvitro , and Rd values. The unbound fraction in plasma (fu ) was determined as 0.880 ± 0.022. Figure 1 shows the profiles of Cp,u and Cisf versus time obtained from the microdialysis studies. The plasma and brain ISF concentrations of tramadol reached a steady state by 120 min after the start of the bolus dose plus constant rate infusion regimen. The average values of total and unbound plasma concentrations, and brain ISF concentration in the steady state were 0.988 ± 0.022, 0.870 ± 0.025, and 2.04 ± 0.38 :M, respectively. The brain ISF concentration was significantly higher than the plasma unbound concentration (Fig. 1a) and the average value of Kp,uu,brain was calculated as 2.30 ± 0.37 (Fig. 1b). Uptake Kinetics of Tramadol by hCMEC/D3 Cells Uptake of tramadol by hCMEC/D3 cells was assessed at various intervals (5, 15, 30, 45 s and 1, 2, and 5 min) to determine the optimal time for uptake studies. The initial cell-tomedium (C/M) ratio of tramadol was approximately 60 :L/(mg protein·min). The C/M was increased with time and reached around 200 :L/mg protein at 5 min (Fig. 2). Tramadol uptake increased linearly with time until 1 min, so the initial uptake rate was assessed at 30 s in subsequent kinetic and inhibition studies. Uptake of tramadol by hCMEC/D3 cells was measured at various concentrations (10–1000 :M) and the kinetic parameters (Km and Vmax ) were assessed by fitting the data to the Michaelis–Menten equation. Initial uptake of tramadol showed concentration-dependency (Fig. 3a). Kinetic analysis provided Kitamura et al., JOURNAL OF PHARMACEUTICAL SCIENCES

4

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

Figure 1. Time courses of unbound tramadol concentration in plasma (•) and brain ISF (◦) in the brain microdialysis study (a). Time course of unbound concentration ratio (Kp,uu,brain ; Cisf, /Cp,u ) of brain and plasma (b). Brain microdialysis was performed and plasma and dialysate samples were obtained at the designated time intervals after bolus infusion followed by a 240 min constant-rate infusion of tramadol and antipyrine (reference compound). Each point represents the mean ± SE of five experiments.

Figure 2. Time course of tramadol uptake by hCMEC/D3 cells. Uptake of tramadol (5 :M) was measured at 37◦ C. Each point represents the mean ± SE of four determinations.

a Km value of 43.7 ± 4.9 :M and a Vmax of 5.71 ± 0.24 nmol/(mg protein·min). The nonsaturable uptake clearance Pdiff was 5.09 ± 0.94 :L/(mg protein·min). The Eadie-Hofstee plot for saturable uptake, calculated by subtraction of nonsaturable uptake from total uptake, approximated a single straight line, indicating involvement of a single saturable process (Fig. 3b). Metabolic Energy and Driving Force of Tramadol Transport To clarify the characteristics of the transport system, uptake of tramadol was carried out in the presence of a metabolic energy inhibitor and in the absence of a sodium ion gradient and membrane potential. Tramadol uptake in hCMEC/D3 cells was significantly inhibited by pretreatment with sodium azide, but Kitamura et al., JOURNAL OF PHARMACEUTICAL SCIENCES

Figure 3. Concentration-dependence of tramadol uptake (a) into hCMEC/D3 cells and the Eadie–Hofstee plot (b). Uptake of tramadol (10, 25, 50, 75, 100, 200, 500, and 1000 :M) was measured at 37◦ C. Each point represents the mean ± SE of four determinations. The solid curve, dotted line, and curve represent estimated total, nonsaturable and saturable uptakes, respectively (a). V and S represent initial uptake velocity (nmol/(mg protein·min)) of the saturable component and concentration of tramadol (b).

Figure 4. Effect of ATP-depletion and sodium replacement with Nmethylglucamine or potassium on tramadol uptake into hCMEC/D3 cells. Tramadol uptake (5 :M, for 30 s) was measured in the absence (control) or presence of 0.1% sodium azide (NaN3 ). Sodium azide was preincubated for 20 min. Tramadol uptake was measured in sodium-containing buffer (control) or sodium-replaced buffer with Nmethylglucamine (NMG) or potassium to disrupt membrane potential (KCl). Each column represents the mean ± SE of four determinations. Asterisks show a significant difference, **p < 0.01 versus control.

was not affected by replacement of extracellular sodium ion with N-methylglucamine+ or potassium chloride (Fig. 4). The C/M ratio of tramadol was decreased in acidic transport buffer (pH 6.4) and significantly increased in alkaline medium (pH 8.4), compared with that at pH 7.4 (Fig. 5). To examine the effect of intracellular pH, the cells were treated with NH4 Cl, because intracellular pH rises in the presence of NH4 Cl (acute DOI 10.1002/jps.24129

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

5

Table 1. Inhibitory Effects of Selected Compounds on Tramadol Uptake by hCMEC/D3 Cells Inhibitor MPP+ TEA Carnitine Morphine Codeine Oxycodone Apomorphine Clonidine Pyrilamine Diphenhydramine Amantadine Memantine Verapamil Quinidine

Concentration (mM) 1 1 1 1 1 1 1 1 1 1 1 1 1 1

Relative Uptake (% of Control) 106.6 95.2 78.8 58.0 29.6 24.5 18.4 40.5 23.3 19.6 32.1 21.7 15.0 16.1

± ± ± ± ± ± ± ± ± ± ± ± ± ±

7.28 4.19 4.88** 3.68** 3.08** 1.61** 1.03** 4.48** 3.47** 2.97** 3.30** 3.90** 1.63** 1.08**

Uptake of tramadol (5 :M) was measured at 37◦ C for 30 s in the transport buffer (pH 7.4) containing each compound (1 mM). Each value is the mean ± SE of four determinations. Asterisks show a significant difference, **p < 0.01 versus control.

Figure 5. Effect of changes of extracellular pH (a) and intracellular pH (b) on tramadol uptake into hCMEC/D3 cells. Tramadol uptake was measured in buffer at buffer pH values of 6.4, 7.4, and 8.4 (a). Tramadol uptake (5 :M, for 30 s) was measured under conditions of intracellular acidification and alkalization induced by NH4 Cl treatment (b). Each column represents the mean ± SE of four determinations. Asterisks show a significant difference, **p < 0.01 versus control or pH 7.4.

treatment), and subsequent removal of NH4 Cl (pretreatment) causes a decrease of intracellular pH.17,18 Acute treatment with NH4 Cl (intracellular alkalization) reduced tramadol uptake to 60% of the control (not significant), whereas pretreatment with NH4 Cl (intracellular acidification) resulted in significant stimulation of tramadol uptake (Fig. 5). Inhibition of Tramadol Uptake Uptake was measured after incubation of cells with tramadol in the presence of selected inhibitors at a concentration of 1 mM. Tramadol uptake was not significantly inhibited by TEA (a classical substrate and/or inhibitor of OCTs) or MPP+ (a classical substrate and/or inhibitor of plasma membrane monoamine transporter as well as OCTs), but was partially inhibited by carnitine (a substrate and inhibitor of OCTN2) (Table 1). The uptake was significantly inhibited (to 58%) by morphine, and also by other cationic compounds, codeine, oxycodone, apomorphine, clonidine, pyrilamine, diphenhydramine, amantadine, memantine, verapamil, and quinidine (to less than 41%). In Lineweaver–Burk plot analyses of inhibitory effects on the uptake of tramadol (Fig. 6), the plots of tramadol uptake in the presence and absence of oxycodone intersected at the ordinate axis. This result indicated that oxycodone competitively inhibited tramadol uptake with a Ki value of 347 :M.

DISCUSSION In the present study, BBB transport of tramadol was characterized by means of both rat microdialysis and in vitro uptake studies using immortalized human brain capillary endothelial DOI 10.1002/jps.24129

Figure 6. Lineweaver–Burk plot of the inhibitory effect of oxycodone on tramadol uptake into hCMEC/D3 cells. Tramadol uptake (10–1000 :M, for 30 s) was measured in the absence (◦) or presence of 500 :M oxycodone (•). Each point represents the mean ± SE of three determinations.

hCMEC/D3 cells. The unbound concentration of tramadol in brain ISF was larger than that in plasma and the Kp,uu,brain value was 2.30 ± 0.37, suggesting active uptake into the brain across the BBB. Tramadol exists as over 98% cationic form at physiological pH,11 and might be actively transported through the BBB by the H+ /OC antiporter, which carries various organic cationic drugs including oxycodone, diphenhydramine, and pyrilamine.3,6,7 This idea was supported by our in vitro uptake study using hCMEC/D3 cells, a human BBB model that functionally expresses H+ /OC antiporter.7 Uptake of tramadol by hCMEC/D3 cells was time-dependent. The y-intersection on Figure 2 was approximately 60 :L/mg protein, suggesting that tramadol is rapidly adsorbed to the cells and/or equilibrated between cells and medium. Uptake of tramadol was concentration-dependent, with Km and Vmax values of 43.7 :M and 5.71 nmol/(mg protein·min), respectively. The saturable component (Vmax /Km ) and the nonsaturable component (Pdiff ) of tramadol transport were calculated to be 131 Kitamura et al., JOURNAL OF PHARMACEUTICAL SCIENCES

6

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

and 5.09 :L/(mg protein·min), respectively. Thus, 96% of the total tramadol uptake is accounted for by the saturable component in the low concentration range. The uptake of tramadol by hCMEC/D3 cells was significantly inhibited by pretreatment with metabolic inhibitor, but was insensitive to loss of extracellular sodium or membrane potential. Further, it showed pH-dependency characteristic of a proton-coupled antiporter. Tramadol uptake was significantly decreased by pyrilamine, oxycodone and diphenhydramine (substrates or inhibitors of the H+ /OC antiporter), but was not inhibited or was only partially inhibited by substrates of OCTs (MPP+ and TEA) and OCTN2 (L-carnitine) (Table 1). The uptake of tramadol was competitively inhibited by oxycodone with a Ki value of 347 :M. The Ki value was comparable with that of diphenhydramine uptake in our previous report in which diphenhydramine uptake by hCMEC/D3 cells was competitively inhibited by oxycodone with a Ki value of 304 :M.7 These results suggest that the BBB transport of tramadol is mainly mediated by the previously reported H+ /OC antiporter-mediated organic cation transport system,6,7 although we cannot deny the possibility that tramadol is transported by any other transporter(s). The uptake was significantly increased at higher extracellular pH (pH 8.4), and after intracellular acidification induced with NH4 Cl. On the other hand, decreased tendency in tramadol uptake was observed at lower extracellular pH (pH 6.4) and after intracellular alkalization. The similar tendency was observed in our previous report.6,7 As the pKa value of tramadol is 9.41,11 the proportion of uncharged tramadol can be estimated as 0.098% at pH 6.4, 0.97% at pH 7.4 and 8.9% at pH 8.4. Compared with the large change of the uncharged fraction, extra- and/or intracellular acidification or alkalization caused relatively small changes in tramadol uptake (Figs. 5a and 5b), suggesting that passive diffusion according to the pH-partition theory could not be solely responsible for tramadol uptake by hCMEC/D3 cells. This is consistent with the fact that tramadol is actively taken up into hCMEC/D3 cells via the H+ /OC antiporter. The H+ /OC antiporter shows no marked species difference between rat and human brain endothelial cells (TR-BBB13 and hCMEC/D3 cells, respectively).7 Thus, the unbound concentration of tramadol in human brain may be estimated to be approximately twofold higher than that in plasma, if the Kp,uu,brain value in human is similar to that measured in rats. From the reported data following a single oral dose of 100 mg, the maximum unbound tramadol concentration in the brain can be estimated to be about 1.5–2.0 :M, taking the Kp,uu,brain value into consideration.22 Such an unbound brain concentration of tramadol should produce fully synergistic effects via monoaminergic modulation and opioid agonism, because it is comparable to the Km values of monoamine transporters, such as NET (0.79 :M) and SERT (0.99 :M), and the :-opioid receptor (2.1 :M) in neuronal cells.1 This conclusion is also supported by the fact that the uptake clearances in TR-BBB13 cells and hCMEC/D3 cells for pyrilamine approximate to the human BBB permeability of [11 C]pyrilamine determined by a human PET study.7,23 Although tramadol possesses moderate affinity for :-opioid receptor, the binding affinity is approximately 6000-fold less than that of morphine.1 The metabolite M1 of tramadol also binds to :-opioid receptor with 300-fold higher affinity than the parent form,24 but its affinity is still much lower than that of morphine. On the other hand, the analgesic potency of tramadol following parenteral administration is only 10-fold less than Kitamura et al., JOURNAL OF PHARMACEUTICAL SCIENCES

that of morphine. The apparent discrepancy between in vitro and in vivo results has been interpreted in terms of synergistic effects of monoaminergic modulation and opioid antagonism.1 That is to say, significant inhibition of neuronal reuptake of norepinephrine and serotonin contributes the antinociceptive effect through descending inhibitory pathways in the CNS.1 It should be noted that the difference of at least fourfold between the Kp,uu,brain of morphine (0.3–0.5)25 and tramadol (2.3) may also contribute to the discrepancy in analgesic potency between in vitro and in vivo.

CONCLUSIONS Our results indicate that tramadol has 2.3-fold higher unbound concentration in the brain than in plasma in rats, suggesting that it is actively taken up into the brain from circulating blood. In vitro uptake study using hCMEC/D3 cells suggested that H+ /OC antiporter plays a major role in active transport of tramadol across the human BBB.

ACKNOWLEDGMENTS We thank Dr. Pierre-Olivier Couraud (Institut Cochin, Paris, France) for supplying hCMEC/D3 cells under license from INSERM. This work was supported in part by a Grant-in-Aid for Scientific Research and by the MEXT-Supported Program for the Strategic Research Foundation at Private Universities provided by The Ministry of Education, Culture, Sports, Science and Technology.

REFERENCES 1. Raffa RB, Friderichs E, Reimann W, Shank RP, Codd EE, Vaught JL. 1992. Opioid and nonopioid components independently contribute to the mechanism of action of tramadol, an “atypical” opioid analgesic. J Pharmacol Exp Ther 260:275–285. 2. Ohtsuki S, Terasaki T. 2007. Contribution of carrier-mediated transport systems to the blood–brain barrier as a supporting and protecting interface for the brain; importance for CNS drug discovery and development. Pharm Res 24:1745–1758. 3. Okura T, Hattori A, Takano Y, Sato T, Hammarlund-Udenaes M, Terasaki T, Deguchi Y. 2008. Involvement of the pyrilamine transporter, a putative organic cation transporter, in blood–brain barrier transport of oxycodone. Drug Metab Dispos 36:2005–2013. 4. Deguchi Y, Miyakawa Y, Sakurada S, Naito Y, Morimoto K, Ohtsuki S, Hosoya K, Terasaki T. 2003. Blood–brain barrier transport of a novel micro 1-specific opioid peptide, H-Tyr-D-Arg-Phe-beta-Ala-OH (TAPA). J Neurochem 84:1154–1161. 5. Shimura T, Tabata S, Terasaki T, Deguchi Y, Tsuji A. 1992. In-vivo blood–brain barrier transport of a novel adrenocorticotropic hormone analogue, ebiratide, demonstrated by brain microdialysis and capillary depletion methods. J Pharm Pharmacol 44:583– 588. 6. Sadiq MW, Borgs A, Okura T, Shimomura K, Kato S, Deguchi Y, Jansson B, Bj¨orkman S, Terasaki T, Hammarlund-Udenaes M. 2011. Diphenhydramine active uptake at the blood–brain barrier and its interaction with oxycodone in vitro and in vivo. J Pharm Sci 100:3912. –3923. 7. Shimomura K, Okura T, Kato S, Couraud PO, Sherrman JM, Terasaki T, Deguchi Y. 2013. Functional expression of a proton-coupled organic cation (H+ /OC) antiporter in human brain capillary endothelial cell line hCMEC/D3, a human blood–brain barrier model. Fluids Barriers CNS 26:10:8. DOI 10.1002/jps.24129

RESEARCH ARTICLE – Pharmacokinetics, Pharmacodynamics and Drug Transport and Metabolism

8. Cisternino S, Chapy H, Andr´e P, Smirnova M, Debray M, Scherrmann JM. 2013. Coexistence of passive and proton antiporter-mediated processes in nicotine transport at the mouse blood–brain barrier. AAPS J 15:299–307. 9. Tega Y, Akanuma S, Kubo Y, Terasaki T, Hosoya K. 2013. Bloodto-brain influx transport of nicotine at the rat blood–brain barrier: Involvement of a pyrilamine-sensitive organic cation transport process. Neurochem Int 62:173–181. 10. Andr´e P, Debray M, Scherrmann JM, Cisternino S. 2009. Clonidine transport at the mouse blood–brain barrier by a new H+ antiporter that interacts with addictive drugs. J Cereb Blood Flow Metab 29:1293– 1304. 11. Azodi-Deilami Saman, Abdoussa Majid, Seyedi Rezvaneh S. 2010. Synthesis and characterization of molecularly imprinted polymer for controlled release of tramadol. Cent Eur J Chem 8:687–695. 12. Frid´en M, Winiwarter S, Jerndal G, Bengtsson O, Wan H, Bredberg U, Hammarlund-Udenaes M, Antonsson M. 2009. Structure-brain exposure relationships in rat and human using a novel data set of unbound drug concentrations in brain interstitial and cerebrospinal fluids. J Med Chem 52:6233–6243. 13. Weksler BB, Subileau EA, Perri`ere N, Charneau P, Holloway K, Leveque M, Tricoire-Leignel H, Nicotra A, Bourdoulous S, Turowski P, Male DK, Roux F, Greenwood J, Romero IA, Couraud PO. 2005. Blood– brain barrier-specific properties of a human adult brain endothelial cell line. FASEB J 19:1872–1874. 14. Shawahna R, Uchida Y, Decl`eves X, Ohtsuki S, Yousif S, Dauchy S, Jacob A, Chassoux F, Daumas-Duport C, Couraud PO, Terasaki T, Scherrmann JM. 2011. Transcriptomic and quantitative proteomic analysis of transporters and drug metabolizing enzymes in freshly isolated human brain microvessels. Mol Pharm 8:1332–1341. 15. L¨oscher W, Potschka H. 2005. Blood–brain barrier active efflux transporters: ATP-binding cassette gene family. NeuroRx 2:86–98. 16. Carl SM, Lindley DJ, Das D, Couraud PO, Weksler BB, Romero I, Mowery SA, Knipp GT. 2010. ABC and SLC transporter expression

DOI 10.1002/jps.24129

7

and pot substrate characterization across the human CMEC/D3 blood– brain barrier cell line. Mol Pharm 7:1057–1068. 17. Boron WF, De Weer P. 1976. Intracellular pH transients in squid giant axons caused by CO2 , NH3 , and metabolic inhibitors. J Gen Physiol 67:91–112. 18. Thomas RC. 1984. Experimental displacement of intracellular pH and the mechanism of its subsequent recovery. J Physiol 354:3–22. 19. Okura T, Kato S, Takano Y, Sato T, Yamashita A, Morimoto R, Ohtsuki S, Terasaki T, Deguchi Y. 2011. Functional characterization of rat plasma membrane monoamine transporter in the blood– brain and blood-cerebrospinal fluid barriers. J Pharm Sci 100:3924– 3938. 20. Deguchi Y, Terasaki T, Kawasaki S, Tsuji A. 1991. Muscle microdialysis as a model study to relate the drug concentration in tissue interstitial fluid and dialysate. J Pharmacobiodyn 14:483– 492. 21. Terasaki T, Deguchi Y, Kasama Y, Pardridge WM, Ysuji A. 1992. Determination of in vivo steady-state unbound drug concentration in the brain interstitial fluid by microdialysis. Int J Pharm 81:143–152. 22. Grond S, Sablotzki A. 2004. Clinical pharmacology of tramadol. Clin Pharmacokinet 43:879–923. 23. Szabo Z, Ravert HT, Gozukara I, Geckle W, Seki C, Sostre S, Peller P, Monsein L, Natarajan TK, Links JM, Wong DF, Dannals RF, Wagner HN Jr. 1993. Noncompartmental and compartmental modeling of the kinetics of carbon-11 labeled pyrilamine in the human brain. Synapse 15:263–275. 24. Gillen C, Haurand M, Kobelt DJ, Wnendt S. 2000. Affinity, potency and efficacy of tramadol and its metabolites at the cloned human mu-opioid receptor. Naunyn Schmiedebergs Arch Pharmacol 362:116– 121. 25. Bostrom E, Hammarlund-Udenaes M, Simonsson USH. 2008. Blood-brain barrier transport helps to explain discrepancies in in vivo potency between oxycodone and morphine. Anesthesiology 108:495– 505.

Kitamura et al., JOURNAL OF PHARMACEUTICAL SCIENCES