ToxSci Advance Access originally published online on March 15, 2007
Toxicological Sciences 2007 97(2):253-264; doi:10.1093/toxsci/kfm057
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Brain Uptake, Pharmacokinetics, and Tissue Distribution in the Rat of Neurotoxic N-Butylbenzenesulfonamide
* Bioanalytical Mass Spectrometry Facility, M305 Wallace Wurth Building, University of New South Wales, Sydney, NSW 2052, Australia
Department of Pharmaceutical Sciences, School of Pharmacy, Texas Tech University Health Sciences Center, 1300 Coulter Boulevard, Amarillo, Texas 79106
Department of Pediatrics and Division of Endocrinology, Metabolism and Diabetes, School of Medicine, University of Colorado Health Sciences Center, Mail Stop 8119 PO Box 6611, Aurora, Colorado 80045
1 To whom correspondence should be addressed. E-mail: netnoggy{at}netscape.net.
Received February 14, 2007; accepted March 5, 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The pharmacokinetics, cerebrovascular permeability, and tissue distribution of the neurotoxic plasticizer N-butylbenzenesulfonamide (NBBS) were determined in rats. A stable isotope–labeled form ([13C6]NBBS) was used to circumvent ubiquitous contamination that was evident whenever the native form was measured. Plasticizer decline in plasma, following an iv dose of 1 mg/kg, was described by a triexponential decay function. NBBS was cleared from plasma at a rate of 25 ml/min/kg, and 24 h after administration, plasma concentrations represented 0.04% of the administered dose. These data suggest rapid elimination and uptake into tissue; however, NBBS was not accumulated by any of the tissues studied (i.e., liver, kidney, muscle, adipose tissue, and brain). Given the critical interest in NBBS neurotoxicity, the brain uptake of [13C6]NBBS was further explored in experiments using the in situ brain perfusion technique. During perfusion with protein-free saline for 15–30 s, the single-pass brain extraction for free [13C6]NBBS was very high (73–100%) with a unidirectional blood-brain barrier transfer constant (Kin) of > 0.08 ml/s/g. No significant differences were found in [13C6]NBBS content among the measured brain regions. Plasma protein binding (70%) only slightly lowered the single-pass brain extraction to 48%. In summary, the results demonstrate that NBBS distributes rapidly to tissues, including brain. Though highly lipophilic with a Log octanol/water partition coefficient of 2.17 ± 0.09, brain:blood ratios (2:1) for NBBS were consistent throughout the experimental duration, with little indication of accumulation.
Key Words: N-butylbenzenesulfonamide; neurotoxin; plasticizer; cerebrovascular; permeability; blood-brain barrier.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
N-butylbenzenesulfonamide (NBBS; CAS registry number 3622-84-2; Fig. 1) is a plasticizer commonly incorporated in films, transparent paper coatings, and nylons. Published reports of the acute toxicity of NBBS are scant, although one report indicates that the plasticizer has a LD50 value in the range of 1725–2050 mg/kg after oral administration to rats and greater than 1150 mg/kg after dermal application to rabbits (the number of animals used to generate these data is unknown; Safety Data Sheet, 1991).
|
Although NBBS is not acutely toxic, the plasticizer appears to cause "slow onset" neurotoxicity. This neurotoxicity is evident following ic and ip administration to rabbits (Strong et al., 1991
) and in rats following ip dosing (Lee et al., 1995). Hyperreflexia, the first sign of neurotoxicity, is evident within 2–3 weeks of iv administration (Strong et al., 1991). External rotation, splaying of the hindlimbs, clonus, hypertonia, and impaired backpedaling then develop progressively (Strong et al., 1991). The development of these symptoms is accompanied by pathologic changes in the spinal cord and brainstem.
NBBS is soluble in water (1.02 g/l at 20°C) and in view of its widespread use, it is predictably found in ground water (Albaiges et al., 1986; Burchill et al., 1983, 1991; Guardiola et al., 1989; Hutchins and Ward, 1984; Noordsij et al., 1985). High concentrations have been reported in sites tested in Australia (95 µg/l; Duffield et al., 1994) and Germany (420–710 µg/l; Schwarzbauer et al., 2002). The plasticizer has been detected in the Delaware River (< 1 µg/l; Sheldon and Hites, 1979) and in effluent discharged from wastewater treatment into the Santa Ana River (< 0.5 µg/l; Gross et al., 2004). NBBS has also been reported in a source of water for Philadelphia although it was not detected in the intake or outfall of sewage in a plant in the same city (Sheldon and Hites, 1979). More recently, the plasticizer was detected at levels appreciably higher than background (sample mean of 260 ng/l) in the waters of the San Francisco estuary but not in sediment or in tissue samples of bivalves (mussels and oysters; Oros, 2002; Oros et al., 2003).
The ubiquitous environmental presence and widespread use of NBBS and the unexplained delay in the onset of neurotoxicity induced by the sulfonamide plasticizer, invokes some obvious questions: how rapidly does the plasticizer gain access to the brain; how does it distribute through the brain and the rest of the body, and how does it induce neurotoxicity? We have performed studies on the cerebrovascular permeability, pharmacokinetics, and tissue distribution of NBBS in an attempt to answer these questions. While these studies characterize the uptake and distribution of the plasticizer in the rat brain, the results have been used to comment on the neurotoxic potential of NBBS in humans.
NBBS is an ideal candidate for quantitative studies based on combined gas chromatography mass spectrometry (GC-MS) because it is volatile and can be analyzed without chemical derivatization (Duffield et al., 1994). Because it is ubiquitous in the environment, the plasticizer is frequently encountered as a low-level background contaminant; consequently, elaborate measures had to be taken to minimize contamination in studies involving the native compound. In order to circumvent the problem of low-level background contamination with native NBBS, we employed a stable isotope–labeled form of the plasticizer (i.e., [13C6]NBBS) for iv administration. In this way, we were able to determine cerebrovascular permeability, pharmacokinetics, and tissue distribution of the plasticizer without the low-level background that would otherwise have complicated analyses.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Synthesis of [13C6]NBBS and [2H9]NBBS.
Unlabeled NBBS was obtained from Pfaltz and Bauer (Waterbury, CT). Stable isotope ring–labeled [13C6]NBBS and deuterium side-chain–labeled [2H9]NBBS were synthesized according to the methods outlined in Duffield et al. (1994)
. The procedure for [13C6]NBBS is as follows. Over 30 min, and with constant stirring, [13C6]benzene (0.5 g, 6 mmol) was added to chlorosulfonic acid (1.3 ml, 20 mmol) in an ice-cooled reaction vial. Chloroform (25 ml) and ice (50 g) were then added. After mixing, the chloroform was separated and dried over anhydrous sodium sulfate. n-Butylamine (0.7 ml in 7 x 0.1 ml aliquots) was then added, and the mixture was allowed to stand (50°C, 45 min). The chloroform was then removed under reduced pressure and the pale, yellow oil (i.e., [13C6]NBBS) was purified by column chromatography over silica gel. The column was equilibrated with acetonitrile-toluene (10:90; 100 ml), and fractions of the crude product were eluted with acetonitrile-toluene (5:95; 8 ml). Fractions containing [13C6]NBBS were then combined and evaporated at reduced pressure (Speedvac; Savant Instruments, Farmingdale, NY). For the synthesis of [2H9]NBBS, native benzene was used as the starting material along with n-[2H9]butylamine. In both instances, identity and isotopic purity were established by GC-MS. Both isotopomers displayed a mass spectrum consistent with the proposed structure and the predicted isotopic incorporation (Fig. 2).
|
Precautions.
In order to ensure that quantification was not invalidated by background contamination, several precautions were adopted: (1) all solvents were either analytical or high-performance liquid chromatography grade and were double distilled and stored in glassware routinely baked at temperatures above 250°C for at least 12 h; (2) work areas were cleaned with methanol prior to use; (3) all animals were given deionized water for 6 days prior to administration of NBBS; (4) all animals were fasted for at least 12 h prior to administration of NBBS, and (5) all animals involved in kinetic and permeability studies received the stable isotope, [13C6]NBBS, iv.
In order to ensure dissolution of the plasticizer in isotonic saline, 0.5 mg/ml or 1.0 mg/ml solutions were prepared in glass vials. These solutions were gently heated for not less than 12 h and allowed to cool before refrigeration. GC-MS was routinely employed to confirm NBBS target concentrations before these solutions were used.
Animal work.
Animal studies detailed in protocols 1 and 2 (below) were approved by the Animal Ethics Committee of the National Institute on Aging, National Institutes of Health, Bethesda, MD. Animal studies detailed in protocol 3 (below) were approved by the Animal Ethics Committee of the University of New South Wales (UNSW), Sydney, Australia. All animal studies comply with the European Communities Council Directive (86/609/EEC) for the use of experimental animals, and all efforts were undertaken to minimize both the number of animals used during the course of this study and their suffering.
Control animals were used and received only isotonic saline or (blank) infusions containing no NBBS. In addition, a sample of blood was routinely drawn from each animal just prior to administration, or infusion, of NBBS. Sampling, processing, and analysis of control samples were identical to that employed for the animals that received NBBS.
Cerebrovascular permeability of NBBS (protocol 1).
The transfer of NBBS between plasma and brain was studied using the in situ brain perfusion technique developed by Takasato et al. (1984) as modified by Smith (2003) and Mandula et al. (2006). Male Sprague-Dawley rats (ca. 400 g; Charles River Laboratories, Kingston, NY) were used in this study. Prior to surgery, the rats were anesthetized with sodium pentobarbital (50 mg/kg; ip). An incision was made to expose the right carotid artery, after which the right pterygopalatine, occipital, and superior thyroid arteries were cauterized and cut. A polyethylene catheter (PE50; filled with heparinised saline; 100 IU/ml) was implanted in the right external carotid artery, and then the right common carotid artery was prepared for ligation just proximal to the bifurcation of the external carotid artery. Following surgery, the heart rate, respiration rate, blood pressure, blood pH, and gases were monitored throughout the experiment to ensure normal physiologic status. Perfusion fluid consisted of bicarbonate-buffered perfusion saline (128mM NaCl, 24mM NaHCO3, 4.2mM KCl, 2.4mM NaH2PO4, 1.5mM CaCl2, 0.9mM MgSO4, 9mM D-glucose; pH = 7.40 ± 0.05) or freshly collected rat serum. All perfusates were filtered, maintained at 37°C, and equilibrated with 95% O2/5% CO2 prior to use.
To start the perfusion, the right common carotid artery was ligated and the left cardiac ventricle was cut. Then, perfusion fluid containing [13C6]NBBS (0.5 mg/ml) was infused for 15–30 s, at a constant rate, into the external carotid artery by a syringe/pump (Model 944 Harvard Apparatus, South Natick, MA) at a rate to achieve a perfusion pressure of 80–120 mmHg. Body temperature was maintained at 36.5 ± 0.5°C with a servo-controlled heating pad linked to a rectal temperature probe (YSI Indicating Controller, Yellow Springs, OH). The perfusion was terminated 15–30 s later by decapitation of the animal, after which, samples of perfusate, right parietal, right frontal, and right occipital cortexes were collected and weighed for plasticizer content determination. Regional blood flow was determined in separate animals from the unidirectional uptake of [3H] diazepam (Mandula et al., 2006; Takasato et al., 1984).
Distribution of NBBS in the central nervous system (protocol 2).
Male Sprague-Dawley rats (ca. 400 g; three per time point; Charles River Laboratories) were used in this study. Two hours prior to administration of the plasticizer, polyethylene catheters (PE50) were implanted in the right femoral artery and vein of each animal. The rats were anesthetized with sodium pentobarbitone (40 mg/kg, ip) and kept warm with a heat lamp during the surgery. Two hours after the surgery, the rats were administered NBBS (iv, 1 mg/kg in isotonic saline). At 1, 5, 15, 30, and 60 min after dosage, three rats were killed by sodium pentobarbitone overdose. Just prior to the administration of sodium pentobarbitone, arterial blood was collected. After death, the brain was dissected and the following areas were collected: right parietal cortex, cerebellum, spinal cord, and cerebrospinal fluid (CSF) from the cisterna magna. All samples were stored frozen (– 20°C) until analysis.
Pharmacokinetics and distribution of NBBS in peripheral tissues (protocol 3).
Female Wistar rats (ca. 200 g body weight; four per time point; Animal Breeding and Holding Unit, UNSW, Sydney, Australia) were used for these studies. Two days prior to administration of the plasticizer, a silastic/polyethylene PE 50 catheter was implanted in the right external jugular vein of each animal. The catheter was exteriorized dorsally at the nape of the neck. Prior to the procedure, each rat was administered a preanesthetic dose of diazepam (3 mg/kg; ip) and then anesthetized with fentanyl (0.1 mg/kg; im) and droperidol (5 mg/kg; im). The rats were allowed to recover for 2 days and fasted for 12 h before and during the study (however, deionized water was freely available). Each rat was given an iv dose (1 mg/kg) of stable isotope ring–labeled [13C6]NBBS. At various times up to 24 h, blood (400 ml) was collected (into microcentrifuge tubes containing heparin) for assay, and the animals were then sacrificed with an iv dose of sodium pentobarbitone (200 mg/kg). Samples of liver, kidney, peripheral fat, and skeletal muscle were removed from the 1, 2, 5, 240, and 480 min animals. Blood samples were centrifuged (13,000 x g; 5 min) and the plasma transferred to glass culture tubes. All samples were stored at – 20°C until assay.
Octanol-water partition analysis and equilibrium dialysis (protocol 4).
The NBBS octanol-water partition coefficient (P) was measured at two different concentrations (0.0225 and 0.045 mg/ml using the method described by Hansch and Leo (1995). The aqueous phase consisted of 0.1M phosphate-buffered saline (PBS) (pH 7.4), and the two phases were mixed for 60 min and then allowed to separate. Following content determination, the Log P value was calculated according to the following formula:
 |
The free fraction of NBBS in rat serum was determined by equilibrium dialysis at 37°C (Mandula et al., 2006). Acrylic equilibrium dialysis cells (1 ml; membrane molecular weight cut-off of 6kDa; Bel-Art Products, Pequannock, NJ) were used. Equilibration of free NBBS was achieved in 4 h as determined with simultaneous samples run without protein (i.e., saline vs. saline). Samples were obtained from both protein and protein-free chambers only after equilibrium was achieved.
Preparation of samples for assay of NBBS.
Plasma was thawed and the internal standard added (50 ng [2H9]NBBS in 50 µl of ethyl acetate). These samples were then extracted into dichloromethane and assayed according to the method described by Duffield et al. (1994).
Tissue samples (including whole blood; 20–80 mg) were mixed with the internal standard and homogenized ultrasonically (Vibracell, Sonic Materials, CT) in a saturated solution of barium hydroxide (250 ml) and ethanol (250 ml). The samples were then centrifuged (13,000 x g, 5 min), and the supernatant was extracted into dichloromethane and assayed as described by Duffield et al. (1994).
Chromatography and mass spectrometry.
A Hewlett-Packard 5890 (series II) gas chromatograph (ic) interfaced to a VG Autospec-Q triple focusing high-resolution mass spectrometer was used for this work. The Autospec-Q was operated at a resolving power of 6000 (10% valley definition) in electron ionization mode (trap current = 500 mA; electron energy = 70 eV). The GC was fitted with a split/splitless capillary inlet and used helium as the carrier gas (inlet pressure = 5 psi). Automated splitless mode injections, with a 60 s purge off interval, were used for sample introduction. Chromatography was performed on a fused silica capillary column (12 m x 0.22 mm i.d.) containing a cross-linked siloxane-carborane copolymer stationary phase (HT5; 0.1 mm film thickness; SGE Scientific, Victoria, Australia). Following sample introduction, the GC was programmed to maintain the oven temperature at 50°C for 1 min and then to increase it to 180°C (at 50°/min). The temperature was maintained at 180°C for another minute and then increased to 300°C (at 20°/min). The temperature was held at 300°C for 5 min to ensure complete elution of other potentially interfering volatile components. The temperatures of the injection port, GC-MS transfer line, and source were held constant at 280°C, 300°C, and 220°C, respectively. Voltage scan–selected ion recording was used for sample NBBS and [13C6]NBBS determinations. Three ions, m/z 170.0276, 172.0401, and 176.0477, representing unlabeled NBBS, [2H9]NBBS, and [13C6]NBBS, respectively, were monitored with an optimized dwell time and delay of 40 and 20 ms, respectively. Perfluorokerosene (m/z 180.9888) served as a check and lock mass to maintain calibration.
The limit of quantification was determined to be 5 pg injected on column (at a signal to noise ratio of 3:1), and the coefficient of variation for replicate standards (n = 3) was 1.4%. Standards (0, 10, 50, 100, 500 pg, 1 ng [13C6]NBBS injected) were prepared by serial dilutions of a solution of [13C6]NBBS (10–1000 ng/mL) in ethyl acetate. A fixed amount of internal standard (50 ng in 50 µl) was added to each tube.
Plasticizer content in unknowns was calculated from the ratio of chromatographic peak areas for ions m/z 176.0477 and 172.0401 ([13C6]NBBS/[2H9]NBBS ion pair) by reference to a calibration curve comprising the six standards. The calibration curve was generated by unweighted least-squares linear regression analysis, and an example of an equation for one of the curves utilized in this study follows:
 |
Precision and extraction efficiency of the assay.
The precision and extraction efficiency of the assay in the picogram range of sample concentrations was assessed by the repeated preparation and analysis of plasma, blood, and tissue samples spiked with [13C6]NBBS. Plasma (n = 4), whole blood (n = 4), and liver (n = 4) were spiked to give final concentrations of 10 ng/ml [13C6]NBBS (plasma and whole blood) and 50 ng/g [13C6]NBBS (liver). The samples were then extracted and analyzed as described above. The recovery of [13C6]NBBS from plasma and whole blood was 72% (SD = 2.9% for plasma and 4.6% for whole blood; compared to a corresponding unextracted standard) and 99% (SD = 8%) for liver (compared to a corresponding unextracted standard).
Cerebrovascular permeability analysis.
The transfer coefficients for NBBS influx into brain, Kin(NBBS), were calculated as follows (Mandula et al., 2006; Smith, 2003; Takasato et al., 1984):
 |
):
 |
Pharmacokinetic analysis.
Plasma NBBS concentrations following iv administration were best described with the following exponential decay function:
 |
: for example, plasma clearance (Clp) was calculated following determination of the area under the mean plasma [13C6]NBBS concentration-time curve, and the steady-state parameters (mean residence time [MRT] and the steady-state volume of distribution [Vss]) were calculated following determination of the area under the first moment. Mean plasma data were used because individual plasma-time curves (for each animal) could not be obtained with any of the protocols adopted in this report; animals were sacrificed at each time point for tissue samples.
Plasma clearances were converted to blood clearances using the correction detailed by Benet et al. (1991), a red blood cell (RBC):plasma NBBS concentration ratio of 3:1, and a rat hematocrit value of 0.46 (Altman and Dittmer, 1974). The RBC:plasma NBBS concentration ratio was calculated following plasticizer determinations in four samples each of packed RBC and plasma obtained 1 min after administration.
Tissue levels were corrected for residual intravascular plasticizer by subtracting the product of the residual plasma volume and plasma [13C6]NBBS from the gross value of the same in tissue for the corresponding time point. The values used for residual plasma volumes were obtained from Altman and Dittmer (1971): 92 ml/g (for kidney); 99 ml/g (for liver), and 4 ml/g (for muscle and fat). Uptake into the various organs, 1 min after administration, was calculated by multiplying tissue NBBS concentrations by organ volumes scaled as a function of rat body weight (scaling relations were obtained from Bischoff et al., 1971).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Analytical Issues
Studies of ubiquitous environmental contaminants, such as NBBS, are difficult to undertake. Before we adopted the use of a stable isotope–labeled form of the plasticizer, a persistent background prevented us from determining the low levels present in plasma following administration. With the introduction of a stable isotope–labeled form of the plasticizer and the implementation of several simple precautions (i.e., double distilling solvents and storing them in glassware, routinely baking glassware at temperatures above 250°C for at least 12 h, and cleaning work areas with methanol prior to benchwork), we were able to undertake precise and accurate content determination of samples for cerebrovascular permeability and pharmacokinetic analyses.
Combined GC-MS is an excellent approach for the analysis of NBBS in a wide variety of sample types; the plasticizer is volatile and can be analyzed without prior chemical derivatization. Furthermore, stable isotope internal standards and accurate mass-selected ion monitoring were employed to ensure high precision and accuracy. The assay for the determination of NBBS sample content was not complicated. NBBS in liquid sample unknowns was extracted into dichloromethane and then subjected to alkaline, acid, and water washes to remove potentially interfering species. Plasticizer in tissue sample unknowns was subjected to a saturated barium hydroxide pre-extraction step, to remove interfering species, prior to extraction into dichloromethane and alkaline, acid, and water washing. This extraction process, coupled with gas chromatography and high-resolution mass spectrometry, ensured good reproducibility.
When the mass spectrometer was operated in low-resolution mode (resolving power of 1000; 10% valley definition), estimates of the peak areas for NBBS, [13C6]NBBS, and [2H9]NBBS were compromised by the presence of coeluting components (with ions of the same nominal mass) from the sample matrix. This situation was amplified by poor peak shapes, especially in tissue where lipophilic [13C6]NBBS had to be extracted from samples high in lipid and protein. This situation could not be improved with preparative chromatography because each additional preparative step caused an increase in background levels of NBBS and reduced sample recovery. As a result, samples were routinely analyzed at a resolving power of 6000 (10% valley definition), which greatly improved peak shapes and peak area determination and eliminated the influence of coeluting species.
Cerebrovascular Permeability of NBBS (Protocol 1)
The amount of plasticizer in the blood vessels of the brain was determined by multiplying brain vascular volume (Vv = 0.008 ml/g; Mandula et al., 2006) by perfusate NBBS concentration. When this product is subtracted from the total measured brain concentration of NBBS, the result is extravascular plasticizer concentration. This vascular correction was not significant because the extravascular NBBS content was, in general, greater than 99% of the total brain NBBS content. The extravascular content of plasticizer was then divided by the product of perfusate NBBS concentration and perfusion time to obtain unidirectional transfer coefficients for influx of plasticizer into the three regions of the brain studied (Kin values; Table 1).
|
Saline perfusion.
NBBS uptake into brain was very rapid, achieving a concentration equal to or exceeding that of the perfusion fluid within 15–30 s. Calculated Kin values were slightly higher at 15 s than at 30 s, but both approximated the mean cerebral perfusion fluid rate (mean cortical flow, F = 0.115 ml/s/g; Table 4). Uptake was sufficiently high that the single-pass extraction of NBBS from saline approached 100%. Calculated NBBS PS values were higher than F. No appreciable differences in the PS, Kin, or extraction were noted among the brain regions studied.
|
Serum.
The 15-s Kin values for NBBS from serum perfusion was 50–60% lower than the values calculated from saline perfusion. The reduction in initial brain NBBS uptake is likely related to the fact that 70% of the NBBS in serum was bound to plasma proteins. Regardless, the uptake from serum was rapid with a mean cortical extraction of 46%.
Distribution of NBBS in the Central Nervous System (Protocol 2)
In these experiments, NBBS was administered iv and then monitored in parietal cortex, cerebellum, spinal cord, CSF, and blood from male Sprague-Dawley rats. Because native plasticizer was used for profiling, a set of tissues from animals not administered NBBS was assayed to correct for background (i.e., each sample run contained tissues from animals that were not administered plasticizer).
Brain and blood.
Background-corrected NBBS content in brain was highest in the first minute following administration: i.e., cortex = 1781 ng/g (n = 3, Fig. 3); cerebellum = 1830 ng/g (n = 3, Table 5); and spinal cord = 1605 ng/g (n = 2, Table 5). Background-corrected NBBS content in blood was also highest in the first minute following administration: 660 ng/ml (n = 3; range 1103–359 ng/ml). Thereafter, NBBS content declined in parallel across cortex, cerebellum, spinal cord, blood, and CSF (Fig. 3, Table 5). NBBS in blood, for example, decreased to 66 ng/ml (n = 3; range = 58–80 ng/ml) 60 min after administration (Fig. 3).
|
|
Average brain to blood NBBS concentration ratios were constant during the experiment and across each of the regions studied: i.e., cortex:blood = 2:1 (SD = 0.5; range = 1–3); cerebellum:blood = 2:1 (SD = 0.6; range = 1–3), and spinal cord:blood = 2:1 (SD = 0.3).
Cerebrospinal fluid.
NBBS content in CSF was highest in the first minute following administration (202 ng/ml; n = 3; Fig. 3). Levels of plasticizer in CSF were lower than corresponding blood levels; the average CSF to blood concentration ratio was 0.28 (SD = 0.04; range = 0.2–0.3). For the duration of the experiment, the concentrations of NBBS in the cortex, cerebellum, and spinal cord were at least fivefold higher than the corresponding CSF concentrations: i.e., cortex:CSF = 6:1 (SD = 1.4; range = 5–9); cerebellum:CSF = 7:1 (SD = 1.9; range = 5–9), and spinal cord:CSF = 7:1 (SD = 0.6; range = 7–8).
Pharmacokinetics and Distribution of NBBS in Peripheral Tissues (Protocol 3)
Pharmacokinetics.
The decline of [13C6]NBBS in plasma was monitored for 24 h following iv administration of a dose of 1 mg/kg. The plasma content of [13C6]NBBS decreased from a mean peak of 603 ng/ml (n = 4), 1 min after administration, to 9.5 ng/ml (n = 4) 24 h after administration (Fig. 4). Plasticizer decline in the rat was polyphasic and was best described by a triexponential decay function. A biexponential decay function could not be fitted to the data satisfactorily because it failed to reach the first time point and resulted in unacceptably large residuals. The polyphasic pattern of decline was characterized by an early and steep drop in plasma concentrations from the first to the second minute following administration; thereafter, plasticizer levels declined gradually. Pharmacokinetic parameters are presented in Table 6.
|
|
The mean half-lives corresponding to the three exponential terms were 0.78, 11, and 1036 min, and the mean total body clearance of NBBS was 25 ml (plasma)/min/kg. The values of Vd (terminal volume of distribution) and Vss (steady-state volume of distribution) are large and slightly exceed 37.5 and 32.8 l/kg, respectively. Plasma clearance was converted to blood clearance because NBBS partitions readily into erythrocytes (RBC:plasma = 3:1; n = 4). As a consequence of this ready partitioning, the amount of plasticizer delivered to the tissues is higher than might be expected from the measured plasma concentrations.
Tissue distribution.
The plasticizer was present in all tissues sampled. In liver, [13C6]NBBS content was highest in the second minute following iv administration (3130 ng/g; n = 4). The levels declined over 8 h to a mean of 3.8 ng/g (n = 4; Fig. 5A). The tissue-plasma ratios for liver were: 3:1 (1 min), 8:1 (2 min), 3:1(5 min), 0.5:1 (240 min), and 0.2:1 (480 min). For kidney, [13C6]NBBS content was highest at 1 min after administration (1338 ng/g; n = 4), and thereafter, levels declined over 8 h to 3.6 ng/g (n = 4; Fig. 5D). The tissue-plasma ratios for kidney were: 2:1 (1 min), 2:1 (2 min), 3:1(5 min), 0.5:1 (240 min), and 0.2:1 (480 min). In both kidney and liver, the decline in plasticizer concentration from the first to the second minute following administration was not as dramatic as the corresponding fall in the plasma profile.
|
In skeletal muscle, [13C6]NBBS reached a maximum 2 min after administration (1044 ng/g; n = 4) and then remained constant until 5 min following administration (Fig. 5B). Thereafter, plasticizer in muscle declined over 8 h to 3.3 ng/g (n = 4). The tissue-plasma ratios for skeletal muscle were: 2:1 (1 min), 3:1 (2 min), 4:1(5 min), 0.6:1 (240 min), and 0.2:1 (480 min). Plasticizer levels in liver, kidney, and skeletal muscle all declined to approximately 10 ng/g at 4 h.
In contrast to liver, kidney, and skeletal muscle, a slight increase in [13C6]NBBS content was observed in perirenal fat 5 min following administration (Fig. 5C). The tissue-plasma plasticizer ratio increased by more than threefold during this time. Fat [13C6]NBBS content declined over 8 h to 2.9 ng/g (from a 5 min maximum of 2374 ng/g; n = 4). At 4 h, plasticizer content was much higher than in liver, kidney, or skeletal muscle. By 8 h, however, NBBS levels in fat were similar to the levels in the other tissues, and these tissue levels were lower than the corresponding plasma NBBS level. Tissue-plasma ratios for fat were: 3:1 (1 min), 4:1 (2 min), 10:1 (5 min), 3:1 (240 min), and 0.2:1 (480 min).
To demonstrate that the patterns of plasticizer decline in the various tissues are not an artifact of the sampling times (i.e., 1, 2, 5, 240, and 480 min), the tissue sampling experiment was repeated over 60 min (sampling times of 1, 5, 15, 30, and 60 min; one animal per time point). The results are shown in Figure 6.
|
NBBS Partition Coefficient and Plasma Protein Binding (Protocol 4)
The equilibrium Log P was determined to be 2.17 ± 0.09 (n = 6) and the free fraction of the plasticizer in rat serum was determined to be 30 ± 0.3% (n = 3).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Prior to the commencement of these studies, key aspects of the pharmacology of NBBS—the cerebrovascular permeability, brain distribution and pharmacokinetics of the sulfonamide plasticizer—had not been characterized. The toxicity of NBBS was, however, well established: (1) rabbits injected ic with doses higher than 10 mg (per animal) developed a progressive spastic myelopathy and (2) oral exposure to water containing NBBS (at a level of 83 mg/ml) was sufficient to elicit progressive neurotoxicity and death in rabbits (Strong et al., 1991
). The significance of these reports was immediately obvious; if the sulfonamide plasticizer is efficiently absorbed into the circulation after oral administration, NBBS contamination of food and water might pose a risk. Cerebrovascular permeability and brain distribution studies were critically important for assessing this risk.
Cerebrovascular Permeability of NBBS
The uptake of NBBS into brain was rapid and limited only by the flow of blood into the central nervous system (CNS). This flow-limited distribution was indicated by unidirectional transfer coefficients (Kin values) approximating the rate of perfusion fluid flow to the brain (>> 0.11 ml/s/g). However, the uptake of NBBS was not exclusively unidirectional during the 30 s of saline perfusion; Kin values decreased slightly after 30 s of saline perfusion. This is an indication of backflux (i.e., movement of NBBS from parenchyma back into the vasculature). Backflux was not unexpected because the concentration of extravascular NBBS had exceeded the concentration of the same in the perfusate. Alternatively, the decrease in Kin could have been caused by NBBS binding to the vascular endothelia. Regardless of the exact mechanism, Kin values for the plasticizer 15 and 30 s into the perfusion are high and indicate that there is no permeability restriction for brain uptake of NBBS.
Because of the lower viscosity of saline (as compared with blood), perfusion flow rates of upto 7 ml/min/g can be achieved with the brain perfusion technique. The 15-s saline Kin values for NBBS reported in this study are actually representative of values that might be achieved with flow rates exceeding 8 ml/min/g. This value exceeds the perfusion flow rate for saline and again attests to the speed with which NBBS is taken up by the brain. The uptake of NBBS does, however, decrease following serum perfusion; in the frontal cortex, for example, the mean Kin value following serum perfusion for 15 s is 65% of the corresponding value following saline perfusion. This reduction in brain uptake of NBBS following serum perfusion is due to the plasticizer binding to plasma proteins. The decrease in cerebrovascular transfer of a drug in a perfusion medium containing plasma proteins is not without precedent; plasma protein binding can decrease cerebrovascular transfer by as much as 99% (Levitan et al., 1983). Brain uptake of the lipophilic chemotherapeutic vinca alkaloids, vinblastine and vincristine, is similarly affected (Greig et al., 1990). Greig and colleagues reported that the permeability surface area products of both compounds were significantly higher in the absence of protein.
Despite the decrease in unidirectional transfer coefficients (Kin values) following serum perfusion, extravascular (parenchymal) plasticizer content remained high across the frontal, parietal, and occipital cortices. Vascular NBBS was extracted efficiently by brain tissue regardless of the limitation posed by plasma proteins. This was mirrored in all the other tissues studied in vivo following iv administration of NBBS.
An estimate of the true permeability surface area product (PS) for the plasticizer can be made using the measured octanol-water partition coefficient of 102.2 at pH 7. The empirical relation between permeability and the octanol-water partition coefficient for a given compound is:
 |
where P is permeability and Pc is the partition coefficient. When the measured partition coefficient is substituted into the empirical relation, the calculated log10P for NBBS is – 2.16 (permeability = 6.85 x 10–3 cm/s). This translates to a permeability surface area product of 0.96/s after the rat brain capillary surface area of 140 cm2/g (i.e., PS = 140(P)) is taken into account. Assuming an average cerebral perfusion fluid flow rate of 0.11 ml/s/g, the calculated single-pass extraction value for NBBS is 99.9% with a Kin of 0.11/s. This calculated value is in close agreement with the combined Kin values obtained after upto 30 s of saline perfusion.
The plasticizer is so highly permeable that PS values for NBBS determined in parietal and frontal cortices following 15 s of serum perfusion exceed corresponding values for antipyrine, an antipyretic and anti-inflammatory agent that is considered to be rapidly penetrating, by a factor of 6.5 (determined by saline perfusion for up to 60 s; Takasato et al., 1984). Permeability surface area products for the other sulfonamide plasticizers and phthalate plasticizers, including diethylhexylphthalate, have yet to be measured.
Distribution of NBBS in the CNS
NBBS profiles for the cortex, cerebellum, and spinal cord indicate that the concentration of plasticizer peaked in the first minute following iv administration. Uptake of NBBS by the various CNS regions sampled was similar; the plasticizer was not preferentially distributed to any one region of the CNS. NBBS in brain and CSF reached distributional equilibrium with plasticizer in blood during the first minute following administration. Subsequent to the attainment of distributional equilibrium, the content of NBBS in cortex, cerebellum, spinal cord, and CSF declined in parallel with blood plasticizer content.
Over 60 min, the concentrations of NBBS in the cortex and cerebellum were at least fivefold higher than the corresponding CSF concentrations. This difference in brain tissue and CSF plasticizer concentrations is most likely due to the low solubility of NBBS in water, higher affinity of the plasticizer for lipophilic brain parenchyma, and the markedly low protein content in CSF (0.15–0.19 mg/ml vs. 60 mg/ml in plasma; Altman and Dittmer, 1971). Because NBBS is a lipid soluble, hydrophobic compound and the predominant constituent of CSF is water (99%; solids 1%; Altman and Dittmer, 1971), the plasticizer preferentially distributes to lipophilic brain tissue.
Passive diffusion of the plasticizers across the blood-CSF barrier is indicated by constant CSF to blood concentration ratios throughout the duration of the experiment. Indeed, the distribution of the plasticizer in CSF is of particular importance because initial studies of the neurotoxicity of NBBS in rabbits utilized ic dosing (Strong et al., 1991). This mode of administration delivers the entire dose of plasticizer directly into the CNS via CSF. All of the administered dose of NBBS would then be available for transfer into the brain because the absence of protein in CSF means that the plasticizer is unbound and active and can diffuse readily across the blood-CSF barrier. Despite the rapid elimination of NBBS from brain, high initial concentrations would result with the ic injections used by Strong and colleagues.
Pharmacokinetics and Tissue Distribution
A comparison of the total plasticizer content of plasma 1 min after an iv dose of NBBS with the amount of plasticizer actually administered, revealed that more than 98% of the dose had been removed from plasma during the first minute. From the first to the second minute following administration, there is a further drop in plasma NBBS concentration. Twenty-four hours after an iv bolus dose (1 mg/kg), the concentration of NBBS in plasma was only 0.04% of the administered dose (based on a rat plasma volume of 8 ml).
This early and rapid decline of the plasma NBBS concentration indicates that the plasticizer is distributed quickly; calculated estimates of plasticizer uptake into the various tissues reveal that 1 min after administration, liver and kidney together received 15% of the dose, blood carried 9%, and muscle accounted for 52% of the dose.
Confirmation of rapid distribution was obtained after an examination of tissue NBBS profiles. The tissue profiles show that the highest levels of NBBS are attained in the 2 min following administration and that these are much higher than corresponding plasma levels. In liver, for example, NBBS levels are approximately eightfold higher than in plasma (at 2 min). Plasticizer concentrations remained higher in the tissues until 4 h after administration when NBBS in plasma clearly exceeded that in all tissues. The exception was adipose tissue where the concentration of NBBS at this time exceeded that of plasma by a factor of three. These data indicate that the loss of plasticizer from adipose tissue (and the approach toward equilibrium), took longer compared to liver, kidney, and skeletal muscle. This is likely due to the higher affinity for NBBS, and poorer perfusion, of adipose tissue: it was only by 8 h that the tissue:plasma NBBS ratio for adipose tissue was comparable to those of liver, kidney, and skeletal muscle.
With the data presented in this report, it can be easily seen that (1) NBBS has high initial tissue-plasma partition ratios; (2) redistribution of NBBS from high perfusion tissues (e.g., liver and kidney) to low perfusion tissues (e.g., adipose tissue) is not significant and cannot alone account for the rapid disappearance of the plasticizer from plasma, and (3) there is no evidence of sustained sequestration by any of the tissues examined. This means that elimination of the plasticizer quite possibly occurs in advance of distribution equilibrium.
Elimination in advance of distribution equilibrium occurs when there is an appreciable difference between Vd (terminal volume of distribution) and Vss (steady-state volume of distribution); the larger the difference, the more extensive is the elimination of the agent (Rowland and Tozer, 1995). For NBBS, the difference between Vd and Vss (approximately 5 l/kg) is appreciable; the disappearance of plasticizer from plasma was so rapid that the concentration of NBBS in the tissues, particularly the slowly equilibrating tissues (such as adipose tissue), was much greater than that in plasma thereby giving rise to the large value for Vd.
Four hours after NBBS administration (8 h after for adipose tissue), the high tissue-plasma concentration ratios have been reduced dramatically, and most of what little is left of the administered plasticizer is in plasma. This effect is readily explained in terms of concentration-dependent plasma protein binding. At high plasma NBBS concentrations, high-affinity binding sites on plasma proteins are saturated and the plasticizer is free to move out of plasma and into tissues; however, as more of the administered NBBS is eliminated with time and the plasma concentration decreases, these high-affinity binding sites increasingly become unsaturated. This would lead to a selective trapping of the plasticizer in plasma, or "the vascular compartment," a time dependent reduction of tissue-plasma concentration ratios and the long terminal phase. This concentration-dependent plasma protein–binding scenario is entirely consistent with the data presented in this report and is also not without precedent. The HIV-1 protease inhibitor, Lopinavir (Kumar et al., 2004) and MK-826, a carbapenem antibiotic (Wong et al., 1998) are two such agents that possess a similar nonlinearity in plasma protein binding.
Are There Risks Associated with Human Exposure to NBBS?
Progressive spastic myelopathy in New Zealand White rabbits only occurs following doses of NBBS greater than 10 mg (ic; Strong et al., 1991). Within 2–3 weeks of the initial administration, hyperreflexia is observed; thereafter, symptoms of progressive neurodegeneration are observed. The neurological deficits apparent with monthly ic treatment of rabbits with doses of upto 100 mg of NBBS (ic) could not, however, be induced more rapidly by repeated ip injections at an increased dosage (200 mg/kg; Strong et al., 1991). In fact, comparable deficits were only induced with ip doses of 300 mg/kg or higher (three times weekly) in rabbits. In rats, repeated ip injections of NBBS (every 6 h; 300 mg/kg) were required (Lee et al., 1995).
The tissues of the CNS demonstrated significant uptake of the plasticizer; each region of the brain studied had an initial extraction that was nearly complete (i.e., approached 100%). While this indicates that there is no permeability restriction for NBBS (i.e., distribution of the plasticizer in brain is flow limited), the rapid establishment of distribution equilibrium between plasticizer in brain and blood, early in the first minute, means that accumulation of NBBS is unlikely. Indeed, over the experimental duration, accumulation was not observed; levels of NBBS in parietal cortex, cerebellum, spinal cord, and CSF fell in parallel with plasticizer in blood. Assuming that the whole-body disposition of NBBS in the rat is representative of the same in other species, accumulation of the plasticizer in the brain and viscera of humans may be unlikely.
The neurotoxicity of NBBS also appears to be route specific. Motor dysfunction was induced with NBBS in the rat only after repeated 6-hourly doses (300 mg/kg, ip), but even then, the effects were of short duration, and the animals appeared to recover when dosing ceased (Lee et al., 1995). With accurate pharmacokinetic data, an assessment can be made of whether an ip dose of 300 mg/kg administered three times per week results in identical or higher brain NBBS concentrations than that achieved by monthly ic doses of upto 100 mg.
The assessment rests on the following assumptions: (1) the volume of CSF is 14% of transcellular fluid (TF) volume; (2) TF volume is 2.5% of total body water (TBW); (3) rabbit TBW is 735 ml/kg (Altman and Dittmer, 1971), and (4) rabbit blood volume is 60 ml/kg (Altman and Dittmer, 1971). The rabbits used by Strong and et al. (1991) weighed 2 kg. Therefore, the values for TBW and blood volume of these rabbits would have been 1470 and 120 ml, respectively. CSF volume would have been 5 ml (1470 x 2.5% x 14%). The final doses in CSF used by Strong and colleagues would have ranged from (10 mg [ic] or 2 mg/ml to 100 mg [ic] or 20 mg/ml).
Our results indicate that following a 1-mg/kg (iv) dose in rats, the concentration of NBBS achieved in CSF, 1 min following administration, is 200 ng/ml (the lowest neurotoxic dose used by Strong and colleagues was 2000 ng/ml; CSF). Assuming that the pharmacokinetics of NBBS in the rabbit and rat are similar (i.e., only ca. 2% of the administered dose remains 1 min following administration), a dose of 300 mg/kg administered ip to 2-kg rabbits would have resulted in a blood concentration of 100 mg/ml (600 mg/120 ml x 2%). The concentration in CSF would have been 30 mg/ml (as the CSF:blood concentration ratio is 0.3). This is approximately the concentration attained by a direct ic injection of 100 mg of NBBS.
This calculation shows that, in absolute terms, more NBBS is required to induce symptoms of neurodegeneration by ip rather than by ic administration; in other words, the toxic potential of the plasticizer is actually reduced as a result of ip administration. This is a good indicator that NBBS may be subject to first-pass metabolism (ip dosing approximates oral dosing in that the liver is not bypassed). Incubation of rat, rabbit, and human liver S9 fractions (post-mitochondrial supernatant) containing NBBS did reveal the presence of a w-1 alcohol (2-hydroxy-N-butylbenzenesulfonamide; unpublished observations); however, further studies of the bioavailability and biotransformation of the plasticizer are warranted.
Of the possible routes that humans could be exposed to NBBS, inhalation is the cause for most concern. Inhalation of vaporized NBBS may allow the plasticizer to bypass organs of biotransformation and elimination and may increase its toxic potential. Once inhaled, NBBS may be able to access susceptible brain tissues via venous drainage of the nasal submucosa or the transport system linking the submucosa to the subarachnoid spaces of the brain (Sakane et al., 1991a,b).
While this work represents the first attempt to characterize the in vivo disposition of any of the sulfonamide plasticizers, it does not provide an explanation for NBBS-mediated progressive spastic myelopathy in rabbits. Human exposure to neurotoxic levels of NBBS is unlikely but not impossible; although surveys for the presence of the sulfonamide plasticizer have returned undetectable levels in selections of plastic packaged drinking waters, fruit juices, milk, and food (unpublished observations), it is not outside the realms of possibility that neurotoxic levels of NBBS could be delivered to the brain via injectable or infusible solutions or by inhalation in industrial settings. Many of the injectable or infusible preparations are packaged and/or stored in plastic containers and have been subjected to a filtration step involving polymer (e.g. nylon) filters, or passage through tubing, containing NBBS. The plasticizer has been reported to leach out of nylon-11 tubing at high levels (764 mg/l; Gilmore et al., 2004). Based on this information, and the possible trapping of NBBS in the vascular compartment, it may be a prudent step for studies to be undertaken on the inhalational bioavailability and biotransformation of the plasticizer.
|
|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Albaiges J, Casado F, Ventura F. Organic indicators of groundwater pollution by a sanitary landfill. Water Res. (1986) 20(9):1153–1159.
Altman PL, Dittmer DS. Blood and Other Body Fluids. [3rd printing with corrections] (1971) Maryland: Federation of American Societies for Experimental Biology.
Altman PL, Dittmer DS. Biology Data Book (1974) Vol. III, 2nd ed. Maryland: Federation of American Societies for Experimental Biology.
Bischoff KB, Dedrick RL, Zaharko DS, Longstreth J. Methotrexate Pharmacokinetics. J. Pharm. Sci (1971) 60(8):1128–1133.[CrossRef][ISI][Medline]
Benet LZ, Mitchell JR, Sheiner LB. Pharmacokinetics: The Dynamics of Drug Absorption, Distribution and Elimination. In: Goodman & Gilman's The Pharmacological Basis of Therapeutics—Gilman AG, Rall TW, Nies AS, Taylor P, eds. (1991) New York: Pergamon Press.
Brambilla A, Broglia L, Nidasio G. N-butylbenzenesulfonamide in drinking water. Boll. Chim. Igien. Parte Sci (1991) 42:779–785.
Burchill P, Herod AA, Marsh KM. Gas chromatography in water Analysis II: Selective detection methods. Water Res (1983) 17(12):1905–1916.
Crone C, Levitt DG. Capillary permeability to small solutes. In Handbook of Physiology, Vol. 4, pp. 411–466 (Microcirculation; Section 2: The cardiovascular system 1) (1984) Bethesda: American Physiological Society.
Duffield P, Bourne D, Tan K, Garruto RM, Duncan MW. Analysis of the neurotoxic plasticiser N-butylbenzenesulfonamide by gas chromatography combined with accurate mass selected ion monitoring. J. Anal. Toxicol (1994) 18:361–368.[ISI][Medline]
Gilmore TJ, Mitroshkov AV, Dresel PE, Sklarew DS. Laboratory investigation into the contribution of contaminants to ground water from equipment materials used in sampling. Ground Water Monit. Remediation (2004) 24(3):88–94.
Greig NH, Soncrant TT, Shetty HU, Momma S, Smith QR, Rapoport SI. Brain uptake and anticancer activities of vincristine and vinblastine are restricted by their low cerebrovascular permeability and binding to plasma constituents in the rat. Cancer Chemother. Pharmacol (1990) 26(4):263–268.[CrossRef][ISI][Medline]
Gross B, Montgomery-Brown J, Naumann A, Reinhard M. Occurrence and fate of pharmaceuticals and alkylphenol ethoxylate metabolites in an effluent dominated river and wetland. Environ. Toxicol. Chem (2004) 23(9):2074–2083.[CrossRef][ISI][Medline]
Guardiola J, Ventura F, Rivera J. Occurrence of industrial organic pollution in a groundwater supply: Screening, monitoring and evaluation of treatment processes. Water Supply (1989) 7:11–16.
Hansch C, Leo A. Exploring QSAR—Fundamentals and Applications in Chemistry and Biology (1995) Washington DC: American Chemical Society.
Hutchins SR, Ward CH. A predictive laboratory study of trace organic contamination of groundwater: Preliminary results. J. Hydrol (1984) 67:223–233.[CrossRef]
Kumar GN, Jayanti VK, Johnson MK, Uchic J, Thomas S, Lee RD, Grabowski BA, Sham HL, Kempf DJ, Denissen JF, et al. Metabolism and disposition of the HIV-1 protease inhibitor Lopinavir (ABT-378) given in combination with ritonavir in rats, dogs and humans. Pharm. Res (2004) 21(9):1622–1630.[CrossRef][ISI][Medline]
Lee WY, Hwang SJ, Cho DJ, Kim JS. Behavioral changes with alterations of choline acetyltransferase immunoreactivities induced by N-butyl benzenesulfonamide. Vet. Hum. Toxicol (1995) 36(6):537–542.
Levitan H, Smith QR, Takasato Y, Ziylan Z, Rapaport SI. Brain uptake of erythrosin-B, a food dye, prevented by binding to plasma proteins in rats. Soc. Neurosci. Abstr. (1983) 9:889. (Abstract).
Mandula H, Parepally JMR, Feng R, Smith QR. Role of site-specific binding to plasma albumin in drug availability to brain. J. Pharmacol. Exp. Ther (2006) 317:667–675.
Noordsij A, Puyker LM, Van Der Gaag MA. The quality of drinking water prepared from bank-filtered river water in the Netherlands. Sci. Total Environ (1985) 47:273–292.[CrossRef]
Oros DR. Identification and Evaluation of Previously Unknown Organic Contaminants in the San Francisco Estuary (1999–2001). RMP Technical Report: SFEI Contribution 75 (2002) Oakland CA: San Francisco Estuary Institute.
Oros DR, Jarman WM, Lowe T, David N, Lowe S, Davis JA. Surveillance for previously unmonitored organic contaminants in the San Francisco Estuary. Mar. Pollut. Bull (2003) 46:1102–1110.[CrossRef][ISI][Medline]
Rowland M, Tozer TN. Pharmacokinetics: Concepts and Applications, 3rd ed (1995) Philadelphia: Lea & Febiger.
Safety Data Sheet (034316/03) for Dellatol-BBS (VPKU 3-3042) (1991) Germany: Bayer AG.
Sakane T, Akizuki M, Yoshida M, Yamashita S, Nadai T, Hashida M, Sezaki H. Transport of cephalexin to the cerebrospinal fluid directly from the nasal cavity. J. Pharm. Pharmacol (1991a) 43:449–451.[ISI][Medline]
Sakane T, Akizuki M, Yamashita S, Nadai T, Hashida M, Sezaki H. The transport of a drug to the cerebrospinal fluid directly from the nasal cavity: The relation to the lipophilicity of the drug. Chem. Pharm. Bull (1991b) 39(9):2456–2458.[Medline]
Schwarzbauer J, Heim S, Brinker S, Littke R. Occurrence and alteration of organic contaminants in seepage and leakage water from a waste deposit landfill. Water Res (2002) 36:2275–2287.[Medline]
Sheldon LS, Hites RA. Sources and movement of organic chemicals in the Delaware River. Environ. Sci. Technol (1979) 13:574–579.
Smith QR. A review of blood-brain barrier transport techniques. Methods Mol. Med (2003) 89:193–208.[Medline]
Smith QR, Takasato Y. Kinetics of amino acid transport at the blood brain barrier studied using an in-situ brain perfusion technique. Ann. N. Y. Acad. Sci (1986) 481:186–201.[Abstract]
Strong MJ, Garruto RM, Wolff AV, Chou SM, Fox SD, Yanagihara R. N-Butyl benzenesulfonamide: A neurotoxic plasticiser inducing a spastic myelopathy in rabbits. Acta Neuropathol (1991) 81:235–241.[CrossRef][Medline]
Takasato Y, Rapaport SI, Smith QR. An in-situ brain perfusion technique to study cerebrovascular transport in the rat. Am. J. Physiol. (1984) 247:H484–H493.[ISI][Medline]
Wong BK, Bruhin PJ, Lin JH. Dose-dependent plasma clearance of MK-826, a cabapenem antibiotic, arising from concentration dependent plasma protein binding in rats and monkeys. J. Pharm. Sci (1998) 88(2):277–280.[CrossRef][ISI]