Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

It is well established that organic anions in the cerebrospinal fluid (CSF) are actively transported into the blood across the epithelial cells of the choroid plexus (Suzuki et al., 1997). This has been established by the observations that organic anions are eliminated from CSF after intracerebroventricular administration and/or during ventriculocisternal perfusion and that the respective organic anions are accumulated in the isolated choroid plexus (Suzuki et al., 1997). Using a series of -lactam antibiotics as model compounds, we previously determined the kinetic parameters governing in vivo elimination from CSF after intracerebroventricular administration and in vitro uptake by the isolated choroid plexus (Ogawa et al., 1994). The similarity in the calculated in vivo and in vitro parameters suggests that the choroid plexus is the predominant organ responsible for the efflux of organic anions from CSF and that the isolated choroid plexus is an excellent tool for predicting the in vivo disposition of ligands in CSF (Ogawa et al., 1994).

Although it has been suggested that many organic anions are accumulated in the isolated choroid plexus in an energy-dependent manner, the precise mechanism for their transport remains unclear. Previously, we found that an outwardly directed Cl gradient stimulated the uptake of benzylpenicillin by the isolated choroid plexus under ATP-depleted conditions and suggested the possible involvement of an anionic exchanger in the uptake of organic anions across the brush-border membrane (Suzuki et al., 1987b). Recently, Angeletti et al. (1997) used reverse transcription-polymerase chain reaction (RT-PCR) and in situ hybridization and found that organic anion-transport protein 1 (oatp1), a transporter responsible for the hepatic uptake of organic anions (Meier et al., 1997; Stieger and Meier, 1998), is expressed on the choroid plexus. Fluorescence confocal microscopy after incubation of the choroid plexus with an antibody against oatp1 revealed that the immunoreactive protein is localized on the brush-border membrane (Angeletti et al., 1997). Because it has also been reported that oatp1 possesses anionic exchange activity (Satlin et al., 1997), oatp1 may be responsible for the previously characterized uptake of organic anions in the choroid plexus, although no functional analysis of this transporter has been performed in choroid plexus.

Moreover, the presence of transporters for cellular extrusion on the basolateral membrane would account for the efficient transcellular transport of organic anions across the choroid plexus from CSF to the blood side. Recently, it has been established that multidrug resistance-associated protein (MRP) family members (MRP1/2) are responsible for the cellular extrusion of organic anions (Keppler and König, 1997; Müller and Jansen, 1997; Paulusma and Oude Elferink, 1997; Kusuhara et al., 1998b; Suzuki and Sugiyama, 1998). In particular, MRP2, also referred to as canalicular multispecific organic anion transporter (cMOAT), is localized on the bile canalicular membrane of hepatocytes and plays an important role in the biliary excretion of organic anions (Keppler and König, 1997; Müller and Jansen, 1997; Paulusma and Oude Elferink, 1997; Kusuhara et al., 1998b; Suzuki and Sugiyama, 1998).

The purpose of our study is to examine the possible role of oatp1 and MRP family members in the transcellular transport of organic anions across the choroid plexus. For this purpose, the transport properties of 17-estradiol 17-D-glucuronide (E₂17G), a dual substrate of the transporters for cellular uptake (oatp1) and cellular excretion (MRP1 and cMOAT/MRP2; Keppler and König, 1997; Meier et al., 1997; Stieger and Meier, 1998; Suzuki and Sugiyama, 1998), in the rat choroid plexus were characterized in vivo and in vitro. Moreover, the expression of MRP family members in the choroid plexus was also examined by RT-PCR and Western blot analyses.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. [³H]E₂17G (45 Ci/mmol) and [1-¹⁴C]n-butanol (1.1 mCi/mmol) were purchased from NEN-Du Pont (Boston, MA). Inulin-[¹⁴C]carboxylic acid (4.5 mCi/mmol) and [-³²P]deoxycytidine triphosphate were obtained from Amersham International (Buckinghamshire, UK). Unlabeled E₂17G, benzylpenicillin sodium salt and probenecid were purchased from Sigma Chemical Co. (St. Louis, MO). All other chemicals were commercial products and of analytical grade. Male Sprague-Dawley rats weighing 220 to 240 g were used in all the experiments, which were carried out according to the guidelines provided by the Institutional Animal Care Committee (Graduate School of Pharmaceutical Sciences, The University of Tokyo).

Uptake of E₂17G by Isolated Choroid Plexus. The uptake of [³H]E₂17G by isolated rat choroid plexus was examined with the centrifugal filtration method, which has been described previously in detail (Suzuki et al., 1987b; Ogawa et al., 1994). Rats were decapitated with a guillotine, and the choroid plexus was isolated from the lateral ventricles. The isolated choroid plexus was incubated at 37°C in 150 µl of artificial CSF, which consisted of 122 mM NaCl, 25 mM NaHCO₃, 10 mM glucose, 3 mM KCl, 1.4 mM CaCl₂, 1.2 mM MgSO₄, 0.4 mM K₂HPO₄, and 10 mM N-(2-hydroxyethylpiperazine)-N-(2-ethanesulfonic acid) (pH 7.3), equilibrated with 95% O₂/5% CO₂. After preincubation for 1 min at 37°C, radiolabeled ligands with or without inhibitors were added simultaneously to initiate uptake. The tissue-to-medium concentration ratio of [³H]E₂17G (10 or 20 µM) was calculated with [¹⁴C]butanol as a cell water marker and was corrected for adherent water space as described previously (Suzuki et al., 1987a,b; Ogawa et al., 1994). ³H and ¹⁴C dpm in the specimens were determined in a liquid scintillation spectrophotometer (LSC-3500; Aloka Co., Tokyo, Japan).

Efflux of E₂17G from CSF. The efflux of [³H]E₂17G after intracerebroventricular administration was studied using the method described previously in detail (Ogawa et al., 1994). Rats were anesthetized with ethylcarbamate (1.5 g/kg), and their heads were fixed in a stereotaxic apparatus. An intracerebroventricular dose of [³H]E₂17G (0.383 µCi/rat) and [¹⁴C]inulin (0.02 µCi/rat), dissolved in artificial CSF, was administered into the left lateral ventricle. In some experiments, probenecid (0.1 mg/rat) was also administered simultaneously. At designated time points, aliquots of CSF (50-100 µl) were withdrawn by cisternal puncture, and their radioactivity was measured (Ogawa et al., 1994). The kinetic parameters for the elimination of substrates were determined by the following equation:
where C_CSF(t) is the CSF concentration of substrates at time t, V_CSF is the volume of CSF, and k_elim is the elimination rate constant. The experimental data were fitted to this equation (Ogawa et al., 1994).

RT-PCR Analysis. RT-PCR was used to detect the expression of MRP1. For each experiment, total RNA was isolated from 30 choroid plexuses by a single-step guanidium thiocyanate procedure, whose concentration was estimated by measuring the UV absorption at 260 nm. A cDNA fragment was amplified from this isolated RNA specimen with degenerate primers prepared form the conserved sequence in the ATP-binding cassette region of human MRP1 (Ito et al., 1996). The sequences of the forward and reverse primers were 5'-dGAGAAGGTCGGCATCGTGGG(AGTC)CG(AGTC) AC(AGTC)GG-3' and 5'-dGTCCACGGCTGC(AGTC)GT(AGTC)GC(TC)TC(AG)TC-3', respectively (Ito et al., 1996). PCR was performed at 94°C for 30 s, 37°C for 30 s, and 72°C for 1 min for 40 cycles with Taq polymerase (Takara Shuzo Co. Ltd.). The sequence of the amplified fragments was determined after insertion into pBluescript II S/K () vector (Stratagene, Inc., La Jolla, CA).

Western Blot Analysis. The lysate of the isolated choroid plexus was prepared by sonicating the tissue in artificial CSF. The same concentration of lung homogenate was also prepared. Fifty micrograms of protein of the lysate and/or homogenate were dissolved in 20 µl of 0.25 M Tris-HCl buffer containing 2% SDS, 30% glycerol and 0.01% bromophenol blue (pH 6.8), without boiling, and loaded onto a 7.5% SDS-polyacrylamide gel electrophoresis with a 4.4% stacking gel. Proteins were transferred electrophoretically to nitrocellulose membranes (Immobilon; Millipore, Bedford, MA) with a blotter (Trans-blot; Bio-Rad, Richmond, CA) at 15V for 1 h. The membranes were blocked with Tris-buffered saline containing 0.05% Tween 20 (TBS-T) and 5% BSA for 1 to 2 h at room temperature. After washing with TBS-T (5 min × 3 times), the membranes were incubated with anti-MRP1 monoclonal antibody (50-fold diluted MRPr1 antibody) in TBS-T containing 5% BSA overnight at 4°C and then washed with TBS-T (5 min × 3 times). The membranes were allowed to bind ¹²⁵I-labeled sheep anti-rat immunoglobulin G antibody for MRPr1 in TBS-T containing 5% BSA for 1 h at room temperature, then placed in contact with an imaging plate for a period ranging from 3 h to overnight after washing with TBS-T (5 min × 3 times). The intensity of specific bands was quantified from a standard curve with a BAS 2000 system. The relative induction ratio was defined as the intensity of a specific band of the treated group to that of the control group.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Uptake of E₂17G by Isolated Choroid Plexus. Figure 1 shows the time profile for the uptake of E₂17G by isolated choroid plexus. Analysis revealed that the initial velocity of the uptake was 7.28 µl · min¹ · µl¹ tissue. The tissue-to-medium concentration ratio of E₂17G was more than 18 at 5 min after initiation of the experiment. This uptake was markedly reduced in the presence of carbonylcyanide p-trifluoromethoxyphenyl hydrazone (FCCP) (Fig. 2), suggesting the involvement of an active transport system. In addition, the uptake was inhibited by probenecid, but not by benzylpenicillin, at a concentration of 300 µM (Fig. 2). The uptake of E₂17G consisted of one saturable and one nonsaturable component (Fig. 3). Kinetic analysis gave a K_m of 3.43 ± 0.96 µM and V_max of 7.77 ± 2.40 pmol · min¹ · µl¹ tissue. The clearance for the uptake associated with the nonsaturable component was 3.40 ± 0.17 µl · min¹ · µl¹ tissue. At a tracer concentration, approximately 40% of the accumulation was mediated by the saturable component.

View larger version (8K): [in this window] [in a new window]   Fig. 1.   Time-profile for the uptake of E217G by isolated choroid plexus. Isolated rat choroid plexus was incubated in medium containing 20 µM [3H]E217G to determine the tissue-to-medium concentration ratio (T/M ratio) as a function of time. Each point and vertical bar represents the mean ± S.E. of three independent experiments.

View larger version (24K): [in this window] [in a new window]   Fig. 2.   Effect of inhibitors on the uptake of E217G by isolated choroid plexus. Isolated rat choroid plexus was incubated in medium containing 20 µM [3H]E217G in the presence and absence of inhibitors for 3 min. Each point and vertical bar represents the mean ± S.E. of three independent experiments. *P < .05, **P < .01.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Saturable uptake of E217G by isolated choroid plexus. Isolated rat choroid plexus was incubated in medium containing 10 µM [3H]E217G in the presence and absence of unlabeled E217G for 3 min. The results are shown as an Eadie-Hofstee plot. Each point and vertical bar represents the mean ± S.E. of three independent experiments. The solid line is the regression line. v and s represent the initial velocity of the uptake and substrate concentration in the medium, respectively.

Elimination of E₂17G from CSF after Intracerebroventricular Administration. The transport of E₂17G across the choroidal epithelium was characterized in vivo. Elimination of E₂17G from CSF after intracerebroventricular administration was much faster than that of inulin (Fig. 4). At 20 min after administration, less than 2% of the administered dose of E₂17G remained in the CSF, whereas the corresponding figure for inulin was 40 to 50% (Fig. 4). The analysis demonstrated a V_CSF of 453 ± 6 and 294 ± 39 µl/rat and a k_elim of 0.167 ± 0.001 and 0.0200 ± 0.0071 min¹ for E₂17G and inulin, respectively. The clearance for elimination from CSF, defined as V_CSF × k_elim, was 76 and 5.9 µl · min¹ · rat¹ for E₂17G and inulin, respectively. Simultaneous intracerebroventricular administration of probenecid reduced the elimination of E₂17G to a level comparable with that of inulin. In the presence of probenecid, the concentrations of E₂17G and inulin were 1.57 ± 0.49 and 2.03 ± 0.52% of the dose/µl CSF at 20 min, respectively. At 20 min after administration, the amount of E₂17G associated with choroid plexus was negligible both in the presence and absence of probenecid.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Time profile for the elimination of E217G from CSF after intracerebroventricular administration. Rats were given [3H]E217G (0.383 µCi/rat) and [14C]inulin (0.02 µCi/rat) by intracerebroventricular administration to determine the cisternal CSF concentration of each compound as a function of time. Each point and vertical bar represents the mean ± S.E. of three independent experiments. The solid line is the regression line. , E217G; , inulin.

Expression of MRP1 in Choroid Plexus. To account for the efficient transcellular transport of E₂17G across the choroid plexus, we assumed there were transporters responsible for the cellular expression of organic anions on the basolateral membrane of the choroid plexus. Because MRP family members are responsible for the elimination of organic anions (Keppler and König, 1997; Müller and Jansen, 1997; Paulusma and Oude Elferink, 1997; Kusuhara et al., 1998b; Suzuki and Sugiyama, 1998), the expression of this series of transporters was examined. RT-PCR with degenerate primers designed for the COOH-terminal ATP-binding cassette region of human MRP1 produced amplification of rat MRP1, whose sequence has been reported by Mayer et al. (1995) and Büchler et al. (1996). Analysis of all fourteen clones obtained after transfection of ligated PCR product revealed that only the MRP1 sequence was amplified by PCR. The expression level of MRP1 in the choroid plexus was quantified by RT-PCR and Western blot analysis and compared with that in the lung, which is one of the tissues highly expressing MRP1 (Figs. 5 and 6). The analysis indicated that the expression of MRP1 is approximately 4 or 5 times higher in terms of both mRNA and protein levels (Figs. 5 and 6).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   RT-PCR analysis to determine the expression level of MRP1 in rat choroid plexus. RNA from rat choroid plexus and lung was subjected to RT-PCR to determine the expression level of MRP1. As a control, the expression of GAPDH was also determined. RT-PCR was performed as described in the text. The abscissa represents the amount of RNA. Each point and vertical bar represents the mean ± S.E. of three independent experiments. Closed symbols, MRP1; open symbols, G3PDH; circles, lung; squares, choroid plexus.

View larger version (60K): [in this window] [in a new window]   Fig. 6.   Western blot analysis to determine the expression level of MRP1 in the rat choroid plexus. The lysate of isolated choroid plexus and lung (6.25 and 12.5 µg of protein, respectively) was subjected to Western blot analysis to determine the expression level of MRP1. MRP1 was detected with MRP r1 antibody, which was visualized with 125I-labeled anti-rat serum sheep antibody. CP, choroid plexus.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

The transport properties of E₂17G in the choroid plexus were studied in vivo and in vitro. In vivo, after intracerebroventricular administration, E₂17G was eliminated rapidly from the CSF in a probenecid-sensitive manner (Fig. 4). Because the amount of E₂17G associated with the choroid plexus was negligible, efficient transcellular transport of this ligand across the choroid plexus from CSF to blood was confirmed.

By use of isolated choroid plexus, transport properties across the brush-border membrane were characterized. The involvement of an active transport system in the uptake of E₂17G was demonstrated by the inhibitory effect of FCCP (Fig. 2). Because the uptake was reduced to 25% of the control in the presence of FCCP, more than 75% of the uptake is probably mediated by a specific mechanism (Fig. 2). Therefore, the saturable uptake of E₂17G (Fig. 3) may be mediated by high- and low-affinity transport system. Because of the limited solubility of E₂17G, it is difficult to detect the presence of a low-affinity system. Kinetic analysis of the saturable uptake of E₂17G revealed that the K_m value for the high-affinity system is 3.4 µM (Fig. 3). Because oatp1 (responsible for the Na⁺-independent uptake of E₂17G into hepatocytes) is also located on the brush-border membrane of the choroid plexus (Angeletti et al., 1997) and because the K_m of E₂17G for oatp1 is approximately 3 µM in cRNA-injected oocytes and cDNA-transfected mammalian cells (Meier et al., 1997), oatp1 probably has some functional significance in the uptake of E₂17G into the choroid plexus. Recently, by use of oatp1 cDNA-transfected cells, it was shown that oatp1 mediates the uptake of organic anions in exchange for the efflux of reduced glutathione (Li et al., 1998). Therefore, excretion of reduced glutathione into CSF may be associated with the cellular uptake of organic anions across the brush-border membrane of the choroid plexus.

Uptake of E₂17G into the isolated choroid plexus was further characterized by examining the inhibitory effect of probenecid and benzylpenicillin (Fig. 2). Although it was probenecid sensitive, benzylpenicillin did not affect the uptake even at 300 µM (Fig. 2). Because we previously demonstrated that the K_m of benzylpenicillin is approximately 60 to 100 µM (Suzuki et al., 1987a), the transporter for benzylpenicillin is different from that for E₂17G. Although our previous study suggested that an anionic exchanger at least partly mediates the transport of benzylpenicillin (Suzuki et al., 1987b), the molecular mechanism still remains to be clarified. Other oatp and oat family members may be involved in the transport of benzylpenicillin across the brush border membrane of the choroid plexus (Sekine et al., 1997; Sweet et al., 1997; Abe et al., 1998).

The clearance for the initial uptake of E₂17G (7.28 µl · min¹ · µl¹ tissue) can be extrapolated to give the in vivo elimination clearance by taking into account the amount of choroid plexus (6 µl/rat). This methodology has been justified by our previous findings with a series of -lactam antibiotics, indicating that the kinetic parameters for in vivo elimination from CSF after intracerebroventricular administration and for in vitro uptake are comparable (Ogawa et al., 1994). The calculated clearance for the active transport was 44 µl/min¹/rat¹, which accounted for 58% of the clearance for elimination from CSF (76 µl · min¹ · rat¹; see Results). In contrast, the contribution of the convectional flow of CSF (3 µl · min¹ · rat¹; Ogawa et al., 1994) to the elimination of E₂17G was minor (4%). In addition, clearance for diffusion into the brain parenchyma followed by transport across the blood-brain barrier can be calculated as 29 µl · min¹ · rat¹ (38% of elimination clearance) by subtracting the sum of 44 and 3 µl · min¹ · rat¹ from 76 µl · min¹ · rat¹. Thus, it is possible that E₂17G is also vectorially transported across the blood-brain barrier by a specific mechanism. The presence of such vectorial transport of organic anions across the blood-brain barrier has already been suggested by analyzing the in vivo data. For example, kinetic analysis of the disposition of cefodizime, a third-generation cephalosporin antibiotic, in the central nervous system based on a spatially distributed model demonstrated that the permeability-surface area product of cefodizime across the blood-brain barrier from brain extracellular fluid to blood was greater than that for the opposite direction (Suzuki et al., 1997). Moreover, by analyzing the amount of ligand remaining in the brain after microinjection into the cerebral hemisphere, the presence of a saturable transport system for the efflux of 1-naphthyl glucuronide and p-aminohippuric acid has been suggested (Leininger et al., 1991; Kakee et al., 1997). Probenecid-sensitive efflux of azidothymidine has also been demonstrated in vivo (Dykstra et al., 1993; Takasawa et al., 1997). As a possible candidate for the transporter responsible for brain extrusion of organic anions, we and others have identified the expression of MRP1 on mRNA and protein levels in freshly isolated rat cerebral endothelial cells (Kusuhara et al., 1998a; Regina et al., 1998). In particular, the functional expression of MRP1 on the luminal membrane of mouse cerebral endothelial-derived cell (MBEC4) monolayer has been demonstrated (Homma et al., 1999). In addition, the expression of MRP5 in isolated rat cerebral endothelial cells has been detected (Suzuki et al., 1999).

If we consider the vectorial transport of E₂17G across the choroidal epithelium in vivo (Fig. 4), the presence of a transporter responsible for the cellular extrusion of organic anions on the basolateral membrane can be postulated. Because it has been demonstrated that MRP family members are endowed with this function (Keppler and König, 1997; Müller and Jansen, 1997; Paulusma and Oude Elferink, 1997; Kusuhara et al., 1998b; Suzuki and Sugiyama, 1998), RT-PCR with the degenerate primers has been used to identify the molecular species responsible. RT-PCR resulted in amplification of only the MRP1 cDNA fragment whose sequence has been described by Mayer et al. (1995) and Büchler et al. (1996). RT-PCR and Western blot analyses indicated that the expression of MRP1 in choroid plexus was four or five times higher than that in the lung, one of the tissues highly expressing this transporter (Figs. 5 and 6). This hypothesis is consistent with the previous observation with fluorescence microscopy. Bresler et al. (1979) demonstrated that incubation of isolated rabbit choroid plexus with fluorescein results in the concentrative accumulation of this ligand in the luminal compartment, which is sensitive to ATP depletion and probenecid. Together with the previous demonstration that human MRP1 accepts fluorescein as a substrate (Lautier et al., 1996; Loe et al., 1996), these findings suggest basolateral expression of MRP1. Moreover, Evers et al. (1996) have demonstrated localization of human MRP1 after its cDNA transfection into polarized epithelial cells (MDCK II cells) via immunohistochemical and functional analyses. Because many kinds of organic anions, including clinically important drugs, have been demonstrated to be taken up via oatp1 and extruded via MRP1 (Meier et al., 1997; Stieger and Meier, 1998; Suzuki and Sugiyama, 1998), it is plausible that the low CSF-to-plasma concentration of anionic compounds after systemic administration is accounted for by efficient transcellular transport across the choroid plexus mediated by these two kinds of transporters.

It has also been established that MRP1 and cMOAT/MRP2 are responsible for the cellular extrusion of many glutathione and glucuronide conjugates (Keppler and König, 1997; Meier et al., 1997; Stieger and Meier, 1998; Suzuki and Sugiyama, 1998). In particular, in the liver, the conjugated metabolites are excreted into bile with the aid of cMOAT/MRP2 (Keppler and König, 1997; Meier et al., 1997; Stieger and Meier, 1998; Suzuki and Sugiyama, 1998). Because the oxidative metabolism and subsequent conjugation are referred to as phase 1 and phase 2 detoxification, respectively, this excretion of conjugated metabolites is referred to as phase 3 detoxification (Ishikawa et al., 1997). The excretion process has functional importance for the detoxification of xenobiotics, synergistically with metabolizing enzymes (Ishikawa et al., 1997). Because it has been demonstrated that the UDP-glucuronosyltransferase activity in choroid plexus toward 1-naphthol and 4-methylumbelliferone is much higher than that in liver (Ghersi-Egea et al., 1994), it is plausible that MRP1 expressed on the basolateral membrane of the choroid plexus also plays an important role in restricting CSF entry of xenobiotics after conjugation. Several endogenous substrates that possess a glutathione moiety (such as leukotriene C₄) can also be eliminated from CSF across the choroid plexus by a process mediated by both oatp1 and MRP1 (Spector and Goetzl, 1985, 1986a, b; Li et al., 1998).

In conclusion, E₂17G, a dual substrate for oatp1 and MRP1, was rapidly eliminated from CSF. Functional analysis, together with RT-PCR and Western blot analyses of the isolated choroid plexus, suggests a role for oatp1 and MRP1 in the transcellular transport of organic anions across the choroidal epithelium. Restricted CSF distribution of some organic anions can be accounted for by the action of these transporters. MRP1 may also have functional significance in restricting CSF entry of xenobiotics, together with the high UDP-glucuronosyltransferase activity associated with this epithelium.