Ž .Brain Research 829 1999 143–150

Research report

L-DOPA transport properties in an immortalised cell line of rat capillarycerebral endothelial cells, RBE 4

Pedro Gomes, P. Soares-da-Silva )

Institute of Pharmacology and Therapeutics, Faculty of Medicine, 4200 Porto, Portugal

Accepted 9 March 1999

Abstract

Ž .The present study aimed to determine the kinetics of L-3,4-dihydroxyphenylalanine L-DOPA uptake in an immortalised cell line ofŽ .rat capillary cerebral endothelial cells clones RBE 4 and RBE 4B , to define the type of inhibition produced by L-5-hydroxytryptophan

Ž . Ž . Ž . Ž . Ž .L-5-HTP , 2-aminobicyclo 2,2,1 -heptane-2-carboxylic acid BHC and N- methylamino -isobutyric acid MeAlB and its sodiumŽ .dependence. Non-linear analysis of the saturation curves for L-DOPA and L-5-HTP revealed in RBE 4 cells K values in mM of 72 andm

Ž . Ž102 and in RBE 4B cells K values in mM of 60 and 118, respectively. IC values for L-5-HTP RBE 4, 1026 mM; RBE 4B, 831m 50. Ž .mM obtained in the presence of a nearly saturating 250 mM concentration of L-DOPA were almost 5-fold those obtained when

Ž . Žnon-saturating 25 mM concentrations of L-DOPA were used. IC values for BHC obtained in the presence of a nearly saturating 25050. Ž .mM concentration of L-DOPA were also 6- to 5-fold those obtained when non-saturating 25 mM concentrations of L-DOPA were used.

Ž .MeAlB up to 2.5 mM was found not to interfere with the uptake of L-DOPA. In RBE 4 cells, V values for L-DOPA uptake weremaxŽ .identical in the absence and the presence of 150 mM L-5-HTP or 150 mM BHC, but K values mM were significantly greaterm

Ž .P-0.05 when L-DOPA uptake was studied in the presence of L-5-HTP or BHC. Similar findings were observed when RBE 4B cellsŽ .were used. Uptake of 250 mM L-DOPA in the absence of sodium in the incubation medium was similar to that observed in the presence

Ž .of increasing concentrations of sodium 20 to 140 mM . It is concluded that RBE 4 and RBE 4B cells are endowed with the L-type aminoacid transporter through which L-DOPA and L-5-HTP can be taken up, and suggested that this immortalised cell line of rat capillarycerebral endothelium might constitute an interesting in vitro model for the study of BBB mechanisms, namely those concerning solute andnutrient transfer across the brain capillary endothelium. q 1999 Elsevier Science B.V. All rights reserved.

Keywords: Brain capillary endothelial cell; L-DOPA; L-5-HTP; Cellular transport

1. Introduction

Therapy in Parkinson’s disease with L-3,4-dihydroxy-Ž .phenylalanine L-DOPA , a direct precursor amino acid of

dopamine, has been in the past decades the ultimate strat-egy to activate the dopaminergic systems in brain. Asimilar approach has been also tempted with L-5-hydroxy-

Ž .tryptophan L-5-HTP in order to improve brain 5-hydroxy-tryptaminergic transmission in neurodegenerative disordersaffecting 5-hydroxytryptamine containing neurones. Thesupply to the brain of immediate precursor amino acids toincrease the amine levels in brain is justified by the factthat neither dopamine nor 5-hydroxytryptamine gets intothe brain across the BBB.

) Corresponding author. Fax: q 351-2-5502402; E-mail:patricio.soares@mail.telepac.pt

In the normal brain, the concentrations of these amineswere demonstrated to be dependent on their precursor

w xamino acids levels 8 . These precursors, namely L-tyrosineand L-tryptophan, are supplied to the brain from theirplasma pool through an active transport system, located atthe BBB. This transport system is always occupied with

Ž .plasma large neutral amino acids LNAAs , which consistof phenylalanine, tyrosine, tryptophan, valine, leucine andisoleucine. Under physiological conditions, therefore,LNAAs compete with each other for transport into the

w xbrain 15,16 . Similarly, the availability of L-DOPA tobrain tissue can be compromised by the presence of high

w x Žlevels of LNAAs in plasma 1 , because they LNNAs and.L-DOPA are believed to share the same transporters to

w xcross the BBB 1,9,14,27 . LNAAs are transported at thelevel of the BBB by the L-type amino acid transporter,which is a sodium-independent and 2-aminobicyclo-Ž . Ž .2,2,1 -heptane-2-carboxylic acid BHC sensitive trans-

0006-8993r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.Ž .PII: S0006-8993 99 01387-6

( )P. Gomes, P. Soares-da-SilÕarBrain Research 829 1999 143–150144

porter. The A-type amino acid transporter has also beenshown to transport phenylalanine, but this transporter ismainly involved in the transfer of small non-essentialamino acids such as alanine and glycine. Differentiationbetween A- and L-type is based on sodium-dependence

Ž .and sensitivity to N- methylamino -isobutyric acidŽ .MeAlB and insensitivity to BHC. The ASC-type aminoacid transporter is a sodium dependent transporter and

w xinsensitive to both BHC and MeAlB 2,20,26 .Although the importance of transport across the BBB is

clear, the available information regarding the transportsystems involved and the function of this barrier is smallcompared with that of the intestine and the renal epithelia.The reason for this lack of knowledge derives mainly frominadequate techniques available to study transport acrossthe BBB. Most of the transport studies have been per-formed using animal models in vivo, which presents limi-tations for the detailed investigation of the BBB at thecellular level. The use of brain capillary endothelial cellsŽ .BCECs in culture have shown that BCECs can rapidlyloose key phenotypic markers of the BBB and undergocellular senescence after a limited number of passages. Inthe present study, we have attempted to define the kineticsof L-DOPA uptake in an immortalised cell line of rat

Žcapillary cerebral endothelial cells clones RBE 4 and RBE.4B and to define the type of inhibition produced by

L-5-HTP, BHC and MeAlB and its sodium dependence.This cell line was obtained by transfection of rat brainmicrovessel endothelial cells with a plasmid containing theE1A adenovirus gene. These cells display a non-trans-formed endothelial phenotype expressing the brain mi-crovessel-associated enzymes g-glutamyl-transpeptidase,alkaline phosphatase, P-glycoprotein and inducible nitric

w xoxide synthase 3,7,12,13,19 and the GLUT1 isoformw xresponsible for glucose transport 18 .

2. Materials and methods

2.1. Cell culture

RBE 4 and RBE 4B clones were kindly supplied by Dr.ŽFrancoise Roux INSERM U. 26, Hopital Fernand Widal,ˆ

.Paris, France and maintained in a humidified atmosphereŽ .of 5% CO –95% air at 378C. RBE 4 passages 25–29 and2

Ž .RBE 4B cells passages 2–6 were grown in MinimumŽ . ŽEssential MediumrHam’s F10 1:1 Sigma, St. Louis,

. y1MO, USA supplemented with 300 ng ml neomycine,Ž . y110% fetal bovine serum Sigma , 1 ng ml basic fibrob-

last growth factor, 100 U mly1 penicillin G, 0.25 mg mly1

y1 Ž .amphotericin B, 100 mg ml streptomycin Sigma and25 mM N-2-hydroxyethylpiperazine-NX-2-ethanosulfonic

Ž .acid HEPES; Sigma . The cell medium was changedevery 2 days, and the cells reached confluence after 3–4days of culture. For subculturing, the cells were dissoci-

Ž .ated with 0.05% trypsin–EDTA Sigma , diluted 1:7 andsubcultured in petri dishes with a 21-cm2 growth areaŽ .Costar, Badhoevedorp, The Netherlands . For uptake stud-ies, the cells were seeded in collagen-treated 24-well plas-

Ž .tic culture clusters internal diameter 16 mm, Costar at ay1 Ž 4 y2 .density of 40,000 cells well 2.0=10 cells cm . For

24 h prior to each experiment, the cell medium was free offetal bovine serum and basic fibroblast growth factor.Experiments were generally performed 2–3 days after cellsreached confluence and 6 days after initial seeding andeach cm2 contained about 50 mg of cell protein.

2.2. Transport studies

On the day of the experiment, the growth medium wasaspirated and the cells were washed with Hanks’ mediumat 48C; thereafter, the cell monolayers were preincubatedfor 30 min in Hanks’ medium at 378C. The Hanks’ medium

Ž .had the following composition mM : NaCl 137, KCl 5,MgSO 0.8, Na HPO 0.33, KH PO 0.44, CaCl 0.25,4 2 4 2 4 2

MgCl 1.0, Tris–HCl 0.15 and sodium butyrate 1.0, pHs2

7.4. In experiments performed in the presence of differentconcentrations of sodium, NaCl was replaced by anequimolar concentration of choline chloride. The incuba-

Ž .tion medium also contained benserazide 50 mM , pargy-Ž . Ž .line 100 mM and tolcapone 1 mM in order to inhibit the

enzymes aromatic L-amino acid decarboxylase, monoamineoxidase and catechol-O-methyltransferase, respectively.During preincubation and incubation, the cells were con-tinuously shaken and maintained at 378C. Apical uptakewas initiated by the addition of 2 ml Hanks’ medium with

Ža given concentration of the substrate L-DOPA or L-5-.HTP . Time course studies were performed in experiments

in which cells were incubated with 1 mM substrate for 1,3, 6 and 12 min. Saturation experiments were performed incells incubated for 6 min with increasing concentrations of

Ž .L-DOPA 10 to 1000 mM . In experiments designed tostudy the effects of L-5-HTP, BHC and MeAlB upon theuptake of L-DOPA, RBE 4 and RBE 4B cells were prein-cubated for 30 min in the presence of the compounds to betested. After preincubation, cells were incubated for 6 minin Hanks’ medium with 25 or 250 mM L-DOPA. Uptakewas terminated by the rapid removal of uptake solution bymeans of a vacuum pump connected to a Pasteur pipettefollowed by a rapid wash with cold Hanks’ medium andthe addition of 250 ml of 0.2 mM perchloric acid. Theacidified samples were stored at 48C before injection intothe high pressure liquid chromatograph for the assay ofL-DOPA.

2.3. Assay of L-DOPA

L-DOPA was quantified by means of high pressureliquid chromatography with electrochemical detection, as

w xpreviously reported 22 . The high-pressure liquid chro-

( )P. Gomes, P. Soares-da-SilÕarBrain Research 829 1999 143–150 145

Fig. 1. Time course of L-DOPA and L-5-HTP accumulation in RBE 4 andRBE 4B cells. Cells were incubated at 378C with 1 mM of either L-DOPA

Ž y1 .or L-5-HTP. The results reflect levels in pmol mg protein of L-DOPAor L-5-HTP accumulated. Each point represents the mean of four experi-ments per group; vertical lines show S.E.M.

Žmatograph system consisted of a pump Gilson model 302;.Gilson Medical Electronics, Villiers le Bel, France con-

Ž .nected to a manometric module Gilson model 802 C andŽa stainless-steel 5-mm ODS column Biophase; Bioanalyti-

.cal Systems, West Lafayette, IN of 25 cm length; sampleswere injected by means of an automatic sample injectorŽ . ŽGilson model 231 connected to a Gilson dilutor model

.401 . The mobile phase was a degassed solution of citricŽ . Ž .acid 0.1 mM , sodium octylsulphate 0.5 mM , sodiumŽ . Ž . Ž .acetate 0.1 M , EDTA 0.17 mM , dibutylamine 1 mM

Ž .and methanol 8% vrv , adjusted to pH 3.5 with perchlo-

Ž . y1ric acid 2 M and pumped at a rate of 1.0 ml min .Detection was carried out electrochemically with a glassycarbon electrode, an AgrAgCl reference electrode and an

Ž .amperometric detector Gilson model 141 ; the detectorcell was operated at 0.75 V. The current produced wasmonitored using the Gilson 712 HPLC software. The lowerlimits for detection of L-DOPA, L-5-HTP, dopamine and5-hydroxytryptamine ranged from 250 to 500 fmol.

2.4. Protein assay

The protein content of monolayers of RBE 4 and RBEw x4B cells was determined by the method of Bradford 4 ,

with human serum albumin as a standard.

2.5. Cell water content

Cell water content was simultaneously measured inw14 xparallel experiments using C inulin as extracellular

marker and tritiated water as total water marker. Intracellu-lar water, obtained by subtracting extracellular water fromtotal water, was expressed as microliters of cell water permilligram of protein. Subsequently, the cells were solubi-

Žlized by 0.1% vrv Triton X-100 dissolved in 5 mM.Tris–HCl, pH 7.4 and radioactivity was measured by

liquid scintillation counting.

2.6. Cell Õiability

ŽCells were preincubated for 30 min in the presence of.L-5-HTP, BHC, MeAlB at 378C and then incubated in the

absence or the presence of L-DOPA for a further 6 min.Subsequently, the cells were incubated at 378C for 2 min

Ž .with trypan blue 0.2% wrv in phosphate buffer. Incuba-tion was stopped by rinsing the cells twice with Hanks’medium and the cells were examined using a Leica micro-scope. Under these conditions, more than 95% of the cellsexcluded the dye.

2.7. Data analysis

K and V values for the uptake of L-DOPA, asm max

determined in saturation experiments, were calculated bynon-linear regression analysis, using the GraphPad Prism

w xstatistics software package 11 . For the calculation of the

Table 1Ž y1 .Mean intracellular concentration mmol l of L-DOPA and L-5-HTP in

cultured REB 4 and RBE 4B cells at 6 min incubation in the presence ofŽ .non-saturating concentration of substrate 1 mM

y1 y1Ž . Ž .L-DOPA mmol l L-5-HTP mmol l

RBE 4 cells 39.5"0.9 27.0"1.0RBE 4B cells 71.0"1.6 52.5"1.8

( )P. Gomes, P. Soares-da-SilÕarBrain Research 829 1999 143–150146

Fig. 2. Accumulation of L-DOPA and L-5-HTP in RBE 4 and RBE 4Bcells. Cells were incubated for 6 min at 378C and increasing concentra-

Ž .tions 10 to 1000 mM of the substrate were applied from the apicalŽ y1 y1.border. The results reflect levels in pmol mg protein 6 min of

accumulated L-DOPA or L-5-HTP. Each point represents the mean of fourexperiments per group; vertical lines show S.E.M.

IC ’s, the parameters of the equation for one site inhibi-50w xtion were fitted to the experimental data 11 . K ’s wereiw xcalculated as defined by Cheng and Prusoff 6 for compet-

itive inhibition. Arithmetic means are given with S.E.M. orgeometric means with 95% confidence values. Statisticalanalysis was done with a One-way analysis of varianceŽ .ANOVA followed by Newman–Keuls test for multiplecomparisons. A P-value less than 0.05 was assumed todenote a significant difference.

2.8. Drugs

Ž . Ž .Drugs used were: BHC Sigma , benserazide Sigma ,Ž .L-b-3,4-dihydroxyphenylalanine Sigma , L-5-HTP

Ž . Ž . Ž .Sigma , MeAlB Sigma , pargyline hydrochloride Sigma ,Žtolcapone kindly donated by late Professor Mose Da´

.Prada, Hoffman La Roche, Basle, Switzerland .

3. Results

To determine the rate constant of uptake, cells wereincubated with 1 mM L-DOPA or 1 mM L-5-HTP for 1, 3,

ŽFig. 3. Inhibition curve of L-DOPA uptake closed squares, 25 mM; open.squares, 250 mM by L-5-HTP in RBE 4 and RBE 4B cells. Symbols

represent means of four experiments per group; vertical lines showS.E.M.

( )P. Gomes, P. Soares-da-SilÕarBrain Research 829 1999 143–150 147

Ž .6 and 12 min Fig. 1 . The accumulation of L-DOPA andL-5-HTP increased linearly with time for several minutes.Though RBE 4B cells accumulate more L-DOPA and

ŽL-5-HTP than RBE 4 cells, the rate of accumulation iny1 y1.pmol mg protein min measured over 12 min was

Ž .identical RBE 4, 12.7"1.7; RBE 4B, 9.4"4.1 . Theintracellular water content of cell monolayers was 7.0"0.7

y1 Ž .ml mg protein ns5 . The intracellular L-DOPA andŽ .L-5-HTP concentration at initial rate of uptake 6 min was

27 to 71 times high that which could be expected bypassive equilibration of the substrate. In fact, at 6 minincubation, mean intracellular concentration of L-DOPA

Ž .and L-5-HTP see Table 1 was several times larger thanŽthe substrate concentration in the incubation medium 1

.mM .Thus, in all subsequent experiments designed to deter-

mine the kinetic parameters of L-DOPA and L-5-HTPuptake, the cells were incubated for 6 min with increasing

Ž .concentrations 10 to 1000 mM of the substrate. Theaccumulation of both L-DOPA and L-5-HTP was found tobe dependent on the concentration used and to be saturable

Ž .at nearly 250 mM Fig. 2 . Accumulated L-DOPA orL-5-HTP were found not to be decarboxylated to theircorresponding amines, dopamine and 5-hydroxytryp-tamine; levels of the two monoamines were below the

Ž .detection limit 250 fmol of the HPLC-EC method used todetect L-DOPA and L-5-HTP. Non-linear analysis of thesaturation curves for L-DOPA and L-5-HTP revealed in

Ž . ŽRBE 4 cells K values in mM of 72 53, 91; 95%m. Ž .confidence values and 102 75, 129 and in RBE 4B cells

Ž . Ž . Ž .K values in mM of 60 46, 74 and 118 84, 152 ,m

respectively.Fig. 3 shows inhibition curves for L-5-HTP obtained in

Ž .the presence of a non-saturating 25 mM and a nearlyŽ .saturating 250 mM concentration of L-DOPA. The inhibi-

tion curve obtained with a saturating concentration ofL-DOPA was shifted to the right from that obtained with25 mM L-DOPA and IC values were almost 5-fold those50

obtained when non-saturating concentrations of the sub-

Table 2IC and K values for inhibition of L-DOPA uptake by L-5-HTP and50 i

Ž .BHC, determined in the presence of non-saturating 25 mM and saturat-Ž .ing 250 mM concentrations of the substrate in cultured REB 4 and RBE

4B cells

Ž . Ž .Inhibitor Substrate IC mM K mM50 i

RBE 4 cellsŽ . Ž .L-5-HTP L-DOPA 25 mM 207 166, 258 153 123, 191Ž . Ž .L-DOPA 250 mM 1026 644, 1636 229 144, 366Ž . Ž .BHC L-DOPA 25 mM 157 146, 169 117 108, 126Ž . Ž .L-DOPA 250 mM 1177 944, 1469 263 211, 329

RBE 4B cellsŽ . Ž .L-5-HTP L-DOPA 25 mM 208 134, 324 155 99, 241Ž . Ž .L-DOPA 250 mM 831 787, 878 186 176, 196Ž . Ž .BHC L-DOPA 25 mM 125 101, 155 88 71, 110Ž . Ž .L-DOPA 250 mM 799 678, 942 155 131, 182

ŽFig. 4. Inhibition curve of L-DOPA uptake closed squares, 25 mM; open.squares, 250 mM by BHC in RBE 4 and RBE 4B cells. Symbols

represent means of four experiments per group; vertical lines showS.E.M.

Ž .strate were used see Table 2 . According to Cheng andw xPrusoff 6 , this is the profile for a competitive type of

inhibition. For the calculation of K values, we then usedi

the expression

K s IC r 1qSrKŽ .i 50 m

where S is the concentration of the substrate and K them

Michaelis–Menten constant, which applies for competitiveinhibitors. Table 2 shows IC and K values for L-5-HTP50 i

and it is interesting to note that for non-saturating concen-trations of the substrate IC values equalled K values,50 i

( )P. Gomes, P. Soares-da-SilÕarBrain Research 829 1999 143–150148

Ž .Fig. 5. Inhibition curve of L-DOPA uptake 250 mM by MeAIB in RBE4 and RBE 4B cells. Symbols represent means of four experiments pergroup; vertical lines show S.E.M.

but for the nearly saturating concentration of the inhibitorIC values were greater than K values.50 i

Fig. 4 shows inhibition curves for BHC obtained in thepresence of 25 and 250 mM L-DOPA. BHC was found toexert a marked inhibitory effect upon L-DOPA uptake inboth RBE 4 and RBE 4B cells. As found for L-5-HTP,inhibition curves obtained for BHC with a saturating con-centration of L-DOPA were shifted to the right from thatobtained with 25 mM L-DOPA and IC values were 6- to50

5-fold those obtained when non-saturating concentrationsŽ .of the substrate were used see Table 2 . As observed for

L-5-HTP, IC values for BHC equalled K values when50 i

non-saturating concentrations of the substrate were used,but for the nearly saturating concentration of the inhibitor

Ž .IC values were greater than K values see Table 2 . In50 i

contrast to that observed with BHC, MeAIB, the inhibitorof the A-type system for LNAAs transport, was found notto change uptake of nearly saturating concentrations of

Ž .L-DOPA in both RBE 4 and RBE 4B cells Fig. 5 .Table 3 shows K and V values for saturationm max

curves for the uptake of L-DOPA in the absence and the

Table 3Ž . Ž y1 y1.K in mM and V in nmol mg protein 6 min values form max

L-DOPA uptake in RBE 4 and RBE 4B cells under control conditions andin the presence of L-5-HTP and BHC

Ž .K mM Vm max

RBE 4 cellsŽ .Control 91 68, 115 21.4"0.56Ž .L-5-HTP 272 248, 296 25.8"0.35Ž .BHC 208 200, 216 19.3"0.10

RBE 4B cellsŽ .Control 56 37, 75 16.68"0.50Ž .L-5-HTP 204 185, 222 19.46"0.24Ž .BHC 254 206, 302 22.23"0.25

presence of a concentration of L-5-HTP or BHC equal toŽ .the K value 150 mM , as determined in inhibition stud-i

ies. As can be observed in this table, V values formax

L-DOPA uptake were identical in the absence and thepresence of L-5-HTP or BHC. By contrast, K values form

L-DOPA uptake in the presence of 150 mM L-5-HTP or150 mM BHC were 2- to 5-fold those observed in controlconditions. In this respect it is interesting to note that RBE4B cells, which were more sensitive to L-5-HTP and BHCin inhibition studies, were also found to present the highestincrease in K values when saturation curves for L-DOPAm

ŽFig. 6. Effect of increasing medium concentrations of sodium 0, 20, 40,. Ž .60, 120 and 140 mM on the uptake of L-DOPA 250 mM in RBE 4 and

RBE 4B cells. Bars represent means of four experiments per group;vertical lines show S.E.M.

( )P. Gomes, P. Soares-da-SilÕarBrain Research 829 1999 143–150 149

uptake were performed in the presence of L-5-HTP orBHC.

The final set of experiments was aimed to test thesodium dependence of L-DOPA uptake in RBE 4 and RBE4B cells. As shown in Fig. 6, uptake of a nearly saturating

Ž .concentration 250 mM of L-DOPA in the absence ofsodium in the incubation medium was similar to thatobserved in the presence of increasing concentrations of

Ž .sodium 20 to 140 mM .

4. Discussion

The results presented here show that RBE 4 and RBE4B cells transport quite efficiently L-DOPA and L-5-HTP

Ž .through the apical luminal cell border and several find-ings demonstrate that this uptake process is a facilitatedmechanism. Firstly, steady-state uptake of non-saturatingconcentrations of L-DOPA and L-5-HTP showed a lineardependence on incubation time. Secondly, at an initial rate

Ž .of uptake 6 min incubation the cellular transport ofL-DOPA and L-5-HTP showed a curvilinear dependence onsubstrate medium concentration, suggesting that the uptakewas saturable. Thirdly, L-5-HTP and BHC markedly inhib-ited the uptake of L-DOPA. The efficiency of the L-DOPAtransport in RBE 4 and RBE 4B cells can be also evi-denced by the ratio of L-DOPA concentration in cellularwater to medium concentration. The intracellular L-DOPA

Ž .and L-5-HTP concentration at initial rate of uptake 6 minŽ .was several times 27 to 71 greater than that which could

be expected by passive equilibration of L-DOPA.Several findings suggest that the type of inhibition by

L-5-HTP and BHC on L-DOPA uptake is of the competi-tive type. Firstly, IC ’s for L-5-HTP and BHC were50

significantly greater when inhibition curves were recordedin the presence of a nearly saturating concentration of thesubstrate. This is in full agreement with the finding thatonly K values, but not V values, for L-DOPA uptakem max

were changed when saturation experiments were per-formed in the presence of a concentration of the inhibitorequal to K values. The most likely explanation for thei

competitive inhibition produced by L-5-HTP upon L-DOPAŽ .uptake is that both compounds share the same transporter s

for uptake, which is probably due to their structural simi-larities. Rat renal tubular epithelial cells and OK cells inculture have also been reported to take up L-5-HTP andL-DOPA with similar kinetics and L-DOPA has been shown

w xto inhibit the uptake of L-5-HTP 10,17 .The sensitivity of L-DOPA uptake to BHC, but not to

MeAIB, supports the view that L-DOPA inward transfer inRBE 4 and RBE 4B cells is promoted neither by the A-nor the ASC-type amino acid transporter, but most proba-bly by the L-type amino acid transporter. Another evidencethat fits well this suggestion is that L-DOPA uptake inRBE 4 and RBE 4B cells is not dependent on sodium. The

L-type, for leucine preferring, amino acid transporter isfacilitative, sodium independent, and blocked by BHCw x2,15,20,26 . This suggestion also agrees with data ob-tained under in vivo experimental conditions where it isshown that L-DOPA transport across the rat BBB is

w xmarkedly inhibited by BHC, but not by MeAIB 26 . Onthe other hand, this may suggest that these cells lack theA-type amino acid transporter, which is also in agreementdata obtained for the BBB under in vivo experimental

ww x xconditions 26 and references within . This would alsow xagree with the observations of Audus and Borchardt 2

while showing in bovine brain microvessel endothelial cellmonolayers that leucine uptake is markedly inhibited byL-DOPA, but not by MeAIB. Another point suggesting thatL-DOPA in RBE 4 and RBE 4B cells is transportedthrough the L-type amino acid transporter concerns thesimilarity of K values for L-DOPA uptake and Km i

values for BHC when acting as an inhibitor for L-DOPAuptake. In this respect, it is interesting to note that theMichaelis-Menten constant for BHC transport across the

Ž .BBB K s0.16 mM under in vivo experimental condi-m

tions has been reported to be slightly lower than that forŽ . w xL-DOPA K s0.34 mM 25 , but similar to that de-m

scribed here in RBE 4 and RBE 4B cells. K values form

leucine uptake in bovine brain microvessel endothelial cellŽ . w xmonolayers 0.18 mM has also been shown 2 similar to

Ž .that observed in RBE 4 cells 72 to 91 mM . The slightlylower K values for L-DOPA and L-5-HTP uptake in RBEm

4B cells vs. RBE 4 cells may not have functional rele-vance, but most probably is related to the greater capacityof this clone to accumulate both substrates when non-

Ž .saturating concentrations are applied see Fig. 1 . How-ever, similarities of K values for L-DOPA uptake inm

RBE 4 and RBE 4B cells and those obtained in rat renalŽ .tubular epithelial cells 135 to 216 mM , LLC-PK cells1

Ž . Ž . Ž123 mM , OK cells 14 to 129 mM and Caco-2 cells 60. w xmM 10,17,21–24 , strongly suggest that the L-DOPA

transporter in BCECs may be similar to that in renal andintestinal epithelial cells. Another type of evidence sug-gesting that L-DOPA uptake at the level of the BBB maybe promoted through the L-type amino acid transporter isthat administration of phenylalanine, a high-protein mealor the intravenous infusion of LNAAs reduces the trans-

w xport of L-DOPA into the brain 1,9,14,27 .The data presented here also give the opportunity to

address the question concerning the relationship betweenIC ’s and K ’s and their utility in the study of a given50 i

inhibitor. It is classically accepted that a competitive in-hibitor does not change V values and increases K ’s,max m

whereas for a competitive inhibitor IC ’s will be greater50

than K ’s when the concentration of the substrate is reach-i

ing saturation. IC ’s will equal K ’s when non-saturating50 i

concentrations of the substrate are used. This is valid for acompetitive inhibitor, but not for non-competitive in-hibitors. For non-competitive inhibition, the affinity of thesubstrate for the transporter or the enzyme is not altered

( )P. Gomes, P. Soares-da-SilÕarBrain Research 829 1999 143–150150

Ž .K ’s do not change , but the V ’s are reduced. In thism max

particular case, K ’s will equal IC ’s, despite the concen-i 50

tration of the substrate. One might conclude, therefore, forthe sake of simplification of the analysis of the inhibitorybehaviour of a given compound that by measuring inhibi-tion curves at non-saturating and saturating concentrationsof the substrate one will obtain the very same type ofinformation which can be obtained by measuring satura-tion curves in the absence and the presence of the in-hibitor. The example reported here also provides someinformation on this and shows that a concentration of theinhibitor equal to the corresponding K value is enough toi

produce the expected changes in K , if the compound ism

behaving as a competitive inhibitor. This line of reasoningdoes not apply to other type of inhibitors, such as tight-bi-

w xnding and slow-binding inhibitors 5 .It is concluded that RBE 4 and RBE 4B cells are

endowed with the L-type amino acid transporter throughwhich L-DOPA and L-5-HTP can be taken up, and sug-gested that this immortalised cell line of rat capillarycerebral endothelium might constitute an interesting invitro model for the study of BBB mechanisms, namelythose concerning solute and nutrient transfer across thebrain capillary endothelium.

Acknowledgements

The present study was supported by grant SAV 123r96.

References

w x1 G.M. Alexander, R.J. Schwartzman, J.R. Grothusen, S.W. Gordon,Effect of plasma levels of large neutral amino acids and degree ofparkinsonism on the blood-to-brain transport of L-DOPA in naive

Ž .and MPTP Parkinsonian monkeys, Neurology 44 1994 1491–1499.w x2 K.L. Audus, R.T. Borchardt, Characteristics of the large neutral

amino acid transport system of bovine brain microvessel endothelialŽ .cell monolayers, J. Neurochem. 47 1986 484–488.

w x3 D.J. Begley, D. Lechardeur, Z.D. Chen, C. Rollinson, M. Bardoul,F. Roux, D. Scherman, N.J. Abbott, Functional expression of P-glycoprotein in an immortalised cell line of rat brain endothelial

Ž .cells, RBE 4, J. Neurochem. 67 1996 988–995.w x4 M.M. Bradford, A rapid method for the quantitation of microgram

quantities of protein utilizing the principle of protein-dye binding,Ž .Anal. Biochem. 72 1976 248–254.

w x5 S. Cha, Tight binding inhibitors: I. Kinetic behavior, Biochem.Ž .Pharmacol. 24 1975 2177–2185.

w x6 Y.-C. Cheng, W.H. Prusoff, Relationship between the inhibitionŽ .constant K and the concentration of the inhibitor which causesi

Ž .50% inhibition IC of an enzymatic reaction, Biochem. Pharma-50Ž .col. 22 1973 3099–3108.

w x7 B. El Hafny, O. Chappey, M. Piciotti, M. Debray, B. Boval, F.Roux, Modulation of P-glycoprotein activity by glial factors andretinoic acid in an immortalized rat brain microvessel endothelial

Ž .cell line, Neurosci. Lett. 236 1997 107–111.

w x8 J.D. Fernstrom, Role of precursor availability in control ofŽ .monoamine biosynthesis in brain, Physiol. Rev. 63 1983 484–546.

w x9 J.P. Frankel, P.A. Kempster, M. Bovingdon, R. Webster, A.J. Lees,G.M. Stern, The effects of oral protein on the absorption of intraduo-denal L-DOPA and motor performance, J. Neurol. Neurosurg. Psy-

Ž .chiatry 52 1989 1063–1067.w x10 P. Gomes, P. Serrao, M.A. Vieira-Coelho, P. Soares-da-Silva, Com-˜

petitive and non-competitive inhibition of L-3,4-dihydroxyphenyl-alanine uptake in Opossum kidney cells, Eur. J. Pharmacol. 332Ž .1997 219–225.

w x Ž11 H.J. Motulsky, P. Spannard, R. Neubig, GraphPad Prism version.1.0 , GraphPad Prism Software, San Diego, USA, 1994.

w x12 M. Naassila, F. Roux, F. Beauge, M. Daoust, Ethanol potentiateslipopolysaccharide- or interleukin-1 b-induced nitric oxide genera-

Ž .tion in RBE 4 cells, Eur. J. Pharmacol. 313 1996 273–277.w x13 M. Nobles, N.J. Abbott, Modulation of the effects of extracellular

w 2q xATP on Ca i in rat brain microvascular endothelial cells, Eur. J.Ž .Pharmacol. 361 1998 119–127.

w x14 J.G. Nutt, W.R. Woodward, J.H. Carter, T.L. Trotman, Influence offluctuations of plasma large neutral amino acids with normal dietson the clinical response to L-DOPA, J. Neurol. Neurosurg. Psychia-

Ž .try 52 1989 481–487.w x15 W.M. Pardridge, Transport of small molecules through the blood–

brain barrier: biology and methodology, Adv. Drug Delivery Rev. 15Ž .1995 5–36.

w x16 W.M. Pardridge, W.H. Oldendorf, Transport of metabolic substratesŽ .through the blood–brain barrier, J. Neurochem. 28 1977 5–12.

´w x17 P.C. Pinto-do-O, P. Soares-da-Silva, Studies on the pharmacology ofthe inward transport of L-DOPA in rat renal tubules, Br. J. Pharma-

Ž .col. 118 1996 741–747.w x18 A. Regina, F. Roux, P.A. Revest, Glucose transport in immortalized

rat brain capillary endothelial cells in vitro: transport activity andŽ .GLUT1 expression, Biochim. Biophys. Acta 1335 1997 135–143.

w x19 F. Roux, O. Durieu-Trautmann, N. Chaverot, M. Claire, P. Mailly,J.M. Bourre, A.D. Strosberg, P.O. Couraud, Regulation of g-gluta-myl-transpeptidase and alkaline phosphatase activities in immortal-ized rat brain microvessel endothelial cells, J. Cell Physiol. 159Ž .1994 101–113.

w x20 M.M. Sanchez del Pino, R.A. Hawkins, D.R. Peterson, Neutralamino acid transport by the blood–brain barrier. Membrane vesicle

Ž .studies, J. Biol. Chem. 267 1992 25951–25957.´w x21 P. Soares-da-Silva, M.H. Fernandes, P.C. Pinto-do-O, Cell inward

transport of L-DOPA and 3-O-methyl-L-DOPA in rat renal tubules,Ž .Br. J. Pharmacol. 112 1994 611–615.

w x22 P. Soares-da-Silva, M.P. Serrao, M.A. Vieira-Coelho, Apical andbasolateral uptake and intracellular fate of dopamine precursor L-

Ž .DOPA in LLC-PK cells, Am. J. Physiol. 274 1998 F243–F251.1w x23 M.A. Vieira-Coelho, P. Soares-da-Silva, Apical and basal uptake of

L-DOPA and L-5-HTP and their corresponding amines, dopamineŽ .and 5-HT, in OK cells, Am. J. Physiol. 272 1997 F632–F639.

w x24 M.A. Vieira-Coelho, P. Soares-da-Silva, Uptake and intracellularfate of L-3,4-dihydroxyphenylalanine, the dopamine precursor, in ahuman intestinal epithelial cells line: Caco-2, Am. J. Physiol. 275Ž .1998 C104–C112.

w x25 L.A. Wade, R. Katzman, 3-O-methyl-DOPA uptake and inhibitionŽ .of L-DOPA at the blood–brain barrier, Life Sci. 17 1975 131–136.

w x26 L.A. Wade, R. Katzman, Synthetic amino acids and the nature ofL-DOPA transport at the blood–brain barrier, J. Neurochem. 25Ž .1975 837–842.

w x27 W.R. Woodward, C.W. Olanow, R.M. Beckner, R.A. Hauser, L.L.Gauger, J.M. Cedarbaum, J.G. Nutt, The effect of L-DOPA infusionswith and without phenylalanine challenges in parkinsonian patients:plasma and ventricular CSF L-DOPA levels and clinical responses,

Ž .Neurology 43 1993 1704–1708.