October 2004 Biol. Pharm. Bull. 27(10) 1489—1496 (2004) 1489 Review New Aspects of the Blood–Brain Barrier Transporters; Its Physiological Roles in the Central Nervous System
Sumio OHTSUKI Department of Molecular Biopharmacy and Genetics, Graduate School of Pharmaceutical Sciences, and New Industry Creation Hatchery Center, Tohoku University; Aoba, Aramaki, Aoba-ku, Sendai 980–8578, Japan: and SORST and CREST, Japan Science and Technology Agency; Japan. Received July 6, 2004
The blood–brain barrier (BBB) segregates the circulating blood from interstitial fluid in the brain, and re- stricts drug permeability into the brain. Our latest studies have revealed that the BBB transporters play impor- tant physiological roles in maintaining the brain milieu. The BBB supplies creatine to the brain for an energy- storing system, and creatine transporter localized at the brain capillary endothelial cells (BCECs) is involved in BBB creatine transport. The BBB is involved in the brain-to-blood efflux transport of the suppressive neuro- transmitter, g-aminobutyric acid, and GAT2/BGT-1 mediates this transport process. BCECs also express sero- tonin and norepinephrine transporters. Organic anion transporter 3 (OAT3) and ASCT2 are localized at the ab- luminal membrane of the BCECs. OAT3 is involved in the brain-to-blood efflux of a dopamine metabolite, a ure- mic toxin and thiopurine nucleobase analogs. ASCT2 plays a role in L-isomer-selective aspartic acid efflux trans- port at the BBB. Dehydroepiandrosterone sulfate and small neutral amino acids undergo brain-to-blood efflux transport mediated by organic anion transporting polypeptide 2 and ATA2, respectively. The BBB transporters are regulated by various factors, ATA2 by osmolarity, taurine transporter by TNF-a, and L-cystine/L-glutamic acid exchange transporter by oxidative stress. Clarifying the physiological roles of BBB transport systems should give us important information allowing the development of better CNS drugs and improving our understanding of the relationship between CNS disorders and BBB function. Key words blood–brain barrier; transporter; brain capillary endothelial cell; creatine; neurotransmitter; neuro-modulator
The blood–brain barrier (BBB) is formed by complex Clarifying the physiological role of the BBB is also neces- tight-junctions of the brain capillary endothelial cells sary to understand the role of the BBB in CNS disorders. (BCECs) and expresses various transport systems. These This is because an alteration in BBB functions may result in characteristics of the BBB make it possible to control selec- a change in CNS conditions leading to CNS disorders. Fur- tive transport across the BBB and to limit the non-selective thermore, analyzing the BBB functions from a different point brain distribution of drugs. Since understanding the mecha- of view could result in the discovery of a new transport sys- nism of BBB transport is important for improving the BBB tem that is involved in drug permeability. In this review, I permeability of CNS-acting drugs, the molecular mechanism shall introduce the recent advances in research into the mole- of the BBB transport has been analyzed mainly with respect cular basis of BBB transport, focusing on its physiological to drug transport. At first, blood-to-brain influx transport was functions. regarded as a system for transporting drugs and nutrients to The BBB Supplies Creation to the Brain The brain the brain. This resulted in the discovery of transport systems consumes 18% of total body energy, while its weight is only for glucose and amino acids.1) Subsequently, the importance 2% of the total body weight. The brain cannot store glucose of efflux transport at the BBB has been recognized after find- for energy synthesis, and the BBB plays a crucial role in glu- ing that P-glycoprotein (P-gp) is involved in the efflux trans- cose supply to the brain by expressing glucose transporter 1 port of xenobiotics.2,3) Due to these studies, the BBB is now (GLUT1) (Fig. 1).4) Besides energy synthesis, an energy- accepted as a complex transport system which is an impor- storing system is necessary for maintain energy homeostasis tant determinant of drug distribution to the brain. in high energy-consuming tissues. We have shown that the In addition to the drug permeability aspects, the physio- BBB also plays an important role in the energy storing logical function of the BBB is important in the central ner- process (Fig. 1).5) vous system (CNS). To date, the BBB transport system has Creatine plays a key role in this energy-storing process. been shown to supply nutrients, such as glucose, lactate, The phosphate-bound energy of ATP is transferred to crea- amino acids and nucleotides, to the brain. However, it is con- tine and stored as phosphocreatine to regenerate ATP from ceivable that the BBB have many other physiological func- ADP (Fig. 1). In the human brain, the creatine level is about tions involving regulation of the CNS milieu. For instance, 180-fold higher than that in plasma.6) Furthermore, the brain the brain is a highly energy-consuming tissue and produces is a key target for creatine deficiency syndrome which in- various metabolites. In particular, metabolism is essential for volves a reduced level of creatine in the body due to impair- the inactivation of neurotransmitters after their secretion ment of creatine synthesis.7) The patients exhibit delayed from the pre-synapses, as well as re-uptake. The metabolites psychomotor development, hypotonia, seizures, and delayed produced in the brain must be removed in order to maintaine myelination. Phosphocreatine is also reduced in the brains of proper neural function. Therefore, the BBB can be thought to mildly demented patients with Alzheimer’s disease.8) There- act as a clearance system to remove various metabolites from fore, the creatine/phosphocreatine shuttle system plays an the CNS. important role in energy homeostasis in the brain to ensure
e-mail: [email protected] © 2004 Pharmaceutical Society of Japan 1490 Vol. 27, No. 10
Fig. 2. Distribution of Creatine into the Brain (A) and Localization of Creatine Transporter (CRT) at the BBB Fig. 1. Network Pathways of Energy Sources and Energy Stores in the (A) Tissue accumulation of [14C]creatine 24 h after intravenous administration CNS (3 mCi/head) in mouse. The apparent tissue-to-plasma concentration ratio (Kp,app) was CK-B, brain-type cytoplasmic creatine kinase; CK-Mi, ubiquitous mitochondrial cre- defined as the amount of [14C]creatine per gram tissue divided by that per milliliter atine kinase; P-creatine, phosphocreatine. Modified from Ohtsuki et al., Farmacia, 39, plasma. Each column represents the mean S.D. value (n 3). (B) Localization of CRT 1057—1061 (2003). in the cerebral cortex of adult mouse brain. Note the intense immunofluorescence on brain capillaries branching in all cortical layers (arrows) with moderate staining in neu- ronal perikarya (arrowheads). Scale bars: 10 mm. Cited from Ohtsuki et al., J. Cereb. Blood Flow Metab., 22, 1327—1335 (2002). proper development and function. Creatine in the body is maintained by biosynthesis mainly in the liver and kidney, and by dietary supplementation. the BBB causes CNS dysfunction due to disruption of energy Therefore, muscle creatine is supplied from the circulating homeostasis. Indeed, GLUT1 mutation has been reported as blood via creatine transporter (CRT).9) In contrast, the crea- GLUT1-deficiency syndrome (GLUT1-DS).14,15) Patients tine supply from the circulating blood to the brain has been with this condition exhibit infantile seizures, delayed devel- reported to be limited, since oral administration of 20 g crea- opment and motility disorders. The mutation in the CRT tine per day for 4 weeks produces only about a 9% increase gene has been also reported in creatine deficiency syndrome in total creatine in the human brain.10) Braissant et al. have as well as mutations in the enzyme involved in creatine also reported that mRNA expression of CRT was not de- biosynthesis.16) In the latter cases, both the brain and plasma tected in brain capillaries, while that of the enzyme involved levels of creatine are reduced, and creatine supplementation in creatine biosynthesis was detected in neural cells.11) improves body creatine levels and symptoms.7) In contrast, in We have been analyzing neurotransmitter transporters by patients with a defect in CRT, the brain levels of creatine are using conditionally immortalized mouse BCECs (TM-BBB), undetectable, while the plasma levels are normal or slightly as an in vitro BBB model.12,13) The CRT cDNA fragment was higher than normal.17) Furthermore, creatine supplementation amplified by RT-PCR from TM-BBB cells possibly due to is not an effective way to improve the brain level of creatine the similarity in the nucleotide sequence between CRT and and patient symptoms. These clinical observations suggest neurotransmitter transporters.5) Furthermore, TM-BBB cells the importance of BBB creatine transport by CRT in CNS exhibited creatine uptake which is significantly inhibited by energy homeostasis. CRT inhibitors, such as b-guanidinopropionate and guanidi- The Brain-to-Blood Efflux Transport of Neurotrans- noacetate. Our immunohistochemical analysis revealed that mitters Neurotransmitter inactivation is a critical process there is intense expression of CRT in mouse BCECs (Fig. 2). for proper neural function. The uptake and metabolism of Furthermore, 24 h after systemic administration, [14C]crea- neurotransmitters by neurons and astrocytes have been well- tine was distributed in the brain, and the apparent tissue-to- characterized. The function of the BBB used to be consid- plasma concentration in the brain was about 30 ml/g tissue ered to be retention of neurotransmitters in the brain. How- (Fig. 2). This value was close to that of endogenous creatine ever, our recent studies have shown that g-aminobutyric acid in the mouse. These results suggest that the BBB has creatine (GABA), a suppressive neurotransmitter, undergoes efflux transport system mediated by CRT which is capable of accu- from the brain across the BBB rather than being retained in mulating creatine in the brain against its concentration gradi- the brain.18) Therefore, the physiological role of the BBB is ent. not merely as a barrier, it also acts as a regulatory interface
The Km value of creatine uptake by TM-BBB cells was for neurotransmitters. found to be 16.2 m M. This suggests that the creatine transport Betaine/GABA transporter-1 (BGT-1; SLC6A12), which at the BBB is saturated by plasma creatine, since the plasma corresponds to GAT2 in the mouse, is involved in the efflux 18) level of creatine is 50—100 m M in the human and 200 m M in transport of GABA at the BBB. TM-BBB cells showed the mouse.9) This saturation of BBB creatine transport ex- [3H]GABA uptake which was Na -, Cl -, and concentration- plains the limited distribution of creatine after its oral admin- dependent with a Km of 679 m M. GAT2/BGT-1 expression in istration. TM-BBB cells and mouse brain capillary-rich fraction has Our report indicated that the BBB plays an important role been detected by RT-PCR and Western blot analysis, while in CNS energy homeostasis not only as a supply pathway for mRNA expression of neuronal and glial GABA transporters the energy source but also as a supply pathway for the energy (GAT1 and 3) was not detected in TM-BBB cells. This result buffer (Fig. 1). This means that a defect in either system at suggests that the GABA transporter expressed at the BBB is October 2004 1491 different from that expressed at neurons and astrocytes. In- deed, the efflux rate of [3H]GABA across the BBB measured by the Brain Efflux Index (BEI) method19) was not inhibited, but increased by nipecotic acid, which is a specific inhibitor of neuronal and glial GABA transporters.20) Immunohisto- chemical analysis of mouse brain revealed that GAT2/BGT-1 is localized at the BCECs.18) Therefore, GAT2/BGT-1 is sug- gested to act as a BBB efflux transporter of GABA (Fig. 3). As far as monoamines are concerned, the uptake of norepi- nephrine and serotonin has been reported using isolated brain capillaries.21,22) TM-BBB cells express norepinephrine trans- porter (NET) and serotonin transporter (SERT). Immunohis- tochemical analysis showed that NET is localized at the ablu- minal membrane of mouse BCECs, and SERT is localized at both the abluminal and luminal membranes (Fig. 3).23) The brain microvasculature is thought to be regulated by monoamines released from adrenergic and serotonergic neu- rons, since adrenergic and serotonergic neurons are in close apposition to the brain capillaries which express adrenergic Fig. 3. The Transporters Involved in the Brain-to-Blood Efflux Transport at the BBB and serotonergic receptors.24—27) The abluminally localized The hatched transporters indicate ATP binding cassette (ABC) transporters. The NET and SERT would function as an inactivation system for transporters in white are unidentified. neurotransmitters around the brain capillaries. Tricyclic antidepressants inhibit the presynapic terminal to excitatory neurotoxicity, while D-Asp needs to be stored as reuptake systems to produce their therapeutic effects by in- a precursor for NMDA. Therefore, the stereo-selective BBB creasing monoamine concentrations in the synaptic clefts. efflux transport appears to have a role in the stereo-selective The tricyclic antidepressants, amitriptyline and imipramine, regulation of amino acids in the brain as far as expression of increase BBB permeability in addition to their main pharma- the distinct functions of L- and D-Asp are concerned. cological effect.28) Hardebo et al. and Spatz et al. have re- Subtypes of excitatory amino acid transporter (EAATs) ported that norepinephrine and serotonin uptake by isolated have been reported to be expressed in isolated bovine brain capillaries is inhibited by these tricyclic antidepres- BCECs.34) Nevertheless, the expression of EAATs at the sants, which inhibit NET and SERT.21,22) Amitriptyline and BBB does not explain the stereoselective BBB efflux trans- imipramine may inhibit the activity of NET and SERT ex- port of Asp, since EAATs transport both Asp isomers.35) We pressed at the BBB, resulting in increased levels of analyzed the stereoselective BBB efflux transport of Asp 36) monoamines around the brain capillaries and altered BBB using TM-BBB cells. TM-BBB cells exhibit L-isomer-se- permeability. lective transport activity, like the in vivo BBB. The L-isomer- The function of luminally localized SERT is still unclear. selective transport by the cells is Na - and pH-dependent, Ganapathy et al. have reported that SERT is localized at the and is inhibited by neutral amino acids, such as L-alanine (L- human placental brush-border membrane facing the maternal Ala) and L-serine (L-Ser). These properties are in good agree- blood.29,30) They have suggested that the clearance of sero- ment with those of the amino acid transport system ASC. To tonin from the intervillous space may be necessary to main- date, ASCT1 and 2 have been identified as members of the tain the uteroplacental blood flow, since serotonin is a potent ASC system.37,38) TM-BBB cells express both ASCT1 and 2 vasoconstrictor.29) Furthermore, in the blood coagulation sys- mRNAs, although the mRNA expression of ASCT2 in the tem, serotonin secreted from platelets enhances blood coagu- cells was found to be 6.7-fold greater than that of ASCT1. 3 lation. Therefore, maintaining a low plasma serotonin level ASCT1- and ASCT2-expressing oocytes exhibited [ H]L-Asp 3 may prevent inappropriate blood coagulation in the cerebral uptake in an Na - and pH-dependent manner, while [ H]D- microvasculature. Based on these pharmacological functions Asp did not undergo such uptake (Fig. 4). Furthermore, im- of serotonin, one possible function of the luminally localized munohistochemical analysis indicated that ASCT2 protein is SERT is clearance of serotonin from the cerebral intravascu- localized on the abluminal membrane of mouse BCECs. lar space to maintain cerebral blood flow. These findings indicate that abluminally localized ASCT2 is The Stereo-Selective Brain-to-Blood Efflux Transport involved in L-isomer-selective Asp efflux transport at the of Acidic Amino Acids Recently, D-amino acids have been BBB (Fig. 3). found to present in mammals. In the brain, D-aspartic acid (D- EAAT1–3 may play a role in the uptake of L-Asp, in addi- Asp) functions as a precursor of NMDA and also influences tion to ASCT2 (Fig. 3), because of the report showing that the secretion of several hormones, such as growth and there is expression of EAATs in the abluminal membrane 31) 34) luteinizing hormone, testosterone, melatonin and oxytocin. fraction of isolated bovine BCECs. ASCT2 transports L- In contrast, L-Asp is an excitatory amino acid as well as L- Asp with a low affinity (Km 9.33 mM), and the Km of L-Asp 32) glutamic acid (L-Glu). We have reported that the BBB can and L-Glu transport by EAATs has been reported to be 2— 35) carry out stereo-selective efflux transport of Asp, transport- 64 m M. TM-BBB cells exhibited L-Asp transport with both 33) ing the L-isomer but not the D-isomer. This L-isomer-selec- low and high affinity, which appears to correspond to the 36) tive Asp efflux transport is consistent with the CNS function transport by ASCT2 and EAATs, respectively. However, D- of each isomer of Asp, since the accumulation of L-Asp leads Asp transport by EAATs at the BBB would be inconsistent 1492 Vol. 27, No. 10
Table1.Effect of Various Compounds on [3H]HVA Uptake by Rat OAT3
Inhibitors No. studied % of controla)
Control 9 100 19 Controlb) 9 100 17 Drugs Probenecid 9 23.4 15.4** PAH620.3 22.1** Indomethacin 10 40.9 27.6** Benzylpenicillin 5 1.01 2.37** Cimetidine 9 24.5 24.3** Baclofen 8 23.2 22.6** Fig. 4. L-Isomer-Selective Aspartic Acid Transport by Mouse ASCT1 (A) 6-Mercaptopurine 7 8.67 14.60** and ASCT2 (B) Valproic acid 7 59.5 15.1 Closed and open columns represent the results for mouse ASCTs-injected oocytes 3 3 Metabolites of monoamine and water-injected oocytes, respectively. [ H]L-Asp (0.455 m M) and [ H]D-Asp neurotransmitters (0.556 m M) uptake with 500 m M of each unlabeled isomer of Asp was determined in the presence or absence of Na at pH 5.5 or 7.4. Each column represents the mean S.E.M. 3,4-Dihydroxyphenylacetic acid 6 50.7 25.6** (n 7—12). Cited from Tetsuka et al., J. Neurochem., 87, 891—901 (2003). Vanillylmandelic acid 6 53.5 21.4* 3,4-Dihydroxymandelic acid 6 62.7 24.7 4-Hydroxy-3-methoxyphenylglycol 8 49.3 34.5** with our previous in vivo study, demonstrating the absence of 5-Hydroxyindoleacetic acid 8 40.5 30.0** 33) brain-to-blood transport of D-Asp. This discrepancy may 5-Methoxytryptophol 7 24.7 19.4** be due to the expression and functional level of each trans- Imidazole-4-acetic acid 6 47.6 24.1** porter at the in vivo BBB and the brain levels of the endoge- 1-Methyl-4-imidazolic acid 7 48.6 35.7** nous substrates. We also cannot rule out the possibility that Sulfate conjugates Estrone sulfate 6 4.35 11.31** the D-Asp efflux transport process may not be present on the Indoxyl sulfate 6 48.9 14.0* luminal side. Therefore, further studies are necessary to clar- Monoamine neurotransmitters ify the contribution of ASCT2 and EAATs and identify the Dopamineb) 7 106 34 luminally localized transporter involved in the efflux trans- Norepinephrineb) 7 106 36 Serotoninb) 8 113 19 port of excitatory amino acids. Histamineb) 6 102 28 The Brain-to-Blood Efflux Transport of Neurotrans- Others mitter Metabolites Homovanillic acid (HVA) is a major Octanoic acid 5 10.6 18.5** metabolite of dopamine. The HVA concentration in blood Quinolinic acid 8 84.6 31.4 and urine is widely used as an indicator of dopaminergic L-Glutamic acid 7 79.7 34.7
3 neuronal activity in the brain. Abnormalities in cerebral The [ H]HVA uptake (200 nM) study was performed at 1 h in the presence or absence dopamine function have been implicated in many mental and of inhibitors. Rat OAT3-mediated transport was obtained by subtracting the transport neurological disorders, such as Parkinsonism and various rate in water-injected oocytes from that in rat OAT3 cRNA-injected oocytes. All inhibitors were used at a concentration of 1 mM. a) Each value represents the 39,40) types of seizures. Furthermore, patients with uremic en- mean S.D. (percent of control). b) The [3H]HVA uptake was investigated using ND96 solution containing 100 m M ascorbic acid, 100 m M pargyline and 10 m M U0521 to cephalopathy or Reye’s syndrome show markedly elevated 41,42) avoid oxidation and metabolism. * p 0.05, significantly different from control. HVA concentrations in the CSF. Therefore, the BBB ef- ** p 0.01, significantly different from control. Cited from Mori et al., J. Cereb. Blood flux transport system for HVA appears to be an important de- Flow Metab., 23, 432—440 (2003). terminant of dopamine turnover in the brain. Previous studies have indicated that the HVA concentra- mouse brain. Thus, abluminally localized OAT3 takes up tion in the brain is increased when probenecid or octanoic HVA from the brain interstitial fluid into BCECs and so acid is administered peripherally.43,44) Brain microdialysis plays a major role in the brain-to-blood efflux transport sys- studies have suggested that the brain-to-blood efflux trans- tem for HVA (Fig. 3). port rate of HVA is faster than the blood-to-brain HVA influx The metabolites of monoamine neurotransmitters include rate across the BBB.45) These results suggest that a many organic anions. We have found that various anionic probenecid- and octanoic acid-sensitive transporter(s) medi- metabolites of neurotransmitters inhibited HVA transport by ates the efflux transport of HVA at the BBB. We have investi- OAT3, while the neurotransmitters themselves had no in- gated the transporter responsible by direct analysis of HVA hibitory effect (Table 1).46) Although compounds that have an efflux using BEI methods.46) [3H]HVA was eliminated from inhibitory effect on a transporter are not always substrates, the brain across the BBB with a half-life of 41.0 min. The ef- the above result raises the possibility that OAT3 mediates the flux rate of [3H]HVA was significantly inhibited by a selec- BBB efflux transport of various neurotransmitter metabolites. tive inhibitor of organic anion transporter 3 (OAT3). Patients with Reye’s syndrome exhibit markedly elevated OAT3 was identified as a brain-expressed member of the HVA concentrations in the CSF, although the dopamine con- OAT family (SLC22A).47) Our group and other researchers centration in the nucleus caudatus is not abnormal.42,52) This have reported that OAT3 is involved in the efflux transport of observation raises the possibility that accumulation of HVA organic anions at the BBB.48—50) Our studies using a Xeno- in the brain may cause encephalopathy. In patients with pus oocyte expression system found that rat and mouse Reye’s syndrome, the octanoic acid concentration in the OAT3 can transport HVA, and this transport activity was in- blood increases markedly with concomitant increases in the hibited by probenecid and octanoic acid (Table 1).46,51) Fur- brain.53) The plasma level of octanoic acid in severe cases of 54) thermore, immunohistochemical analysis revealed that OAT3 Reye’s syndrome reaches about 1 mM, and we have shown is localized on the abluminal membrane of BCECs in rat and that octanoic acid (1 mM) significantly inhibits (by 90%) October 2004 1493 rOAT3-mediated HVA transport (Table 1). Therefore, it was System A is a transport system for small neutral amino proposed that the increased level of octanoic acid in the acids that accepts L-Ala, L-proline (L-Pro), glycine (Gly), and plasma and brain of patients inhibits the brain-to-blood efflux a-methylaminoisobutyric acid (MeAIB) as substrates. Gly transport of HVA mediated by OAT3. modulates the function of NMDA receptors, and L-Pro inter- The BBB efflux transport mediated by OAT3 is a physio- acts with NMDA and AMPA receptors. Furthermore, small logically multi-functional transport system. We have reported neutral amino acids function as osmolytes, and induction of that an endogenous uremic toxin (indoxyl sulfate (IS)) and system A activity by hypertonicity has been reported in a va- drugs (6-mercaptopurine (6-MP), 6-thioguanine (6-TG)) are riety of different cells.67) It has been suggested that system A excreted from the rat brain by OAT3-mediated BBB efflux is present on the abluminal side of the BBB following an up- transport (Fig. 3).50,55) During maintenance chemotherapy for take study using isolated rat brain capillaries and isolated acute lymphoblastic leukemia, CNS relapses often occur due abluminal membrane vesicles from bovine BCECs.68,69) We to penetration and proliferation of leukemic cells in the brain, have used the BEI method to show that brain-to-blood efflux 3 3 because of the limited distribution of thiopurine nucleobase transport of [ H]L-Pro and [ H]Gly takes place across the in analogs, such as 6-MP and 6-TG.56) Microdialysis studies vivo BBB.70) 3 have shown that the brain-to-blood efflux of 6-MP across the [ H]L-Pro uptake by conditionally immortalized rat BBB is 20-fold greater than the brain-to-blood influx.57) BCECs (TR-BBB)71) is an Na -dependent process with high- Thus, the OAT3-mediating BBB efflux transport plays a cru- affinity and low-affinity saturable components (Km 425 m M 70) 3 cial role in limiting the effects of thiopurine nucleobase and 10.8 mM). The mode of inhibition of [ H]L-Pro uptake analogs in the brain. by amino acids is consistent with the involvement of system The involvement of OAT3 in drug efflux at the BBB shows A. Three Na -dependent small neutral amino acid trans- the importance of analyzing human OAT3 as far as drug de- porters have been identified as system A isoforms, namely, velopment is concerned. Human OAT3 has already been ATA1 (GlnT; SAT1; SLC38A1), ATA2 (SAT2; SA1; identified in kidney based on the homology with rat OAT1.58) SLC38A2), and ATA3 (SLC38A4).72—76) All three isoforms Its amino acid sequence is 75% and 73% identical to those of of system A have been found to be expressed in the rat brain rat and mouse OAT3, respectively. Since the function and lo- capillary-rich fraction and TR-BBB cells, and ATA2 mRNA calization of human OAT3 at the BBB have not been ana- was present in 93-fold and 2140-fold greater levels than lyzed yet, further studies are necessary to clarify the role of ATA1 and ATA3 mRNAs in TR-BBB cells, respectively. This human OAT3 at the BBB. result suggests that ATA2 is the responsible transporter for The Brain-to-Blood Efflux Transport of Neuro-Modu- system A efflux transport at the BBB (Fig. 3). lators Neuronal function is regulated by various neuro- Under hypertonic conditions (450 mOsm/kg), ATA2 modulators which must be maintained at low levels for the mRNA in TR-BBB cells is induced by up to 373% concomi- proper function of neuro-modulators. The BBB efflux trans- tantly with activation of [3H]MeAIB uptake.70) This suggests port system is thought to be involved in the clearance of that system A at the BBB is regulated by the osmolarity neuro-modulators, like neurotransmitters and their metabo- around the BBB. Osmo-regulation in the brain is important lites. for maintaining a constant milieu in the CNS, since hyper- Dehydroepiandrosterone sulfate (DHEAS) is a neuroster- tonicity causes brain edema. Therefore, osmo-regulation of oid which can interact with GABA type A receptors and ATA2 may contribute to the regulation of osmolarity in the sigma receptors to increase memory and learning ability and brain and the cell volume in BCECs in order to maintain a to protect neurons against excitatory amino acid-induced stable environment in the brain and to permit the BBB to neurotoxicity.59) We have reported that DHEAS is eliminated function under pathological conditions. from the brain across the BBB by using the BEI method, and Regulation of BBB Functions by CNS Conditions To an inhibition study showed that oatp2 was involved in this ef- control the CNS milieu as a “CNS supporting and protecting flux transport.60) In addition, Gao et al. have reported that system”, the function of the BBB should be regulated by the oatp2 is localized on both the luminal and abluminal mem- factors reflecting the conditions of the CNS and the circulat- brane of rat BCECs.61) Thus, oatp2 expressed at the BBB ing blood. The expression of transporter of energy sources, plays a role in the BBB efflux transport of DHEAS (Fig. 3). GLUT1 and MCT1, at the BBB changes during develop- Compared with the OAT family, oatp family members ment,4,77) and is thought to be regulated by dietary condi- accept bulky organic anions.62) Interesting substrates of the tions. As noted in the former section, ATA2 in BCECs is reg- oatp family, as far as CNS function is concerned, include opi- ulated by osmolarity.70) Studies to identify transporters and oid peptides and thyroid hormones. These compounds need tight-junction proteins at the BBB are now in progress. After to cross the BBB to express their CNS functions. Oatp2 is a the molecular identification, the regulation of these mole- bidirectional transporter and has been reported to be involved cules will play a key part in clarifying the physiological and in the entry of a cyclic opioid pentapeptide, [D-penicil- pathological role of the BBB in the CNS. lamine2,5]enkephalin (DPDPE).63) In addition to oatp2, re- Taurine (2-aminoethanesulfonic acid) is one of the abun- cent studies have shown that OATP-A, oatp3 and oatp14 are dant free sulfur-containing b-amino acids in the CNS and is expressed at the human, mouse and rat BBB, respective- thought to play a role as an osmoregulator and neuro-modu- ly.64—66) Therefore, further identification of the subtypes of lator.78,79) Taurine also plays an important role in the develop- the oatp family expressed at the BBB and clarification of ment of CNS,80) and its brain levels in children is about 4- their contribution to the BBB transport system will help in- fold greater than that in adults. The level of an enzyme re- crease our understanding of BBB function and drug perme- quired for taurine biosynthesis is low in humans, primates ability. and cats.80) Therefore, dietary taurine is important for main- 1494 Vol. 27, No. 10
Table2. The Uptake of [3H]Taurine by TR-BBB Cells under Different protect the CNS function. The results regarding the regula- Conditions tion of TAUT and xCT suggest that one of the physiological roles of BBB regulation is protecting the CNS against dam- Treatment Uptake of [3H]taurine (% of control) age and degeneration. Control 100 2 TNF-a 121 4* CONCLUSION LPS 103 6 DEM 94.1 6.7 Understanding the molecular basis of the BBB is impor- Taurine 10.4 0.5** tant for the development of new CNS drugs. However, BBB Cells were treated with TNF-a (20 ng/ml), LPS (10 ng/ml), DEM (100 m M) or taurine research into drug transport can only clarify limited aspects (50 mM) for 24 h, respectively. After treatment, taurine uptake was measured after a 5 min incubation with [3H]taurine. Each value represents the mean S.E.M. (n 4). of BBB function. As shown in this review, recent BBB re- * p 0.01, ** p 0.0001 significantly different from the control. Cited from Kang et al., search from the viewpoint of physiological function throws J. Neurochem., 83, 1188—1195 (2003). new light on the novel functions of the BBB, including new transporting and regulation systems. These studies show that tain taurine levels in the body. A study using primary cul- the BBB is one of the important brain systems for maintain- tured bovine BCECs has shown that there is a taurine trans- ing the CNS milieu. Furthermore, BBB function is closely port system in both the luminal and abluminal membranes of related to CNS disorders. This means that studies from the the BCECs.81) Therefore, this taurine transport system at the viewpoint of neuroscience are necessary for future BBB re- BBB may supply taurine to the brain, and be involved in the search. In addition, the new analytical technologies, espe- maintenance of taurine levels in the brain. cially genetic technologies, should be developed for BBB re- We have analyzed the regulation of taruine transport at the search, e.g. BBB-selective gene transfer and silencing, pro- BBB focusing on the functions of osmoregulation and neuro- teomics of BBB proteins and imaging (histological and func- protection.82) Taurine transporter (TAUT) was detected in tional) of BBB function. This multidisciplinary approach to TR-BBB cells, suggesting the involvement of TAUT in tau- BBB research will markedly increase our knowledge of BBB rine transport at the BBB. Hypertonic conditions induced the function, which will help improve drug delivery to the brain. uptake activity of taurine and mRNA expression of TAUT in Furthermore, the new knowledge about the physiological and TR-BBB cells. Therefore, regulation of TAUT would play a pathological functions of the BBB may open up new path- role in CNS osmoregulation in conjunction with ATA2. ways for drug development and offer improved therapeutic Taurine exerts a neuroprotective effect against excitotoxic strategies for CNS disorders. agents and oxidative stress, and its level in brain interstitial fluid (ISF) is elevated in ischemia.83—85) We hypothesized Acknowledgements I am most grateful to Drs T. that taurine transport at the BBB is involved in the changes Terasaki, K. Hosoya, H. Takanaga, and S. Hori for their ad- in taurine ISF levels under pathological conditions. We have vice, encouragement and criticism. I wish to thank Drs H. tested TNF-a, LPS and diethyl maleate (DEM), and found Asaba, K. Tetsuka, M. Tomi and S. Mori, and all graduate that, of these compounds, TNF-a induced the uptake of tau- and undergraduate students who are involved in this study for rine and mRNA expression of TAUT in TR-BBB cells (Table their creative work and discussions, and Ms. N. Funayama 2).82) TNF-a is induced in the brain after cerebral ischemia for her secretarial assistance. This study was supported, in and traumatic brain injury.86) Therefore, taurine transport at part, by Grants-in-Aid for Scientific Research (B) and for the BBB can be up-regulated by TNF-a in response to CNS Young Scientists (B) from Japan Society for the Promotion cell damage. We have also reported that taurine uptake and of Science (JSPS). It was also supported in part by the Ue- TAUT expression in TR-BBB cells is suppressed by excess hara Memorial Foundation, the Nissan Science Foundation, taurine (Table 2).82) It has been reported that the taurine the Tokyo Biochemical Research Foundation and the Indus- transport system is present in both the luminal and abluminal trial Technology Research Grant Programs in ’00 and ’03 membranes of the BCEC, and the luminal uptake rate is from the New Energy and the Industrial Technology Devel- about 2-fold greater than that of the abluminal uptake.81) opment Organization (NEDO) of Japan. Therefore, in order to understand the role of TAUT regula- tion in BCECs, it is important to see how changes in luminal REFERENCES and/or abluminal taurine transport are affected by regulation of TAUT in BCECs. 1) Cornford E. M., Mol. Physiol., 7, 219—259 (1985). 2) Schinkel A. H., Smit J. J., van Tellingen O., Beijnen J. H., Wagenaar We have also shown that blood-to-brain system xc trans- E., van Deemter L., Mol C. A., van der Valk M. A., Robanus-Maandag 87) port is induced by 12-h DEM infusion, and this induction E. C., te Riele H. P., Berns A. J. M., Borst P., Cell, 77, 491—502 is due to the up-regulation of xCT,88) which consists of sys- (1994). tem x forming a complex with 4F2hc.88) System x is an 3) Tsuji A., Terasaki T., Takabatake Y., Tenda Y., Tamai I., Yamashima c c T., Moritani S., Tsuruo T., Yamashita J., Life Sci., 51, 1427—1437 L-cystine and L-Glu exchange transport system. Supplying L- (1992). cysteine, which is reduced from L-cystine, to cells is one of 4) Cornford E. M., Hyman S., Pardridge W. M., J. Cereb. Blood Flow 89) the rate-limiting steps of glutathione biosynthesis. Glu- Metab., 13, 841—854 (1993). tathione provides protection against free radicals, peroxides 5) Ohtsuki S., Tachikawa M., Takanaga H., Shimizu H., Watanabe M., and other toxic compounds in the CNS,90) and DEM is used Hosoya K., Terasaki T., J. Cereb. Blood Flow Metab., 22, 1327—1335 (2002). as a reagent to deplete intracellular glutathione to induce ox- 6) Marescau B., Deshmukh D. R., Kockx M., Possemiers I., Qureshi I. idative stress. Therefore, the up-regulation of xCT at the A., Wiechert P., De Deyn P. P., Metabolism, 41, 526—532 (1992). BBB would lead to an increase in brain glutathione levels to 7) Schulze A., Hess T., Wevers R., Mayatepek E., Bachert P., Marescau October 2004 1495
B., Knopp M. V., De Deyn P. P., Bremer H. J., Rating D., J. Pediatr., 43) Emanuelsson B. M., Paalzow L., Sunzel M., J. Pharm. Pharmacol., 131, 626—631 (1997). 39, 705—710 (1987). 8) Pettegrew J. W., Panchalingam K., Klunk W. E., McClure R. J., Muenz 44) Kim C. S., Roe C. R., Mann J. D., Breese G. R., J. Neurochem., 58, L. R., Neurobiol. Aging, 15, 117—132 (1994). 1499—1503 (1992). 9) Wyss M., Kaddurah-Daouk R., Physiol. Rev., 80, 1107—1213 (2000). 45) Morrison P. F., Morishige G. M., Beagles K. E., Heyes M. P., J. Neu- 10) Dechent P., Pouwels P. J., Wilken B., Hanefeld F., Frahm J., Am. J. rochem., 72, 2135—2144 (1999). Physiol., 277, R698—704 (1999). 46) Mori S., Takanaga H., Ohtsuki S., Deguchi T., Kang Y. S., Hosoya K., 11) Braissant O., Henry H., Loup M., Eilers B., Bachmann C., Brain Res. Terasaki T., J. Cereb. Blood Flow Metab., 23, 432—440 (2003). Mol. Brain. Res., 86, 193—201 (2001). 47) Kusuhara H., Sekine T., Utsunomiya-Tate N., Tsuda M., Kojima R., 12) Hosoya K., Tetsuka K., Nagase K., Tomi M., Saeki S., Ohtsuki S., Cha S. H., Sugiyama Y., Kanai Y., Endou H., J. Biol. Chem., 274, Takanaga H., Yanai N., Obinata M., Kikuchi A., Okano T., Terasaki T., 13675—13680 (1999). AAPS Pharm. Sci., 2, E27 (2000). 48) Sugiyama D., Kusuhara H., Shitara Y., Abe T., Meier P. J., Sekine T., 13) Terasaki T., Ohtsuki S., Hori S., Takanaga H., Nakashima E., Hosoya Endou H., Suzuki H., Sugiyama Y., J. Pharmacol. Exp. Ther., 298, K., Drug Discov. Today, 8, 944—954 (2003). 316—322 (2001). 14) De Vivo D. C., Trifiletti R. R., Jacobson R. I., Ronen G. M., Behmand 49) Kikuchi R., Kusuhara H., Sugiyama D., Sugiyama Y., J. Pharmacol. R. A., Harik S. I., N. Engl. J. Med., 325, 703—709 (1991). Exp. Ther., 306, 51—58 (2003). 15) Seidner G., Alvarez M. G., Yeh J. I., O’Driscoll K. R., Klepper J., 50) Ohtsuki S., Asaba H., Takanaga H., Deguchi T., Hosoya K., Otagiri Stump T. S., Wang D., Spinner N. B., Birnbaum M. J., De Vivo D. C., M., Terasaki T., J. Neurochem., 83, 57—66 (2002). Nat. Genet., 18, 188—191 (1998). 51) Ohtsuki S., Kikkawa T., Mori S., Hori S., Takanaga H., Otagiri M., 16) Salomons G. S., van Dooren S. J., Verhoeven N. M., Cecil K. M., Ball Terasaki T., J. Pharmacol. Exp. Ther., (2004). W. S., Degrauw T. J., Jakobs C., Am. J. Hum. Genet., 68, 1497—1500 52) Lloyd K. G., Davidson L., Price K., McClung H. J., Gall D. G., Neu- (2001). rology, 27, 985—988 (1977). 17) Cecil K. M., Salomons G. S., Ball W. S., Jr., Wong B., Chuck G., Ver- 53) Oldendorf W. H., Am. J. Physiol., 224, 1450—1453 (1973). hoeven N. M., Jakobs C., DeGrauw T. J., Ann. Neurol., 49, 401—404 54) Mamunes P. D., De Vries G. H., Miller C. D., David R. B., “Reye’s (2001). Syndrome,” ed. by Pollack D., Grune & Stratton, New York, 1975, pp. 18) Takanaga H., Ohtsuki S., Hosoya K., Terasaki T., J. Cereb. Blood Flow. 245—254. Metab., 21, 1232—1239 (2001). 55) Mori S., Ohtsuki S., Takanaga H., Kikkawa T., Kang Y. S., Terasaki T., 19) Kakee A., Terasaki T., Sugiyama Y., J. Pharmacol. Exp. Ther., 277, J. Neurochem., 90, 931—941 (2004). 1550—1559 (1996). 56) Jacqz-Aigrain E., Nafa S., Medard Y., Bessa E., Lescoeur B., Vilmer 20) Kakee A., Takanaga H., Terasaki T., Naito M., Tsuruo T., Sugiyama E., Eur. J. Clin. Pharmacol., 53, 71—74 (1997). Y., J. Neurochem., 79, 110—118 (2001). 57) Deguchi Y., Yokoyama Y., Sakamoto T., Hayashi H., Naito T., Yamada 21) Hardebo J. E., Owman C., Acta Physiol. Scand., 108, 223—229 S., Kimura R., Life Sci., 66, 649—662 (2000). (1980). 58) Race J. E., Grassl S. M., Williams W. J., Holtzman E. J., Biochem. Bio- 22) Spatz M., Maruki C., Abe T., Rausch W. D., Abe K., Merkel N., Brain phys. Res. Commun., 255, 508—514 (1999). Res., 220, 214—219 (1981). 59) Wolf O. T., Kirschbaum C., Brain Res. Brain Res. Rev., 30, 264—288 23) Wakayama K., Ohtsuki S., Takanaga H., Hosoya K., Terasaki T., Neu- (1999). rosci Res., 44, 173—180 (2002). 60) Asaba H., Hosoya K., Takanaga H., Ohtsuki S., Tamura E., Takizawa 24) Rennels M. L., Nelson E., Am. J. Anat., 144, 233—241 (1975). T., Terasaki T., J. Neurochem., 75, 1907—1916 (2000). 25) Di Carlo V., Neurosci. Lett., 51, 295—302 (1984). 61) Gao B., Stieger B., Noe B., Fritschy J. M., Meier P. J., J. Histochem. 26) Kobayashi H., Magnoni M. S., Govoni S., Izumi F., Wada A., Trabuc- Cytochem., 47, 1255—1264 (1999). chi M., Experientia, 41, 427—434 (1985). 62) Hagenbuch B., Meier P. J., Biochim. Biophys. Acta, 1609, 1—18 27) Cohen Z., Bonvento G., Lacombe P., Hamel E., Prog. Neurobiol., 50, (2003). 335—362 (1996). 63) Dagenais C., Ducharme J., Pollack G. M., Neurosci. Lett., 301, 155— 28) Preskorn S. H., Hartman B. K., Raichle M. E., Clark H. B., J. Pharma- 158 (2001). col. Exp. Ther., 213, 313—320 (1980). 64) Sugiyama D., Kusuhara H., Taniguchi H., Ishikawa S., Nozaki Y., 29) Ganapathy V., Prasad P. D., Ganapathy M. E., Leibach F. H., J. Phar- Aburatani H., Sugiyama Y., J. Biol. Chem., 278, 43489—43495 macol. Exp. Ther., 294, 413—420 (2000). (2003). 30) Balkovetz D. F., Tiruppathi C., Leibach F. H., Mahesh V. B., Ganapa- 65) Gao B., Hagenbuch B., Kullak-Ublick G. A., Benke D., Aguzzi A., thy V., J. Biol. Chem., 264, 2195—2198 (1989). Meier P. J., J. Pharmacol. Exp. Ther., 294, 73—79 (2000). 31) Wolosker H., Panizzutti R., De Miranda J., Neurochem. Int., 41, 327— 66) Ohtsuki S., Takizawa T., Takanaga H., Hori S., Hosoya K., Terasaki T., 332 (2002). J. Neurochem., 90, 743—749 (2004). 32) Zorumski C. F., Olney J. W., Pharmacol. Ther., 59, 145—162 (1993). 67) Alfieri R. R., Petronini P. G., Bonelli M. A., Caccamo A. E., Cavaz- 33) Hosoya K., Sugawara M., Asaba H., Terasaki T., J. Neurochem., 73, zoni A., Borghetti A. F., Wheeler K. P., Biochem. Biophys. Res. Com- 1206—1211 (1999). mun., 283, 174—178 (2001). 34) O’Kane R. L., Martinez-Lopez I., DeJoseph M. R., Vina J. R., 68) Betz A. L., Goldstein G. W., Science, 202, 225—227 (1978). Hawkins R. A., J. Biol. Chem., 274, 31891—31895 (1999). 69) Sanchez del Pino M. M., Peterson D. R., Hawkins R. A., J. Biol. 35) Palacin M., Estevez R., Bertran J., Zorzano A., Physiol. Rev., 78, Chem., 270, 14913—14918 (1995). 969—1054 (1998). 70) Takanaga H., Tokuda N., Ohtsuki S., Hosoya K., Terasaki T., Mol. 36) Tetsuka K., Takanaga H., Ohtsuki S., Hosoya K., Terasaki T., J. Neu- Pharmacol., 61, 1289—1296 (2002). rochem., 87, 891—901 (2003). 71) Hosoya K., Takashima T., Tetsuka K., Nagura T., Ohtsuki S., Takanaga 37) Arriza J. L., Kavanaugh M. P., Fairman W. A., Wu Y. N., Murdoch G. H., Ueda M., Yanai N., Obinata M., Terasaki T., J. Drug. Target., 8, H., North R. A., Amara S. G., J. Biol. Chem., 268, 15329—15332 357—370 (2000). (1993). 72) Varoqui H., Zhu H., Yao D., Ming H., Erickson J. D., J. Biol. Chem., 38) Utsunomiya-Tate N., Endou H., Kanai Y., J. Biol. Chem., 271, 275, 4049—4054 (2000). 14883—14890 (1996). 73) Sugawara M., Nakanishi T., Fei Y. J., Huang W., Ganapathy M. E., 39) Giroud M., Dumas R., Dauvergne M., D’Athis P., Rochette L., Beley Leibach F. H., Ganapathy V., J. Biol. Chem., 275, 16473—16477 A., Bralet J., Epilepsia, 31, 178—181 (1990). (2000). 40) Le Witt P. A., Galloway M. P., “Therapy of Parkinson’s Disease,” ed. 74) Reimer R. J., Chaudhry F. A., Gray A. T., Edwards R. H., Proc. Natl. by Koller W. C., Paulson G., Marcel Dekker, New York, 1990, pp. Acad. Sci. U.S.A., 97, 7715—7720 (2000). 63—93. 75) Yao D., Mackenzie B., Ming H., Varoqui H., Zhu H., Hediger M. A., 41) Sullivan P. A., Murnaghan D., Callaghan N., Kantamaneni B. D., Cur- Erickson J. D., J. Biol. Chem., 275, 22790—22797 (2000). zon G., J. Neurol. Neurosurg. Psychiatry, 41, 581—588 (1978). 76) Sugawara M., Nakanishi T., Fei Y. J., Martindale R. G., Ganapathy M. 42) Shaywitz B. A., Venes J., Cohen D. J., Bowers M. B., Neurology, 29, E., Leibach F. H., Ganapathy V., Biochim. Biophys. Acta, 1509, 7—13 467—472 (1979). (2000). 1496 Vol. 27, No. 10
77) Pellerin L., Pellegri G., Martin J. L., Magistretti P. J., Proc. Natl. Acad. 84) Boldyrev A. A., Johnson P., Wei Y., Tan Y., Carpenter D. O., Neurosci. Sci. U.S.A., 95, 3990—3995 (1998). Lett., 263, 169—172 (1999). 78) Tuz K., Ordaz B., Vaca L., Quesada O., Pasantes-Morales H., J. Neu- 85) Matsumoto K., Lo E. H., Pierce A. R., Halpern E. F., Newcomb R., J. rochem., 79, 143—151 (2001). Cereb. Blood Flow Metab., 16, 114—124 (1996). 79) Oja S. S., Saransaari P., Metab. Brain Dis., 11, 153—164 (1996). 86) Shohami E., Ginis I., Hallenbeck J. M., Cytokine Growth Factor Rev., 80) Schuller-Levis G. B., Park E., FEMS Microbiol. Lett., 226, 195—202 10, 119—130 (1999). (2003). 87) Hosoya K., Saeki S., Terasaki T., Microvasc. Res., 62, 136—142 81) Tamai I., Senmaru M., Terasaki T., Tsuji A., Biochem. Pharmacol., 50, (2001). 1783—1793 (1995). 88) Tomi M., Hosoya K., Takanaga H., Ohtsuki S., Terasaki T., Invest. 82) Kang Y. S., Ohtsuki S., Takanaga H., Tomi M., Hosoya K., Terasaki T., Ophthalmol Vis. Sci., 43, 774—779 (2002). J. Neurochem., 83, 1188—1195 (2002). 89) Bannai S., Tateishi N., J. Membr. Biol., 89, 1—8 (1986). 83) French E. D., Vezzani A., Whetsell W. O., Jr., Schwarcz R., Adv. Exp. 90) Meister A., Anderson M. E., Annu. Rev. Biochem., 52, 711—760 Med. Biol., 203, 349—362 (1986). (1983).