Tryptophan metabolites such as kynurenate (KYNA), xanthurenate (XA), and quinolinate are considered to have an important impact on many physiological processes, especially brain function. Many of these metabolites are secreted with the urine. Because organic anion transporters (OATs) facilitate the renal secretion of weak organic acids, we investigated whether the secretion of bioactive tryptophan metabolites is mediated by OAT1 and OAT3, two prominent members of the OAT family. Immunohistochemical analyses of the mouse kidneys revealed the expression of OAT1 to be restricted to the proximal convoluted tubule (representing S1 and S2 segments), whereas OAT3 was detected in almost all parts of the nephron, including macula densa cells. In the mouse brain, OAT1 was found to be expressed in neurons of the cortex cerebri and hippocampus as well as in the ependymal cell layer of the choroid plexus. Six tryptophan metabolites, including the bioactive substances KYNA, XA, and the serotonin metabolite 5-hydroxyindol acetate inhibited [3H]p-aminohippurate (PAH) or 6-carboxyfluorescein (6-CF) uptake by 50–85%, demonstrating that these compounds interact with OAT1 as well as with OAT3. Half-maximal inhibition of mOAT1 occurred at 34 μM KYNA and 15 μM XA, and it occurred at 8 μM KYNA and 11.5 μM XA for mOAT3. Quinolinate showed a slight but significant inhibition of [3H]PAH uptake by mOAT1 and no alteration of 6-CF uptake by mOAT3. [14C]-Glutarate (GA) uptake was examined for both transporters and demonstrated differences in the transport rate for this substrate by a factor of 4. Trans-stimulation experiments with GA revealed that KYNA and XA are substrates for mOAT1. Our results support the idea that OAT1 and OAT3 are involved in the secretion of bioactive tryptophan metabolites from the body. Consequently, they are crucial for the regulation of central nervous system tryptophan metabolite concentration.
- macula densa
- transforming growth factor
- N-methyl-d-aspartate receptor
the kidneys are the target organ for the secretion of substances of endogenous as well as of exogenous origin. Since the identification of the organic anion transporter (OAT) family in 1997, the focus of research has been on understanding the contribution of these transport proteins to renal secretion of xenobiotics and drugs (for review, see Refs. 8, 20, 52). The extrarenal expression of these transporter proteins in the brain, especially in the choroid plexus and the blood-brain barrier (12, 33, 36), the eyes (6), the skeletal muscle (49), and also several organs in different stages of embryo development (35) indicates that OATs may play a broader role in physiological processes than previously thought. Recently, we documented that the neurotransmitter metabolites homovanillate (HVA) and vanilline mandelate (VMA), which are diagnostic markers for brain activity in terms of neurodegeneration (32), are substrates of OAT1 (6). Moreover, rat OAT3 transports HVA (28) and human OAT1 and OAT3 transport VMA (1), supporting the concept that both transporters are involved in the elimination of these endogenous metabolites from the brain.
Tryptophan is an essential amino acid, which, if not used for protein synthesis, is further metabolized via the kynurenine pathway (Fig. 1) (for review, see Refs. 40, 44). This metabolic pathway, which was detected in several organs such as the liver, the kidneys (2), the eyes (37), and the brain (45), results after one main branch in the synthesis of the neurotransmitter serotonin, which can be further metabolized to 5-hydroxyindol acetate (5-HIAA) or the neuroprotective substance melatonin. The alternative branch of the kynurenine pathway finally generates NAD+ (vitamin B3) during protein shortage. More important, a variety of neuroactive as well as physiologically relevant metabolites such as kynurenate (KYNA), picolinate (PICA), xanthurenate (XA), anthranilate (ANTRA), 3-hydroxykynurenine (3-HK), and quinolinate (QUIN) are derived from the common major metabolite of tryptophan, l-kynurenine. Many of these metabolites are routinely found in the urine (3). XA is well known as a clinical marker for vitamin B6 deficiencies, in which urinary XA levels are elevated after a tryptophan load (27). KYNA and QUIN both are known to act on the glycine-B site of N-methyl-d-aspartate (NMDA) receptors, whereas KYNA alone is known to act on α7-nicotinic acetylcholine receptors. The neuroprotective/antiexcitatory effects of KYNA and the neurodegenerative/excitatory effects of QUIN in the brain have been attributed to their actions on the above-mentioned receptors (41, 46). An imbalance of these substances, which was observed in several neurodegenerative diseases such as schizophrenia, Alzheimer’s disease, and epilepsy, is thought to be an early event in chronic brain inflammation (16).
At present, transport proteins involved in the handling of these substances are not characterized at the molecular level. However, probenecid-sensitive transport systems facilitate the secretion of these substances from the brain (26, 29), suggesting that organic anion transporters 1 and 3 (OAT1 and OAT3) may be candidate transporters of bioactive tryptophan metabolites. OAT1 is known as the high-affinity basolateral p-aminohippurate (PAH) transporter mediating the renal secretion of a variety of hydrophilic endogenous metabolites, xenobiotics, and drugs in exchange for the dicarboxylate α-ketoglutarate (α-KG) (8, 9). With the identification of OAT3 from rat kidneys in 1999, a second transporter was determined to be involved in the renal secretion of PAH and of other organic anions (OAs) on the basolateral side of proximal tubule cells (22). In the meantime, several OAT3 orthologs from human (11), rabbit (53), mouse (19, 34), and pig (15) have been identified and functionally characterized. These studies show that OAT3 shares several substrates with OAT1 but prefers lipophilic substrates with estrone sulfate (ES) and dehydroepiandosterone sulfate (DHEAS) as the major endogenous substrates (8). Recently, it was shown that OAT3, like OAT1, operates as an OA/α-KG exchanger (7, 47, 53).
Because tryptophan metabolites must be eliminated from the brain via the blood into the urine, the goals of this study were to 1) determine the localization of mOAT1 and mOAT3 in kidneys and brain and 2) to explore the interaction of mOAT1 and mOAT3 with these metabolites.
MATERIALS AND METHODS
Materials used included fetal bovine serum, trypsin, and PBS, which were obtained from Invitrogen (Groningen, The Netherlands). Buffer ingredients and unlabeled substrates such as 5-HIAA, ANTRA, glutarate, probenecid, KYNA, 3-HK, QUIN, PICA, and XA were purchased from Sigma-Aldrich (Deisenhofen, Germany). 6-Carboxyfluorescein (6-CF) was purchased from Molecular Probes (Leiden, The Netherlands). [3H]PAH (3.25 Ci/mmol) was provided by PerkinElmer Life Sciences (Boston, MA). [14C]glutarate (GA) (glutaric acid [1,5-14C]−; 30.8 mCi/mmol) was obtained from ICN (Costa Mesa, CA).
mOAT1 (IMAGp998P219857Q2; GenBank accession no. BF788173) and mOAT3 (IMAGp998D159848Q2; GenBank accession no. BF783835) were obtained from the Resource Center/Primary Database and cloned into the pSPORT6 expression vector. Both clones were sequence verified using an automated sequencer (ABI, Weiterstadt, Germany). Sequence analysis was performed using an online search engine [MAP; available at http://genome.cs.mtu.edu/map.html (17)].
Animal tissue was prepared as previously described (24) according to the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals, and our experimental protocol was approved by the Institutional Ethics Committee.
Polyclonal rabbit anti-OAT1 and anti-OAT3 antibodies (OAT11-A, OAT31-A; affinity-purified immune serum) and the respective peptides (OAT11-P, OAT31-P) were purchased from Alpha Diagnostic (San Antonio, TX). The CY3-labeled (GARCY3) secondary antibody was supplied by Jackson ImmunoResearch Laboratories (West Grove, PA). Immunohistochemistry was performed on the mouse brain. After 24-h incubation of the sections with the first antibody, a second biotinylated anti-rabbit antibody was applied for 2 h. Antibody binding was visualized using the avidin-biotin-peroxidase complex (Vectastain) and diaminobenzidine (DAB; Sigma). To obtain 4-μm frozen sections, tissue slices from mouse kidneys and brain stored in PBS containing 0.02% NaN3 were infiltrated with 30% sucrose in PBS overnight, embedded in optimum cutting temperature (OCT) medium (Tissue-Tek, Sakura, Japan), frozen at −25°C, and cut in a Leica CM 1850 cryostat (Leica Instruments, Nussloch, Germany). The sections were collected on Superfrost Plus microscope slides (Fisher Scientific, Pittsburgh, PA) and rehydrated in PBS for 10 min. Subsequently, the sections were pretreated for 5 min with 1% SDS in PBS to expose potential cryptic antigenic sites. SDS was removed by performing extensive washing with PBS. All sections were double-stained on OAT1. Nonspecific binding of antibodies was prevented by incubating the sections with 1% bovine serum albumin (in PBS) for 15 min, followed by one of the polyclonal anti-OAT1 antibodies (diluted with 100 μg/ml PBS) at 4°C overnight (for 12–14 h). After two wash steps with high-salt PBS (PBS containing 2.7% NaCl) and two wash steps with regular PBS (5 min each), incubation with GARCY3 (1.6 μg/ml in PBS) at room temperature (RT) was performed for 60 min. The procedure was terminated by performing wash steps with high-salt and regular PBS (twice for 5 min each) and mounting the tissue slices in a fluorescence-fading retardant (Vectashield; Vector Laboratories, Burlingame, CA).
The preparations were examined and photographed with an Opton III RS fluorescence microscope (Opton Feintechnik, Oberkochen, Germany) using a Spot RT Slider camera and software (Diagnostic Instruments, Sterling Heights, MI).
Cell culture and transport measurements.
The green monkey kidney cell line COS-7 was cultured in plastic flasks or petri dishes (Sarstedt, Nümbrecht, Germany) in DMEM (Invitrogen, Groningen, The Netherlands) with 580 mg/l glutamine, 110 mg/l Na+-pyruvate, and 10% heat-inactivated fetal calf serum (FCS) or in fibroblast complete medium (PAA, Cölbe, Germany) in a 5% CO2 atmosphere at 37°C. Five micrograms of mOAT1 or mOAT3 plasmid DNA were transiently transfected into COS-7 cells using electroporation (GenePulser II; Bio-Rad, Munich, Germany) at 250 V and 300 μF. Twenty-four hours after being transfected, the cells were harvested and plated into 24-well plastic dishes (Sarstedt) at a density of 2 × 105 cells/well. Transport assays were performed 48 h posttransfection in mammalian Ringer solution (in mM: 130 NaCl, 4 KCl, 1 CaCl2, 1 Mason, 1 NaHPO, 20 HEPES, and 18 glucose, pH 7.4) essentially as described previously (5). The cells were washed twice with 500 ml of buffer and incubated at RT in buffer containing 0.25 μM [3H]PAH or 2 and 5 μM 6-CF for 2 min (mOAT1) or 15 min (mOAT3). For GA uptake, 1.8 μM substrate was used for 5 min at RT. In some experiments, the test solutions included additional substances as described in the figure legends. The incubation was stopped and the extracellular tracer was removed by washing the monolayer twice with 750 ml of ice-cold PBS. For trans-stimulation experiments, the cells were preloaded for 2–3 h with 1.8 μM GA, and efflux was determined after 2 min at RT in the supernatant and correlated to the total load of the cells derived from the sum of the radioactivity of the supernatant and the cells. Cells were lysed in 0.5 ml of 1 N NaOH. To assess intracellular 6-CF accumulation, fluorescence was measured in a fluorescence spectrophotometer (Hitachi, Tokyo, Japan) at 492-nm excitation/512-nm emission. The [3H] and [14C] content were determined using liquid scintillation counting (Canberra-Packard, Dreieich, Germany). For determination of the KYNA or XA concentrations necessary to block 50% of 6-CF uptake (IC50), the following equation (Eq. 1) was used and fitted by performing nonlinear regression analysis with SigmaPlot 2001 software (SPSS, Chicago, IL): (1) In Eq. 1, ν is the rate of 6-CF uptake in the presence of the inhibitor (KYNA or XA), ν0 is the rate of 6-CF uptake in the absence of the inhibitor (set to 100%), I is the inhibitor concentration, and nH is the Hill coefficient, representing the cooperativity between the tested substances and the transporter.
Student’s t-test was applied to calculate statistical significance at P < 0.05. Graphic layouts were created with Microsoft Excel software (Microsoft, Unterschleissheim, Germany) and SigmaPlot 2001 software (SPSS Science, Chicago, IL), respectively.
Expression of mOAT1 and mOAT3 proteins in kidneys and brain.
Expression of mOAT1 and mOAT3 was determined by immunohistochemistry. Murine OAT1 was detected at the basolateral side of proximal convoluted tubules (PCT) corresponding to tubule segments S1 and S2 (Fig. 2). No OAT1 was detectable in the glomerulus (G) and in the proximal straight tubules (PSTs) representing tubule segment S3 (Fig. 2, B and D). In contrast to OAT1, OAT3 could be detected at the basolateral side of all renal epithelia except the glomerulus (Fig. 3). Interestingly, we detected OAT3 expression for the first time at the basolateral side of macula densa cells (see Fig. 3E). To investigate OAT1 expression in the mouse brain, we used sections of the cortex and the hippocampus. OAT1 was detectable in neurons and their dendrites in the cortex cerebri and the hippocampus (Fig. 4). Preincubation of OAT1 antiserum with the peptide resulted in negative staining (Fig. 4, B and D). OAT3, however, was not detectable in the mouse brain.
Inhibition studies on mOAT1- and mOAT3-mediated transport activity using several tryptophan metabolites.
mOAT1 and mOAT3 were transiently transfected into COS-7 cells, and the uptake of labeled 0.25 μM PAH or 2 or 5 μM 6-CF was investigated in the absence or presence of 1 mM tryptophan metabolites, namely, KYNA, XA, ANTRA, PICA, 3-HK, QUIN, and 5-HIAA. For comparison, the well-known OAT substrates PAH and GA and the inhibitor probenecid (Prob) were tested in parallel. For all tested substances, we observed a highly significant reduction of mOAT1-mediated uptake of PAH or 6-CF with a remaining uptake compared with noninhibited transport for Prob (7.9%), KYNA (8.9%), 3-HK (11.6%), QUIN (63.8%), ANTRA (19.8%), 5-HIAA (10.9%), PICA (53%), XA (15.5%), PAH (22.9%), and GA (22.6%) (Fig. 5, A and B). In the case of mOAT3, we obtained an almost identical inhibition profile, showing highly significant effects of 5-HIAA, ANTRA, and PICA on PAH transport but no alteration of mOAT3-mediated 6-CF uptake for QUIN (Fig. 6, A and B).
GA uptake and efflux mediated by mOAT1 and mOAT3.
To test whether one of the tryptophan metabolites is a substrate for mOAT1 or mOAT3, we first examined GA uptake of OAT1- or OAT3-transfected COS-7 cells for 5 min at RT (Fig. 7). At a concentration of 1.8 μM, GA uptake differs by a factor of 4 between mOAT1 (10.83 pmol·mg−1·5 min−1) and mOAT3 (2.68 pmol·mg−1·5 min−1). This uptake was almost abolished by 1 mM PAH or 1 mM GA as observed for both transporters. GA efflux was tested for mOAT1 and mOAT3. After transfected COS-7 cells were preloaded with 1.8 μM GA for 2–4 h, efflux of GA was determined for 2 min at RT using mammalian Ringer solution as a control and 500 μM GA, XA, or KYNA as countersubstrates. GA, XA, and KYNA significantly trans-stimulated the efflux of GA in mOAT1-transfected COS-7 cells (Fig. 8). In mOAT3-transfected cells, however, no trans-stimulation of GA efflux for the tested compounds was detectable (data not shown).
IC50 determination for KYNA and XA.
Increasing concentrations of unlabeled KYNA inhibited 6-CF transport with an IC50 of 34 μM for mOAT1 and 8 μM for mOAT3 (Fig. 9, A and B), respectively. Using increasing concentrations of XA, we observed IC50 values of 15 μM for mOAT1 and 11.5 μM for mOAT3 (Fig. 10, A and B).
To address the question whether OAT1 and OAT3 are involved in the elimination process of tryptophan metabolites from the brain via the kidneys into the urine, we first investigated the localization of murine OAT1 and OAT3 in kidneys and brain using immunohistochemical techniques. Consistent with previous findings in the rat (24, 50) and human (30) kidneys, we detected mOAT1 in cells of the kidney cortex and the outer stripe of the medulla representing the S1 and S2 segments of the proximal tubule. The inner stripe as well as the papilla (S3 segment) were negative for OAT1 staining. In the S1 and S2 segments, labeling was found at the basolateral membrane domain, which is in agreement with OAT1 function in these cells (10). In rat and rabbit kidneys, maximal PAH secretion occurs in the S2 segment (42). Given the OAT1 distribution in mouse kidney, it appears that PAH secretion in proximal tubules may be high in S1 and S2 and low in S3 segments.
Renal OAT3 expression has a different pattern compared with OAT1. We detected OAT3 protein in many parts of the mouse nephron, such as the proximal tubule (S1, S2, and S3), the distal tubule, and the collecting duct, which is in good agreement with recent findings for the rat and the human OAT3 (21, 24, 30). At all locations, immunoreactivity was detected at the basolateral cell side. An unexpected finding was the localization of OAT3 at the basolateral side of macula densa cells. These cells are involved in the tubuloglomerular feedback (TGF) regulation of arteriolar blood flow and subsequently of renin secretion. In case of a low luminal NaCl concentration at the macula densa, these cells release prostaglandin E2 (PGE2) at the basolateral side, resulting in the release of renin from granular cells and finally in the adjustment of systemic and renal blood flow (39). Importantly, mOAT3 as well as hOAT1–hOAT4 were shown to transport PGE2 with high affinities, e.g., Km of hOAT3 for PGE2 was documented to be 0.345 μM (18, 19). Thus OAT3 may be directly involved in the TGF signaling process by transporting PGE2. OAT3 was recently investigated regarding its function as an OA exchanger mediating the uptake of OAs into the proximal tubule cell in exchange for α-KG (7, 48, 53). The possibility that OAT3 facilitates PGE2 efflux on the basolateral side of macula densa cells is supported by the notion of OAT3-mediated bidirectional transport of some substrates, namely GA (our present study and Ref. 7; regarding urate, see Ref. 53). We have demonstrated a substantial increased uptake (5 times greater than control) in OAT3-mediated glutarate using a five times lower GA concentration than the physiological concentration of ∼10 μM α-KG noted for the extracellular compartment (43). When expressed in Xenopus laevis oocytes, hOAT1 and hOAT3 showed GA Km values of 3.8 and 23 μM, respectively (Bakhiya N, personal communication, June 2004). Consequently, we speculate that in a direct competition of intracellular α-KG with PGE2 in view of the high internal-external α-KG gradient, the >60 times higher affinity of OAT3 for PGE2 favors basolateral PGE2 efflux rather than α-KG release. Moreover, the given extracellular α-KG concentration and the Km value for glutarate are reasonable to assume α-KG as the counterion for OAT3-mediated PGE2 release.
Herein we describe for the first time the expression of OAT1 in neurons and their dendrites in the cortex cerebri and hippocampus. In 1968, Ashcroft et al. (4) demonstrated that probenecid inhibits the secretion of dopamine and serotonin metabolites such as HVA and 5-HIAA in the dog brain. Pritchard et al. (36) provided the first molecular evidence for the expression of OAT1 in the choroid plexus of the rat brain. Recently, Alebouyeh et al. (1) demonstrated the expression of human OAT1 and OAT3 in the ependymal epithelium of the choroid plexus facilitating the efflux of neurotransmitter metabolites, especially of VMA from the brain into the cerebrospinal fluid. OAT3 was also noted to be expressed in the blood-brain barrier, where it may be responsible for the elimination of substrates benzylpenicillin (PCG) and HVA, for example (31). Mori et al. (28) determined a Km of 274 μM HVA in Xenopus laevis oocytes expressing rOAT3. The physiological concentration of HVA in the cerebrospinal fluid of patients ranges from 10 to 30 ng/ml CSF (equivalent to 110–170 nM) (38). This high Km value detected for rOAT3 leaves the question open regarding whether OAT3 contributes to HVA excretion from the brain under physiological conditions in vivo. Sweet et al. (48) generated OAT3-knockout mice and were able to show a reduced uptake of PAH and fluorescein into the choroid plexus tissue. However, they concluded that several other mechanisms must compensate for the lack of OAT3 function, because the mice did not show any pathophysiological phenotype.
The physiological role of OAT1 detected in neurons remains open. Recently, we showed that rabbit OAT1 translocates the neurotransmitter metabolites HVA and VMA (6). Further investigations revealed a species-specific uptake pattern for neurotransmitter metabolites. For the human OAT1, we detected a significant trans-stimulation of fluorescein uptake by 5-HIAA, suggesting that it is a transported substrate for hOAT1 (Bahn A, unpublished data). Therefore, one possible function of OAT1 in neuronal cells might be the release of the serotonin metabolite 5-HIAA after reuptake and metabolization of serotonin by the monoaminoxidase (MAO) to 5-HIAA within the neuron. Consequently, OAT1 might be involved in the regulation of the serotonin/5-HIAA level in neurons.
Tryptophan metabolites are synthesized in different organs of the body, especially in astrocytes and microglial cells in the brain, where they are considered to modulate brain function (44, 45). Because nothing was known on the molecular mechanisms of tryptophan metabolite transport in the body, we investigated their interaction with mOAT1 and mOAT3. The main metabolites of this pathway are QUIN, PICA, and nicotinamide (vitamin B3). Many tryptophan metabolites such as KYNA, ANTRA, or XA, which are generated in side branches of this pathway, are usually excreted with the urine (3, 13, 40). It is for this reason that the tryptophan metabolic pathway is also called kynurenine pathway (for review, see Ref. 40). We transfected COS-7 cells transiently with mOAT1 or mOAT3 and measured the uptake of [3H]PAH (mOAT1) and 6-CF (mOAT1 and mOAT3) in the absence and presence of different tryptophan metabolites and the neurotransmitter metabolite 5-HIAA. All tested substances, excluding QUIN, showed remarkable inhibition of PAH or 6-CF uptake by 50–85% for mOAT1 and mOAT3, suggesting that these substances substantially interact with both transporters. QUIN exhibited a slight reduction of PAH transport for mOAT1, indicating a relatively weak interaction with OAT1, whereas it lacked any interaction with mOAT3. Inhibition of PAH or 6-CF uptake does not prove that the tested compounds are actually translocated. Because radiolabeled metabolites were not available for a direct demonstration of transport mediated by mOAT1 and mOAT3, we tried to provide indirect evidence by showing trans-stimulation of PAH or 6-CF uptake by intracellularly loaded test compounds. However, these experiments did not provide any information about one of the metabolites to be a substrate for OAT1 or OAT3. Alternatively, we used labeled glutarate as a tracer. GA was taken up by both mOAT1 and mOAT3, although at different rates, with mOAT1 showing a fourfold higher transport rate, consistent with the above-mentioned difference in GA affinity determined for the human orthologs. Preloading mOAT1-expressing COS-7 cells with GA and measuring GA efflux revealed a significant trans-stimulation by GA (used as a control), KYNA, and XA. On the basis of this result, we conclude that mOAT1 transports KYNA an XA. We were not able to detect a significant trans-stimulation of mOAT3 mediated transport (data not shown), which may be due to the fact that in the cells, GA is not accumulated to the same extent as mOAT1-expressing COS-7 cells. Although we detected a substantial interaction of tryptophan metabolites with OAT3, the question remains open whether these substances are transported by OAT3.
To estimate and compare the possible involvement of mOAT1 and mOAT3 in the handling of KYNA and XA in the body, we determined the IC50 values on 6-CF uptake. Increasing concentrations of KYNA as well as XA reduced 6-CF in the low micromolar range for both transporters, suggesting that they both participate in the elimination of these substances from the brain.
The important neuroprotective and neuromodulatory role of KYNA was initially explored by Stone et al. (46) more than two decades ago. Meanwhile, therapeutic concepts are the focus of pharmacological research to explore and test the modification of the KYNA pathway by inhibiting key enzymes toward a higher synthesis rate of endogenous KYNA to treat patients with neurodegenerative diseases such as schizophrenia (44).
In vivo KYNA levels are elevated, e.g., during fetal development, and they decrease substantially after birth (51). The high fetal concentrations of KYNA are thought to be due to the chronic hypoxia in utero. Along that line, Pavlova et al. (35) recently described a high level of OAT1 expression in the brain during mouse embryogenesis, whereas it is only slightly expressed at stage E14 in the kidneys. This finding is consistent with the fact that higher amounts of KYNA have to be eliminated from the brain. Interestingly, in the developing fetus, the kidneys are of minor importance for excretion, which instead occurs via the maternal circulation. In the adult brain, KYNA was confirmed to modulate NMDA receptors, showing a higher affinity for the NR2A subunit of NMDA receptors (25). Hippocampal NMDA receptors consist mostly of a basic NR1 subunit and two types of NR2 subunits (NR2A and NR2B). Recent studies on the mechanism of long-term potentiation (LTP) and long-term depression (LTD) made clear that the NR2A subunit is mainly involved in the LTP process (23) and therefore is important in the context of synaptic plasticity. Consequently, OATs, especially OAT1 and OAT3, as tested in this study may contribute to these processes by adjusting the level of KYNA and other tryptophan metabolite levels in the brain.
In summary, we have shown the expression of mOAT1 and mOAT3 in kidneys and brain. Moreover, we demonstrate a substantial interaction of various bioactive tryptophan metabolites with both transporters. Especially KYNA and XA were explored as new endogenous substrates for OAT1, supporting the idea of an important role for OAT1 in brain function.
We thank A. Hillemann and S. Petzke for excellent technical assistance and A. Nolte (Dept. of Biochemistry, Universität Göttingen) for nucleotide sequencing.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society