- Email: [email protected]
Cl−-dependent neurotransmitter transporter activity
ANALYTICAL BIOCHEMISTRY Analytical Biochemistry 321 (2003) 31–37 www.elsevier.com/locate/yabio
Development of a scintillation proximity assay for analysis of Naþ /Cl -dependent neurotransmitter transporter activity Jacinta B. Williams,a Pierre J. Mallorga,a Wei Lemaire,a David L. Williams,a Sang Na,a Smita Patel,b Jeffrey P. Conn,a,1 Douglas J. Pettibone,a Christopher Austin,a,2 and Cyrille Sura,* a
b
Department of Neuroscience West Point, Merck & Co. Inc., West Point, PA 19486, USA Neuroscience Research Centre, Merck Sharp & Dohme, Terlings Park, Harlow, Essex CM20 2QR, UK Received 14 February 2003
Abstract Human placental choriocarcinoma (JAR) cells endogenously expressing glycine transporter type 1a (GlyT1a) have been cultured in 96-well scintillating microplates to develop a homogenous screening assay for the detection of GlyT1 antagonists. In these microplates uptake of [14 C]glycine was time dependent and saturable with a Michaelis–Menten constant (Km ) of 27 3 lM. The GlyT1 transport inhibitors sarcosine, ALX-5407, and Org-24598 were tested and shown to block [14 C]glycine uptake with expected IC50 values of 37.5 4.6 lM, 2.8 0.6 nM, and 6.9 0.9 nM, respectively. The [14 C]glycine uptake process was sensitive to membrane Naþ gradient as blockade of membrane Naþ /Kþ -ATPase by ouabain or Naþ exchanger by benzamil-disrupted glycine accumulation in JAR cells. Glycine influx was not affected by concentration of dimethyl sulfoxide up to 2%. The versatility of this technological approach was further confirmed by the characterization of a saturable [14 C]taurine uptake in JAR cells. Taurine transport was of high affinity with a Km of 10.2 1.7 lM and fully inhibited by ALX-5407 (IC50 ¼ 522 83 nM). The developed assay is homogenous, rapid, versatile and amenable to automation for the discovery of new neurotransmitter transporter inhibitors. Ó 2003 Elsevier Inc. All rights reserved. Keywords: Glycine; Taurine; Transporter; Cytostar-T scintillating microplates; Proximity assay; JAR cells
Neurotransmitter transporters are essential for synaptic transmission in the central nervous system. The Naþ /Cl -dependent neurotransmitter transporter family is composed of several members such as monoamines (dopamine, norepinephrine, serotonin) and inhibitory transmitters (GABA,3 glycine) transporters [1]. Two glycine transporters, GlyT1 and GlyT2, have
*
been cloned from various species and shown to have only 50% homology at the amino acid level and to be differentially expressed in rodent central nervous system [2–8]. GlyT2 has a high expression restricted to areas innervated by glycinergic fibers such as spinal cord and brainstem, whereas GlyT1 distribution is more widespread and includes brain regions such as cortex,
Corresponding author. Fax: +1-215-652-2075. E-mail address: [email protected] (C. Sur). 1 Present address: Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN 37232-6600, USA. 2 Present address: National Institute of Health, Bethesda, MD 20892, USA. 3 Abbreviations used: ALX-5407, (N-[3-(40 -fluorophenyl)-3-(40 -phenylphenoxy)propyl])sarcosine; DAPI, 4,6-diamidino-2-phenylindole; DIDS, 4,40 diisothiocyanatostilbene-2,20 -disulfonic acid; DMA, 5-N 0 N-dimethyl-amiloride; DMSO, dimethyl sulfoxide; FITC, fluorescein isothiocyanate; GABA, c-aminobutyric acid; GlyT, glycine transporter; JAR, human placental choriocarcinoma cells; MeAIB, methylaminoisobutyric acid; NMDA, N-methylD -aspartate; Org-24598, N-methyl-N-[3-[(4-trifluoromethyl)phenoxy]-3-phenyl-propyl]glycine; PBS, phosphate-buffered saline; SPA, scintillation proximity assay. 0003-2697/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0003-2697(03)00431-7
32
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
thalamus, hippocampus, striatum, and hypothalamus [3–6]. This differential expression pattern has led to the suggestion that GlyT2 is involved in the reuptake of glycine at glycinergic synapses, whereas GlyT1 is implicated in controlling glycine concentration at excitatory synapses harboring NMDA receptors [3–5]. Recent electrophysiology studies in rat hippocampus and brain stem slices have indeed demonstrated that modulation of GlyT1 activity impacts NMDA-dependent excitatory postsynaptic potentials [9,10]. These observations thus suggest that blockade of glycine uptake by GlyT1 may increase NMDA receptor activity and offer a pharmacological approach to the treatment of disorders proposed to be due to NMDA hypofunction such as schizophrenia [11]. Presently, methods to study neurotransmitter uptake in cultured cells, involve multiple steps such as addition of radiolabeled probe, aspiration of reaction medium, washing and solubilization of cells, and addition of scintillation cocktail [12–15]. Furthermore, these approaches allow the determination of only one time point for each well of the plate. To simplify the detection of neurotransmitter transport, we established a homogenous high-throughput uptake assay with realtime measurement of neurotransmitter accumulation into cells. To this end we benefited from the technology of Cytostar-T scintillating microplates from Amersham. Cells are grown on a scintillant-containing plastic base that is activated only by molecules of radioactive tracer that are in close proximity (i.e., molecules bound to the base or accumulated into cells) (Fig. 1). The light emitted by the scintillating base can be detected with a multiwell scintillation counter and this operation can be repeated at multiple time points [16]. This report demonstrates the use of scintillating microplates technology in establishing a homogenous and versatile assay for the detection of [14 C]glycine and [14 C]taurine transport by endogenously expressed Naþ / Cl -dependent transporters in human choriocarcinoma cells.
Materials and methods Materials [14 C]glycine (112.7 mCi/mmol) and [14 C]taurine (119.2 mCi/mmol) were obtained from NEN Life Science Product Inc. (Boston, MA). All chemicals were purchased from Sigma–Aldrich (St Louis, MO) except L - and D serine, D -cycloserine, taurine, and c-aminobutyric acid which were purchased from Tocris (Ellisville, MO). ALX5407, Org-24598, and 4-benzyloxy-3,5-dimethoxy-N[(1-di-methylaminocyclopentyl)methyl]benzamide were synthesized at Merck & Co Inc. according to published structures [14,15,17]. Cell culture Human placental choriocarcinoma cells (ATCC No. HTB-144) were from the American Type Culture Collection (Manassas, VA). JAR cells were cultured in 96-well Cytostar plates in RPMI 1640 medium (Life Technologies, MD) containing 10% fetal calf serum in the presence of penicillin (100 lg/ml) and streptomycin (100 lg/ml) as previously reported [18]. Cells were plated at a density of 1.3 104 cells/well except as otherwise indicated and grown at 37 °C in a humidified atmosphere of 5% CO2 for 40–48 h unless otherwise indicated. Uptake measurement Culture medium was removed from Cytostar plate and JAR cells were incubated with 30 ll of TB1A buffer (120 mM NaCl, 2 mM KCl, 1 mM CaCl2 , 1 mM MgCl2, 10 mM Hepes, 5 mM L -alanine, pH 7.5 adjusted with 1 M Tris base) with or without drug for 1 min. Then 30 ll of [14 C]glycine diluted in TB1A was added to each well to give a final concentration of 10 lM unless otherwise specified. After incubation at room temperature for the desired time, usually 2–3 h, sealed 96-well Cytostar plates were counted on a Top Count (Packard). Nonspecific uptake of [14 C]glycine was determined in the presence of 10 mM cold glycine. [14 C]taurine uptake experiments were performed according to the same protocol except that 10 mM cold taurine was used to determine nonspecific uptake. Immunocytochemistry
Fig. 1. Principle of the scintillation proximity assay used to measured [14 C]glycine uptake. Human JAR cells were plated on 96-well Cytostar plates (Amersham) that have a scintillating matrix at their base. Only radioactive molecules in contact or in very close proximity of this base (i.e., in the cells) will stimulate the scintillant and produce light that can be detected by the photomultipliers of the Top Count.
GlyT1 expression in JAR cells was visualized with selective goat anti-GlyT1 polyclonal antibody (Chemicon, CA). Briefly, JAR cells cultured in Chamber Slide (Nalge Nunc, IL) were rinsed and fixed with fresh paraformaldehyde (4%) for 2 h at room temperature. Cells were washed with phosphate-buffered saline (PBS, pH 7.4) and incubated overnight at 4 °C with GlyT1 antibody diluted
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
33
1/1000–1/5000 in PBS containing 0.25% (w/v) Triton X100 and 0.12% (w/v) gelatin [19]. Bound primary antibodies were detected with a donkey anti-goat antibody coupled to fluorescein isothiocyanate (FITC) (Jackson Immunoresearch, PA). Sections were counterstained with 4,6-diamidino-2-phenylindole (DAPI, Molecular Probes, CA) and examined microscopically and images were digitally acquired using a Nikon E1000 epifluorescence microscope and Micromax charge-coupled device camera (Princeton Instruments, San Diego, CA). RT-PCR Cells were grown to confluence in a 75-cm2 tissue culture flask. Cells were lysed directly in the flask using 2 ml Trizol Reagent (Life Technologies Cat. No. 15596). RNA was then phenol/chloroform extracted according to current protocols. Total RNA was used at a final concentration of 20 ng/lL to generate cDNA with an oligo d(T) primer and Multiscribe Reverse Transcriptase (Applied Biosystems); 5 lL of the resulting cDNA was added to a PCR using 0.025 U/lL Amplitaq Gold (Applied Biosystems) in 5.5 mM MgCl2 and isotype-specific primers at a final concentration of 200 nM. Primer sequences were as follows: huGlyT1a F0 : 50 -ATGGTAGGAAAAGGTGCCAA AGGGATGCTG-30 ; huGlyT1bc F0 : 50 -ATGGCCGCGGCTCATGGACC TGTGGCC-30 ; and huGlyT1abc R0 : 50 -GAGGTATGGGAAGCGCCA GACATTG-30 . The amplified cDNA was separated on a 1.5% Agar/ TBE gel and visualized with ethidium bromide. Data analysis All experiments were performed with Cytostar (Amersham, NJ) except for a few preliminary experiments carried out with Flash Plates scintillating microplates (Perkin–Elmer, MA). Experiments were performed at least in triplicate and data were analyzed by nonlinear regression analysis using Prism software (GraphPad, CA).
Results Characterization of GlyT1 expression in JAR cells Expression of GlyT1 in human choriocarcinoma cells was visualized with a specific antibody. As shown in Fig. 2A, every JAR cell exhibited a green fluorescent signal associated with cytoplasmic membrane and cytoplasm, indicating the presence of GlyT1 in these cellular compartments. Several variants of human GlyT1
Fig. 2. Expression of GlyT1 in human placental choriocarcinoma cells. (A) Immunofluorescent labeling with a specific GlyT1 antibody revealed GlyT1 expression (FITC, green) in cultured JAR cells. Omission of primary antibody (insert) resulted in absence of GlyT1 signal. JAR cells nuclei were visualized with DAPI (blue). (B) RT-PCR analysis of mRNA expressed by JAR cells with primers specific for GlyT1 isoform variants indicated that GlyT1a but not GlyT1b,c isoform is present in the human choriocarcinoma placental cells; bp, basepair. (C) Initial experiments with [3 H]glycine demonstrated that 95 1% (n ¼ 6) of JAR glycine uptake is Naþ dependent, and methylaminoisobutyric acid (MeAIB) and L -alanine inhibited uptake to 74 10% (n ¼ 6) and 77 4% (n ¼ 16) of control values, respectively. Reciprocally, sarcosine at 1 mM decreased transport to 18 4% (n ¼ 7) of control value, whereas the GlyT2 inhibitor 4-benzyloxy-3,5-dimethoxy-N-[(1-di-methylaminocyclopentyl)methyl]benzamide did not antagonize glycine transport (111 3% of control, n ¼ 3). Note that GlyT2 compound data were generated with [14 C]glycine. Data are the mean SE of n determinations.
(GlyT1a,b, and c) have been reported and shown to arise from both different promoter usage and alternative splicing [2]. Results from RT-PCR experiments using primers in exon 1 and exon 2 to specifically detect GlyT1a
34
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
and GlyT1b,c, respectively, showed a single band with GlyT1a probe and no signal with GlyT1b,c primer (Fig. 2B). Initial pharmacological studies with [3 H]glycine carried out as reported by Liu and collaborators [18] indicated that glycine uptake by JAR cells is reduced to 5 1% (n ¼ 6) of control value when Naþ is replaced by Liþ , and methylaminoisobutyric acid (MeAIB) and L alanine inhibited uptake to 74 10% (n ¼ 6) and 77 4% (n ¼ 16) of control values, respectively (Fig. 2C). In addition, glycine transport was reduced to 18 4% (n ¼ 7) of control uptake in the presence of 1 mM sarcosine (Fig. 2B), a selective GlyT1 transport inhibitor [2–4] whereas 4-benzyloxy-3,5-dimethoxy-N-[(1-di-methylaminocyclopentyl)methyl]benzamide, a potent and selective GlyT2 antagonist (hGlyT2 IC50 : 16 nM) [17] did not significantly reduce glycine uptake by JAR cell at concentration up to 10 lM (111 4% of control, n ¼ 3; Fig. 2). Altogether, these results corroborated a previous report [18] that GlyT1 and system A are the major glycine transport systems and account for 75 and 20% of total glycine uptake by JAR cells, respectively. Thus, all future [14 C]glycine uptake experiments were performed in the presence of 5 mM L -alanine to block any contribution from system A to total glycine transport.
Characterization of [14 C]glycine uptake The kinetics of [14 C]glycine (10 lM) transport by JAR cells were determined by counting a given 96-well plate at multiple time points. As shown in Fig. 4A, specific uptake was linear over time, whereas total uptake began to plateau at 3 h. The unusual stability of the nonspecific signal over time is a direct consequence of the scintillating microplate technology as it is mainly due to molecules of [14 C]glycine colliding with the scintillating base of the well. Future uptake experiments were carried out between 2 and 3 h to generate a large and stable specific signal. Uptake of increasing concentration of glycine revealed that JAR cellsÕ GlyT1 activity is saturable with a Michaelis–Menten constant Km of 26.8 3 lM and maximal velocity (Vmax ) of 20 cpm/well/min in 96-well Cytostar microplates (Fig. 4B).
Effects of cell density and culture time on [14 C]glycine uptake As a first step in establishing a scintillation proximity assay (SPA) for [14 C]glycine transport, we determined the optimal culture conditions of JAR cells in 96-well microplates. As shown in Fig. 3, JAR cells were plated at densities ranging from 103 to 3 104 cells per well and [14 C]glycine uptake was measured after 28 and 48 h. The largest specific signal was obtained with JAR cells plated at 104 cells/well and cultured for 2 days (Fig. 3); so these conditions were chosen for subsequent experiments.
Fig. 3. Effect of cell density on [14 C]glycine uptake. JAR cells were plated at the indicated density in 96-well plates and the uptake of 10 lM [14 C]glycine in the presence of alanine (5 mM) was measured during 75 min after 28 and 48 h of culture. Data are the mean SE of four determinations.
Fig. 4. Characterization of [14 C]glycine uptake in SPA assay. (A) Accumulation of [14 C]glycine (10 lM) in JAR cells was followed over time by counting the 96-well plate with a Top Count at the indicated time points. The total signal started to plateau after 3 h and at 4 h specific uptake accounts for around 50% of signal. Note that nonspecific signal is constant over time as it results from extracellular [14 C]glycine molecules colliding with the scintillating base. Data were generated in the presence of alanine (5 mM) and are the mean SE of four wells from a representative plate. (B) Saturation experiments in the presence of alanine (5 mM) showed that [14 C]glycine uptake by JAR cells grown in 96-well Cytostar plates is saturable. Michaelis– Menten analysis indicated that Km ¼ 26.8 3.0 lM and Vmax ¼ 140 20 cpm/well/min. Data are the mean SE of four independent experiments. In A and B, nonspecific uptake was determined in the presence of 10 mM glycine.
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
Pharmacological profile of [14 C]glycine uptake To further validate our SPA, the effects of various amino acids and compounds known to interact with GlyT1 were investigated. As reported in Table 1, close congeners of glycine were weak inhibitors of [14 C]glycine uptake mediated by GlyT1, whereas amino acids and neurotransmitters such as L -proline, D -serine, taurine, and GABA were inactive. In contrast, known potent and selective GlyT1 inhibitors such as ALX-5407 [15] and Org-24598 [14] fully antagonized [14 C]glycine uptake by JAR cells in 96-well Cytostar plates with affinities of 2.8 0.6 and 6.9 0.9 nM, respectively (Fig. 5). Glycine and sarcosine, another substrate of GlyT1, blocked [14 C]glycine transport with potencies of 34.0 2.5 and 37.5 4.6 lm, respectively (Fig. 5). Furthermore, under
Table 1 Pharmacological profile of [14 C]glycine uptake by JAR cells cultured in SPA assay Compounds (1 mM)
[14 C]glycine uptake (% of control)
Glycine methyl ester Glycine ethyl ester L -Serine D -Serine D -Cycloserine L -Proline Taurine GABA MelA
22 5 24 7 76 3 109 17 98 14 94 14 99 9 97 8 111 12
35
these experimental conditions the potent and selective human GlyT2 inhibitor [17] 4-benzyloxy-3,5-dimethoxyN-[(1-di-methylaminocyclopentyl)methyl]benzamide did not antagonize [14 C]glycine transport by JAR cells (Fig. 5). Effect of Cl and Naþ flux blockers on [14 C]glycine uptake The transport of glycine by GlyT1 is known to be dependent on Naþ and Cl gradients across cell cytoplasmic membranes so the effect of compounds known to interact with these ionic fluxes were investigated. Blockade of Kþ /Cl cotransporter by furosemide (10 lM) slightly reduced [14 C]glycine uptake by 13%. Similarly, antagonism of Naþ /HCO3 cotransporter and Naþ /Hþ antiporter by 4,40 -diisothiocyanatostilbene2,20 -disulfonic acid (DIDS) and 5-N 0 N-dimethyl-amiloride (DMA), respectively, induced a modest reduction of GlyT1 activity (Fig. 6). In contrast, ouabain and especially benzamil strongly inhibited glycine transport by JAR cells through their blockade of Naþ /Kþ ATPase and Naþ /Ca2þ exchanger, respectively (Fig. 6). Another important aspect to consider when developing a screening assay is the impact of DMSO concentration. As shown in Fig. 6, [14 C]glycine uptake by JAR cells remained unaffected by concentrations of DMSO up to 2% and decreased by around 50% by a 5% DMSO solution.
Data are the mean SE of four experiments performed in the presence of 5 mM alanine.
Fig. 5. Pharmacology of [14 C]glycine uptake by human JAR cells. Experiments performed in the presence of alanine (5 mM) demonstrated that ALX-5407 (d), Org-24598 (s), sarcosine (), and glycine (j) fully inhibit transport of 10 lM [14 C]glycine, whereas taurine (m) and GlyT2 antagonist 4-benzyloxy-3,5-dimethoxy-N-[(1-di-methylaminocyclopentyl)methyl]benzamide (r) were inactive. Nonlinear regression analysis revealed that ALX-5407 and Org-24598 are high affinity antagonists with IC50 values of 2.8 0.6 and 6.9 0.9 nM, respectively. Glycine and sarcosine inhibited transport with lower potency as indicated by IC50 values of 34.0 2.5 and 37.5 4.6 lM, respectively. Nonspecific uptake was determined in the presence of 10 mM glycine and data are the mean SE of three experiments.
Fig. 6. Effect of Cl and Naþ flux inhibitors on [14 C]glycine uptake. Compounds known to block Cl transport such as furosemide (10 lM) did not strongly antagonize the uptake of [14 C]glycine (87 4% of control) by JAR cells. Similarly, antagonism of Naþ /HCO3 cotransporter by 4,40 -diisothiocyanatostilbene-2,20 -disulfonic acid (DIDS) produced a modest reduction of GlyT1 activity. In contrast inhibition of Naþ /Kþ ATPase and Naþ /Hþ and Naþ /Ca2þ exchangers by ouabain (10 lM) and benzamil (10 lM), respectively, reduced [14 C]glycine accumulation in JAR cells to 56 6 and 5 1% of control, respectively. The Naþ /Hþ antiport inhibitor DMA (10 lM) was less potent and reduced [14 C]glycine to 78 5% of control level. Non specific uptake was determined in the presence of 10 mM glycine. Data are the mean SE of six determinations performed in the presence of 5 mM alanine.
36
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
Characterization of [14 C]taurine uptake by JAR cells
Discussion
As GlyT1 belongs to a large family of Naþ /Cl - dependent neurotransmitter transporters, we investigated the versatility of our Cytostar SPA assay by studying the activity of Naþ /Cl -dependent taurine transporter expressed in JAR cells [20]. As shown in Fig. 7, JAR cells exhibited saturation kinetics of [14 C]taurine uptake with an apparent Km of 10.2 1.7 lM and a Vmax in Cytostar microplates of 174 25 cpm/well/min. Pharmacological experiments revealed that ALX-5407 and Org-24598 antagonized [14 C]taurine uptake with potencies of 522 83 nM and 14.9 1.5 lM, respectively (Fig. 7). Taurine inhibited its own transport with a Ki of 9.2 2.3 lM, whereas glycine and sarcosine were inactive (Fig. 7). Consistent with [14 C]glycine uptake results, taurine transporter activity was impaired by ouabain (50 4% at 10 lM, n ¼ 5) and benzamil (15 1% at 10 lM, n ¼ 5) and by DMSO concentration of more than 2% (not shown).
Since its development [16] the Cytostar-T scintillating microplate has been used to detect the cellular accumulation of various radioactive tracers such as [14 C]thymidine, [45 Ca], or [14 C]glycocholate to study cell growth, ionotropic glutamate 4 receptor activity, or bile acid transport, respectively [16,21,22]. We have now used this proximity assay technology to measure [14 C]glycine transport in human placenta choriocarcinoma cells. In agreement with previous work [18] our results confirm that 95% of glycine transport into JAR cells is a Naþ dependent process mediated mostly by an active system A and GlyT1. We also demonstrated that JAR cells express only the GlyT1a isoform, an observation consistent with the reported lack of GlyT1c in human placenta [2]. The lack of activity of the potent and selective GlyT2 inhibitor 4-benzyloxy-3,5-dimethoxy-N-[(1-di-methylaminocyclopentyl)methyl]benzamide [17] indicates that this transporter is not expressed in JAR cells or is expressed at a level below the sensitivity of our uptake assay. Following isolation of GlyT1 activity in JAR cells by addition of L -alanine in transport buffer [18], we showed that [14 C]glycine uptake was dependent on cell density and culture time, whereas temperature (22 °C versus 37 °C) had little influence (not shown). The validity of the SPA was established by pharmacological studies that showed GlyT1-mediated [14 C]glycine uptake to be saturable and of high affinity with a Km value consistent with previous reports in JAR cells (15 lM; [18]) and recombinant systems (70–117 lM; [2,3,23]). The rank order of potency of various amino acids and neurotransmitters agreed with the reported effects or lack thereof of these molecules at GlyT1 transporter [2,4,18]. Finally, full and concentration-dependent inhibition of [14 C]glycine uptake in JAR cells by the potent and selective GlyT1 inhibitors ALX-5407 [14] and Org-24598 [15] clearly demonstrated the validity of our SPA. The versatility of this SPA was demonstrated by our ability to examine taurine transporter activity in JAR cells following minor modification to the uptake protocol ([14 C]taurine instead of [14 C]glycine). Kinetic constant for [14 C]taurine uptake determined in SPA (Km : 10 lM) was comparable to values obtained in JAR cells (2.3 lM; [20]) and in Hela cells expressing cloned human taurine transporter (5.9 lM; [24]). Since taurine and glycine transporters belong to the same family of Naþ / Cl -dependent neurotransmitter transporters, our taurine assay can be used for counterscreening purposes and to determine the selectivity of GlyT1 inhibitors. Using this approach we showed that glycine, sarcosine, and Org-24598 were weak blockers of taurine uptake, whereas ALX-5407 was only 70-fold selective for GlyT1 compared to taurine transporter. Finally, our [14 C]taurine uptake SPA allowed specificity assessment of new transporter inhibitors if one assumes that compounds
Fig. 7. Characterization of a SPA for [14 C]taurine uptake by human JAR cells. (A) Saturation experiments showed that specific [14 C]taurine uptake (gray triangles) by JAR cells grown in 96-well Cytostar plates is saturable and accounts for 90% of total taurine uptake (black triangles). Michaelis–Menten analysis indicated that Km ¼ 10.2 1.7 lM and Vmax ¼ 174 25 cpm/well/min. Data are the mean SE of four experiments, and nonspecific uptake (white triangles) was determined in the presence of 10 mM taurine. (B) Competition experiments demonstrated that ALX-5407 (d) fully inhibits transport of 10 lM [14 C]taurine with a IC50 ¼ 522 83 nM whereas taurine (m) and Org24598 (s) were less potent with IC50 values of 21.1 5.3 and 14.9 1.5 lM, respectively. Sarcosine () and glycine (j) did not significantly block [14 C]taurine uptake by JAR cells. Nonspecific uptake was determined in the presence of 10 mM taurine, and data are the mean SE of three or four experiments.
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
inhibiting [14 C]glycine and [14 C]taurine with similar potency are likely to do so through disruption of Naþ gradient and/or other vital cellular functions. In conclusion, we showed that using Cytostar-T scintillating microplates it is possible to accurately and efficiently investigate Naþ /Cl -dependent neurotransmitter transporter activities. This coupled to the fact that this scintillation proximity assay is amenable to automation for screening of large chemical collections should facilitate the discovery of new selective transporter inhibitors. References [1] N. Nelson, The family of Naþ /Cl- neurotransmitter transporters, J. Neurochem. 71 (1998) 1785–1803. [2] K.M. Kim, S.F. Kingsmore, H. Han, T.L. Yang-Feng, N. Godinot, M.F. Seldin, M.G. Caron, B. Giros, Cloning of the human glycine transporter type 1: molecular and pharmacological characterization of novel isoform variants and chromosomal localization of the gene in the human and mouse genomes, Mol. Pharmacol. 45 (1994) 608–617. [3] B. Borowsky, E. Mezey, B.J. Hoffman, Two glycine transporter variants with distinct localization in the CNS and peripheral tissues are encoded by a common gene, Neuron 10 (1993) 851– 863. [4] K.E. Smith, L.A. Borden, P.R. Hartig, T. Branchek, R.L. Weinsbank, Cloning and expression of a glycine transporter reveal colocalization with NMDA receptors, Neuron 8 (1992) 927–935. [5] F. Zafra, J. Gomeza, L. Olivares, C. Aragon, C. Gimenez, Regional distribution and developmental variation of the glycine transporters GLYT1 and GLYT2 in the rat CNS, Eur. J. Neurosci. 7 (1995) 1342–1352. [6] F. Zafra, C. Aragon, L. Olivares, N.C. Danbolt, C. Gimenez, J. Storm-Mathisen, Glycine transporters are differentially expressed among CNS cells, J. Neurosci. 15 (1995) 3952–3969. [7] Q.-R. Liu, B. Lopez-Corcuera, S. Mandiyan, H. Nelson, N. Nelson, Cloning and expression of a spinal cord- and brainspecific glycine transporter with novel structural features, J. Biol. Chem. 268 (1993) 22802–22808. [8] J.A. Morrow, I.T. Collie, D.R. Dunbar, G.B. Walker, M. Shahid, D.R. Hill, Molecular cloning and functional expression of the human glycine transporter GlyT2 and chromosomal localisation of the gene in the human genome, FEBS Lett. 439 (1998) 334–340. [9] R. Bergeron, T.M. Meyer, J.T. Coyle, R.W. Greene, Modulation of N-methyl-D -aspartate receptor function by glycine transport, Proc. Natl. Acad. Sci. USA 95 (1998) 15730–15734. [10] A.J. Berger, S. Dieudonne, P. Ascher, Glycine uptake governs glycine site occupancy at NMDA receptors of excitatory synapses, J. Neurophysiol. 80 (1998) 3336–3340. [11] G. Tsai, J.T. Coyle, Glutamatergic mechanisms in schizophrenia, Annu. Rev. Pharmacol. Toxicol. 42 (2002) 165–179.
37
[12] K. Sakata, K. Sato, P. Schloss, H. Betz, S. Shimada, M. Tohyama, Characterization of glycine release mediated by glycine transporter 1 stably expressed in HEK-293 cells, Mol. Brain Res. 49 (1997) 89–94. [13] C. Sur, H. Betz, P. Schloss, A single serine residue controls the cation dependence of substrate transport by the rat serotonin transporter, Proc. Natl. Acad. Sci. USA 94 (1997) 7639– 7644. [14] A. Brown, I. Carlyle, J. Clark, W. Hamilton, S. Gibson, G. McGarry, S. McEachen, D. Rae, S. Thorn, G. Walker, Discovery and SAR of Org 24598-A selective glycine uptake inhibitor, Bioorg. Med. Chem. Lett. 11 (2001) 2007–2009. [15] B.N. Atkinson, S.C. Bell, M. De Vivo, L.R. Kowalski, S.M. Lechner, V.I. Ognyanov, C.-S. Tham, C. Tsai, J. Jia, D. Ashton, M.A. Klitenick, ALX 5407: a potent, selective inhibitor of the hGlyT1 glycine transporter, Mol. Pharmacol. 60 (2001) 1414– 1420. [16] R. Graves, R. Davies, P. Owen, M. Clynes, I. Cleary, G.O. Beirne, An homogeneous assay for measuring the uptake and efflux of radiolabelled drugs in adherent cells, J. Biochem. Biophys. Methods 34 (1997) 177–187. [17] W.L. Caulfield, I.T. Collie, R.S. Dickins, O. Epemolu, R. McGuire, D.R. Hill, G. McVey, J.R. Morphy, Z. Rankovic, H. Sundaram, The first potent and selective inhibitors of the glycine transporter type 2, J. Med. Chem. 44 (2001) 2679– 2682. [18] W. Liu, F.H. Leibach, V. Ganapathy, Characterization of the glycine transport system GLYT1 in human placental choriocarcinoma cells (JAR), Biochim. Biophys. Acta 1194 (1994) 176– 184. [19] C. Sur, H. Betz, P. Schloss, Immunocytochemical detection of the serotonin transporter in rat brain, Neuroscience 73 (1996) 217– 231. [20] P. Kulanthaivel, D.R. Cool, S. Ramamoorthy, V.B. Mahesh, F.H. Leibach, V. Ganapathy, Transport of taurine and its regulation by protein kinase C in the JAR human placental choriocarcinoma cell line, Biochem. J. 277 (1991) 53–58. [21] A. Cushing, M.J. Price-Jones, R. Graves, A.J. Harris, K.T. Hughes, D. Bleakman, D. Lodge, Measurement of calcium flux through ionotropic glutamate receptors using Cytostar-T scintillating microplates, J. Neurosci. Methods 90 (1999) 33– 36. [22] H. Bonge, S. Hallen, J. Fryklund, J.E. Sjostrom, Cytostar-T scintillating microplate assay for measurement of sodium-dependent bile acid uptake in transfected HEK-293 cells, Anal. Biochem. 282 (2000) 94–101. [23] H.J. Herdon, F.M. Godfrey, A.M. Brown, S. Coulton, J.R. Evans, W.J. Cairns, Pharmacological assessment of the role of the glycine transporter GlyT-1 in mediating high-affinity glycine uptake by rat cerebral cortex and cerebellum synaptosomes, Neuropharmacology 41 (2001) 88–96. [24] S. Ramamoorthy, F.H. Leibach, V.B. Mahesh, H. Han, T. Yang-Feng, R.D. Blakely, V. Ganapathy, Functional characterization and chromosomal localization of a cloned taurine transporter from human placenta, Biochem. J. 800 (1994) 893–900.
Development of a scintillation proximity assay for analysis of Naþ /Cl -dependent neurotransmitter transporter activity Jacinta B. Williams,a Pierre J. Mallorga,a Wei Lemaire,a David L. Williams,a Sang Na,a Smita Patel,b Jeffrey P. Conn,a,1 Douglas J. Pettibone,a Christopher Austin,a,2 and Cyrille Sura,* a
b
Department of Neuroscience West Point, Merck & Co. Inc., West Point, PA 19486, USA Neuroscience Research Centre, Merck Sharp & Dohme, Terlings Park, Harlow, Essex CM20 2QR, UK Received 14 February 2003
Abstract Human placental choriocarcinoma (JAR) cells endogenously expressing glycine transporter type 1a (GlyT1a) have been cultured in 96-well scintillating microplates to develop a homogenous screening assay for the detection of GlyT1 antagonists. In these microplates uptake of [14 C]glycine was time dependent and saturable with a Michaelis–Menten constant (Km ) of 27 3 lM. The GlyT1 transport inhibitors sarcosine, ALX-5407, and Org-24598 were tested and shown to block [14 C]glycine uptake with expected IC50 values of 37.5 4.6 lM, 2.8 0.6 nM, and 6.9 0.9 nM, respectively. The [14 C]glycine uptake process was sensitive to membrane Naþ gradient as blockade of membrane Naþ /Kþ -ATPase by ouabain or Naþ exchanger by benzamil-disrupted glycine accumulation in JAR cells. Glycine influx was not affected by concentration of dimethyl sulfoxide up to 2%. The versatility of this technological approach was further confirmed by the characterization of a saturable [14 C]taurine uptake in JAR cells. Taurine transport was of high affinity with a Km of 10.2 1.7 lM and fully inhibited by ALX-5407 (IC50 ¼ 522 83 nM). The developed assay is homogenous, rapid, versatile and amenable to automation for the discovery of new neurotransmitter transporter inhibitors. Ó 2003 Elsevier Inc. All rights reserved. Keywords: Glycine; Taurine; Transporter; Cytostar-T scintillating microplates; Proximity assay; JAR cells
Neurotransmitter transporters are essential for synaptic transmission in the central nervous system. The Naþ /Cl -dependent neurotransmitter transporter family is composed of several members such as monoamines (dopamine, norepinephrine, serotonin) and inhibitory transmitters (GABA,3 glycine) transporters [1]. Two glycine transporters, GlyT1 and GlyT2, have
*
been cloned from various species and shown to have only 50% homology at the amino acid level and to be differentially expressed in rodent central nervous system [2–8]. GlyT2 has a high expression restricted to areas innervated by glycinergic fibers such as spinal cord and brainstem, whereas GlyT1 distribution is more widespread and includes brain regions such as cortex,
Corresponding author. Fax: +1-215-652-2075. E-mail address: [email protected] (C. Sur). 1 Present address: Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN 37232-6600, USA. 2 Present address: National Institute of Health, Bethesda, MD 20892, USA. 3 Abbreviations used: ALX-5407, (N-[3-(40 -fluorophenyl)-3-(40 -phenylphenoxy)propyl])sarcosine; DAPI, 4,6-diamidino-2-phenylindole; DIDS, 4,40 diisothiocyanatostilbene-2,20 -disulfonic acid; DMA, 5-N 0 N-dimethyl-amiloride; DMSO, dimethyl sulfoxide; FITC, fluorescein isothiocyanate; GABA, c-aminobutyric acid; GlyT, glycine transporter; JAR, human placental choriocarcinoma cells; MeAIB, methylaminoisobutyric acid; NMDA, N-methylD -aspartate; Org-24598, N-methyl-N-[3-[(4-trifluoromethyl)phenoxy]-3-phenyl-propyl]glycine; PBS, phosphate-buffered saline; SPA, scintillation proximity assay. 0003-2697/$ - see front matter Ó 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0003-2697(03)00431-7
32
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
thalamus, hippocampus, striatum, and hypothalamus [3–6]. This differential expression pattern has led to the suggestion that GlyT2 is involved in the reuptake of glycine at glycinergic synapses, whereas GlyT1 is implicated in controlling glycine concentration at excitatory synapses harboring NMDA receptors [3–5]. Recent electrophysiology studies in rat hippocampus and brain stem slices have indeed demonstrated that modulation of GlyT1 activity impacts NMDA-dependent excitatory postsynaptic potentials [9,10]. These observations thus suggest that blockade of glycine uptake by GlyT1 may increase NMDA receptor activity and offer a pharmacological approach to the treatment of disorders proposed to be due to NMDA hypofunction such as schizophrenia [11]. Presently, methods to study neurotransmitter uptake in cultured cells, involve multiple steps such as addition of radiolabeled probe, aspiration of reaction medium, washing and solubilization of cells, and addition of scintillation cocktail [12–15]. Furthermore, these approaches allow the determination of only one time point for each well of the plate. To simplify the detection of neurotransmitter transport, we established a homogenous high-throughput uptake assay with realtime measurement of neurotransmitter accumulation into cells. To this end we benefited from the technology of Cytostar-T scintillating microplates from Amersham. Cells are grown on a scintillant-containing plastic base that is activated only by molecules of radioactive tracer that are in close proximity (i.e., molecules bound to the base or accumulated into cells) (Fig. 1). The light emitted by the scintillating base can be detected with a multiwell scintillation counter and this operation can be repeated at multiple time points [16]. This report demonstrates the use of scintillating microplates technology in establishing a homogenous and versatile assay for the detection of [14 C]glycine and [14 C]taurine transport by endogenously expressed Naþ / Cl -dependent transporters in human choriocarcinoma cells.
Materials and methods Materials [14 C]glycine (112.7 mCi/mmol) and [14 C]taurine (119.2 mCi/mmol) were obtained from NEN Life Science Product Inc. (Boston, MA). All chemicals were purchased from Sigma–Aldrich (St Louis, MO) except L - and D serine, D -cycloserine, taurine, and c-aminobutyric acid which were purchased from Tocris (Ellisville, MO). ALX5407, Org-24598, and 4-benzyloxy-3,5-dimethoxy-N[(1-di-methylaminocyclopentyl)methyl]benzamide were synthesized at Merck & Co Inc. according to published structures [14,15,17]. Cell culture Human placental choriocarcinoma cells (ATCC No. HTB-144) were from the American Type Culture Collection (Manassas, VA). JAR cells were cultured in 96-well Cytostar plates in RPMI 1640 medium (Life Technologies, MD) containing 10% fetal calf serum in the presence of penicillin (100 lg/ml) and streptomycin (100 lg/ml) as previously reported [18]. Cells were plated at a density of 1.3 104 cells/well except as otherwise indicated and grown at 37 °C in a humidified atmosphere of 5% CO2 for 40–48 h unless otherwise indicated. Uptake measurement Culture medium was removed from Cytostar plate and JAR cells were incubated with 30 ll of TB1A buffer (120 mM NaCl, 2 mM KCl, 1 mM CaCl2 , 1 mM MgCl2, 10 mM Hepes, 5 mM L -alanine, pH 7.5 adjusted with 1 M Tris base) with or without drug for 1 min. Then 30 ll of [14 C]glycine diluted in TB1A was added to each well to give a final concentration of 10 lM unless otherwise specified. After incubation at room temperature for the desired time, usually 2–3 h, sealed 96-well Cytostar plates were counted on a Top Count (Packard). Nonspecific uptake of [14 C]glycine was determined in the presence of 10 mM cold glycine. [14 C]taurine uptake experiments were performed according to the same protocol except that 10 mM cold taurine was used to determine nonspecific uptake. Immunocytochemistry
Fig. 1. Principle of the scintillation proximity assay used to measured [14 C]glycine uptake. Human JAR cells were plated on 96-well Cytostar plates (Amersham) that have a scintillating matrix at their base. Only radioactive molecules in contact or in very close proximity of this base (i.e., in the cells) will stimulate the scintillant and produce light that can be detected by the photomultipliers of the Top Count.
GlyT1 expression in JAR cells was visualized with selective goat anti-GlyT1 polyclonal antibody (Chemicon, CA). Briefly, JAR cells cultured in Chamber Slide (Nalge Nunc, IL) were rinsed and fixed with fresh paraformaldehyde (4%) for 2 h at room temperature. Cells were washed with phosphate-buffered saline (PBS, pH 7.4) and incubated overnight at 4 °C with GlyT1 antibody diluted
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
33
1/1000–1/5000 in PBS containing 0.25% (w/v) Triton X100 and 0.12% (w/v) gelatin [19]. Bound primary antibodies were detected with a donkey anti-goat antibody coupled to fluorescein isothiocyanate (FITC) (Jackson Immunoresearch, PA). Sections were counterstained with 4,6-diamidino-2-phenylindole (DAPI, Molecular Probes, CA) and examined microscopically and images were digitally acquired using a Nikon E1000 epifluorescence microscope and Micromax charge-coupled device camera (Princeton Instruments, San Diego, CA). RT-PCR Cells were grown to confluence in a 75-cm2 tissue culture flask. Cells were lysed directly in the flask using 2 ml Trizol Reagent (Life Technologies Cat. No. 15596). RNA was then phenol/chloroform extracted according to current protocols. Total RNA was used at a final concentration of 20 ng/lL to generate cDNA with an oligo d(T) primer and Multiscribe Reverse Transcriptase (Applied Biosystems); 5 lL of the resulting cDNA was added to a PCR using 0.025 U/lL Amplitaq Gold (Applied Biosystems) in 5.5 mM MgCl2 and isotype-specific primers at a final concentration of 200 nM. Primer sequences were as follows: huGlyT1a F0 : 50 -ATGGTAGGAAAAGGTGCCAA AGGGATGCTG-30 ; huGlyT1bc F0 : 50 -ATGGCCGCGGCTCATGGACC TGTGGCC-30 ; and huGlyT1abc R0 : 50 -GAGGTATGGGAAGCGCCA GACATTG-30 . The amplified cDNA was separated on a 1.5% Agar/ TBE gel and visualized with ethidium bromide. Data analysis All experiments were performed with Cytostar (Amersham, NJ) except for a few preliminary experiments carried out with Flash Plates scintillating microplates (Perkin–Elmer, MA). Experiments were performed at least in triplicate and data were analyzed by nonlinear regression analysis using Prism software (GraphPad, CA).
Results Characterization of GlyT1 expression in JAR cells Expression of GlyT1 in human choriocarcinoma cells was visualized with a specific antibody. As shown in Fig. 2A, every JAR cell exhibited a green fluorescent signal associated with cytoplasmic membrane and cytoplasm, indicating the presence of GlyT1 in these cellular compartments. Several variants of human GlyT1
Fig. 2. Expression of GlyT1 in human placental choriocarcinoma cells. (A) Immunofluorescent labeling with a specific GlyT1 antibody revealed GlyT1 expression (FITC, green) in cultured JAR cells. Omission of primary antibody (insert) resulted in absence of GlyT1 signal. JAR cells nuclei were visualized with DAPI (blue). (B) RT-PCR analysis of mRNA expressed by JAR cells with primers specific for GlyT1 isoform variants indicated that GlyT1a but not GlyT1b,c isoform is present in the human choriocarcinoma placental cells; bp, basepair. (C) Initial experiments with [3 H]glycine demonstrated that 95 1% (n ¼ 6) of JAR glycine uptake is Naþ dependent, and methylaminoisobutyric acid (MeAIB) and L -alanine inhibited uptake to 74 10% (n ¼ 6) and 77 4% (n ¼ 16) of control values, respectively. Reciprocally, sarcosine at 1 mM decreased transport to 18 4% (n ¼ 7) of control value, whereas the GlyT2 inhibitor 4-benzyloxy-3,5-dimethoxy-N-[(1-di-methylaminocyclopentyl)methyl]benzamide did not antagonize glycine transport (111 3% of control, n ¼ 3). Note that GlyT2 compound data were generated with [14 C]glycine. Data are the mean SE of n determinations.
(GlyT1a,b, and c) have been reported and shown to arise from both different promoter usage and alternative splicing [2]. Results from RT-PCR experiments using primers in exon 1 and exon 2 to specifically detect GlyT1a
34
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
and GlyT1b,c, respectively, showed a single band with GlyT1a probe and no signal with GlyT1b,c primer (Fig. 2B). Initial pharmacological studies with [3 H]glycine carried out as reported by Liu and collaborators [18] indicated that glycine uptake by JAR cells is reduced to 5 1% (n ¼ 6) of control value when Naþ is replaced by Liþ , and methylaminoisobutyric acid (MeAIB) and L alanine inhibited uptake to 74 10% (n ¼ 6) and 77 4% (n ¼ 16) of control values, respectively (Fig. 2C). In addition, glycine transport was reduced to 18 4% (n ¼ 7) of control uptake in the presence of 1 mM sarcosine (Fig. 2B), a selective GlyT1 transport inhibitor [2–4] whereas 4-benzyloxy-3,5-dimethoxy-N-[(1-di-methylaminocyclopentyl)methyl]benzamide, a potent and selective GlyT2 antagonist (hGlyT2 IC50 : 16 nM) [17] did not significantly reduce glycine uptake by JAR cell at concentration up to 10 lM (111 4% of control, n ¼ 3; Fig. 2). Altogether, these results corroborated a previous report [18] that GlyT1 and system A are the major glycine transport systems and account for 75 and 20% of total glycine uptake by JAR cells, respectively. Thus, all future [14 C]glycine uptake experiments were performed in the presence of 5 mM L -alanine to block any contribution from system A to total glycine transport.
Characterization of [14 C]glycine uptake The kinetics of [14 C]glycine (10 lM) transport by JAR cells were determined by counting a given 96-well plate at multiple time points. As shown in Fig. 4A, specific uptake was linear over time, whereas total uptake began to plateau at 3 h. The unusual stability of the nonspecific signal over time is a direct consequence of the scintillating microplate technology as it is mainly due to molecules of [14 C]glycine colliding with the scintillating base of the well. Future uptake experiments were carried out between 2 and 3 h to generate a large and stable specific signal. Uptake of increasing concentration of glycine revealed that JAR cellsÕ GlyT1 activity is saturable with a Michaelis–Menten constant Km of 26.8 3 lM and maximal velocity (Vmax ) of 20 cpm/well/min in 96-well Cytostar microplates (Fig. 4B).
Effects of cell density and culture time on [14 C]glycine uptake As a first step in establishing a scintillation proximity assay (SPA) for [14 C]glycine transport, we determined the optimal culture conditions of JAR cells in 96-well microplates. As shown in Fig. 3, JAR cells were plated at densities ranging from 103 to 3 104 cells per well and [14 C]glycine uptake was measured after 28 and 48 h. The largest specific signal was obtained with JAR cells plated at 104 cells/well and cultured for 2 days (Fig. 3); so these conditions were chosen for subsequent experiments.
Fig. 3. Effect of cell density on [14 C]glycine uptake. JAR cells were plated at the indicated density in 96-well plates and the uptake of 10 lM [14 C]glycine in the presence of alanine (5 mM) was measured during 75 min after 28 and 48 h of culture. Data are the mean SE of four determinations.
Fig. 4. Characterization of [14 C]glycine uptake in SPA assay. (A) Accumulation of [14 C]glycine (10 lM) in JAR cells was followed over time by counting the 96-well plate with a Top Count at the indicated time points. The total signal started to plateau after 3 h and at 4 h specific uptake accounts for around 50% of signal. Note that nonspecific signal is constant over time as it results from extracellular [14 C]glycine molecules colliding with the scintillating base. Data were generated in the presence of alanine (5 mM) and are the mean SE of four wells from a representative plate. (B) Saturation experiments in the presence of alanine (5 mM) showed that [14 C]glycine uptake by JAR cells grown in 96-well Cytostar plates is saturable. Michaelis– Menten analysis indicated that Km ¼ 26.8 3.0 lM and Vmax ¼ 140 20 cpm/well/min. Data are the mean SE of four independent experiments. In A and B, nonspecific uptake was determined in the presence of 10 mM glycine.
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
Pharmacological profile of [14 C]glycine uptake To further validate our SPA, the effects of various amino acids and compounds known to interact with GlyT1 were investigated. As reported in Table 1, close congeners of glycine were weak inhibitors of [14 C]glycine uptake mediated by GlyT1, whereas amino acids and neurotransmitters such as L -proline, D -serine, taurine, and GABA were inactive. In contrast, known potent and selective GlyT1 inhibitors such as ALX-5407 [15] and Org-24598 [14] fully antagonized [14 C]glycine uptake by JAR cells in 96-well Cytostar plates with affinities of 2.8 0.6 and 6.9 0.9 nM, respectively (Fig. 5). Glycine and sarcosine, another substrate of GlyT1, blocked [14 C]glycine transport with potencies of 34.0 2.5 and 37.5 4.6 lm, respectively (Fig. 5). Furthermore, under
Table 1 Pharmacological profile of [14 C]glycine uptake by JAR cells cultured in SPA assay Compounds (1 mM)
[14 C]glycine uptake (% of control)
Glycine methyl ester Glycine ethyl ester L -Serine D -Serine D -Cycloserine L -Proline Taurine GABA MelA
22 5 24 7 76 3 109 17 98 14 94 14 99 9 97 8 111 12
35
these experimental conditions the potent and selective human GlyT2 inhibitor [17] 4-benzyloxy-3,5-dimethoxyN-[(1-di-methylaminocyclopentyl)methyl]benzamide did not antagonize [14 C]glycine transport by JAR cells (Fig. 5). Effect of Cl and Naþ flux blockers on [14 C]glycine uptake The transport of glycine by GlyT1 is known to be dependent on Naþ and Cl gradients across cell cytoplasmic membranes so the effect of compounds known to interact with these ionic fluxes were investigated. Blockade of Kþ /Cl cotransporter by furosemide (10 lM) slightly reduced [14 C]glycine uptake by 13%. Similarly, antagonism of Naþ /HCO3 cotransporter and Naþ /Hþ antiporter by 4,40 -diisothiocyanatostilbene2,20 -disulfonic acid (DIDS) and 5-N 0 N-dimethyl-amiloride (DMA), respectively, induced a modest reduction of GlyT1 activity (Fig. 6). In contrast, ouabain and especially benzamil strongly inhibited glycine transport by JAR cells through their blockade of Naþ /Kþ ATPase and Naþ /Ca2þ exchanger, respectively (Fig. 6). Another important aspect to consider when developing a screening assay is the impact of DMSO concentration. As shown in Fig. 6, [14 C]glycine uptake by JAR cells remained unaffected by concentrations of DMSO up to 2% and decreased by around 50% by a 5% DMSO solution.
Data are the mean SE of four experiments performed in the presence of 5 mM alanine.
Fig. 5. Pharmacology of [14 C]glycine uptake by human JAR cells. Experiments performed in the presence of alanine (5 mM) demonstrated that ALX-5407 (d), Org-24598 (s), sarcosine (), and glycine (j) fully inhibit transport of 10 lM [14 C]glycine, whereas taurine (m) and GlyT2 antagonist 4-benzyloxy-3,5-dimethoxy-N-[(1-di-methylaminocyclopentyl)methyl]benzamide (r) were inactive. Nonlinear regression analysis revealed that ALX-5407 and Org-24598 are high affinity antagonists with IC50 values of 2.8 0.6 and 6.9 0.9 nM, respectively. Glycine and sarcosine inhibited transport with lower potency as indicated by IC50 values of 34.0 2.5 and 37.5 4.6 lM, respectively. Nonspecific uptake was determined in the presence of 10 mM glycine and data are the mean SE of three experiments.
Fig. 6. Effect of Cl and Naþ flux inhibitors on [14 C]glycine uptake. Compounds known to block Cl transport such as furosemide (10 lM) did not strongly antagonize the uptake of [14 C]glycine (87 4% of control) by JAR cells. Similarly, antagonism of Naþ /HCO3 cotransporter by 4,40 -diisothiocyanatostilbene-2,20 -disulfonic acid (DIDS) produced a modest reduction of GlyT1 activity. In contrast inhibition of Naþ /Kþ ATPase and Naþ /Hþ and Naþ /Ca2þ exchangers by ouabain (10 lM) and benzamil (10 lM), respectively, reduced [14 C]glycine accumulation in JAR cells to 56 6 and 5 1% of control, respectively. The Naþ /Hþ antiport inhibitor DMA (10 lM) was less potent and reduced [14 C]glycine to 78 5% of control level. Non specific uptake was determined in the presence of 10 mM glycine. Data are the mean SE of six determinations performed in the presence of 5 mM alanine.
36
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
Characterization of [14 C]taurine uptake by JAR cells
Discussion
As GlyT1 belongs to a large family of Naþ /Cl - dependent neurotransmitter transporters, we investigated the versatility of our Cytostar SPA assay by studying the activity of Naþ /Cl -dependent taurine transporter expressed in JAR cells [20]. As shown in Fig. 7, JAR cells exhibited saturation kinetics of [14 C]taurine uptake with an apparent Km of 10.2 1.7 lM and a Vmax in Cytostar microplates of 174 25 cpm/well/min. Pharmacological experiments revealed that ALX-5407 and Org-24598 antagonized [14 C]taurine uptake with potencies of 522 83 nM and 14.9 1.5 lM, respectively (Fig. 7). Taurine inhibited its own transport with a Ki of 9.2 2.3 lM, whereas glycine and sarcosine were inactive (Fig. 7). Consistent with [14 C]glycine uptake results, taurine transporter activity was impaired by ouabain (50 4% at 10 lM, n ¼ 5) and benzamil (15 1% at 10 lM, n ¼ 5) and by DMSO concentration of more than 2% (not shown).
Since its development [16] the Cytostar-T scintillating microplate has been used to detect the cellular accumulation of various radioactive tracers such as [14 C]thymidine, [45 Ca], or [14 C]glycocholate to study cell growth, ionotropic glutamate 4 receptor activity, or bile acid transport, respectively [16,21,22]. We have now used this proximity assay technology to measure [14 C]glycine transport in human placenta choriocarcinoma cells. In agreement with previous work [18] our results confirm that 95% of glycine transport into JAR cells is a Naþ dependent process mediated mostly by an active system A and GlyT1. We also demonstrated that JAR cells express only the GlyT1a isoform, an observation consistent with the reported lack of GlyT1c in human placenta [2]. The lack of activity of the potent and selective GlyT2 inhibitor 4-benzyloxy-3,5-dimethoxy-N-[(1-di-methylaminocyclopentyl)methyl]benzamide [17] indicates that this transporter is not expressed in JAR cells or is expressed at a level below the sensitivity of our uptake assay. Following isolation of GlyT1 activity in JAR cells by addition of L -alanine in transport buffer [18], we showed that [14 C]glycine uptake was dependent on cell density and culture time, whereas temperature (22 °C versus 37 °C) had little influence (not shown). The validity of the SPA was established by pharmacological studies that showed GlyT1-mediated [14 C]glycine uptake to be saturable and of high affinity with a Km value consistent with previous reports in JAR cells (15 lM; [18]) and recombinant systems (70–117 lM; [2,3,23]). The rank order of potency of various amino acids and neurotransmitters agreed with the reported effects or lack thereof of these molecules at GlyT1 transporter [2,4,18]. Finally, full and concentration-dependent inhibition of [14 C]glycine uptake in JAR cells by the potent and selective GlyT1 inhibitors ALX-5407 [14] and Org-24598 [15] clearly demonstrated the validity of our SPA. The versatility of this SPA was demonstrated by our ability to examine taurine transporter activity in JAR cells following minor modification to the uptake protocol ([14 C]taurine instead of [14 C]glycine). Kinetic constant for [14 C]taurine uptake determined in SPA (Km : 10 lM) was comparable to values obtained in JAR cells (2.3 lM; [20]) and in Hela cells expressing cloned human taurine transporter (5.9 lM; [24]). Since taurine and glycine transporters belong to the same family of Naþ / Cl -dependent neurotransmitter transporters, our taurine assay can be used for counterscreening purposes and to determine the selectivity of GlyT1 inhibitors. Using this approach we showed that glycine, sarcosine, and Org-24598 were weak blockers of taurine uptake, whereas ALX-5407 was only 70-fold selective for GlyT1 compared to taurine transporter. Finally, our [14 C]taurine uptake SPA allowed specificity assessment of new transporter inhibitors if one assumes that compounds
Fig. 7. Characterization of a SPA for [14 C]taurine uptake by human JAR cells. (A) Saturation experiments showed that specific [14 C]taurine uptake (gray triangles) by JAR cells grown in 96-well Cytostar plates is saturable and accounts for 90% of total taurine uptake (black triangles). Michaelis–Menten analysis indicated that Km ¼ 10.2 1.7 lM and Vmax ¼ 174 25 cpm/well/min. Data are the mean SE of four experiments, and nonspecific uptake (white triangles) was determined in the presence of 10 mM taurine. (B) Competition experiments demonstrated that ALX-5407 (d) fully inhibits transport of 10 lM [14 C]taurine with a IC50 ¼ 522 83 nM whereas taurine (m) and Org24598 (s) were less potent with IC50 values of 21.1 5.3 and 14.9 1.5 lM, respectively. Sarcosine () and glycine (j) did not significantly block [14 C]taurine uptake by JAR cells. Nonspecific uptake was determined in the presence of 10 mM taurine, and data are the mean SE of three or four experiments.
J.B. Williams et al. / Analytical Biochemistry 321 (2003) 31–37
inhibiting [14 C]glycine and [14 C]taurine with similar potency are likely to do so through disruption of Naþ gradient and/or other vital cellular functions. In conclusion, we showed that using Cytostar-T scintillating microplates it is possible to accurately and efficiently investigate Naþ /Cl -dependent neurotransmitter transporter activities. This coupled to the fact that this scintillation proximity assay is amenable to automation for screening of large chemical collections should facilitate the discovery of new selective transporter inhibitors. References [1] N. Nelson, The family of Naþ /Cl- neurotransmitter transporters, J. Neurochem. 71 (1998) 1785–1803. [2] K.M. Kim, S.F. Kingsmore, H. Han, T.L. Yang-Feng, N. Godinot, M.F. Seldin, M.G. Caron, B. Giros, Cloning of the human glycine transporter type 1: molecular and pharmacological characterization of novel isoform variants and chromosomal localization of the gene in the human and mouse genomes, Mol. Pharmacol. 45 (1994) 608–617. [3] B. Borowsky, E. Mezey, B.J. Hoffman, Two glycine transporter variants with distinct localization in the CNS and peripheral tissues are encoded by a common gene, Neuron 10 (1993) 851– 863. [4] K.E. Smith, L.A. Borden, P.R. Hartig, T. Branchek, R.L. Weinsbank, Cloning and expression of a glycine transporter reveal colocalization with NMDA receptors, Neuron 8 (1992) 927–935. [5] F. Zafra, J. Gomeza, L. Olivares, C. Aragon, C. Gimenez, Regional distribution and developmental variation of the glycine transporters GLYT1 and GLYT2 in the rat CNS, Eur. J. Neurosci. 7 (1995) 1342–1352. [6] F. Zafra, C. Aragon, L. Olivares, N.C. Danbolt, C. Gimenez, J. Storm-Mathisen, Glycine transporters are differentially expressed among CNS cells, J. Neurosci. 15 (1995) 3952–3969. [7] Q.-R. Liu, B. Lopez-Corcuera, S. Mandiyan, H. Nelson, N. Nelson, Cloning and expression of a spinal cord- and brainspecific glycine transporter with novel structural features, J. Biol. Chem. 268 (1993) 22802–22808. [8] J.A. Morrow, I.T. Collie, D.R. Dunbar, G.B. Walker, M. Shahid, D.R. Hill, Molecular cloning and functional expression of the human glycine transporter GlyT2 and chromosomal localisation of the gene in the human genome, FEBS Lett. 439 (1998) 334–340. [9] R. Bergeron, T.M. Meyer, J.T. Coyle, R.W. Greene, Modulation of N-methyl-D -aspartate receptor function by glycine transport, Proc. Natl. Acad. Sci. USA 95 (1998) 15730–15734. [10] A.J. Berger, S. Dieudonne, P. Ascher, Glycine uptake governs glycine site occupancy at NMDA receptors of excitatory synapses, J. Neurophysiol. 80 (1998) 3336–3340. [11] G. Tsai, J.T. Coyle, Glutamatergic mechanisms in schizophrenia, Annu. Rev. Pharmacol. Toxicol. 42 (2002) 165–179.
37
[12] K. Sakata, K. Sato, P. Schloss, H. Betz, S. Shimada, M. Tohyama, Characterization of glycine release mediated by glycine transporter 1 stably expressed in HEK-293 cells, Mol. Brain Res. 49 (1997) 89–94. [13] C. Sur, H. Betz, P. Schloss, A single serine residue controls the cation dependence of substrate transport by the rat serotonin transporter, Proc. Natl. Acad. Sci. USA 94 (1997) 7639– 7644. [14] A. Brown, I. Carlyle, J. Clark, W. Hamilton, S. Gibson, G. McGarry, S. McEachen, D. Rae, S. Thorn, G. Walker, Discovery and SAR of Org 24598-A selective glycine uptake inhibitor, Bioorg. Med. Chem. Lett. 11 (2001) 2007–2009. [15] B.N. Atkinson, S.C. Bell, M. De Vivo, L.R. Kowalski, S.M. Lechner, V.I. Ognyanov, C.-S. Tham, C. Tsai, J. Jia, D. Ashton, M.A. Klitenick, ALX 5407: a potent, selective inhibitor of the hGlyT1 glycine transporter, Mol. Pharmacol. 60 (2001) 1414– 1420. [16] R. Graves, R. Davies, P. Owen, M. Clynes, I. Cleary, G.O. Beirne, An homogeneous assay for measuring the uptake and efflux of radiolabelled drugs in adherent cells, J. Biochem. Biophys. Methods 34 (1997) 177–187. [17] W.L. Caulfield, I.T. Collie, R.S. Dickins, O. Epemolu, R. McGuire, D.R. Hill, G. McVey, J.R. Morphy, Z. Rankovic, H. Sundaram, The first potent and selective inhibitors of the glycine transporter type 2, J. Med. Chem. 44 (2001) 2679– 2682. [18] W. Liu, F.H. Leibach, V. Ganapathy, Characterization of the glycine transport system GLYT1 in human placental choriocarcinoma cells (JAR), Biochim. Biophys. Acta 1194 (1994) 176– 184. [19] C. Sur, H. Betz, P. Schloss, Immunocytochemical detection of the serotonin transporter in rat brain, Neuroscience 73 (1996) 217– 231. [20] P. Kulanthaivel, D.R. Cool, S. Ramamoorthy, V.B. Mahesh, F.H. Leibach, V. Ganapathy, Transport of taurine and its regulation by protein kinase C in the JAR human placental choriocarcinoma cell line, Biochem. J. 277 (1991) 53–58. [21] A. Cushing, M.J. Price-Jones, R. Graves, A.J. Harris, K.T. Hughes, D. Bleakman, D. Lodge, Measurement of calcium flux through ionotropic glutamate receptors using Cytostar-T scintillating microplates, J. Neurosci. Methods 90 (1999) 33– 36. [22] H. Bonge, S. Hallen, J. Fryklund, J.E. Sjostrom, Cytostar-T scintillating microplate assay for measurement of sodium-dependent bile acid uptake in transfected HEK-293 cells, Anal. Biochem. 282 (2000) 94–101. [23] H.J. Herdon, F.M. Godfrey, A.M. Brown, S. Coulton, J.R. Evans, W.J. Cairns, Pharmacological assessment of the role of the glycine transporter GlyT-1 in mediating high-affinity glycine uptake by rat cerebral cortex and cerebellum synaptosomes, Neuropharmacology 41 (2001) 88–96. [24] S. Ramamoorthy, F.H. Leibach, V.B. Mahesh, H. Han, T. Yang-Feng, R.D. Blakely, V. Ganapathy, Functional characterization and chromosomal localization of a cloned taurine transporter from human placenta, Biochem. J. 800 (1994) 893–900.