- Email: [email protected]
Aequorin luminescence-based assay for 5-hydroxytryptamine (serotonin) type 3 receptor characterization
ANALYTICAL BIOCHEMISTRY Analytical Biochemistry 368 (2007) 185–192 www.elsevier.com/locate/yabio
Aequorin luminescence-based assay for 5-hydroxytryptamine (serotonin) type 3 receptor characterization Jutta Walstab a, Sandra Combrink a, Michael Bru¨ss a, Manfred Go¨thert a, Beate Niesler b, Heinz Bo¨nisch a,* b
a Institute of Pharmacology and Toxicology, University of Bonn, 53113 Bonn, Germany Department of Human Molecular Genetics, University of Heidelberg, 69120 Heidelberg, Germany
Received 16 March 2007 Available online 8 June 2007
Abstract The classical electrophysiological method to measure the function of the 5-hydroxytryptamine (serotonin) type 3 (5-HT3) receptor, a cation-permeable ligand-gated ion channel, is time-consuming and not suitable for high-throughput screening. Therefore, we have optimized the conditions for a sensitive assay suitable to measure 5-HT3 receptor responses in cell suspension based on aequorin bioluminescence caused by Ca2+ influx. The assay, carried out in 96-well plates, was applied for the pharmacological characterization of 5-HT3 receptors on human embryonic kidney (HEK) 293 cells transiently coexpressing apoaequorin and either the human homopentameric 5-HT3A receptor or the human heteromeric 5-HT3A/B receptor in the same subset of cells. Thus, the luminescence signal originates exclusively from transfected cells, leading to a high signal/noise ratio, a major advantage compared with fluorescence techniques using Ca2+sensitive dyes. The potencies of two 5-HT3A receptor agonists and two antagonists as well as the potency and efficacy of serotonin at the heteromeric 5-HT3A/B receptor were comparable to those reported using other functional methods. In conclusion, the aequorin assay described here provides a convenient and highly sensitive method for functional characterization of 5-HT3 receptors that is well suited for high-throughput screening. 2007 Elsevier Inc. All rights reserved. Keywords: 5-HT3 receptor; Aequorin luminescence; HEK293 cells; Ca2+ influx; Ligand-gated ion channels
The 5-hydroxytryptamine (serotonin) type 3 (5-HT3)1 receptor belongs to the superfamily of Cys-loop ligandgated ion channels (LGICs), also including nicotinic acetylcholine receptors (nAChRs), glycine, and c-aminobutyric acid type A (GABAA) receptors [1]. These receptors are pentameric proteins that are composed of different sub*
Corresponding author. Fax: +49 228 735404. E-mail address: [email protected] (H. Bo¨nisch). 1 Abbreviations used: 5-HT3, 5-hydroxytryptamine (serotonin) type 3; LGIC, ligand-gated ion channel; nAChR, nicotinic acetylcholine receptor; GABAA, c-aminobutyric acid type A; HEK, human embryonic kidney; h5-HT3, human 5-HT3; cDNA, complementary DNA; DMEM, Dulbecco’s modified Eagle’s medium; FCS, fetal calf serum; DMSO, dimethyl sulfoxide; mCPBG, meta-chlorophenylbiguanide; RLU, relative light units; CICR, Ca2+-induced Ca2+ release; pEC50, log EC50; pIC50, log IC50; FLIPR, fluorometric imaging plate reader. 0003-2697/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2007.06.007
units in most cases. The channel pore is permeable for either cations such as Na+, K+, and Ca2+ (5-HT3 and nAChRs) or the anion Cl (glycine and GABAA receptor). The 5-HT3 receptors occur as hetero- and homopentameric functional complexes [2]. Homomeric 5-HT3A and heteromeric 5-HT3A/B receptors are well characterized [3,4], whereas the modulation of 5-HT3 receptor function by the recently cloned subunits 5-HT3C,D,E [5] is still unknown. Therefore, the availability of a robust and technically simple method for the functional characterization of 5-HT3 receptors of different subunit stoichiometries was highly desirable. The functional characteristics of ligand-gated ion channels usually are determined by time-consuming electrophysiological techniques. Based on the Ca2+ influx through the receptor channel pore on activation by the
186
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
agonist [6,7], receptor function can be monitored by fluorimetric techniques using Ca2+-sensitive dyes or by luminescent techniques using Ca2+-sensitive photoproteins such as aequorin. Whereas fluorescent dyes have already been applied for the measurement of 5-HT3 receptor function [4,7,8], the aequorin assay has not yet been used for this purpose. Aequorin, a photoprotein of the jellyfish Aequorea victoria, is composed of apoaequorin and the hydrophobic prosthetic group coelenterazine. With the chromophore cofactor coelenterazine reconstituted apoaequorin (= holoaequorin) emits blue light on binding of at least two Ca2+ ions in an oxidation reaction. This leads to the formation of apoaequorin, coelenteramide, and CO2 [9,10]. The light emission can be detected by conventional luminometers. Aequorin-based assays have already been applied for the functional characterization of various G protein-coupled receptors [11–13] and, in a single report, for ligand-gated nAChRs [14]. The current investigation aimed to establish a reproducible and highly sensitive aequorin-based luminescence assay in a 96-well plate format for the characterization of human 5-HT3 receptors expressed in human embryonic kidney (HEK) 293 cells. Materials and methods Expression constructs The human 5-HT3A (h5-HT3A) and h5-HT3B subunitencoding complementary DNAs (cDNAs) from HTR3A and HTR3B (GenBank Accession Nos. AJ003079 and AF080582, respectively) were cloned into the expression vector pcDNA3 (Invitrogen, Karlsruhe, Germany). The fidelity of the cDNA sequences was verified by sequencing on a Licor L4200 sequencer (MWG Biotech, Ebersberg, Germany) by using the Deaza GTP Cycle Sequencing Kit (Amersham, Freiburg, Germany). The aequorin cDNA (GenBank Accession No. L29571) originally was derived from cytAEQ/pcDNA1 (Molecular Probes/Invitrogen, Karlsruhe, Germany) and subcloned into HindIII/XbaIdigested pcDNA 3.1/zeo(+) (Invitrogen). Cell culture and transfection HEK293 cells (American Type Culture Collection, Manassas, VA, USA) were grown as monolayers in 175cm2 tissue culture flasks in Dulbecco’s modified Eagle’s medium (DMEM)/Ham’s F12 (1:1) supplemented with 10% fetal calf serum (FCS), 100 U/ml penicillin, and 100 lg/ml streptomycin in a humidified atmosphere containing 5% CO2 at 37 C. Cells were seeded in either 75-cm2 or 25-cm2 cell culture flasks in DMEM/Ham’s F12 (1:1) + 10% FCS to obtain a cell density of 40–70% for transient transfection the following day. Transfection was performed by lipofection with TransIT-293 Transfection Reagent (Mobitec, Go¨ttingen, Germany) using either 15 or 5 lg of total DNA. The following mixtures of cDNAs were used: (i) for the 5-HT3A
subunit, 20% h5-HT3A cDNA and 80% apoaequorin cDNA; (ii) for cotransfection with the h5-HT3B subunit, 10% h5-HT3A cDNA, 10% h5-HT3B cDNA, and 80% apoaequorin cDNA; in controls without h5-HT3B cDNA, salmon sperm DNA was used instead. Cells were used 48 h posttransfection. Aequorin luminescence assay Cell preparation Cells were harvested by centrifugation (180g, 4 min) 48 h posttransfection and resuspended in 1.5 ml (75-cm2 flask) or 0.5 ml (25-cm2 flask) DMEM/Ham’s F12 (1:1) + 0.1% bovine serum albumin. Coelenterazine h was added to give a final concentration of 5 lM from a 5mM stock dissolved in ethanol (stored at 80 C). To load the cells with coelenterazine h, the cell suspension was incubated for 2.5 h at room temperature (22 ± 1 C) in the dark. After loading, the cells were harvested by centrifugation (45g, 3 min) and resuspended in assay buffer containing 150 mM NaCl, 1.8 mM CaCl2, 5.4 mM KCl, 10 mM Hepes, and 20 mM D-glucose at pH 7.4 to obtain an approximate cell density of 1 to 3 · 106 cells/ml. Counts of cells were performed using a hemacytometer (Brand, Wertheim, Germany). An incubation time of 20 min at room temperature followed. Instruments Luminescence was measured using a Centro LB 960 luminometer (Berthold Technologies, Bad Wildbad, Germany) equipped with an autoinjector to allow rapid injection and simultaneous reading. Luminescence was recorded at a sampling rate of 2 Hz by means of the supplied Microwin software in the kinetic mode. Aequorin assay For the measurement of agonist concentration-dependent responses, a white 96-well plate (Nunc, Wiesbaden, Germany) with 80 ll of the cell suspension per well was placed into the luminometer. Prior to injection of the agonist, baseline luminescence was recorded for 8 s. Following autoinjection of 20 ll agonist solution to the cells, light emission was measured for up to 60 s. For antagonist concentration-dependent responses, 60 ll of the cell suspension was preincubated with 20 ll of the respective antagonist solution in a white 96-well plate at room temperature for 15 min to reach an equilibrium. The subsequent steps were identical to those for the measurement of the agonist concentration-dependent responses. Each drug concentration was measured in triplicate or quadruplicate. For recording 5-HT concentration-dependent responses in the presence or absence of ryanodine (dissolved in dimethyl sulfoxide [DMSO]), cells were preincubated with 100 lM ryanodine or assay buffer (containing DMSO) for 30 min before starting the experiment. At the end of those experiments, in which 5-HT maximum responses
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
were recorded, cells were lysed by the addition (autoinjection) of 100 ll of cell lysis solution (0.2% [v/v] aqueous Triton X-100 solution containing 100 mM CaCl2) and remaining aequorin luminescence was recorded for 15 s to obtain the maximum possible Ca2+ response. Compounds and solutions Coelenterazine h was obtained from Nanolight (Pinetop, AZ, USA). 5-HT (serotonin), meta-chlorophenylbiguanide (mCPBG), and ondansetron hydrochloride were obtained from Sigma (Munich, Germany). Azasetron (Y-25130), ryanodine, and thapsigargin were obtained from Biotrend (Cologne, Germany). Caffeine was purchased from Merck (Darmstadt, Germany). Ryanodine and thapsigargin solutions were prepared daily from DMSO stocks (stored as aliquots at 20 C). The other drug solutions were prepared daily from aqueous stocks (stored as aliquots at 20 C). Data analysis The raw luminescent data traces were exported as text files to GraphPad Prism software (version 4.0, San Diego, CA, USA). Peak values for the concentration-dependent response curves were obtained by subtraction of baseline luminescence from the agonist-induced peak maximum luminescence. In agonist maximum response experiments, the peak luminescence (RLUpeak) was normalized against total aequorin luminescence (RLUmax) after cell lysis to control for differences in transfection efficiency and cell number (i.e., fractional luminescence: RLUpeak/[RLUpeak + RLUmax], where RLU is relative light units). The concentration-dependent response curves, as well as the corresponding constants (EC50, IC50, and Hill slope), were calculated by means of the Prism software. Data are given as means ± SEM. Statistical analysis was performed with unpaired Student’s t test. Differences were considered as significant at P < 0.05. Results Assay principle HEK293 cells were chosen to establish a functional assay for the rapid determination of functional properties of h5-HT3 receptors composed of different subunits and naturally occurring variants of the receptor. In initial experiments with cells transiently transfected with h5HT3A receptor and apoaequorin cDNAs, it could be shown that 5-HT elicits fast responses with a luminescence maximum after 3 to 8 s (Fig. 1A). 5-HT-induced luminescence reached baseline values after 30 to 50 s due to receptor desensitization, which is a typical feature of ligand-gated ion channels. For this reason, the use of a luminometer with an integrated injector is necessary to record light emission without any delay simultaneously with autoinjection
187
of the agonist. Baseline luminescence measured for 8 s prior to injection of the agonist was subtracted from the agonist-induced peak luminescence. Baseline luminescence amounted to 2.6 ± 0.4% (n = 17) of the 5-HT (10 lM)induced luminescence signal. Increasing 5-HT concentrations led to increasing luminescence peaks and steeper curves with shorter time-to-peak values. Taking this into account, light emission needed to be recorded over time periods of at least 35 and 60 s in the cases of high and low agonist concentrations, respectively. To test whether the light signal originates exclusively from aequorin luminescence induced by activation of the heterologously expressed h5-HT3A receptors, we measured 5-HT (10 lM) effects in coelenterazine h-loaded HEK293 cells transfected only with apoaequorin cDNA or only with h5-HT3A cDNA. In both cases, no luminescence signal could be recorded (data not shown). The lack of a 5-HT response indicates that the luminescence signal originates solely from aequorin and that HEK293 cells do not express 5-HT receptors natively, which could disturb the measurement of 5-HT3 receptor responses. Because activation of 5-HT3 receptors induces cell depolarization [15–19], HEK293 cells were tested for the involvement of voltage-gated calcium ion channels in the 5-HT-induced Ca2+ luminescence. We depolarized HEK293 cells by autoinjection of a 50-mM KCl solution. No K+-induced luminescence was observed (data not shown). To check the possibility that Ca2+-induced Ca2+ release (CICR) from intracellular stores (mediated primarily by ryanodine receptors) could influence the agonistinduced luminescence signal, 5-HT responses were recorded in the presence of ryanodine. Therefore, cells were pretreated either with a high concentration (100 lM) of ryanodine (to block ryanodine receptors) or with buffer containing only its solvent. The log EC50 (pEC50) value and Hill coefficient amounted to 5.75 ± 0.02 and 3.14 ± 0.53, respectively, in the presence of ryanodine and 5.77 ± 0.04 and 3.15 ± 0.08, respectively, in the absence of ryanodine (n = 3 each). Thus, ryanodine had no effect on the concentration-dependent response relationship for 5-HT. However, in the presence of ryanodine, the maximum peak response to 10 lM 5-HT was reduced to 77.8 ± 3.7% of controls (n = 6). Similar results were obtained in one experiment with cells pretreated with 10 mM caffeine and 2.5 lM thapsigargin to deplete intracellular Ca2+ stores (data not shown). Hence, the 5-HTinduced luminescence signal is due predominantly to influx of Ca2+ ions through the pore of the 5-HT3A receptor channel. The sensitivity of the aequorin assay was tested by stimulating decreasing numbers of HEK293 cells transiently transfected with apoaequorin and h5-HT3A receptor cDNA with 5-HT (10 lM) (Fig. 1B). It is evident that 5-HT-induced peak luminescence decreased with decreasing cell numbers and that the peak response was related linearly to the cell number (Fig. 1C), indicating that the assay is reliable over a broad number of cells per well.
188
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
Fig. 1. (A) Time courses of luminescence induced by increasing 5-HT concentrations in coelenterazine h-loaded HEK293 cells transiently transfected with h5-HT3A receptor subunit and apoaequorin cDNAs. Shown is the result of a representative experiment. Luminescence was measured as relative light units (RLU) at a sampling rate of 2 Hz from 8 s before until 35 to 60 s after the addition of 5-HT. (B) Luminescence induced by 5-HT (10 lM) in coelenterazine h-loaded HEK293 cells of varying density in the suspension. The cells were transiently transfected with h5-HT3A receptor subunit and apoaequorin cDNAs. Shown are mean curves calculated from triplicate measurements of luminescence in a representative experiment containing the cell numbers indicated in the figure. (C) Linear dependence of the 5-HT-induced peak luminescence values (means ± SEM) on the number of cells (see panel B). SEM values of triplicate peak luminescence values were less than 7.4% of the respective mean values. (D) Luminescence induced by 5-HT (10 lM) in coelenterazine h-loaded HEK293 cells transiently transfected with h5-HT3A receptor subunit and apoaequorin cDNAs at different ratios. Shown are peak values for the respective h5-HT3A/aequorin cDNA ratios determined in triplicate in a representative experiment (means ± SEM of the three measurements). (E) Luminescence induced by lysis of HEK293 cells transiently transfected with apoaequorin cDNA and loaded with coelenterazine h at increasing concentrations. Cell lysis was induced by the addition of aqueous Triton X-100 solution containing CaCl2 (see Materials and methods). Cells were loaded with increasing concentrations of coelenterazine h and incubated for 2.5 h at room temperature prior to the assay. Shown are means ± SEM of triplicate determinations in a representative experiment for each condition. (F) Luminescence induced by lysis (see panel E) of HEK293 cells transiently transfected with apoaequorin cDNA and loaded with coelenterazine h (5 lM) at different temperatures and loading times. Shown are means ± SEM of quadruplicate determinations for each experimental condition.
Optimization of transfection and coelenterazine h loading protocol To determine the amounts of h5-HT3A receptor and apoaequorin cDNA that yield the highest peak luminescence values, HEK293 cells were transfected with different receptor/apoaequorin cDNA ratios. Transfection of cells with increasing apoaequorin and decreasing receptor cDNA amounts led to enhanced 5-HT (10 lM)-induced peak luminescence values; a h5-HT3A/apoaequorin cDNA ratio of 1:4 yielded approximately a 5.5 times greater light emission as compared with equal amounts of cDNAs for both proteins (Fig. 1D). For further optimization, HEK293 cells transiently expressing apoaequorin were incubated for a time period of 3.5 h at room temperature with different concentrations (1–10 lM) of coelenterazine h. The maximum luminescence response, depending on the total amount of reconstituted aequorin, increased with rising coelenterazine h concentrations to reach a maximum at 7.5 lM (Fig. 1E). In subsequent experiments, 5 lM coelenterazine h, which induced nearly 80% of the maximum luminescence value, was used. In the next step samples of apoaequorin-expressing cells were loaded with coelenterazine h at two different temperatures (37 C and room temperature) for different times. Incubation at 37 C caused lower luminescence peaks after
cell lysis, and thus lower reconstituted aequorin amounts, than did incubation at room temperature (Fig. 1F). An incubation time of 3.5 h at room temperature yielded only a slightly increased total amount of aequorin as compared with an incubation time of 2 h; therefore, a loading time of 2.5 h at room temperature was chosen for all further experiments. Potency of 5-HT3 receptor ligands at the homopentameric h5-HT3A receptor Experiments performed with 5-HT (i.e., the physiological agonist at 5-HT receptors) and with the selective 5-HT3 receptor agonist mCPBG yielded concentrationdependent increases in aequorin luminescence. The agonist-induced peak luminescence resulted in concentration-dependent response curves (Fig. 2A) characterized by a pEC50 value of 5.76 ± 0.03 (mean EC50 = 1.77 lM, n = 11) for 5-HT and 5.76 ± 0.02 (mean EC50 = 1.75 lM, n = 7) for mCPBG, with Hill slopes for 5-HT and mCPBG amounting to 3.19 ± 0.26 and 2.76 ± 0.19, respectively. Responses to a maximum 5-HT concentration (10 lM) were concentration-dependently inhibited by the 5-HT3 receptor antagonists ondansetron and azasetron with log IC50 (pIC50) values of 9.00 ± 0.02 (n = 4) and 8.96 ± 0.03 (n = 5), respectively (Fig. 2B). Corrections of
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
189
Fig. 3. Concentration-dependent response curves for 5-HT on coelenterazine h-loaded HEK293 cells transiently transfected with the cDNAs of apoaequorin and either the h5-HT3A receptor subunit alone (black) or both the h5-HT3A and h5-HT3B receptor subunits (gray). 5-HT-induced luminescence is expressed as a percentage of maximum luminescence produced by the agonist at the respective homopentameric h5-HT3A or heteromeric h5-HT3A/B receptor. Shown are means ± SEM of 4 to 11 independent transfections. pEC50 values at the homopentameric h5-HT3A receptor and the heteromeric h5-HT3A/B receptor amounted to 5.75 ± 0.02 and 5.45 ± 0.08, respectively (P < 0.05).
Fig. 2. Luminescence induced by increasing concentrations of 5-HT3 receptor agonists (A) or by 5-HT (10 lM) in the presence of increasing concentrations of 5-HT3 receptor antagonists (B) in coelenterazine h-loaded HEK293 cells transiently transfected with h5-HT3A receptor subunit and apoaequorin cDNAs. (A) Concentration-dependent response curves for the agonists 5-HT and mCPBG. Agonist-induced luminescence is expressed as a percentage of maximum peak luminescence produced by the respective agonist. (B) Inhibition of 5-HT (10 lM)-induced luminescence by increasing concentrations of the 5-HT3 receptor antagonists ondansetron and azasetron (Y-25130). The antagonist was present 15 min before and during the application of 5-HT. Data are expressed as percentages of the 5-HT (10 lM) response in the absence of the antagonist. Shown are means ± SEM of 4 to 11 independent transfections.
these values according to Cheng and Prusoff [20] results in pKi (Ki) values of 9.82 (0.15 nM) and 9.78 (0.17 nM) for ondansetron and azasetron, respectively. 5-HT potency and efficacy at the heteromeric h5-HT3A/B receptor Coexpression of the h5-HT3B subunit together with the h5-HT3A subunit led to a slightly reduced pEC50 value for 5-HT as compared with that at the homopentameric h5-HT3A receptor (5.45 ± 0.08 [mean EC50 = 3.55 lM] vs. 5.76 ± 0.03 [mean EC50 = 1.77 lM], P < 0.05). The Hill coefficient of the concentration-dependent response curve at the heteromeric h5-HT3A/B receptor was also reduced (1.45 ± 0.29 vs. 3.19 ± 0.26, P < 0.01) (Fig. 3). To examine whether the maximum 5-HT-induced luminescence response (Emax) at the heteromeric h5-HT3A/B receptor differs from that at the homopentameric h5-HT3A receptor, the receptors were stimulated with 500 lM 5-HT and luminescence was recorded. The Emax value (i.e., fractional luminescence [see Materials and methods]) for 5-HT at the heteromeric h5-HT3A/B receptor was increased signifi-
cantly to 165.9 ± 21.3% (n = 6, P < 0.05) of the Emax at the homopentameric h5-HT3A receptor. Discussion The aim of the study was to establish a functional assay for the characterization of 5-HT3 receptors using aequorinbased luminescence. We adapted a sensitive and reliable method that was optimized for the high-throughput analysis of Ca2+ influx through 5-HT3 receptors composed of varying subunits. The standard method for the characterization of 5-HT3 receptors and of other ligand-gated ion channels is the patch clamp analysis. It is the only method suitable to analyze the kinetic properties of the channel. However, it is very time-consuming and, thus, not suited for a fast and high-throughput analysis of the function of 5-HT3 receptors composed of different subunits. Other methods applied for the measurement of 5-HT3 receptors are, for example, the [14C]guanidinium influx [21] and the use of Ca2+-sensitive fluorescent dyes [4,7]. The first method mentioned has major disadvantages such as the use of radioactivity and a low signal/noise ratio. In addition, the artificial buffer used in these experiments, where Na+ ions are replaced by larger organic cations such as choline and N-methyl-D-glucamine that cannot pass the channel pore [22], represents artificial conditions that may falsify the results. Although the determination of Ca2+ influx by measuring fluorescence on binding of Ca2+ ions to Ca2+-sensitive dyes has been used successfully for the characterization of 5-HT3 receptors in stably transfected cells, there are some drawbacks, especially in the case of transient transfections. It is not possible to load the dye selectively into transfected cells; this causes a high background fluorescence because the signal originates from the whole cell population (this is shown, e.g.,
4.3 5.7 ND ND ND ND 9.3 ND 0.8 ND Note. ND, not determined. a Ki values were calculated according to the equation of Cheng and Prusoff [20]. b In this study, cells were transfected with mouse 5-HT3A receptor. c Ki value (lM) is given.
ND 0.1 ND ND ND ND ND 0.1 0.2 0.2 Antagonists Azasetron Ondansetron
ND ND
0.4c 0.3c 1.1 0.2 ND 0.2 0.3 0.5 8.5 5.9 8.6 ND 3.0 ND 2.1 0.8 3.4 ND 2.9 2.5 1.8 1.8 Agonists 5-HT mCPBG
Fura-2 fluorescence (FlexStation) [8]b Fura-2 fluorescence (single cells) [7]b FLIPR assay [4] Excised outside-out patch clamp [28] Excised outsideout patch clamp [25,30] Whole cell patch clamp [32] Whole cell patch clamp [29]b Whole cell patch clamp [6] Whole cell patch clamp [3] Aequorin assay (current study) Compound
EC50 value (lM) for agonists (unless stated otherwise) and Ki valuesa for antagonists (nM)
at heterologously expressed a1-adrenergic receptors [23]). For this reason, we applied an aequorin luminescencebased Ca2+ assay whose use as a detector of Ca2+ fluctuations in living cells is well established [12]. This assay takes advantage of the fact that after transient cotransfection of the cells with apoaequorin and receptor cDNA, the encoded proteins are expressed in the same subset of transfected cells, leading to a very high signal/noise ratio because luminescence originates only from transfected cells. On the other hand, the total amount of cellular aequorin measured at the end of the experiment by cell lysis in the presence of a high external Ca2+ concentration can be used as an internal control for transfection efficiency. HEK293 cells were chosen for the heterologous expression of 5-HT3 receptors because they do not express serotonin receptors natively. We proved this in cells transfected only with the apoaequorin cDNA. In these cells, a high concentration (10 lM) of the agonist 5-HT caused no increase in luminescence over the baseline (data not shown). Furthermore, HEK293 cells were examined for the expression of voltage-gated calcium channels that can be opened by depolarization with KCl. No K+-induced luminescence could be recorded (data not shown), indicating the lack of expression of such calcium channels in HEK293 cells that confirms the results of previous studies [7,24]. Furthermore, CICR from intracellular stores contributed to only 20% of the 5-HT-induced maximum response. Hence, it may be concluded that the luminescence signal is due predominantly to Ca2+ influx through the 5-HT3 receptor channel. In initial experiments with the agonist 5-HT, we observed that with increasing 5-HT concentrations, the luminescence curves became steeper and reached baseline values within a shorter period of time. Although much faster, the same relationship of rising and decay rate of the recorded currents on the 5-HT concentration was obtained in patch-clamp experiments [25]. To improve reproducibility of our luminescence-based assay, we first optimized parameters such as the loading procedure of the cells with the cofactor coelenterazine and obtained satisfying results with a loading time of 2.5 h with 5 lM coelenterazine. Furthermore, the sensitivity of the assay was enhanced by using the synthetic cofactor coelenterazine h. Aequorins containing the h form of coelenterazine have been reported to be more sensitive to Ca2+ and exhibit a 16 times greater luminescence intensity as compared with aequorin reconstituted with native coelenterazine [26,27]. In addition, the amounts of cDNAs encoding apoaequorin and the 5-HT3 receptor were optimized to obtain high luminescence peaks on the application of the agonist. We observed that the height of the luminescence peak depends strongly on the amount of apoaequorin cDNA used for transfection; the higher the ratio of apoaequorin cDNA over receptor cDNA, the greater the maximum 5-HT-induced luminescence response. This is consistent
[3H]GR65630 binding [28]
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
Table 1 Potencies or affinities of 5-HT3 receptor ligands obtained with different methods at the h5-HT3A receptor heterologously expressed in HEK293 cells
190
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
with results reported by Sheu and coworkers [12], who observed that the signal is proportional to the amount of aequorin in the cell. This optimization is especially useful for the characterization of 5-HT3 receptor isoforms that are expressed in a low density on the cell surface. The EC50 values for the two receptor agonists, 5-HT and mCPBG, were not different from those determined previously for 5-HT or mCPBG with other techniques at the h5-HT3A receptor in HEK293 cells (Table 1). For both agonists, the Hill coefficients were similar to those obtained with other techniques such as [14C]guanidinium influx [31] and fluorescence determination with Ca2+-sensitive dyes performed either on single cells [7] or with a fluorometric imaging plate reader (FLIPR) assay [4]. The 5-HT (10 lM)-induced responses were dose-dependently blocked by the specific 5-HT3 receptor antagonists ondansetron and azasetron, and the Ki values were very similar to the values obtained previously (Table 1). In previous studies [3,4,32], different pharmacological properties of 5-HT at heteromeric h5-HT3A/B and homopentameric h5-HT3A receptors have been reported. In the current study, with the aequorin-based assay, we could confirm that the affinity of 5-HT is lower at the h5-HT3A/B receptor than at the homopentameric h5-HT3A receptor and that the Hill slope of the dose–response curve is smaller with the heteromeric h5-HT3A/B receptor than with the homopentameric h5-HT3A receptor, indicating a loss of cooperativity for 5-HT at the heteromeric h5-HT3A/B receptor. In addition, the maximum response (Emax) for 5-HT was approximately 1.7-fold greater at the heteromeric h5-HT3A/B receptor than at the homopentameric h5-HT3A receptor; this is compatible with the proposed higher single channel conductance of the heteromeric h5-HT3A/B receptor. Thus, the aequorin assay described here is well suited to study, for example, the pharmacological properties of 5-HT3 receptors containing further subunits described recently [5]. In conclusion, we have reported the establishment of optimized conditions for an aequorin luminescence-based assay in 96-well plates for the characterization of the 5-HT3 receptor. This functional assay provides a highly sensitive and rapid method appropriate for high-throughput screening of Ca2+-permeable ligand-gated ion channels. Acknowledgment This research was supported by the Deutsche Forschungsgemeinschaft (DFG, BR 1741/2-1). References [1] V. Derkach, A. Surprenant, R.A. North, 5-HT3 receptors are membrane ion channels, Nature 339 (1989) 706–709. [2] J.A. Peters, T.G. Hales, J.J. Lambert, Molecular determinants of single-channel conductance and ion selectivity in the Cys-loop family: insights from the 5-HT3 receptor, Trends Pharmacol. Sci. 26 (2005) 587–594.
191
[3] P.A. Davies, M. Pistis, M.C. Hanna, J.A. Peters, J.J. Lambert, T.G. Hales, E.F. Kirkness, The 5-HT3B subunit is a major determinant of serotonin-receptor function, Nature 397 (1999) 359–363. [4] A.E. Dubin, R. Huvar, M.R. D’Andrea, J. Pyati, J.Y. Zhu, K.C. Joy, S.J. Wilson, J.E. Galindo, C.A. Glass, L. Luo, M.R. Jackson, T.W. Lovenberg, M.G. Erlander, The pharmacological and functional characteristics of the serotonin 5-HT3A receptor are specifically modified by a 5-HT3B receptor subunit, J. Biol. Chem. 274 (1999) 30799–30810. [5] B. Niesler, B. Frank, J. Kapeller, G.A. Rappold, Cloning, physical mapping, and expression analysis of the human 5-HT3 serotonin receptor-like genes HTR3C, HTR3D, and HTR3E, Gene 310 (2003) 101–111. [6] A.M. Brown, A.G. Hope, J.J. Lambert, J.A. Peters, Ion permeation and conduction in a human recombinant 5-HT3 receptor subunit (h5HT3A), J. Physiol. 507 (1998) 653–665. [7] A.C. Hargreaves, S.C. Lummis, C.W. Taylor, Ca2+ permeability of cloned and native 5-hydroxytryptamine type 3 receptors, Mol. Pharmacol. 46 (1994) 1120–1128. [8] K.L. Price, S.C. Lummis, FlexStation examination of 5-HT3 receptor function using Ca2+- and membrane potential-sensitive dyes: advantages and potential problems, J. Neurosci. Methods 149 (2005) 172–177. [9] O. Shimomura, F.H. Johnson, Regeneration of the photoprotein aequorin, Nature 256 (1975) 236–238. [10] O. Shimomura, Luminescence of aequorin is triggered by the binding of two calcium ions, Biochem. Biophys. Res. Commun. 211 (1995) 359–363. [11] S. An, T. Bleu, Y. Zheng, Transduction of intracellular calcium signals through G protein-mediated activation of phospholipase C by recombinant sphingosine 1-phosphate receptors, Mol. Pharmacol. 55 (1999) 787–794. [12] Y.A. Sheu, L.J. Kricka, D.B. Pritchett, Measurement of intracellular calcium using bioluminescent aequorin expressed in human cells, Anal. Biochem. 209 (1993) 343–347. [13] M.D. Ungrin, L.M. Singh, R. Stocco, D.E. Sas, M. Abramovitz, An automated aequorin luminescence-based functional calcium assay for G-protein-coupled receptors, Anal. Biochem. 272 (1999) 34–42. [14] M.S. Karadsheh, M.S. Shah, X. Tang, R.L. Macdonald, J.A. Stitzel, Functional characterization of mouse a4b2 nicotinic acetylcholine receptors stably expressed in HEK293T cells, J. Neurochem. 91 (2004) 1138–1150. [15] D.I. Wallis, R.A. North, The action of 5-hydroxytryptamine on single neurones of the rabbit superior cervical ganglion, Neuropharmacology 17 (1978) 1023–1028. [16] J.R. Fozard, Neuronal 5-HT receptors in the periphery, Neuropharmacology 23 (1984) 1473–1486. [17] J. Azami, J.R. Fozard, A.A. Round, D.I. Wallis, The depolarizing action of 5-hydroxytryptamine on rabbit vagal primary afferent and sympathetic neurones and its selective blockade by MDL 72222, Naunyn Schmiedebergs Arch. Pharmacol. 328 (1985) 423–429. [18] A. Round, D.I. Wallis, The depolarizing action of 5-hydroxytryptamine on rabbit vagal afferent and sympathetic neurones in vitro and its selective blockade by ICS 205-930, Br. J. Pharmacol. 88 (1986) 485–494. [19] J.A. Peters, H.M. Malone, J.J. Lambert, Recent advances in the electrophysiological characterization of 5-HT3 receptors, Trends Pharmacol. Sci. 13 (1992) 391–397. [20] Y. Cheng, W.H. Prusoff, Relationship between the inhibition constant (K1) and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction, Biochem. Pharmacol. 22 (1973) 3099–3108. [21] H. Bo¨nisch, M. Barann, J. Graupner, M. Go¨thert, Characterization of 5-HT3 receptors of N1E-115 neuroblastoma cells by use of the influx of the organic cation [14C]-guanidinium, Br. J. Pharmacol. 108 (1993) 436–442. [22] M. Barann, K. Schmidt, M. Go¨thert, B.W. Urban, H. Bo¨nisch, Influence of sodium substitutes on 5-HT-mediated effects at mouse 5-HT3 receptors, Br. J. Pharmacol. 142 (2004) 501–508.
192
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
[23] M. Brini, R. Marsault, C. Bastianutto, J. Alvarez, T. Pozzan, R. Rizzuto, Transfected aequorin in the measurement of cytosolic Ca2+ concentration ([Ca2+]c): a critical evaluation, J. Biol. Chem. 270 (1995) 9896–9903. [24] M.E. Williams, P.F. Brust, D.H. Feldman, S. Patthi, S. Simerson, A. Maroufi, A.F. McCue, G. Velicelebi, S.B. Ellis, M.M. Harpold, Structure and functional expression of an omega-conotoxin-sensitive human N-type calcium channel, Science 257 (1992) 389–395. [25] M. Barann, W. Meder, Z. Dorner, M. Bru¨ss, H. Bo¨nisch, M. Go¨thert, B.W. Urban, Recombinant human 5-HT3A receptors in outside-out patches of HEK 293 cells: basic properties and barbiturate effects, Naunyn Schmiedebergs Arch. Pharmacol. 362 (2000) 255–265. [26] M.R. Knight, N.D. Read, A.K. Campbell, A.J. Trewavas, Imaging calcium dynamics in living plants using semi-synthetic recombinant aequorins, J. Cell Biol. 121 (1993) 83–90. [27] O. Shimomura, Y. Kishi, S. Inouye, The relative rate of aequorin regeneration from apoaequorin and coelenterazine analogues, Biochem. J. 296 (1993) 549–551.
[28] D. Kurzwelly, M. Barann, A. Kostanian, S. Combrink, H. Bo¨nisch, M. Go¨thert, M. Bru¨ss, Pharmacological and electrophysiological properties of the naturally occurring Pro391Arg variant of the human 5-HT3A receptor, Pharmacogenetics 14 (2004) 165–172. [29] A.D. Spier, S.C. Lummis, The role of tryptophan residues in the 5hydroxytryptamine-3 receptor ligand binding domain, J. Biol. Chem. 275 (2000) 5620–5625. [30] M. Barann, M. Bruess, M. Brinkmann, I. Linden, M. Lyutenska, M. Schneider, J. Walkembach, M. Wittmann, 5-HT3 receptors and emesis, Pharmacol. Rep. 58 (2006) 253–255. [31] M. Bru¨ss, M. Barann, M. Hayer-Zillgen, T. Eucker, M. Go¨thert, H. Bo¨nisch, Modified 5-HT3A receptor function by co-expression of alternatively spliced human 5-HT3A receptor isoforms, Naunyn Schmiedebergs Arch. Pharmacol. 362 (2000) 392–401. [32] G. Hapfelmeier, C. Tredt, R. Haseneder, W. Zieglgansberger, B. Eisensamer, R. Rupprecht, G. Rammes, Co-expression of the 5-HT3B serotonin receptor subunit alters the biophysics of the 5-HT3 receptor, Biophys. J. 84 (2003) 1720–1733.
Aequorin luminescence-based assay for 5-hydroxytryptamine (serotonin) type 3 receptor characterization Jutta Walstab a, Sandra Combrink a, Michael Bru¨ss a, Manfred Go¨thert a, Beate Niesler b, Heinz Bo¨nisch a,* b
a Institute of Pharmacology and Toxicology, University of Bonn, 53113 Bonn, Germany Department of Human Molecular Genetics, University of Heidelberg, 69120 Heidelberg, Germany
Received 16 March 2007 Available online 8 June 2007
Abstract The classical electrophysiological method to measure the function of the 5-hydroxytryptamine (serotonin) type 3 (5-HT3) receptor, a cation-permeable ligand-gated ion channel, is time-consuming and not suitable for high-throughput screening. Therefore, we have optimized the conditions for a sensitive assay suitable to measure 5-HT3 receptor responses in cell suspension based on aequorin bioluminescence caused by Ca2+ influx. The assay, carried out in 96-well plates, was applied for the pharmacological characterization of 5-HT3 receptors on human embryonic kidney (HEK) 293 cells transiently coexpressing apoaequorin and either the human homopentameric 5-HT3A receptor or the human heteromeric 5-HT3A/B receptor in the same subset of cells. Thus, the luminescence signal originates exclusively from transfected cells, leading to a high signal/noise ratio, a major advantage compared with fluorescence techniques using Ca2+sensitive dyes. The potencies of two 5-HT3A receptor agonists and two antagonists as well as the potency and efficacy of serotonin at the heteromeric 5-HT3A/B receptor were comparable to those reported using other functional methods. In conclusion, the aequorin assay described here provides a convenient and highly sensitive method for functional characterization of 5-HT3 receptors that is well suited for high-throughput screening. 2007 Elsevier Inc. All rights reserved. Keywords: 5-HT3 receptor; Aequorin luminescence; HEK293 cells; Ca2+ influx; Ligand-gated ion channels
The 5-hydroxytryptamine (serotonin) type 3 (5-HT3)1 receptor belongs to the superfamily of Cys-loop ligandgated ion channels (LGICs), also including nicotinic acetylcholine receptors (nAChRs), glycine, and c-aminobutyric acid type A (GABAA) receptors [1]. These receptors are pentameric proteins that are composed of different sub*
Corresponding author. Fax: +49 228 735404. E-mail address: [email protected] (H. Bo¨nisch). 1 Abbreviations used: 5-HT3, 5-hydroxytryptamine (serotonin) type 3; LGIC, ligand-gated ion channel; nAChR, nicotinic acetylcholine receptor; GABAA, c-aminobutyric acid type A; HEK, human embryonic kidney; h5-HT3, human 5-HT3; cDNA, complementary DNA; DMEM, Dulbecco’s modified Eagle’s medium; FCS, fetal calf serum; DMSO, dimethyl sulfoxide; mCPBG, meta-chlorophenylbiguanide; RLU, relative light units; CICR, Ca2+-induced Ca2+ release; pEC50, log EC50; pIC50, log IC50; FLIPR, fluorometric imaging plate reader. 0003-2697/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2007.06.007
units in most cases. The channel pore is permeable for either cations such as Na+, K+, and Ca2+ (5-HT3 and nAChRs) or the anion Cl (glycine and GABAA receptor). The 5-HT3 receptors occur as hetero- and homopentameric functional complexes [2]. Homomeric 5-HT3A and heteromeric 5-HT3A/B receptors are well characterized [3,4], whereas the modulation of 5-HT3 receptor function by the recently cloned subunits 5-HT3C,D,E [5] is still unknown. Therefore, the availability of a robust and technically simple method for the functional characterization of 5-HT3 receptors of different subunit stoichiometries was highly desirable. The functional characteristics of ligand-gated ion channels usually are determined by time-consuming electrophysiological techniques. Based on the Ca2+ influx through the receptor channel pore on activation by the
186
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
agonist [6,7], receptor function can be monitored by fluorimetric techniques using Ca2+-sensitive dyes or by luminescent techniques using Ca2+-sensitive photoproteins such as aequorin. Whereas fluorescent dyes have already been applied for the measurement of 5-HT3 receptor function [4,7,8], the aequorin assay has not yet been used for this purpose. Aequorin, a photoprotein of the jellyfish Aequorea victoria, is composed of apoaequorin and the hydrophobic prosthetic group coelenterazine. With the chromophore cofactor coelenterazine reconstituted apoaequorin (= holoaequorin) emits blue light on binding of at least two Ca2+ ions in an oxidation reaction. This leads to the formation of apoaequorin, coelenteramide, and CO2 [9,10]. The light emission can be detected by conventional luminometers. Aequorin-based assays have already been applied for the functional characterization of various G protein-coupled receptors [11–13] and, in a single report, for ligand-gated nAChRs [14]. The current investigation aimed to establish a reproducible and highly sensitive aequorin-based luminescence assay in a 96-well plate format for the characterization of human 5-HT3 receptors expressed in human embryonic kidney (HEK) 293 cells. Materials and methods Expression constructs The human 5-HT3A (h5-HT3A) and h5-HT3B subunitencoding complementary DNAs (cDNAs) from HTR3A and HTR3B (GenBank Accession Nos. AJ003079 and AF080582, respectively) were cloned into the expression vector pcDNA3 (Invitrogen, Karlsruhe, Germany). The fidelity of the cDNA sequences was verified by sequencing on a Licor L4200 sequencer (MWG Biotech, Ebersberg, Germany) by using the Deaza GTP Cycle Sequencing Kit (Amersham, Freiburg, Germany). The aequorin cDNA (GenBank Accession No. L29571) originally was derived from cytAEQ/pcDNA1 (Molecular Probes/Invitrogen, Karlsruhe, Germany) and subcloned into HindIII/XbaIdigested pcDNA 3.1/zeo(+) (Invitrogen). Cell culture and transfection HEK293 cells (American Type Culture Collection, Manassas, VA, USA) were grown as monolayers in 175cm2 tissue culture flasks in Dulbecco’s modified Eagle’s medium (DMEM)/Ham’s F12 (1:1) supplemented with 10% fetal calf serum (FCS), 100 U/ml penicillin, and 100 lg/ml streptomycin in a humidified atmosphere containing 5% CO2 at 37 C. Cells were seeded in either 75-cm2 or 25-cm2 cell culture flasks in DMEM/Ham’s F12 (1:1) + 10% FCS to obtain a cell density of 40–70% for transient transfection the following day. Transfection was performed by lipofection with TransIT-293 Transfection Reagent (Mobitec, Go¨ttingen, Germany) using either 15 or 5 lg of total DNA. The following mixtures of cDNAs were used: (i) for the 5-HT3A
subunit, 20% h5-HT3A cDNA and 80% apoaequorin cDNA; (ii) for cotransfection with the h5-HT3B subunit, 10% h5-HT3A cDNA, 10% h5-HT3B cDNA, and 80% apoaequorin cDNA; in controls without h5-HT3B cDNA, salmon sperm DNA was used instead. Cells were used 48 h posttransfection. Aequorin luminescence assay Cell preparation Cells were harvested by centrifugation (180g, 4 min) 48 h posttransfection and resuspended in 1.5 ml (75-cm2 flask) or 0.5 ml (25-cm2 flask) DMEM/Ham’s F12 (1:1) + 0.1% bovine serum albumin. Coelenterazine h was added to give a final concentration of 5 lM from a 5mM stock dissolved in ethanol (stored at 80 C). To load the cells with coelenterazine h, the cell suspension was incubated for 2.5 h at room temperature (22 ± 1 C) in the dark. After loading, the cells were harvested by centrifugation (45g, 3 min) and resuspended in assay buffer containing 150 mM NaCl, 1.8 mM CaCl2, 5.4 mM KCl, 10 mM Hepes, and 20 mM D-glucose at pH 7.4 to obtain an approximate cell density of 1 to 3 · 106 cells/ml. Counts of cells were performed using a hemacytometer (Brand, Wertheim, Germany). An incubation time of 20 min at room temperature followed. Instruments Luminescence was measured using a Centro LB 960 luminometer (Berthold Technologies, Bad Wildbad, Germany) equipped with an autoinjector to allow rapid injection and simultaneous reading. Luminescence was recorded at a sampling rate of 2 Hz by means of the supplied Microwin software in the kinetic mode. Aequorin assay For the measurement of agonist concentration-dependent responses, a white 96-well plate (Nunc, Wiesbaden, Germany) with 80 ll of the cell suspension per well was placed into the luminometer. Prior to injection of the agonist, baseline luminescence was recorded for 8 s. Following autoinjection of 20 ll agonist solution to the cells, light emission was measured for up to 60 s. For antagonist concentration-dependent responses, 60 ll of the cell suspension was preincubated with 20 ll of the respective antagonist solution in a white 96-well plate at room temperature for 15 min to reach an equilibrium. The subsequent steps were identical to those for the measurement of the agonist concentration-dependent responses. Each drug concentration was measured in triplicate or quadruplicate. For recording 5-HT concentration-dependent responses in the presence or absence of ryanodine (dissolved in dimethyl sulfoxide [DMSO]), cells were preincubated with 100 lM ryanodine or assay buffer (containing DMSO) for 30 min before starting the experiment. At the end of those experiments, in which 5-HT maximum responses
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
were recorded, cells were lysed by the addition (autoinjection) of 100 ll of cell lysis solution (0.2% [v/v] aqueous Triton X-100 solution containing 100 mM CaCl2) and remaining aequorin luminescence was recorded for 15 s to obtain the maximum possible Ca2+ response. Compounds and solutions Coelenterazine h was obtained from Nanolight (Pinetop, AZ, USA). 5-HT (serotonin), meta-chlorophenylbiguanide (mCPBG), and ondansetron hydrochloride were obtained from Sigma (Munich, Germany). Azasetron (Y-25130), ryanodine, and thapsigargin were obtained from Biotrend (Cologne, Germany). Caffeine was purchased from Merck (Darmstadt, Germany). Ryanodine and thapsigargin solutions were prepared daily from DMSO stocks (stored as aliquots at 20 C). The other drug solutions were prepared daily from aqueous stocks (stored as aliquots at 20 C). Data analysis The raw luminescent data traces were exported as text files to GraphPad Prism software (version 4.0, San Diego, CA, USA). Peak values for the concentration-dependent response curves were obtained by subtraction of baseline luminescence from the agonist-induced peak maximum luminescence. In agonist maximum response experiments, the peak luminescence (RLUpeak) was normalized against total aequorin luminescence (RLUmax) after cell lysis to control for differences in transfection efficiency and cell number (i.e., fractional luminescence: RLUpeak/[RLUpeak + RLUmax], where RLU is relative light units). The concentration-dependent response curves, as well as the corresponding constants (EC50, IC50, and Hill slope), were calculated by means of the Prism software. Data are given as means ± SEM. Statistical analysis was performed with unpaired Student’s t test. Differences were considered as significant at P < 0.05. Results Assay principle HEK293 cells were chosen to establish a functional assay for the rapid determination of functional properties of h5-HT3 receptors composed of different subunits and naturally occurring variants of the receptor. In initial experiments with cells transiently transfected with h5HT3A receptor and apoaequorin cDNAs, it could be shown that 5-HT elicits fast responses with a luminescence maximum after 3 to 8 s (Fig. 1A). 5-HT-induced luminescence reached baseline values after 30 to 50 s due to receptor desensitization, which is a typical feature of ligand-gated ion channels. For this reason, the use of a luminometer with an integrated injector is necessary to record light emission without any delay simultaneously with autoinjection
187
of the agonist. Baseline luminescence measured for 8 s prior to injection of the agonist was subtracted from the agonist-induced peak luminescence. Baseline luminescence amounted to 2.6 ± 0.4% (n = 17) of the 5-HT (10 lM)induced luminescence signal. Increasing 5-HT concentrations led to increasing luminescence peaks and steeper curves with shorter time-to-peak values. Taking this into account, light emission needed to be recorded over time periods of at least 35 and 60 s in the cases of high and low agonist concentrations, respectively. To test whether the light signal originates exclusively from aequorin luminescence induced by activation of the heterologously expressed h5-HT3A receptors, we measured 5-HT (10 lM) effects in coelenterazine h-loaded HEK293 cells transfected only with apoaequorin cDNA or only with h5-HT3A cDNA. In both cases, no luminescence signal could be recorded (data not shown). The lack of a 5-HT response indicates that the luminescence signal originates solely from aequorin and that HEK293 cells do not express 5-HT receptors natively, which could disturb the measurement of 5-HT3 receptor responses. Because activation of 5-HT3 receptors induces cell depolarization [15–19], HEK293 cells were tested for the involvement of voltage-gated calcium ion channels in the 5-HT-induced Ca2+ luminescence. We depolarized HEK293 cells by autoinjection of a 50-mM KCl solution. No K+-induced luminescence was observed (data not shown). To check the possibility that Ca2+-induced Ca2+ release (CICR) from intracellular stores (mediated primarily by ryanodine receptors) could influence the agonistinduced luminescence signal, 5-HT responses were recorded in the presence of ryanodine. Therefore, cells were pretreated either with a high concentration (100 lM) of ryanodine (to block ryanodine receptors) or with buffer containing only its solvent. The log EC50 (pEC50) value and Hill coefficient amounted to 5.75 ± 0.02 and 3.14 ± 0.53, respectively, in the presence of ryanodine and 5.77 ± 0.04 and 3.15 ± 0.08, respectively, in the absence of ryanodine (n = 3 each). Thus, ryanodine had no effect on the concentration-dependent response relationship for 5-HT. However, in the presence of ryanodine, the maximum peak response to 10 lM 5-HT was reduced to 77.8 ± 3.7% of controls (n = 6). Similar results were obtained in one experiment with cells pretreated with 10 mM caffeine and 2.5 lM thapsigargin to deplete intracellular Ca2+ stores (data not shown). Hence, the 5-HTinduced luminescence signal is due predominantly to influx of Ca2+ ions through the pore of the 5-HT3A receptor channel. The sensitivity of the aequorin assay was tested by stimulating decreasing numbers of HEK293 cells transiently transfected with apoaequorin and h5-HT3A receptor cDNA with 5-HT (10 lM) (Fig. 1B). It is evident that 5-HT-induced peak luminescence decreased with decreasing cell numbers and that the peak response was related linearly to the cell number (Fig. 1C), indicating that the assay is reliable over a broad number of cells per well.
188
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
Fig. 1. (A) Time courses of luminescence induced by increasing 5-HT concentrations in coelenterazine h-loaded HEK293 cells transiently transfected with h5-HT3A receptor subunit and apoaequorin cDNAs. Shown is the result of a representative experiment. Luminescence was measured as relative light units (RLU) at a sampling rate of 2 Hz from 8 s before until 35 to 60 s after the addition of 5-HT. (B) Luminescence induced by 5-HT (10 lM) in coelenterazine h-loaded HEK293 cells of varying density in the suspension. The cells were transiently transfected with h5-HT3A receptor subunit and apoaequorin cDNAs. Shown are mean curves calculated from triplicate measurements of luminescence in a representative experiment containing the cell numbers indicated in the figure. (C) Linear dependence of the 5-HT-induced peak luminescence values (means ± SEM) on the number of cells (see panel B). SEM values of triplicate peak luminescence values were less than 7.4% of the respective mean values. (D) Luminescence induced by 5-HT (10 lM) in coelenterazine h-loaded HEK293 cells transiently transfected with h5-HT3A receptor subunit and apoaequorin cDNAs at different ratios. Shown are peak values for the respective h5-HT3A/aequorin cDNA ratios determined in triplicate in a representative experiment (means ± SEM of the three measurements). (E) Luminescence induced by lysis of HEK293 cells transiently transfected with apoaequorin cDNA and loaded with coelenterazine h at increasing concentrations. Cell lysis was induced by the addition of aqueous Triton X-100 solution containing CaCl2 (see Materials and methods). Cells were loaded with increasing concentrations of coelenterazine h and incubated for 2.5 h at room temperature prior to the assay. Shown are means ± SEM of triplicate determinations in a representative experiment for each condition. (F) Luminescence induced by lysis (see panel E) of HEK293 cells transiently transfected with apoaequorin cDNA and loaded with coelenterazine h (5 lM) at different temperatures and loading times. Shown are means ± SEM of quadruplicate determinations for each experimental condition.
Optimization of transfection and coelenterazine h loading protocol To determine the amounts of h5-HT3A receptor and apoaequorin cDNA that yield the highest peak luminescence values, HEK293 cells were transfected with different receptor/apoaequorin cDNA ratios. Transfection of cells with increasing apoaequorin and decreasing receptor cDNA amounts led to enhanced 5-HT (10 lM)-induced peak luminescence values; a h5-HT3A/apoaequorin cDNA ratio of 1:4 yielded approximately a 5.5 times greater light emission as compared with equal amounts of cDNAs for both proteins (Fig. 1D). For further optimization, HEK293 cells transiently expressing apoaequorin were incubated for a time period of 3.5 h at room temperature with different concentrations (1–10 lM) of coelenterazine h. The maximum luminescence response, depending on the total amount of reconstituted aequorin, increased with rising coelenterazine h concentrations to reach a maximum at 7.5 lM (Fig. 1E). In subsequent experiments, 5 lM coelenterazine h, which induced nearly 80% of the maximum luminescence value, was used. In the next step samples of apoaequorin-expressing cells were loaded with coelenterazine h at two different temperatures (37 C and room temperature) for different times. Incubation at 37 C caused lower luminescence peaks after
cell lysis, and thus lower reconstituted aequorin amounts, than did incubation at room temperature (Fig. 1F). An incubation time of 3.5 h at room temperature yielded only a slightly increased total amount of aequorin as compared with an incubation time of 2 h; therefore, a loading time of 2.5 h at room temperature was chosen for all further experiments. Potency of 5-HT3 receptor ligands at the homopentameric h5-HT3A receptor Experiments performed with 5-HT (i.e., the physiological agonist at 5-HT receptors) and with the selective 5-HT3 receptor agonist mCPBG yielded concentrationdependent increases in aequorin luminescence. The agonist-induced peak luminescence resulted in concentration-dependent response curves (Fig. 2A) characterized by a pEC50 value of 5.76 ± 0.03 (mean EC50 = 1.77 lM, n = 11) for 5-HT and 5.76 ± 0.02 (mean EC50 = 1.75 lM, n = 7) for mCPBG, with Hill slopes for 5-HT and mCPBG amounting to 3.19 ± 0.26 and 2.76 ± 0.19, respectively. Responses to a maximum 5-HT concentration (10 lM) were concentration-dependently inhibited by the 5-HT3 receptor antagonists ondansetron and azasetron with log IC50 (pIC50) values of 9.00 ± 0.02 (n = 4) and 8.96 ± 0.03 (n = 5), respectively (Fig. 2B). Corrections of
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
189
Fig. 3. Concentration-dependent response curves for 5-HT on coelenterazine h-loaded HEK293 cells transiently transfected with the cDNAs of apoaequorin and either the h5-HT3A receptor subunit alone (black) or both the h5-HT3A and h5-HT3B receptor subunits (gray). 5-HT-induced luminescence is expressed as a percentage of maximum luminescence produced by the agonist at the respective homopentameric h5-HT3A or heteromeric h5-HT3A/B receptor. Shown are means ± SEM of 4 to 11 independent transfections. pEC50 values at the homopentameric h5-HT3A receptor and the heteromeric h5-HT3A/B receptor amounted to 5.75 ± 0.02 and 5.45 ± 0.08, respectively (P < 0.05).
Fig. 2. Luminescence induced by increasing concentrations of 5-HT3 receptor agonists (A) or by 5-HT (10 lM) in the presence of increasing concentrations of 5-HT3 receptor antagonists (B) in coelenterazine h-loaded HEK293 cells transiently transfected with h5-HT3A receptor subunit and apoaequorin cDNAs. (A) Concentration-dependent response curves for the agonists 5-HT and mCPBG. Agonist-induced luminescence is expressed as a percentage of maximum peak luminescence produced by the respective agonist. (B) Inhibition of 5-HT (10 lM)-induced luminescence by increasing concentrations of the 5-HT3 receptor antagonists ondansetron and azasetron (Y-25130). The antagonist was present 15 min before and during the application of 5-HT. Data are expressed as percentages of the 5-HT (10 lM) response in the absence of the antagonist. Shown are means ± SEM of 4 to 11 independent transfections.
these values according to Cheng and Prusoff [20] results in pKi (Ki) values of 9.82 (0.15 nM) and 9.78 (0.17 nM) for ondansetron and azasetron, respectively. 5-HT potency and efficacy at the heteromeric h5-HT3A/B receptor Coexpression of the h5-HT3B subunit together with the h5-HT3A subunit led to a slightly reduced pEC50 value for 5-HT as compared with that at the homopentameric h5-HT3A receptor (5.45 ± 0.08 [mean EC50 = 3.55 lM] vs. 5.76 ± 0.03 [mean EC50 = 1.77 lM], P < 0.05). The Hill coefficient of the concentration-dependent response curve at the heteromeric h5-HT3A/B receptor was also reduced (1.45 ± 0.29 vs. 3.19 ± 0.26, P < 0.01) (Fig. 3). To examine whether the maximum 5-HT-induced luminescence response (Emax) at the heteromeric h5-HT3A/B receptor differs from that at the homopentameric h5-HT3A receptor, the receptors were stimulated with 500 lM 5-HT and luminescence was recorded. The Emax value (i.e., fractional luminescence [see Materials and methods]) for 5-HT at the heteromeric h5-HT3A/B receptor was increased signifi-
cantly to 165.9 ± 21.3% (n = 6, P < 0.05) of the Emax at the homopentameric h5-HT3A receptor. Discussion The aim of the study was to establish a functional assay for the characterization of 5-HT3 receptors using aequorinbased luminescence. We adapted a sensitive and reliable method that was optimized for the high-throughput analysis of Ca2+ influx through 5-HT3 receptors composed of varying subunits. The standard method for the characterization of 5-HT3 receptors and of other ligand-gated ion channels is the patch clamp analysis. It is the only method suitable to analyze the kinetic properties of the channel. However, it is very time-consuming and, thus, not suited for a fast and high-throughput analysis of the function of 5-HT3 receptors composed of different subunits. Other methods applied for the measurement of 5-HT3 receptors are, for example, the [14C]guanidinium influx [21] and the use of Ca2+-sensitive fluorescent dyes [4,7]. The first method mentioned has major disadvantages such as the use of radioactivity and a low signal/noise ratio. In addition, the artificial buffer used in these experiments, where Na+ ions are replaced by larger organic cations such as choline and N-methyl-D-glucamine that cannot pass the channel pore [22], represents artificial conditions that may falsify the results. Although the determination of Ca2+ influx by measuring fluorescence on binding of Ca2+ ions to Ca2+-sensitive dyes has been used successfully for the characterization of 5-HT3 receptors in stably transfected cells, there are some drawbacks, especially in the case of transient transfections. It is not possible to load the dye selectively into transfected cells; this causes a high background fluorescence because the signal originates from the whole cell population (this is shown, e.g.,
4.3 5.7 ND ND ND ND 9.3 ND 0.8 ND Note. ND, not determined. a Ki values were calculated according to the equation of Cheng and Prusoff [20]. b In this study, cells were transfected with mouse 5-HT3A receptor. c Ki value (lM) is given.
ND 0.1 ND ND ND ND ND 0.1 0.2 0.2 Antagonists Azasetron Ondansetron
ND ND
0.4c 0.3c 1.1 0.2 ND 0.2 0.3 0.5 8.5 5.9 8.6 ND 3.0 ND 2.1 0.8 3.4 ND 2.9 2.5 1.8 1.8 Agonists 5-HT mCPBG
Fura-2 fluorescence (FlexStation) [8]b Fura-2 fluorescence (single cells) [7]b FLIPR assay [4] Excised outside-out patch clamp [28] Excised outsideout patch clamp [25,30] Whole cell patch clamp [32] Whole cell patch clamp [29]b Whole cell patch clamp [6] Whole cell patch clamp [3] Aequorin assay (current study) Compound
EC50 value (lM) for agonists (unless stated otherwise) and Ki valuesa for antagonists (nM)
at heterologously expressed a1-adrenergic receptors [23]). For this reason, we applied an aequorin luminescencebased Ca2+ assay whose use as a detector of Ca2+ fluctuations in living cells is well established [12]. This assay takes advantage of the fact that after transient cotransfection of the cells with apoaequorin and receptor cDNA, the encoded proteins are expressed in the same subset of transfected cells, leading to a very high signal/noise ratio because luminescence originates only from transfected cells. On the other hand, the total amount of cellular aequorin measured at the end of the experiment by cell lysis in the presence of a high external Ca2+ concentration can be used as an internal control for transfection efficiency. HEK293 cells were chosen for the heterologous expression of 5-HT3 receptors because they do not express serotonin receptors natively. We proved this in cells transfected only with the apoaequorin cDNA. In these cells, a high concentration (10 lM) of the agonist 5-HT caused no increase in luminescence over the baseline (data not shown). Furthermore, HEK293 cells were examined for the expression of voltage-gated calcium channels that can be opened by depolarization with KCl. No K+-induced luminescence could be recorded (data not shown), indicating the lack of expression of such calcium channels in HEK293 cells that confirms the results of previous studies [7,24]. Furthermore, CICR from intracellular stores contributed to only 20% of the 5-HT-induced maximum response. Hence, it may be concluded that the luminescence signal is due predominantly to Ca2+ influx through the 5-HT3 receptor channel. In initial experiments with the agonist 5-HT, we observed that with increasing 5-HT concentrations, the luminescence curves became steeper and reached baseline values within a shorter period of time. Although much faster, the same relationship of rising and decay rate of the recorded currents on the 5-HT concentration was obtained in patch-clamp experiments [25]. To improve reproducibility of our luminescence-based assay, we first optimized parameters such as the loading procedure of the cells with the cofactor coelenterazine and obtained satisfying results with a loading time of 2.5 h with 5 lM coelenterazine. Furthermore, the sensitivity of the assay was enhanced by using the synthetic cofactor coelenterazine h. Aequorins containing the h form of coelenterazine have been reported to be more sensitive to Ca2+ and exhibit a 16 times greater luminescence intensity as compared with aequorin reconstituted with native coelenterazine [26,27]. In addition, the amounts of cDNAs encoding apoaequorin and the 5-HT3 receptor were optimized to obtain high luminescence peaks on the application of the agonist. We observed that the height of the luminescence peak depends strongly on the amount of apoaequorin cDNA used for transfection; the higher the ratio of apoaequorin cDNA over receptor cDNA, the greater the maximum 5-HT-induced luminescence response. This is consistent
[3H]GR65630 binding [28]
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
Table 1 Potencies or affinities of 5-HT3 receptor ligands obtained with different methods at the h5-HT3A receptor heterologously expressed in HEK293 cells
190
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
with results reported by Sheu and coworkers [12], who observed that the signal is proportional to the amount of aequorin in the cell. This optimization is especially useful for the characterization of 5-HT3 receptor isoforms that are expressed in a low density on the cell surface. The EC50 values for the two receptor agonists, 5-HT and mCPBG, were not different from those determined previously for 5-HT or mCPBG with other techniques at the h5-HT3A receptor in HEK293 cells (Table 1). For both agonists, the Hill coefficients were similar to those obtained with other techniques such as [14C]guanidinium influx [31] and fluorescence determination with Ca2+-sensitive dyes performed either on single cells [7] or with a fluorometric imaging plate reader (FLIPR) assay [4]. The 5-HT (10 lM)-induced responses were dose-dependently blocked by the specific 5-HT3 receptor antagonists ondansetron and azasetron, and the Ki values were very similar to the values obtained previously (Table 1). In previous studies [3,4,32], different pharmacological properties of 5-HT at heteromeric h5-HT3A/B and homopentameric h5-HT3A receptors have been reported. In the current study, with the aequorin-based assay, we could confirm that the affinity of 5-HT is lower at the h5-HT3A/B receptor than at the homopentameric h5-HT3A receptor and that the Hill slope of the dose–response curve is smaller with the heteromeric h5-HT3A/B receptor than with the homopentameric h5-HT3A receptor, indicating a loss of cooperativity for 5-HT at the heteromeric h5-HT3A/B receptor. In addition, the maximum response (Emax) for 5-HT was approximately 1.7-fold greater at the heteromeric h5-HT3A/B receptor than at the homopentameric h5-HT3A receptor; this is compatible with the proposed higher single channel conductance of the heteromeric h5-HT3A/B receptor. Thus, the aequorin assay described here is well suited to study, for example, the pharmacological properties of 5-HT3 receptors containing further subunits described recently [5]. In conclusion, we have reported the establishment of optimized conditions for an aequorin luminescence-based assay in 96-well plates for the characterization of the 5-HT3 receptor. This functional assay provides a highly sensitive and rapid method appropriate for high-throughput screening of Ca2+-permeable ligand-gated ion channels. Acknowledgment This research was supported by the Deutsche Forschungsgemeinschaft (DFG, BR 1741/2-1). References [1] V. Derkach, A. Surprenant, R.A. North, 5-HT3 receptors are membrane ion channels, Nature 339 (1989) 706–709. [2] J.A. Peters, T.G. Hales, J.J. Lambert, Molecular determinants of single-channel conductance and ion selectivity in the Cys-loop family: insights from the 5-HT3 receptor, Trends Pharmacol. Sci. 26 (2005) 587–594.
191
[3] P.A. Davies, M. Pistis, M.C. Hanna, J.A. Peters, J.J. Lambert, T.G. Hales, E.F. Kirkness, The 5-HT3B subunit is a major determinant of serotonin-receptor function, Nature 397 (1999) 359–363. [4] A.E. Dubin, R. Huvar, M.R. D’Andrea, J. Pyati, J.Y. Zhu, K.C. Joy, S.J. Wilson, J.E. Galindo, C.A. Glass, L. Luo, M.R. Jackson, T.W. Lovenberg, M.G. Erlander, The pharmacological and functional characteristics of the serotonin 5-HT3A receptor are specifically modified by a 5-HT3B receptor subunit, J. Biol. Chem. 274 (1999) 30799–30810. [5] B. Niesler, B. Frank, J. Kapeller, G.A. Rappold, Cloning, physical mapping, and expression analysis of the human 5-HT3 serotonin receptor-like genes HTR3C, HTR3D, and HTR3E, Gene 310 (2003) 101–111. [6] A.M. Brown, A.G. Hope, J.J. Lambert, J.A. Peters, Ion permeation and conduction in a human recombinant 5-HT3 receptor subunit (h5HT3A), J. Physiol. 507 (1998) 653–665. [7] A.C. Hargreaves, S.C. Lummis, C.W. Taylor, Ca2+ permeability of cloned and native 5-hydroxytryptamine type 3 receptors, Mol. Pharmacol. 46 (1994) 1120–1128. [8] K.L. Price, S.C. Lummis, FlexStation examination of 5-HT3 receptor function using Ca2+- and membrane potential-sensitive dyes: advantages and potential problems, J. Neurosci. Methods 149 (2005) 172–177. [9] O. Shimomura, F.H. Johnson, Regeneration of the photoprotein aequorin, Nature 256 (1975) 236–238. [10] O. Shimomura, Luminescence of aequorin is triggered by the binding of two calcium ions, Biochem. Biophys. Res. Commun. 211 (1995) 359–363. [11] S. An, T. Bleu, Y. Zheng, Transduction of intracellular calcium signals through G protein-mediated activation of phospholipase C by recombinant sphingosine 1-phosphate receptors, Mol. Pharmacol. 55 (1999) 787–794. [12] Y.A. Sheu, L.J. Kricka, D.B. Pritchett, Measurement of intracellular calcium using bioluminescent aequorin expressed in human cells, Anal. Biochem. 209 (1993) 343–347. [13] M.D. Ungrin, L.M. Singh, R. Stocco, D.E. Sas, M. Abramovitz, An automated aequorin luminescence-based functional calcium assay for G-protein-coupled receptors, Anal. Biochem. 272 (1999) 34–42. [14] M.S. Karadsheh, M.S. Shah, X. Tang, R.L. Macdonald, J.A. Stitzel, Functional characterization of mouse a4b2 nicotinic acetylcholine receptors stably expressed in HEK293T cells, J. Neurochem. 91 (2004) 1138–1150. [15] D.I. Wallis, R.A. North, The action of 5-hydroxytryptamine on single neurones of the rabbit superior cervical ganglion, Neuropharmacology 17 (1978) 1023–1028. [16] J.R. Fozard, Neuronal 5-HT receptors in the periphery, Neuropharmacology 23 (1984) 1473–1486. [17] J. Azami, J.R. Fozard, A.A. Round, D.I. Wallis, The depolarizing action of 5-hydroxytryptamine on rabbit vagal primary afferent and sympathetic neurones and its selective blockade by MDL 72222, Naunyn Schmiedebergs Arch. Pharmacol. 328 (1985) 423–429. [18] A. Round, D.I. Wallis, The depolarizing action of 5-hydroxytryptamine on rabbit vagal afferent and sympathetic neurones in vitro and its selective blockade by ICS 205-930, Br. J. Pharmacol. 88 (1986) 485–494. [19] J.A. Peters, H.M. Malone, J.J. Lambert, Recent advances in the electrophysiological characterization of 5-HT3 receptors, Trends Pharmacol. Sci. 13 (1992) 391–397. [20] Y. Cheng, W.H. Prusoff, Relationship between the inhibition constant (K1) and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction, Biochem. Pharmacol. 22 (1973) 3099–3108. [21] H. Bo¨nisch, M. Barann, J. Graupner, M. Go¨thert, Characterization of 5-HT3 receptors of N1E-115 neuroblastoma cells by use of the influx of the organic cation [14C]-guanidinium, Br. J. Pharmacol. 108 (1993) 436–442. [22] M. Barann, K. Schmidt, M. Go¨thert, B.W. Urban, H. Bo¨nisch, Influence of sodium substitutes on 5-HT-mediated effects at mouse 5-HT3 receptors, Br. J. Pharmacol. 142 (2004) 501–508.
192
Assay for 5-HT3 receptor characterization / J. Walstab et al. / Anal. Biochem. 368 (2007) 185–192
[23] M. Brini, R. Marsault, C. Bastianutto, J. Alvarez, T. Pozzan, R. Rizzuto, Transfected aequorin in the measurement of cytosolic Ca2+ concentration ([Ca2+]c): a critical evaluation, J. Biol. Chem. 270 (1995) 9896–9903. [24] M.E. Williams, P.F. Brust, D.H. Feldman, S. Patthi, S. Simerson, A. Maroufi, A.F. McCue, G. Velicelebi, S.B. Ellis, M.M. Harpold, Structure and functional expression of an omega-conotoxin-sensitive human N-type calcium channel, Science 257 (1992) 389–395. [25] M. Barann, W. Meder, Z. Dorner, M. Bru¨ss, H. Bo¨nisch, M. Go¨thert, B.W. Urban, Recombinant human 5-HT3A receptors in outside-out patches of HEK 293 cells: basic properties and barbiturate effects, Naunyn Schmiedebergs Arch. Pharmacol. 362 (2000) 255–265. [26] M.R. Knight, N.D. Read, A.K. Campbell, A.J. Trewavas, Imaging calcium dynamics in living plants using semi-synthetic recombinant aequorins, J. Cell Biol. 121 (1993) 83–90. [27] O. Shimomura, Y. Kishi, S. Inouye, The relative rate of aequorin regeneration from apoaequorin and coelenterazine analogues, Biochem. J. 296 (1993) 549–551.
[28] D. Kurzwelly, M. Barann, A. Kostanian, S. Combrink, H. Bo¨nisch, M. Go¨thert, M. Bru¨ss, Pharmacological and electrophysiological properties of the naturally occurring Pro391Arg variant of the human 5-HT3A receptor, Pharmacogenetics 14 (2004) 165–172. [29] A.D. Spier, S.C. Lummis, The role of tryptophan residues in the 5hydroxytryptamine-3 receptor ligand binding domain, J. Biol. Chem. 275 (2000) 5620–5625. [30] M. Barann, M. Bruess, M. Brinkmann, I. Linden, M. Lyutenska, M. Schneider, J. Walkembach, M. Wittmann, 5-HT3 receptors and emesis, Pharmacol. Rep. 58 (2006) 253–255. [31] M. Bru¨ss, M. Barann, M. Hayer-Zillgen, T. Eucker, M. Go¨thert, H. Bo¨nisch, Modified 5-HT3A receptor function by co-expression of alternatively spliced human 5-HT3A receptor isoforms, Naunyn Schmiedebergs Arch. Pharmacol. 362 (2000) 392–401. [32] G. Hapfelmeier, C. Tredt, R. Haseneder, W. Zieglgansberger, B. Eisensamer, R. Rupprecht, G. Rammes, Co-expression of the 5-HT3B serotonin receptor subunit alters the biophysics of the 5-HT3 receptor, Biophys. J. 84 (2003) 1720–1733.