Abstract
Trace amine-associated receptor 1 (TAAR1) has been implicated in the behavioral effects of amphetamine-type stimulant drugs in rodents. TAAR1 has also been suggested as a target for novel medications to treat psychostimulant addiction. We previously reported that binding affinities at TAAR1 can differ between structural analogs of psychostimulants, and species differences have been observed. In this study, we complement our previous findings with additional substances and the determination of functional activation potencies. In summary, we present here pharmacological in vitro profiles of 101 psychoactive substances at human, rat, and mouse TAAR1. p-Tyramine, β-phenylethylamine, and tryptamine were included as endogenous comparator compounds. Functional cAMP measurements and radioligand displacement assays were conducted with human embryonic kidney 293 cells that expressed human, rat, or mouse TAAR1. Most amphetamines, phenethylamine, and aminoindanes exhibited potentially physiologically relevant rat and mouse TAAR1 activation (EC50 < 5 µM) and showed full or partial (Emax < 80%) agonist properties. Cathinone derivatives, including mephedrone and methylenedioxypyrovalerone, exhibited weak (EC50 = 5–10 µM) to negligible (EC50 > 10 µM) binding properties at TAAR1. Pipradrols, including methylphenidate, exhibited no affinity for TAAR1. We found considerable species differences in activity at TAAR1 among the highly active ligands, with a rank order of rat > mouse > human. This characterization provides information about the pharmacological profile of psychoactive substances. The species differences emphasize the relevance of clinical studies to translationally complement rodent studies on the role of TAAR1 activity for psychoactive substances.
Introduction
Trace amine-associated receptor 1 (TAAR1) is a relatively recently discovered G protein–coupled receptor (Borowsky et al., 2001; Bunzow et al., 2001), which is expressed in monoaminergic brain regions and throughout the limbic system (Borowsky et al., 2001; Lindemann et al., 2008; Espinoza et al., 2015). TAAR1 is thought to play a role in regulating the limbic network, reward circuits, cognitive processes, and mood states and has been proposed as a pharmacological target for the treatment of mental disorders (Wolinsky et al., 2007; Lindemann et al., 2008; Miller, 2011; Revel et al., 2013) and psychostimulant dependence (Di Cara et al., 2011; Pei et al., 2014; Cotter et al., 2015; Jing and Li, 2015). TAAR1 is stimulated by endogenous ligands, including β-phenylethylamine (β-PEA), p-tyramine, tryptamine, and 3-iodothyronamine (Scanlan et al., 2004; Zucchi et al., 2006). Many psychoactive compounds, including amphetamine and phenethylamine derivatives, also bind to TAAR1 (Bunzow et al., 2001; Wainscott et al., 2007; Simmler et al., 2013; Reese et al., 2014). The activation of TAAR1 results in elevations in intracellular cAMP (Bunzow et al., 2001; Xie and Miller, 2007).
Amphetamines have structural similarity to the endogenous ligand β-PEA and were initially identified as TAAR1 ligands (Bunzow et al., 2001). We previously reported that many novel psychoactive substances are also ligands of rat and mouse TAAR1 (Simmler et al., 2013, 2014a,b; Rickli et al., 2015a,b,c). However, several novel psychoactive substances do not bind to TAAR1, and little has been reported on the activation of human TAAR1. The pharmacological and toxicological actions of novel psychoactive substances are also of interest because of the emergence of hundreds of these substances, referred to as “legal highs” or “research chemicals.” These chemical compounds are recreationally used but have poorly known pharmacological properties.
TAAR1 is implicated in the control of neuronal firing frequency and is thus likely to contribute to psychoactive and abuse-related drug effects. Ex vivo electrophysiology experiments that used slices from TAAR1 knockout (KO) mice (Lindemann et al., 2008) or pharmacological TAAR1 blockade (Bradaia et al., 2009) suggest that TAAR1 is constitutively active to control dopamine (DA) and serotonin [5-hydroxytryptamine (5-HT)] tone.
Compared with wild-type (WT) mice, TAAR1 KO mice were shown to consume more ethanol and be more susceptible to its sedating effects (Lynch et al., 2013). The TAAR1 partial agonist RO5203648 [(S)-4-(3,4-dichlorophenyl)-4,5-dihydrooxazol-2-amine dihydrochloride] reduced cocaine self-administration and cocaine-induced hyperlocomotion in rats (Revel et al., 2012b). Both selective TAAR1 partial agonists and selective TAAR1 full agonists reduced cocaine self-administration and the reinstatement of drug-seeking behavior in rats (Pei et al., 2014, 2015) and decreased cocaine-mediated intracranial self-stimulation (Pei et al., 2015). Reductions of hyperlocomotion, self-administration, and reinstatement by TAAR1 partial agonism have also been reported for methamphetamine (Cotter et al., 2015; Jing and Li, 2015). These studies established TAAR1 as a promising target for therapeutics to treat substance use disorders, regardless of the TAAR1 binding properties of the abused substances themselves. By directly interacting with TAAR1, psychoactive substances may also modulate their own pharmacological effects. For example, amphetamine induces markedly more striatal monoamine release in TAAR1 KO mice than in WT mice (Lindemann et al., 2008). Methamphetamine and amphetamine increase locomotor activity to a greater extent in TAAR1 KO mice compared with WT mice (Achat-Mendes et al., 2012). TAAR1 also plays a role in contingent oral methamphetamine intake (Harkness et al., 2015). Similar to amphetamine and methamphetamine, 3,4-methylenedioxymethamphetamine (MDMA) significantly increased extracellular striatal DA and 5-HT levels to a greater extent in TAAR1 KO mice compared with WT mice (Di Cara et al., 2011). TAAR1 KO mice are hypersensitive to psychoactive substances that are also TAAR1 ligands. By contrast, TAAR1 overexpression in mice reduced locomotor activity in response to amphetamine (Revel et al., 2012a). Because MDMA, methamphetamine, and amphetamine are TAAR1 ligands, they possibly autoinhibit their own effects on neurotransmitter release. Di Cara et al. (2011) supported the concept of the autoregulation of TAAR1-activating psychostimulants, showing that the TAAR1 ligand o-phenyl-3-iodotyramine decreased the DA release response to p-chloroamphetamine, which is a psychostimulant that is inactive at TAAR1, in WT mice but not in TAAR KO mice.
Because TAAR1 might be significantly involved in the mode of action of many psychoactive drugs, we determined the TAAR1 binding and activation properties of a series of mostly novel substances and found considerable differences in TAAR1 binding properties within and between substance classes. Our data set provides evidence of significant species differences in ligand/receptor interactions between rodent and human TAAR1.
Materials and Methods
Chemicals.
The compounds were purchased from Lipomed (Arlesheim, Switzerland) or Cayman Chemicals (Ann Arbor, MI) as racemic mixtures, with the exception of d-amphetamine, d-methamphetamine, (+)-ephedrine, and (−)-ephedrine. A list of generic or full chemical names is provided in (Supplemental Table 1). 5-EAPB, diclophensine, diphenidine, ethylphenidate, methoxphenidine, and N-methyl-2-AI were obtained from the Forensic Institute Zürich (Zürich, Switzerland). Naphyrone and MDAI were synthesized in our laboratory as reported previously (Simmler et al., 2013, 2014b). Radiochemicals (3H-isotopes) were purchased from PerkinElmer (Schwerzenbach, Switzerland), with the exception of [3H]RO5166017 [(S)-4-[(ethyl-phenyl-amino)-methyl]-4,5-dihydro-oxazol-2-ylamine], which was synthesized at Roche (Basel, Switzerland).
Cell Culture and Membrane Preparation.
Human embryonic kidney 293 cells that stably expressed human, rat, or mouse TAAR1 were used as described previously (Revel et al., 2011). All of the cell lines were maintained at 37°C and 5% CO2 in high-glucose Dulbecco’s modified Eagle’s medium that contained 10% fetal calf serum (heat-inactivated for 30 minutes at 56°C), 1% penicillin/streptomycin, and 375 μg/ml Geneticin (Gibco, Zug, Switzerland). For membrane preparation, the cells were released from culture flasks using trypsin/EDTA, harvested, washed twice with ice-cold phosphate-buffered saline (PBS; without Ca2+ and Mg2+), pelleted at 1000 × g for 5 minutes at 4°C, frozen, and stored at −80°C. Frozen pellets were suspended in buffer A [20 ml HEPES-NaOH (20 mM, pH 7.4) that contained 10 mM EDTA] and homogenized with a Polytron (PT 6000; Kinematica, Luzern, Switzerland) at 14,000 rpm for 20 seconds. The homogenate was centrifuged for 30 minutes at 48,000 × g at 4°C. The supernatant was removed and discarded, and the pellet was resuspended in buffer A using the Polytron (20 seconds at 14,000 rpm). The centrifugation and removal of the supernatant was repeated, and the final pellet was resuspended in buffer A and homogenized using the Polytron. Typically, 2-ml aliquots of membrane portions were stored at −80°C. With each new membrane batch, the dissociation constant (Kd) was determined by a saturation curve.
Radioligand Binding Assay.
For the competitive binding assays, the TAAR1 agonist [3H]RO5166017 was used as a TAAR1 radioligand at a concentration equal to Kd values, which was usually around 0.7 nM (mouse TAAR1) and 2.3 nM (rat TAAR1). Nonspecific binding was defined as the amount of radioligand bound in the presence of 10 µM RO5166017. Compounds were tested at a broad range of concentrations (10 pM to 10 μM) in duplicate. Compounds (20 µl/well) were transferred to a 96-deep-well plate (TreffLab, Degersheim, Switzerland), and 180 µl binding buffer (20 mM HEPES-NaOH, 10 mM MgCl2, and 2 mM CaCl2, pH 7.4), 300 µl radioligand, and 500 µl membranes (resuspended at 60 µg protein/ml) were added. The plates were incubated at 4°C for 90 minutes. Incubations were terminated by rapid filtration through Unifilter-96 plates (Packard Instrument Company, PerkinElmer) and glass filters GF/C (PerkinElmer) presoaked for 1 hour in polyethylenimine (0.3%) and washed three times with 1 ml cold binding buffer. After the addition of 45 µl Microscint 40 (PerkinElmer), the Unifilter-96 plate was sealed. After 1 hour, radioactivity was counted using a TopCount Microplate Scintillation Counter (Packard Instrument Company). IC50 values were determined by calculating nonlinear regression curves for a one-site model using at least three independent 10-point concentration-response curves, run in duplicate, for each compound. Ki (affinity) values, which correspond to the dissociation constants, were determined using the Cheng–Prusoff equation. Ki values are presented as means ± S.D. (in micromoles). For reasons of integrity, Table 1 includes several Ki values that we have previously published as indicated by the references in the table.
Binding affinities, activation potencies, and efficacy of psychoactive substances and endogenous comparator compounds at rat, mouse, and human TAAR1
Values are given as means ± S.D. Ki values of human TAAR1 were not determined because of the lack of a reliable radioligand needed for the in vitro assay. EC50 was determined for substances with relevant binding (Ki value < 10 µM). Ki values from our previous publications are included and referenced. The generic or full chemical names for abbreviated substances are listed in Supplemental Table 1.
Functional TAAR1 Activity.
Substances were tested for binding affinity at rat and mouse TAAR1 as described above. If relevant binding was observed, then we also determined potencies for receptor activation and maximal efficacy at rat, mouse, and human TAAR1 to characterize the compounds as full or partial agonists (Emax < 80%). The endogenous TAAR1 ligands β-PEA, p-tyramine, and tryptamine served as reference substances for comparisons of affinity values and functional potency and efficacy. cAMP measurements were performed as described previously (Revel et al., 2011). In brief, cells that expressed rat or mouse TAAR1 were plated on 96-well plates (BIOCOAT 6640; Becton Dickinson, Allschwil, Switzerland) and incubated for 20 hours at 37°C. Prior to stimulation of the cells with a broad concentration range of agonists for 30 minutes at 37°C, the cells were washed with PBS and preincubated with PBS that contained 1 mM 3-isobutyl-1-methylxanthine for 10 minutes at 37°C and 5% CO2. Stimulation with 0.2% dimethylsulfoxide was set as the basal level, and the effect of 30 μM β-PEA was set as the maximal response. Subsequently, the cells were lysed, and cAMP assays were performed according to the manufacturer’s instructions (cAMP kit; Upstate/Millipore, Schaffhausen, Switzerland). Finally, the plates were read with a luminometer (1420 Multilabel Counter; PerkinElmer), and the amount of cAMP was calculated. The results were obtained from at least three independent experiments. Experiments were run in duplicate or triplicate. EC50 values are presented as means ± S.D. (in micromoles). The Emax value for the functional activity data at TAAR1 describes the degree of functional activity compared with 100% for the endogenous ligand and full agonist β-PEA.
Results
The binding affinity values (Ki), receptor activation potencies (EC50), and maximal efficacy (Emax) of 104 substances are summarized in Table 1. These substances represent the collection of compounds in our laboratory that have been used to characterize the in vitro pharmacology of novel designer drugs (for review, see Liechti, 2015). The substances for these studies were chosen based on the availability of pure chemical compounds and the reported abuse of these substances. For a subset of compounds, we have previously published rat and mouse TAAR1 affinities as indicated by references in Table 1, but no human, rat, and mouse activity data. All substances were grouped according to their basic chemical structure (Fig. 1) as phenethylamines, amphetamines, cathinones, ephedrines, tryptamines, aminoindanes, pipradrols, and piperazines. A few psychoactive substances, such as cocaine and lysergic acid diethylamide, were added but not classified because of the lack of common basic structures.
Chemical structures of basic compounds (in bold) and location for derivatization (indicated by “R”) for compounds included in the data set. Residues for each compound are specified in Fig. 2 and Supplemental Table 1. Structural analogs were grouped according to the basic chemical structures for presentation of the data in tables and heat maps (Fig. 2). The generic or full chemical names for abbreviated substances are listed in Supplemental Table 1.
TAAR1 Binding and Functional TAAR1 Activation.
We found marked differences in affinities at TAAR1 and the functional activation of TAAR1 among the various psychoactive substances. The ligand properties varied considerably within substance classes, with the exception of cathinones and pipradrols, which generally exhibited none to very weak binding to TAAR1. We also observed species differences in TAAR1 activation. At the human TAAR1, only 19 substances had EC50 values that indicated functional activation (<10 µM), whereas the EC50 values were < 10 µM for 52 and 68 substances at the mouse and rat TAAR1, respectively. Therefore, below we present the properties of the psychoactive compounds compared with the endogenous TAAR1 ligands β-PEA, p-tyramine, and tryptamine separately for each species.
Human TAAR1.
β-PEA and p-tyramine activated the human TAAR1 with EC50 values of 0.26 and 0.99 µM, respectively, and showed full agonistic properties (Emax = 104% and 91%, respectively). Amphetamine, 7-APB, and 2-AI had potency (EC50 = 0.6–2.8 μM) and agonist efficacy (Emax > 89%) that were comparable to β-PEA and p-tyramine at human TAAR1. Methamphetamine exhibited an EC50 of 5.3 µM and 70% efficacy. The endogenous ligand tryptamine exhibited weak activation of human TAAR1, with an EC50 of 21 µM (Emax = 73%). Generally, most of the psychoactive compounds that were tested were weak human TAAR1 ligands, and none of them were more potent than the endogenous ligand β-PEA.
Rat TAAR1.
At the rat TAAR1, β-PEA had a Ki of 0.24 µM and showed full agonism, with an EC50 of 0.11 µM, whereas p-tyramine was more potent, with a Ki of 0.06 µM and EC50 of 0.03 µM, with full agonist properties (Emax = 94%). The third endogenous ligand tested, tryptamine, showed similar affinity (Ki = 0.13 µM) to β-PEA and p-tyramine and slightly lower functional activity (EC50 = 0.41 µM). Several of the screened amphetamines (5-APB, 6-APB, 7-APB, and 4-fluoroamphetamine) and phenethylamines (2C-P, 2C-T2, 2C-T4, and 2C-T7) and the aminoindane 5-IAI had affinities and EC50 values that were comparable to p-tyramine, the most potent of the three endogenous ligands at rat TAAR1, but none of these had a more potent EC50 than p-tyramine. The majority of these potent novel psychoactive substances exhibited full agonist properties (Emax > 80%) at rat TAAR1. The Emax values of 4-fluoroamphetamine and 2C-T4 were 78% and 67%, suggesting partial agonism.
The piperazine m-CPP and several amphetamines (4-APB, 4-fluoromethamphetamine, 5-IT, 4-methylamphetamine, and 4-MTA), phenethylamines (2C-B, 2C-B-Fly, 2C-C, 2C-E, 2C-I, and 2C-N), and aminoindanes (2-AI, N-methyl-2-AI, and MDAI) were comparable to the less potent endogenous ligands β-PEA and tryptamine in their affinities and functional potencies at rat TAAR1. The majority of these compounds were partial agonists. Amphetamine and the well known amphetamine derivatives methamphetamine, MDMA, and MDA were slightly less active than the structurally related endogenous ligand β-PEA.
Mouse TAAR1.
β-PEA and p-tyramine had equal affinities at mouse TAAR1, with Ki values of 0.31 and 0.38 µM, respectively, and full agonist properties, with EC50 values of 0.2 and 0.28 µM, respectively. Various amphetamines (amphetamine, 6-APDB, 4-fluoroamphetamine, 5-IT, MDA, and PMA) and one phenethylamine (2C-P) showed similar binding affinities and functional potencies to these endogenous TAAR1 ligands, whereas some amphetamines were even more potent (5-APB, 6-APB, 7-APB, 4-methylamphetamine, and 4-MTA), with mostly full agonist properties. The endogenous ligand tryptamine was slightly weaker than β-PEA and p-tyramine, and many phenethylamines (25E-NB2OMe, 25P-NB2OMe, 25T7-NB2OMe, 2C-E, and 2C-T7), amphetamines (4-APB, 4-fluoroamphetamine, 5-IT, methamphetamine, and PMMA), and aminoindanes (2-AI, N-methyl-2-AI, and MDAI) were similarly active as β-PEA and p-tyramine, although all of them were partial agonists (Emax = 36%–78%), with the exception of PMMA and MDAI (Emax = 82% and 99%, respectively). Interestingly, binding affinity did not always predict functional potency, such as with the aminoindanes, which showed functional activities similar to amphetamine but exhibited much lower binding affinities than amphetamine.
Differences in Activation Potencies at Human versus Rat and Mouse TAAR1.
Our results suggest significant species differences in TAAR1 affinities and activation potencies for most of the substances with relevant binding properties in the rat. Importantly, although the endogenous ligands p-tyramine and tryptamine activated TAAR1 with a potency rank order of rat > mouse > human, like many psychostimulant compounds, β-PEA had similar EC50 values across the three species. This is relevant because a comparison of activation potencies across species with in vitro assays could be biased by assay-specific variables, such as expression levels of the transporters in the respective cell lines. However, β-PEA can serve as a reference compound for species comparisons. Wainscott et al. (2007) also reported comparable EC50 values for β-PEA at rat and human TAAR1 and species differences for other compounds. To quantify the extent of species differences in TAAR1 activation potencies, we calculated human/rat EC50 ratios and human/mouse EC50 ratios for substances with low to submicromolar (<10 µM) potencies for human TAAR1 activation. The human/rat ratios ranged from 171 to 2.4, demonstrating the lower activity of the compounds at human versus rat TAAR1 (Table 2). This broad range of ratios showed that the extent of species differences varied substantially between compounds. The endogenous ligand tryptamine exhibited a high human/rat ratio (51). Methamphetamine and amphetamine presented relatively low human/rat ratios of 6.2 and 4.2, respectively, whereas the human/rat ratio for MDMA was significantly higher (35). The human/mouse ratios were lower than their respective human/rat ratios for all substances, with human/rat ratios ≥ 8.9. Calculations of ratios for substances that were inactive at human TAAR1 (EC50 > 10 µM) were not meaningful, but substantial differences between human/rat and human/mouse ratios were observed among the substances that were active at rat and mouse TAAR1.
EC50 ratios calculated for all substances with EC50 values < 10 µM for the human TAAR1
Ratios are ranked according to the human/rat ratio. Values > 1 indicate lower potency at the human TAAR1 versus the rat or mouse TAAR1.
Species differences and differences across substances in TAAR1 activation potencies are presented as a heat map in Fig. 2, in which the substances were sorted according to their EC50 values for the activation of rat TAAR1. Clearly, there was an overall rank order of rat > mouse > human across the psychoactive substances. Figure 2 also shows that TAAR1 binding and activation was greater for certain substance classes than for others. Amphetamines, cathinones, and phenethylamines represented the three largest substance classes in our screen. Amphetamines and phenethylamines were well represented among the potent TAAR1 ligands, with EC50 values within a range that could be physiologically relevant (30 nM to 5 µM), whereas the activity of cathinone derivatives was low (EC50 > 5 µM, except for cathinone). None of the pipradrols exhibited significant binding properties. Interestingly, tryptamine derivatives were weak agonists or did not bind to TAAR1 at all, although tryptamine itself is an endogenous rat and mouse TAAR1 ligand, with activation potency that is comparable to β-PEA and full efficacy.
Heat map illustrating the diversity of TAAR1 activity between individual substances and between human, rat, and mouse TAAR1. The substances are sorted according to their rat TAAR1 activity (EC50 values). The compounds were split into the 52 more active (A) and the 51 less active (B) rat TAAR1 ligands. Next to the substance names, the respective substance classes are specified by color code and basic chemical structures are defined by numbers. The residues R2–R11 refer to the chemical structures presented in Fig. 1. Underlined residues indicate ring structures between two locations for derivatization. The generic or full chemical names for abbreviated substances are listed in Supplemental Table 1.
Discussion
For this in vitro study, we determined the binding affinities and activation potencies of a large set of psychoactive compounds at the human, rat, and mouse TAAR1 in heterologous expression systems. We also characterized the ligands as full or partial agonists. None of the active compounds had full antagonist properties. As indicated by our previous studies (Simmler et al., 2013, 2014a), cathinone derivatives stood out as poor TAAR1 ligands. Most of the other psychoactive compounds were potent to moderate rat and mouse TAAR1 agonists but exhibited generally weak or no activity at human TAAR1. The active compounds showed full or partial TAAR1 agonist properties, with generally no distinct patterns related to their chemical structure.
To our knowledge, our screening is the most extensive published data set to date, which included 101 psychoactive substances and 3 endogenous ligands as comparator compounds. The in vitro pharmacology of comparator compounds and some psychoactive substances (amphetamine, methamphetamine, MDMA) were previously determined by us and others (Borowsky et al., 2001; Bunzow et al., 2001; Reese et al., 2007; Wainscott et al., 2007; Lindemann et al., 2008; Lewin et al., 2011). The replication of those data for this study was an important validation of our assays. Furthermore, because we determined binding affinities for some novel psychoactive substances in earlier studies, we included these data in this study. As a result, all data determined by our laboratory are conveniently summarized here in Table 1.
Species differences in TAAR1 activation between rodent and human receptors have been reported previously for phenethylamine analogs (Wainscott et al., 2007), p-tyramine, and methamphetamine (Reese et al., 2007). These compounds have consistently exhibited lower potencies for human TAAR1 activation than for rodent TAAR1 activation. Structure-activity correlations for β-PEA derivatives with regard to human TAAR1 activation have shown that bulky residues on the amine or phenyl ring reduced ligand potency (Lewin et al., 2008). Reduced human TAAR1 activity could be expected for novel psychoactive substances for which substantial derivatization is typical. Together with previous reports on species differences, the data provide evidence that many psychoactive substances are considerably less potent at human TAAR1 than at mouse or rat TAAR1. In rodents, psychoactive compounds could reduce neuronal firing via TAAR1 activation, comparable to the endogenous TAAR1 ligand p-tyramine, which has been shown to reduce the DA neuron firing rate (Bradaia et al., 2009). Consequently, the psychoactive TAAR1 ligands likely exert autoregulatory effects on their TAAR1-independent effects, such as reducing drug-induced DA release in the striatum (Di Cara et al., 2011). Given that studies in rodents have reported autoregulatory effects of the psychostimulant TAAR1 ligands amphetamine, methamphetamine, and MDMA (Lindemann et al., 2008; Di Cara et al., 2011; Achat-Mendes et al., 2012; Harkness et al., 2015), these species differences at TAAR1 could be relevant to the translational validity of preclinical studies. Particularly for novel psychoactive substances with large TAAR1 species differences, the abuse liability that is evaluated in rodent models may actually underestimate the risk for addiction that is posed by the drugs in humans.
Species differences in TAAR1/ligand interactions have been predicted from sequence alignment studies that compared the critical residues for the binding of β-PEA, showing that amino acids that correspond to the critical residues differ between rat, mouse, and human TAAR1 (Kratochwil et al., 2011). Site-directed mutagenesis studies have identified two locations in TAAR1 transmembrane domains 6 and 7, where amino acid substitutions markedly reduce or increase methamphetamine TAAR1 activation potencies in the rat and mouse TAAR1 (Reese et al., 2014). Docking studies with a homology model for the human TAAR1 (Cichero et al., 2013, 2014) could serve to further elucidate the essential amino acids that are required for ligand binding and discover structural determinants for the TAAR1 activity or inactivity of psychoactive substances.
Importantly, TAAR1 is a promising target for therapeutic drugs for the treatment of substance use disorders, regardless of species differences in the direct TAAR1 agonism properties of psychoactive substances. TAAR1 agonists that have been reported in the literature are similarly potent at both human and rat TAAR1. Furthermore, the efficacy that has been reported in animal models is comparable to the efficacy that has been reported in in vitro expression systems, which may provide a basis for predicting effective doses in humans. The TAAR1 partial agonist RO5203648 effectively reduced cocaine self-administration and relapse to drug-seeking behavior in rats (Revel et al., 2012b; Pei et al., 2014), although cocaine is not a TAAR1 ligand itself. TAAR1 partial agonism markedly reduced cocaine-induced DA overflow in the nucleus accumbens (Pei et al., 2014). Because TAAR1 is involved in the constitutive regulation of neuronal firing (Bradaia et al., 2009), pharmacological TAAR1 activation with a therapeutic drug may regulate neuronal firing and result in hyposensitivity to drugs, similar to the overexpression of TAAR1 in a transgenic mouse model (Revel et al., 2012a). Moreover, the efficacy of these potentially therapeutic compounds could be even greater in humans than in rodents. In rodents, but not or less so in humans, TAAR1-mediated negative feedback that is induced by the abused substances could be present and attenuate the extent of therapeutic drug effects.
Ex vivo electrophysiology experiments in the ventral tegmental area and dorsal raphe nuclei showed that both partial agonists and antagonists enhanced DA and 5-HT neuron firing rates in WT mice (Bradaia et al., 2009; Revel et al., 2012b), whereas full agonists like p-tyramine decreased firing rates (Revel et al., 2011, 2012b). However, both full and partial agonists have been shown to be protective against the rewarding and reinforcing effects of the psychostimulant cocaine (Pei et al., 2015), but the opposing effects of full and partial agonists on firing rates suggest that psychoactive substances that are full agonists would exert effects that are different from partial agonists. Data on the full or partial agonist properties of TAAR1 ligands are thus important. Whereas full agonists such as amphetamine induce negative feedback to blunt their own effect on DA and 5-HT systems (Lindemann et al., 2008; Di Cara et al., 2011), partial agonists might increase their effect on neuronal signal transmission by increasing firing rates via TAAR1. This is an assumption that would be based on findings with selective TAAR1 ligands and requires further investigation.
In our data set and based on data reported previously by two different laboratories (Reese et al., 2007; Wainscott et al., 2007), the activation potencies of β-PEA at rat, human, and mouse TAAR1 exhibited similar EC50 values between species, whereas p-tyramine was more potent at rat TAAR1, followed by mouse and human TAAR1. Given that these data were generated in independent laboratories that used different assay conditions and expression systems, the similarities of the pharmacological profiles suggest good consistency of the data and support the validity of the comparisons between species.
In this study, we simply determined activity at specific targets, which is common with interpretations of in vitro data, and we did not take into account that the processes that allow a substance to interact with TAAR1 in vivo depend on more variables than solely substance/receptor interactions. The location of TAAR1 expression is mostly intracellular in neurons (Miller, 2011) and also in glial cells (Cisneros and Ghorpade, 2014). Because the substances need to reach the location of expression of TAAR1 to bind to the receptor, the intracellular availability of the ligands is also relevant. Certain psychoactive substances, such as amphetamine derivatives, are substrates of monoaminergic transporters and carried into the cell (Zaczek et al., 1991). These substrate-type substances, therefore, might be more likely available to intracellular TAAR1 than substances that are not transporter substrates, including, for example, cocaine, MDPV, other pyrovalerone cathinones, methylphenidate, and other pipradrols (Simmler et al., 2013, 2014b). One limitation of our study is that we did not consider stereoselectivity of the compounds by screening racemic mixtures for most substances. As with activity at other psychostimulant targets, such as monoaminergic reuptake transporters, TAAR1 has a stereoselective binding site, and the assessment of racemates could underestimate the activity of the more active isomer (Lewin et al., 2011).
In conclusion, we provide an extensive data set on the ligand properties of psychoactive substances at TAAR1. With differences between activity at rodent and human TAAR1, we provide evidence of significant species differences in interactions between TAAR1 and psychoactive drugs, which could be relevant to the translational validity of preclinical studies to clinical applications.
Acknowledgments
The authors thank Lipomed for providing the 2C and NBOMe drugs at no cost, Michael Arends for text editing, and Roger Norcross for helpful discussions.
Authorship Contributions
Participated in research design: Hoener, Liechti.
Conducted experiments: Buchy, Chaboz.
Performed data analysis: Simmler, Buchy, Chaboz, Hoener.
Wrote or contributed to the writing of the manuscript: Simmler, Hoener, Liechti.
Footnotes
- Received October 5, 2015.
- Accepted January 19, 2016.
This research was supported by the Federal Office of Public Health [Grant 13.006497] and F. Hoffmann-La Roche Ltd. and the University of Basel [Translational Medicine Hub Innovation Fund].
↵
This article has supplemental material available at jpet.aspetjournals.org.
Abbreviations
- β-PEA
- β-phenylethylamine
- DA
- dopamine
- 5-HT
- 5-hydroxytryptamine (serotonin)
- KO
- knockout
- MDMA
- 3,4-methylenedioxymethamphetamine
- PBS
- phosphate-buffered saline
- RO5166017
- (S)-4-[(ethyl-phenyl-amino)-methyl]-4,5-dihydro-oxazol-2-ylamine
- RO5203648
- (S)-4-(3,4-dichlorophenyl)-4,5-dihydrooxazol-2-amine dihydrochloride
- TAAR1
- trace amine-associated receptor 1
- WT
- wild type
- Copyright © 2016 by The American Society for Pharmacology and Experimental Therapeutics
