Journal of Neuroscience Research 88:1962–1969 (2010)

Normal Thermoregulatory Responses to 3-iodothyronamine, Trace Amines and -like Psychostimulants in Associated Receptor 1 Knockout Mice

Helen N. Panas,1 Laurie J. Lynch,1 Eric J. Vallender,1 Zhihua Xie,1 Guo-lin Chen,1 Spencer K. Lynn,2 Thomas S. Scanlan,3 and Gregory M. Miller1* 1Division of Neuroscience, Harvard Medical School/NEPRC, Southborough, Massachusetts 2Molecular Pharmacology Laboratory, McLean Hospital, Harvard Medical School, Belmont, Massachusetts and Interdisciplinary Affective Science Laboratory, Department of Psychology, Boston College, Chestnut Hill, Massachusetts 3Department of Physiology & Pharmacology, Oregon Health and Science University, Portland, Oregon

3-Iodothyronamine (T1AM) is a metabolite of thyroid hor- 3-Iodothyronamine (T1AM) is an endogenous de- mone. It is an at trace amine-associated receptor rivative of thyroid hormone and itself a possible chemi- 1 (TAAR1), a recently identified receptor involved in mono- cal messenger (Scanlan et al., 2004). In various studies in aminergic regulation and a potential novel therapeutic tar- which the biological response to T1AM has been meas- get. Here, T1AM was studied using rhesus monkey ured, changes in temperature, heart rate and cardiac out- TAAR1 and/or human transporter (DAT) co- put have been observed. Most strikingly, T1AM pro- transfected cells, and wild-type (WT) and TAAR1 knock- duces a profound hypothermic response when adminis- out (KO) mice. The IC50 of T1AM competition for binding tered in vivo (Scanlan et al., 2004). The robust of the DAT-specific radio-ligand [3H]CFT was highly similar hypothermic action of T1AM suggests that the thyron- in DAT cells, WT striatal synaptosomes and KO striatal amine could serve as a neuroprotectant, and that its re- synaptosomes (0.72–0.81 lM). T1AM inhibition of 10 nM ceptor could be a potential therapeutic target for cryo- 3 [ H]dopamine uptake (IC50:WT,1.46 0.5 lM; KO, 1.2 6 genic agents. In this regard, T1AM has been assessed for 3 0.4 lM) or 50 nM [ H] uptake (IC50:WT,4.56 its potential neuroprotective action in a stroke injury 0.6 lM; KO, 4.7 6 1.1 lM) in WT and KO synaptosomes model. Treatment with T1AM was found to reduce was also highly similar. Unlike other TAAR1 that infarct volume by about one third when tested in a mid- are DAT substrates, TAAR1 signaling in response to T1AM dle cerebral artery occlusion stroke model in mice was not enhanced in the presence of DAT as determined (Doyle et al., 2007). The study by Doyle et al. (2007) by CRE-luciferase assay. In vivo, T1AM induced robust was the first to use endogenous chemicals including hypothermia in WT and KO mice equivalently and dose dependently (maximum change degrees Celsius: 50 mg/ Additional Supporting Information may be found in the online version kg at 60 min: WT 26.0 6 0.4, KO 25.6 6 1.0; and 25 of this article. mg/kg at 30 min: WT 22.7 6 0.4, KO 23.0 6 0.2). Other The first two authors contributed equally to this work. TAAR1 agonists including beta–phenylethylamine (b-PEA), Contract grant sponsor: National Institute on Drug Abuse; Contract MDMA (3,4-methylenedioxymethamphetamine) and meth- grant sponsor: National Institute of Diabetes and Digestive Kidney Dis- amphetamine also induced significant, time-dependent eases; Contract grant sponsor: National Center for Research Resources; thermoregulatory responses that were alike in WT and KO Contract grant number: RR00168; Contract grant number: DA022323 mice. Therefore, TAAR1 co-expression does not alter (to G.M.M.); Contract grant number: DA016606 (to G.M.M.); Contract T1AM binding to DAT in vitro nor T1AM inhibition of grant number: DA025802 (to G.M.M.); Contract grant number: DA025697 (to G.M.M.); Contract grant number: DK52798 (to T.S.S.). [3H]monoamine uptake ex vivo, and TAAR1 agonist- induced thermoregulatory responses are TAAR1-inde- *Correspondence to: Gregory M. Miller, New England Primate pendent. Accordingly, TAAR1-directed compounds will Research Center, Harvard Medical School, One Pine Hill Drive, South- likely not affect thermoregulation nor are they likely to be borough, MA. E-mail: [email protected] cryogens. VC 2010 Wiley-Liss, Inc. Received 22 October 2009; Revised 30 November 2009; Accepted 2 December 2009 Key words: TAAR1; thyronamine; psychostimulant; Published online 12 February 2010 in Wiley InterScience (www. thermoregulation; monoamine interscience.wiley.com). DOI: 10.1002/jnr.22367

' 2010 Wiley-Liss, Inc. TAAR1 and Thermoregulation 1963

T1AM to induce prophylactic hypothermia to treat MATERIALS AND METHODS stroke, and to suggest a therapeutic use of T1AM as a Animals cryogen. The body’s varied and rapid response to T1AM, as well as the potential use of thyronamine as a Animals were approximately 3 months old at the time cryogenic agent has made determining the mechanism of of the studies and only males were tested. TAAR1 KO and action of T1AM as well as its biological role a relevant WT mouse colonies were established at the New England Pri- goal of recent experimental work. mate Research Center (Southborough, MA) from six pairs of Central to recent studies of T1AM is the specula- heterozygous mice received as a gift from Lundbeck Research USA, Inc. (Paramus, NJ). All animals were maintained on a tion of the involvement of trace amine associated recep- 6 8 tor 1 (TAAR1) in its mechanism of action. TAAR1 is a 12-hr light/dark schedule at a room temperature of 22 1 C G protein-coupled receptor that is activated by a wide with free access to food and water. Animal care was in ac- spectrum of compounds, including common biogenic cordance with the Guide for the Care and Use of Laboratory amines, trace amines, and amphetamine-like psychosti- Animals (National Research Council, National Academy mulant drugs, and plays an important role in modulating Press, 1996) and all procedures were conducted in accordance brain monoamine systems (reviewed in Xie and Miller, with the Animal Experimentation Protocol approved by the 2009a). T1AM has been found to be a potent agonist of Harvard Medical Area Standing Committee on Animals. rat and mouse TAAR1 (EC50: 14 nM and 112 nM, Synaptosome Preparation respectively) and to stimulate cAMP production in rat and mouse TAAR1-expressing cells in a dose dependent Synaptosomes were prepared from adult male TAAR1 manner (Scanlan et al., 2004). It is also recognized that KO and WT mice as described previously (Xie and Miller, 2008). Briefly, freshly dissected striatal tissue was homogenized T1AM is a monoamine transporter inhibitor, making 3 this TAAR1 agonist different from other TAAR1 ago- in cold 0.32 M sucrose (10 volume) using a Kontes Pellet Pestle with 15 up and down strokes. The mixture was centri- nists such as endogenous monoamines (e.g., trace amines 8 and common biogenic amines) and exogenous ampheta- fuged at low speed (1,000g, 10 min, 4 C) and the supernatant carefully removed and centrifuged at high speed (10,000g, mine-like psychostimulants (e.g., amphetamine and 8 ) that are monoamine transporter sub- 20 min, 4 C) to yield the pelleted synaptosomes. The pellet strates (Snead et al., 2007; Xie et al., 2007). Accordingly, was resuspended in ice-cold uptake buffer (25 mM HEPES, 120 mM NaCl, 5 mM KCl, 2.5 mM CaCl2, 1.2 mM the demonstrated effects of T1AM could be mediated by l TAAR1 or could be a result of altered monoamine con- MgSO4,1 M pargyline, 2 mg/ml glucose, 0.2 mg/mL ascor- centration via blockade of monoamine transporters, or bic acid, pH 7.5) and used immediately in assays. All reagents some other mechanism. Also, TAAR1 is under current and the pestles used in the preparation of synaptosomes were investigation as a potential target for novel therapeutics purchased from Sigma-Aldrich (St. Louis, MO). which may regulate monoaminergic function, and so determining whether TAAR1 mediates thermoregula- Cell Culture tory responses is important with regard to medications Cell lines were generated as described previously (Xie development. et al., 2008). Cell lines were cultured in untreated tissue cul- Accordingly, we took advantage of a TAAR1 ture dishes (Greiner America, Inc., Lake Mary, FL) in knockout mouse model to directly address the question DMEM supplemented with 10% fetal bovine serum, 100 of whether TAAR1 mediates the thermoregulatory units/ml penicillin, 100 lg/ml streptomycin and 0.1 mM response of T1AM in vivo, and compared responses of nonessential amino acids. Cell transfections were performed T1AM to two trace amines, tyramine and beta-phenyle- using calcium phosphate (250 mM calcium chloride and 23 thylamine (b-PEA), and two amphetamine-like psychos- HBS buffer [50 mM HEPES, 1.5 mM Na2HPO4 (Fisher Sci- timulants, methamphetamine and MDMA, which also entific, Pittsburgh, PA), 280 mM NaCl, 10 mM KCl, 12 mM affect thermoregulation. We verified the interaction of glucose, pH 7.05]). With the exception of reagents noted T1AM with TAAR1 and monoamine transporters in above, all reagents were obtained from Invitrogen Corpora- transfected cells and striatal synaptosomes prepared from tion, Carlsbad, CA. wild-type (WT) and TAAR1 knockout mice (KO). We measured cAMP production using a CRE-Luciferase Dual Luciferase Reporter Assays (CRE-Luc) reporter assay in TAAR1 and TAAR1/ Cells were plated into 48-well plates (75,000 cells/well DAT cells after exposure to T1AM and assessed whether in 0.5 mL DMEM) 24 hr prior to transfection. On the fol- DAT could potentiate TAAR1 signaling in response to lowing day the cells were transfected with CRE-Luc, a re- T1AM, as is the case for other TAAR1 agonists that are porter control construct pGL4.73 (cAMP irresponsive) and ei- DAT substrates (Miller et al., 2005; Xie et al., 2007). In ther a rhesus TAAR1 expression construct or pcDNA3.1 con- striatal synaptosomes, we measured IC50 values for trol vector added to keep the amount of DNA used T1AM binding to DAT as well as for [3H]dopamine and equivalent between the cells (Miller et al., 2005; Xie et al., [3H]serotonin uptake by the DAT and SERT, respec- 2007). After 24 hr, the transfection media was replaced with tively. A comparison of mouse synaptosomal [3H]sero- 0.5 mL of serum-free DMEM immediately followed by treat- tonin uptake blockade, using both WT and KO mice, ment with T1AM (3-iodothyronamine; synthesized according was also performed. to the procedure described in Hart et al., 2006) or b-PEA

Journal of Neuroscience Research 1964 Panas et al.

(Sigma-Aldrich) and incubated for 18 hr. Alternatively, cells For the synaptosomal binding experiments, equal were immediately exposed to various DAT blockers. The volumes of serially diluted T1AM and [3H]CFT were mixed, DAT blockers were CFT (WIN 35428; Organix, Inc., followed by the addition of the synaptosomes in solution. Woburn, MA), GBR 12909 (Sigma-Aldrich), HCl This mixture was incubated for 2 hr at room temperature and (Burroughs-Welcome Co., Research Park, NC), cocaine the reaction was then terminated by centrifugation (12,500g (NIDA, Bethesda, MD), indatraline (Sigma-Aldrich) and for 15 min) and aspiration of the supernatant. The remaining HCl (Sigma-Aldrich). All DAT blockers, pellet was solubilized in 1% SDS and the radioactivity T1AM and b-PEA were tested at 10 lM. A 10 mM stock so- counted. The experimental points were done in triplicate. lution was made fresh daily for each compound, and diluted The baseline drug was cocaine at a final concentration of 1:10 into DMEM for a 1mM solution. 5 lL of this solution 30 lM. was added into 495 lL of DMEM in each well for a final When synaptosomes were used for [3H]dopamine concentration of 10 lM. GBR 12909 is believed to be less uptake experiments, serially diluted T1AM was added to the stable in solutions so it was made and used immediately. All synaptosomes and allowed to incubate for 15 min at room 10 mM stock solutions were made with water with the temperature. [3H]dopamine (PerkinElmer Life and Analytical exception of T1AM (0.1% DMSO at final dilution) and CFT Sciences; 48.1 Ci/mMol; 10 nM final) was added to the mix- (0.1% ethanol at final dilution). All blockers were incubated ture and incubated for 15 min at room temperature. The with the cells for 15 min with the exception of CFT, which uptake was terminated by plunging the vials into an ice water was incubated for 1 hr. Cells were then treated with T1AM bath. After chilling for at least 2 min, the vials were centri- or b-PEA for 18 hr. The passive lysis buffer (PLB) and lucif- fuged as described above. The baseline drug was indatraline at erase assay substrate reagents were prepared following the a final concentration of 10 lM. manufacturer’s protocol (Promega Corporation, Madison, For the synaptosomal [3H]serotonin uptake experiments, WI). Following drug treatments, media was removed from serially diluted T1AM was added to the synaptosomes and the cells and the cells were lysed by the addition of 100 lLof allowed to incubate for 30 min. [3H]serotonin (PerkinElmer 13 PLB into each well and shaking the plate for 1 hr at Life and Analytical Sciences; 28.1 Ci/mMol; 50 nM final) was room temperature. An aliquot of the lysate (20 lL) from each added and the mixture was incubated for 15 min at room well was transferred into wells of an opaque 96-well micro- temperature. The uptake was terminated by plunging the vials plate (PerkinElmer, Shelton, CT). Luciferase substrate reagents into an ice water bath. After chilling for at least 2 min, the (25 lL) were injected into each well and luciferase levels were vials were centrifuged as described above. The baseline drug measured as relative light units (RLUs) for 12 sec on a Wallac was fluoxetine (Sigma-Aldrich) at a final concentration of 1420 multilabel counter, Victor 3V (PerkinElmer). To analyze 10 lM. the data, a ratio of the Firefly RLU reading and the Renilla RLU reading for each well was divided by the similar ratio Body Temperature Measurement for the baseline wells and changed to a percentage. In experi- All animals were relocated from the colony into a be- ments utilizing DAT blockers, the baseline was derived from havioral study holding room and habituated to the new envi- cells treated with the DAT blocker alone. An increase in the ronment for at least 72 hr prior to testing. During this time, value above baseline indicates cAMP accumulation in response all animals were weighed, marked for identification, and had to the drug exposure. their lower backs shaved (Oster Turbo A5 Clippers with an Elite CryogenX-AgION size 50 blade, McMinnville, TN). Binding and Uptake Assays On the testing day, each animal was individually housed in a new, bedding free, open clean cage and relocated to a testing Cell lines used in binding assays were grown to 80% area for 1 hr prior to receiving a single intraperitoneal (i.p.) confluence in cell media, rinsed with PBS and removed from injection of 50 mg/kg T1AM, 25 mg/kg T1AM, 50 mg/kg the plate by repeated PBS wash. The cells collected were cen- b-PEA, 50mg/kg tyramine (Sigma-Aldrich), 25 mg/kg (6) trifuged at 1000g for 5 min and the supernatant removed. The MDMA (NIDA), 3 mg/kg (1) methamphetamine (Sigma- cells were resuspended and brought to a concentration of 1.25 Aldrich) or saline. Body temperature readings were captured 3 106 cells/mL in buffer (25 mM HEPES, 120 mM NaCl, 5 using a wireless infra-red thermometer (La Crosse Technology mM KCl, 2.5 mM CaCl , 1.2 mM MgSO ,1lM pargyline, 2 4 Ltd., La Crescent, MN) over the shaved back area of the 2 mg/mL glucose, 0.2 mg/mL ascorbic acid, pH 7.5) and mouse at 5 min prior to the drug injection and at post injec- used immediately in assays. In radio-receptor binding assays, tion times: 15, 30, 45, 60, 75, 90, 120, 150, 180, 240, and equal volumes of serially diluted T1AM and [3H]CFT (Perkin- 320 min. Room temperature of 22 6 18C was continuously Elmer Life and Analytical Sciences, Boston, MA; 85.9 Ci/ monitored throughout the experiment. The research assistant mMol; 1 nM final) were mixed, followed by the addition of the obtaining the temperature readings was blind to the treatment cells in solution. This mixture was incubated for 2 hr at room conditions of the animals. temperature with the reaction terminated by centrifugation (12,500g, 15 min) and the supernatant aspirated. The remaining pellet was solubilized in 1% SDS and the radioactivity counted. Data Analysis The experimental points were done in triplicate. Non-specific A one-way ANOVA was used to test for significant dif- binding was determined using cocaine at a final concentration ferences between doses of T1AM or b-PEA in HEK cells of 30 lM. transfected with CRE-Luc and pGL4.73 only. A t-test was

Journal of Neuroscience Research TAAR1 and Thermoregulation 1965

Fig. 1. Activation of TAAR1 by T1AM and comparison to b-PEA. TAAR1-DAT cells were exposed to various DAT blockers at (10 T1AM and b-PEA activation of TAAR1 and TAAR1-DAT cells lM) or to vehicle (no DAT blocker), and were then treated with was determined by a CRE-Luc reporter assay and are reported as T1AM (10 lM) or b-PEA (10 lM). Pretreatment with DAT block- percent change in relative light units (RLU). A: T1AM (10 lM) ers had no significant effect on the response to 10 lM T1AM in resulted in a 680% increase in RLU in TAAR1 cells versus a 760% TAAR1-DAT cells, whereas all DAT blockers significantly attenu- RLU increase in TAAR1-DAT cells, whereas b-PEA (10 lM) ated the enhanced response to 10 lM b-PEA in the TAAR1-DAT resulted in a 531% increase in RLU in TAAR1 cells and 1750% cells (n 5 3). All data shown are values of mean 6 SEM. **: P < increase in RLU response in TAAR1-DAT cells (n 5 3). B,C: 0.01; ***: P < 0.001. used to compare the effect of 10 lM T1AM or 10 lM b- ously tested (e.g., b-PEA, tyramine, dopamine, metham- PEA on TAAR1 transfected HEK cells either with or without phetamine) are also substrates for the DAT, and the a DAT co-transfection. One-way ANOVA with Dunnett’s TAAR1 receptor is largely intracellularly sequestered and correction for multiple testing was used to compare the effect associated with an intracellular membrane fraction (Miller of DAT blockers on TAAR1-transfected cells in the presence et al., 2005; Xie et al., 2008b). Accordingly, we assessed or absence of DAT. Binding and uptake assays in cells and whether the presence of DAT would enhance TAAR1 synaptosomes vary in radiation counting between cells and signaling in response to T1AM. T1AM (10 lM) exposure synaptosome preparations from different sources, and accord- resulted in a 760% increase in TAAR1-DAT cells relative ingly, data are normalized to the maximum level or the base- to vehicle treatment, which did not significantly differ line and expressed as percentage values in each set of assays. from the response of T1AM in the TAAR1 cells. In con- Data for the drug effects on thermoregulation were normal- trast and similar to our previous findings (Xie and Miller, ized to the saline controls prior to statistical analyses. A two- 2008), b-PEA (10 lM) resulted in a 531% increase in way ANOVA with repeated measures was employed to deter- RLU in TAAR1 cells versus a 1750% increase in RLU in mine if there was a significant difference in thermoregulatory TAAR1-DAT cells (P < 0.001; Fig. 1A). The data dem- responses over all time points between WT and KO mice and onstrate a significant enhancement by DAT of TAAR1 across different drug treatments. For each drug treatment in signaling in response to b-PEA but not T1AM. each set of mice, a one way ANOVA followed by Bonferroni Next, we assessed whether a series of DAT blockers post-hoc analysis was also used to determine significance of could alter TAAR1 signaling in response to T1AM or b- each time point versus the baseline temperature reading taken PEA in TAAR1-DAT cells. None of the DAT blockers at the 25 time point. All results were finalized as mean 6 used are agonists at TAAR1 nor cause any elevation in SEM of the indicated number of observations. CRE-Luc expression in TAAR1cells (data not shown). TAAR1 and TAAR1-DAT cells were exposed to the vari- RESULTS ous DAT blockers at a concentration of 10 lMorvehicle (noDATblocker)andwerethentreatedwith10lM T1AM Activation of TAAR1 and l b b T1AM or 10 M -PEA for 18 hr. Pretreatment with Comparison to -PEA DAT blockers had no significant effect on the response to Neither T1AM nor b-PEA significantly stimulated 10 lMT1AMinTAAR1-DATcells(Fig.1B),whereasall CRE-Luc expression in HEK cells transfected with CRE- DAT blockers significantly attenuated the enhanced response Luc and pGL-4.73 across concentrations tested (100 nM, to 10 lM b-PEA in the TAAR1-DAT cells, to the same 1 lMand10lM; data not shown). T1AM (10 lM) level as in TAAR1 cells without DAT (P < 0.01, Fig. 1C). treatment resulted in a 680% increase in RLU in TAAR1 cells relative to vehicle treatment, demonstrating that T1AM acts as an agonist at rhesus monkey TAAR1 (Fig. T1AM Binding to DAT and Blockade of 1A). Previous work in our lab has demonstrated a robust DAT and SERT in Striatal Synaptosomes enhancement of TAAR1 signaling when DAT is also T1AM binding to DAT was measured by competi- present, presumably because the TAAR1 agonists previ- tive binding versus [3H]CFT in stable human DAT-

Journal of Neuroscience Research 1966 Panas et al.

Fig. 2. T1AM binding and uptake in cells and synaptosomes. A: for KO striatal synaptosomes (n 5 2). C: The IC50 of T1AM versus T1AM binding to the dopamine transporter was measured by com- [3H]serotonin was 4.5 6 0.6 3 10-6 M for WT and 4.6 6 1.1 3 3 -6 petitive binding versus [ H]CFT in DAT/HEK cells. The IC50 was 10 M for KO striatal synaptosomes (n 5 3). All data shown are val- -7 3 8.0 6 1.2 3 10 M(n5 2). B: The IC50 of T1AM versus [ H]do- ues of mean 6 SEM. pamine was 1.4 6 0.5 3 10-6 M for WT and 1.2 6 0.4 3 10-6 M transfected cells and in WT and KO striatal synaptosomes. rebound effect that was not observed in the WT mice -7 The IC50 was determined to be 8.0 6 1.2 3 10 M(n (60, 75 and 90 min time points; Fig. 3B), though this 5 2; each point done in triplicate; Fig. 2A), 7.2 3 10-7 did not reach statistical significance. Also, the hyperther- M in WT synaptosomes and 8.1 3 10-7 MinKOsynap- mic response to both MDMA and methamphetamine tosomes (n 5 1; each point done in triplicate; Supp. Info treatments reached significance at earlier time points in 3 Fig. 1). The IC50 of T1AM inhibition of [ H]dopamine the KO mice (15 min and 30 min, respectively) com- or [3H]serotonin uptake at 15 min was also measured in pared with the WT mice (30 min and 75 min, respec- WT and KO striatal synaptosomes. The IC50 of T1AM tively). Saline-treated mice showed a gradual decrease in inhibition of [3H]dopamine was 1.4 6 0.5 3 10-6 Min body temperature over time and maintained this lower WT synaptosomes and 1.2 6 0.4 3 10-6 MinKOsyn- temperature for the entire duration of the experiment aptosomes (n 5 2; each point done in triplicate; Fig. 2B). (Fig. 3F). This was likely due to the absence of bedding 3 The IC50 of T1AM inhibition of [ H]serotonin was 4.5 in the test cage, and appears to be a typical finding in 6 0.6 3 10-6 M in WT synaptosomes and 4.6 6 1.1 3 mice (Fantegrossi et al., 2008). Accordingly, we chose to 10-6 M in KO synaptosomes (n 5 2; each point done in present data for the drug effects (Fig. 3A–E) as normal- triplicate; Fig. 2C). ized to the saline controls (Fig. 3F) rather than the actual temperature change recorded in each mouse. Thermoregulatory Responses of T1AM, b-PEA, Tyramine, MDMA and Methamphetamine in WT and KO Mice DISCUSSION WT and KO mice were exposed to a single intra- The data presented here demonstrate that T1AM is peritoneal injection of drug or saline and body tempera- an agonist at the rhesus monkey TAAR1 receptor and, ture was measured over time. WT and KO mice did not unlike other TAAR1 agonists that we have tested to significantly differ in body temperature at the baseline date using the CRE-luciferase assay, reporter expression temperature measurement collected at the 25 time point indicative of cAMP accumulation is unaffected by the (WT: 36.3 6 0.78C, n 5 71; KO: 36.2 6 0.78C, n 5 presence of DAT. We find that T1AM binds to the 70). Both WT and KO mice showed a dose-dependent human DAT in transfected cells, and to the mouse DAT hypothermic response to T1AM, without a significant in WT synaptosomes and KO synaptosomes with similar difference in the magnitude or duration of this response potency. We also demonstrate that T1AM blocks (Fig. 3A). Similarly, the thermoregulatory response char- [3H]dopamine uptake and [3H]serotonin uptake in WT acteristics for other tested drugs were not significantly and KO striatal synaptosomes with similar potency. And, different between the WT and KO mice (Fig. 3B–E). we show that the time-dependent thermoregulatory Further analyses by one way ANOVA were performed responses to the TAAR1 agonists T1AM, b-PEA, tyra- on the normalized data for each drug treatment to deter- mine, MDMA and methamphetamine are highly similar mine whether the drug treatments caused significant and do not significantly differ in WT and KO mice, thermoregulatory responses relative to the baseline tem- indicating that TAAR1 is not involved in mediation of perature measurement collected at the 25 time point. thermoregulation and that the KO mice do not have a All drug treatments except for tyramine showed signifi- compensatory phenotype with regard to pharmacologi- cant effects on thermoregulation (summarized in Supp. cally-induced thermoregulatory responses. Accordingly, Info. Table I). Whereas WT mice demonstrated a rapid the robust cryogenic action of T1AM is not mediated and short-lived hyperthermic response to 50 mg/kg b- via TAAR1, nor is the hyperthermic response to psy- PEA, the KO mice displayed an additional hypothermic chostimulants.

Journal of Neuroscience Research TAAR1 and Thermoregulation 1967

Fig. 3. Thermoregulatory responses of T1AM, b -PEA, tyramine, with repeated measures was used to determine if there was a signifi- MDMA, methamphetamine and saline in WT and KO mice. Wild cant difference in the thermoregulatory responses over all time points type (WT) and TAAR1 knockout (KO) male mice were treated between WT and KO mice and across different drug treatments. with 25 mg/kg or 50 mg/kg T1AM (A), 50 mg/kg b -PEA (B), 50 There were no significant main or interactive effects. For each drug mg/kg tyramine (C), 25 mg/kg MDMA (D), 3 mg/kg methamphet- treatment, a one way ANOVA was used to determine statistical sig- amine (E) or saline (F). The number of mice in each group is indi- nificance of individual time points versus the baseline (-5) tempera- cated. Body temperature readings were captured using a wireless ture (detailed in Supp. Info. Table I). All data shown are values of infra-red thermometer held over the shaved back area of each mouse mean 6 SEM. *, P < 0.05 or lower versus the 25 time point for at 5 min prior to the drug injection (-5) and at 15, 30, 45, 60, 75, both WT and KO analyses except as indicated: in A, ns refers to KO 90, 120, 150, 180, 240, and 320 min post injection. Data was nor- data; in D and E, ns refers to WT data. See Supporting Information malized to the saline controls (shown in F). A two-way ANOVA Table I for more detail.

TAAR1 mediation in the physiological effects of Also, TAAR1 mRNA is expressed in the mouse arcuate T1AM was initially suggested by the finding that the nucleus (ARC) and has been detected in human and rank order of potency of T1AM and related analogs at rhesus monkey hypothalamus (Borowsky et al., 2001; TAAR1 matched their potency to decrease body tem- Branchek and Blackburn, 2003; Lindemann et al., 2008), perature in vivo (Scanlan et al., 2004; Hart et al., 2006). brain areas important in the regulation of food intake

Journal of Neuroscience Research 1968 Panas et al. and energy balance. Dhillo et al. (2009) have recently brain synaptosomal preparations, whereas T1AM inhib- investigated the putative role of T1AM in appetite regu- ited transport at the human dopamine transporter (DAT) lation and energy expenditure and found that following and the human (NET) but failed to do intracerebroventricular administration of T1AM into the so at the human serotonin transporter (SERT) in experi- ARC in rats, a three-fold increase in food consumption ments utilizing monoamine transporter-transfected HeLa was observed, along with a significant increase in c-fos cells (Snead et al., 2007). We also demonstrated that expression, indicating the ARC neurons were activated T1AM can block the SERT in WT and KO mice, clari- by T1AM to release cAMP, consistent with the possibil- fying its role as a monoamine transporter blocker. Fur- ity of mediation via TAAR1 receptor activation. Addi- ther, the current data suggest that T1AM may differ tionally, recent work has shown a connection between from other TAAR1 agonists in its ability to access the TAAR1, T1AM and hyperglycemia, suggesting a possi- receptor in that T1AM activation of TAAR1 is not ble control of metabolism (Klieverik et al., 2009) and in- enhanced in the presence of DAT (Xie et al., 2007, sulin secretion (Regard et al., 2007). These findings war- 2008a). Particularly, our recent work has focused on ranted our assessment of the role of TAAR1 in T1AM- methamphetamine, which produces very large enhance- induced thermoregulatory responses in the TAAR1 ments in TAAR1 signaling in the presence of DAT. We knockout mice. have speculated that this enhanced signaling could con- In agreement with our current findings, Frascarelli tribute to adaptive cellular responses that are associated et al. (2008) found that the rank order of potency of with addiction (Xie and Miller, 2009b). Accordingly, drugs for negative inotropic effects on isolated rat hearts cellular actions of thyronamine that are mediated via as measured by cardiac output did not match that from TAAR1 could be predicted to be similar to other cAMP production of either TAAR1 or TAAR4 in cells, TAAR1 agonists in cells devoid of monoamine trans- and suggested that the effect on heart function is medi- porters, but different at monoaminergic cells where it ated by a different TAAR subtype other than TAAR1 will block monoamine transporters and also lack the or TAAR4. It was suggested that a possible target of enhanced signaling properties of the substrates agonists. T1AM in the rat heart is TAAR8a, where the mRNA Its ability to block monoamine transporters suggests that content of TAAR8a is 58 times that of TAAR1 (Chiel- it would permit elevations in extracellular monoamine lini et al., 2007). The TAAR family of receptors is concentrations which in turn could produce downstream rather large, and many are still orphan receptors, so it is effects different from those of other TAAR1 agonists possible that other TAAR family members are sensitive that are monoamine transporter substrates. to T1AM and drive the thyronamine-mediated effects It has been reported that an i.p. injection of b-PEA on hypothermia as well. Other results demonstrate that a in mice produces an initial hyperthermia followed by a primary target of T1AM is the central nervous system. prolonged hypothermia (Mantegazza and Riva, 1963; Small doses of T1AM administered intracerebroventricu- Jackson, 1975), although the hypothermic rebound effect larly can mirror larger systemic administration (Klieverik at 50 mg/kg is quite subtle (Jackson, 1975). Our data et al., 2009). In studies using the Djungarian hamster reveals some indication of this effect in the KO but not Phodopus sungorus, T1AM administration brought about a the WT mice, though a more systematic investigation rapid decrease in metabolic rate, body temperature, re- that directly addresses this question is required to reveal spiratory quotient, ketonuria and loss of body fat indicat- whether there are any significant differences between ing a change in fuel utilization from carbohydrate to how WT and KO mice respond to b-PEA with regard lipid (Braulke et al., 2008). These authors suggested that to thermoregulation. We also observed that the hyper- the hypothermia is a result of reduced metabolic rate thermic response to both MDMA and methamphetamine and that T1AM results in depressed metabolism by treatments reached significance at earlier time points in decreasing carbohydrate utilization and increasing lipid the KO mice compared with the WT mice, perhaps in- metabolism. Another possible target of T1AM might be dicative of a greater sensitivity to in the a2A adrenergic receptors. The hyperglycemia induced in KO mice. Lastly, we observed that tyramine did not cause WT mice by T1AM is lost in a2A adrenergic null mice significant thermoregulatory responses in either WT or and blocked in WT mice that are treated with the a2A KO mice, exemplifying that not all TAAR1 agonists adrenergic antagonist yohimbine, although the concen- cause significant changes in thermoregulation. tration of T1AM needed to activate the receptor is in Accordingly, data presented here demonstrates that the micromolar range, a level high above that measured TAAR1 does not mediate the thermoregulatory in tissue (Regard et al., 2007). responses of T1AM nor those of b-PEA or ampheta- Nevertheless, our data demonstrate that, similar to mine-like psychostimulants. The very robust cryogenic rodent, T1AM is an agonist at rhesus monkey TAAR1. actions of T1AM suggest that it may be a therapeuti- We also confirm that T1AM is a blocker at the human cally-relevant cryogen, and its other physiological effects and mouse DAT as has been reported previously warrant further investigation into its biological role and (Scanlan et al., 2004; Snead et al., 2007). It had been mechanism of action. Importantly, compounds under shown that T1AM (10 lM) significantly inhibited development currently that target TAAR1 will likely [3H]dopamine, [3H]norepinephrine and [3H]serotonin not be limited in potential therapeutic efficacy by ther- transport by 77.6%, 72.1% and 42.2% respectively, in rat moregulatory side effects, nor are they likely to be cryo-

Journal of Neuroscience Research TAAR1 and Thermoregulation 1969 gens useful as neuroprotective agents in stroke. Never- Jackson DM. 1975. The effect of beta-phenylethylamine on temperature theless, T1AM produces robust thermoregulatory in mice and its possible mechanisms of action. Arzneimittelforschung responses and is an agonist at TAAR1, warranting fur- 25:776–780. ther investigation of its endogenous functions as well as Klieverik LP, Janssen SF, van Riel A, Foppen E, Bisschop PH, Serlie its usefulness as a therapeutic agent. MJ, Boelen A, Ackermans MT, Sauerwein HP, Fliers E, Kalsbeek A. 2009. Thyroid hormone modulates glucose production via a sympa- thetic pathway from the hypothalamic paraventricular nucleus to the ACKNOWLEDGMENTS liver. Proc Natl Acad Sci U S A 106:5966–5971. We thank Lundbeck Research USA, Inc. for gen- Lindemann L, Meyer CA, Jeanneau K, Bradaia A, Ozmen L, Blue- erously providing heterozygous TAAR1 knockout foun- thmann H, Bettler B, Wettstein JG, Borroni E, Moreau JL, Hoener der mice for establishment of our mouse colony, the MC. 2008. Trace amine-associated receptor 1 modulates dopaminergic NEPRC veterinary staff and animal care workers, and activity. J Pharmacol Exp Ther 324:948–956. the NEPRC Primate Genetics Core for assistance in Mantegazza P, Riva M. 1963. Amphetamine-like activity of b-phene- genotyping mice. thylamine after a monoamine oxidase inhibitor in vivo. J Pharm Phar- macol 15:472–478. REFERENCES Miller GM, Verrico CD, Jassen A, Konar M, Yang H, Panas H, Bahn M, Johnson R, Madras BK. 2005. Primate trace amine receptor 1 mod- Borowsky B, Adham N, Jones KA, Raddatz R, Artymyshyn R, Ogoza- ulation by the dopamine transporter. J Pharmacol Exp Ther 313:983– lek KL, Durkin MM, Lakhlani PP, Bonini JA, Pathirana S, Boyle N, 994. Pu X, Kouranova E, Lichtblau H, Ochoa FY, Branchek TA, Gerald C. National Research Council. 1996. Guide for the Care and Use of Labo- 2001. Trace amines: identification of a family of mammalian G protein- ratory Animals. National Academy Press, Washington, DC. coupled receptors. Proc Natl Acad Sci U S A 98:8966–8971. Regard JB, Kataoka H, Cano DA, Camerer E, Yin L, Zheng YW, Scan- Branchek TA, Blackburn TP. 2003. Trace amine receptors as targets for lan TS, Hebrok M, Coughlin SR. 2007. Probing cell type-specific novel therapeutics: legend, myth and fact. Curr Opin Pharmacol 3:90– functions of Gi in vivo identifies GPCR regulators of insulin secretion. 97. J Clin Invest 117:4034–4043. Braulke LJ, Klingenspor M, DeBarber A, Tobias SC, Grandy DK, Scan- Scanlan TS, Suchland KL, Hart ME, Chiellini G, Huang Y, Kruzich PJ, lan TS, Heldmaier G. 2008. 3-Iodothyronamine: a novel hormone Frascarelli S, Crossley DA, Bunzow JR, Ronca-Testoni S, Lin ET, controlling the balance between glucose and lipid utilisation. J Comp Hatton D, Zucchi R, Grandy DK. 2004. 3-Iodothyronamine is an en- Physiol B 178:167–177. dogenous and rapid-acting derivative of thyroid hormone. Nat Med Chiellini G, Frascarelli S, Ghelardoni S, Carnicelli V, Tobias SC, 10:638–642. DeBarber A, Brogioni S, Ronca-Testoni S, Cerbai E, Grandy DK, Snead AN, Santos MS, Seal RP, Miyakawa M, Edwards RH, Scanlan Scanlan TS, Zucchi R. 2007. Cardiac effects of 3-iodothyronamine: a TS. 2007. Thyronamines inhibit plasma membrane and vesicular mono- new aminergic system modulating cardiac function. FASEB J 21:1597– amine transport. ACS Chem Biol 2:390–398. 1608. Dhillo WS, Bewick GA, White NE, Gardiner JV, Thompson EL, Bata- Xie Z, Miller GM. 2008. Beta-phenylethylamine alters monoamine trans- veljic A, Murphy KG, Roy D, Patel NA, Scutt JN, Armstrong A, Gha- porter function via trace amine-associated receptor 1: implication for tei MA, Bloom SR. 2009. The thyroid hormone derivative 3-iodothyr- modulatory roles of trace amines in brain. J Pharmacol Exp Ther onamine increases food intake in rodents. Diabetes Obes Metab 325:617–628. 11:251–260. Xie Z, Miller GM. 2009a. Trace amine-associated receptor 1 as a mono- Doyle KP, Suchland KL, Ciesielski TM, Lessov NS, Grandy DK, Scanlan aminergic modulator in brain. Biochem Pharmacol (In Press). TS, Stenzel-Poore MP. 2007. Novel thyroxine derivatives, thyron- Xie Z, Miller GM. 2009b. A receptor mechanism for methamphetamine amine and 3-iodothyronamine, induce transient hypothermia and action in dopamine transporter regulation in brain. J Pharmacol Exp marked neuroprotection against stroke injury. Stroke 38:2569–2576. Ther 330:316–325. Fantegrossi WE, Ciullo JR, Wakabayashi KT, De La Garza R 2nd, Tray- Xie Z, Westmoreland SV, Bahn ME, Chen GL, Yang H, Vallender EJ, nor JR, Woods JH. 2008. A comparison of the physiological, behav- Yao WD, Madras BK, Miller GM. 2007. Rhesus monkey trace amine- ioral, neurochemical and microglial effects of methamphetamine and associated receptor 1 signaling: enhancement by monoamine transport- 3,4-methylenedioxymethamphetamine in the mouse. Neuroscience ers and attenuation by the D2 autoreceptor in vitro. J Pharmacol Exp 151:533–543. Ther 321:116–127. Frascarelli S, Ghelardoni S, Chiellini G, Vargiu R, Ronca-Testoni S, Xie Z, Westmoreland SV, Miller GM. 2008a. Modulation of monoamine Scanlan TS, Grandy DK, Zucchi R. 2008. Cardiac effects of trace transporters by common biogenic amines via trace amine-associated re- amines: pharmacological characterization of trace amine-associated ceptor 1 and monoamine autoreceptors in human embryonic kidney receptors. Eur J Pharmacol 10:587:231–236. 293 cells and brain synaptosomes. J Pharmacol Exp Ther 325:629–640. Hart ME, Suchland KL, Miyakawa M, Bunzow JR, Grandy DK, Scanlan Xie Z, Vallender EJ, Yu N, Kirstein SL, Yang H, Bahn ME, Westmore- TS. 2006. Trace amine-associated receptor agonists: synthesis and evalu- land SV, Miller GM. 2008b. Cloning, expression, and functional analy- ation of thyronamines and related analogues. J Med Chem 49:1101– sis of rhesus monkey trace amine-associated receptor 6: evidence for 1112. lack of monoaminergic association. J Neurosci Res 86:3435–3446.

Journal of Neuroscience Research