| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Dipartimento di Scienze dell’Uomo e dell’Ambiente, University of Pisa, Pisa, Italy;
Departments of Pharmaceutical Chemistry and Cellular and Molecular Pharmacology, University of California San Francisco, San Francisco, California, USA;
Departments of Physiology and Pharmacology and Cell and Developmental Biology, Oregon Health and Science University, Portland, Oregon, USA; and
Dipartimento di Farmacologia, University of Florence, Florence, Italy
1Correspondence: Dip. di Scienze dell’Uomo e dell’Ambiente, via Roma 55, 56126 Pisa, Italy. E-mail: r.zucchi{at}med.unipi.it
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: thyronamines • signal transduction • G protein-coupled receptors • myocardial function
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Endogenously produced T1AM has been detected in human, rat, mouse, and guinea pig blood as well as other tissues (1)
. Nanomolar concentrations of T1AM can activate trace amine associated receptor 1 (TAAR1), one member of a novel family of related G-protein coupled receptors (GPCRs). An extensive and comprehensive search for TAAR1-like sequences in several nucleotide sequence databases led to the identification of 17 different rat TAAR genes (2
3
4)
. It has been proposed (1)
that T1AM is a novel signaling molecule that can rapidly influence several physiological manifestations of thyroid hormone action, including body temperature, heart rate, and cardiac output. Unlike thyroid hormone, it has been hypothesized that T1AM exerts its influence via nongenomic effectors that include the TAAR1 GPCR and perhaps other TAAR subtypes as well. Here we characterize the cardiac effects of T1AM and provide biochemical evidence that it likely represents a new aminergic system modulating cardiac function. In particular, we describe the complete hemodynamic response to T1AM in the isolated working rat heart preparation and report for the first time the expression of various TAAR subtype transcripts, the existence of specific T1AM binding sites, the associated transduction pathways, and the presence of endogenous T1AM in rat cardiac tissue.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. [125I]T1AM (specific activity 2208 Ci/mmol) was synthesized by a procedure described elsewhere (6)
. The [125I]NaI required for the synthesis was obtained from New England Nuclear (Milan, Italy). Monoclonal 4G10 anti-phosphotyrosine antibodies were obtained from Upstate (Charlottesville, VA, USA), while secondary antibodies conjugated with horseradish peroxidase were purchased from Bio-Rad Laboratories (Milan, Italy). All other reagents were from Sigma-Aldrich (St. Louis, MO, USA) or from Invitrogen Life Technologies (Paisley, UK).
Animals and perfusion technique
This investigation conforms to the Declaration of Helsinki and the Guiding Principles in the Care and Use of Animals. The project was approved by the Animal Care and Use committee of the University of Pisa. Male Wistar rats (275–300 g body wt), fed with standard diet, were anesthetized with a mixture of ether and air. After injection of 1000 U sodium heparin in the femoral vein, the heart was quickly excised and perfused according to the working heart technique, as described previously (7)
. The height of the atrial and aortic cannulae was set at 20 and 100 cm, respectively. Unless otherwise specified, the perfusion buffer included (mM): 118 NaCl, 25 NaHCO3, 4.5 KCl, 1.2 KH2PO4, 1.2 MgSO4, 1.5 CaCl2, and 11 glucose. Perfusions were carried out using 200 ml of recirculating buffer, which was equilibrated with a mixture of O2 (95%) and CO2 (5%). Temperature was kept between 36.8 and 37°C, and the pH was 7.4. Powerlab/200 (ADInstruments, Castle Hill, Australia) was used for data acquisition. A few hearts were paced through an LE 12006 stimulator (LSI Letica, Barcelona, Spain) using a platinum electrode inserted into the right auricle, while the other electrode was tied to the aortic cannula.
Cardiomyocyte experiments
Single left ventricular myocytes were isolated from Wistar rats (200–250 g) as described previously (8)
. Cells were placed in an experimental bath on the platform of an inverted microscope (Nikon Diaphot TMD, Kawasaki, Japan). Action potential was recorded by means of the perforated-patch technique (9)
using a patch amplifier (Axopatch 200B, Molecular Devices, Sunnyvale, CA, USA) Cell length and shortening [expressed in arbitrary units (AU)] were measured by means of a Video Dimension Analyzer; the average of three consecutive twitches was used in order to minimize beat-to-beat variability. Signals were digitized via a DAC/ADC interface (Digidata 1200B, Molecular Devices) and acquired by means of pClamp software (vers. 9, Molecular Devices). The patch-clamped cell was superfused with Tyrode’s solution by means of a temperature-controlled microsuperfusor, which allowed rapid changes in the solution bathing the cell; the temperature was set to 36 ± 0.5°C. Patch-clamp pipettes, prepared from glass capillary tubes (Borosilicate GC150T, Harvard App., Kent, UK) by means of a two-stage puller (Flaming/Brown micropipette puller, model P-97, Sutter Inst., Novato, CA, USA), had a resistance of 1.5–2.5 M
when filled with the internal solution. The output signal was amplified (x100), digitized (Digidata 1200B, Axon Instruments, Union City, CA, USA), and acquired through the pClamp software. The composition of the solutions employed was as follows (in mM): Tyrode’s solution: 140 NaCl, 5.4 KCl, 1.5 CaCl2, 1.2 MgCl2, 10 glucose, 5 HEPES-NaOH, pH 7.35; and Pipette solution: 125 KMeSO4, 25 KCl, 1 EGTA, 0.13 amphotericin B, and 5 HEPES-KOH, pH 7.0.
Gene expression
The qualitative expression of target sequences was determined by reverse transcription-polymerase chain reaction (RT-PCR). Total RNA was extracted from rat ventricular tissue prepared from freshly dissected heart using TRIzol reagent, as described previously (10)
and quantified spectrophotometrically. DNAseI-treated RNA (1 µg) was retrotranscribed using the iScript cDNA synthesis kit (Bio-Rad Laboratories, Hercules, CA, USA) according to the instructions of the manufacturer. PCR reactions were performed in medium (total vol. of 25 µl) with the following composition: 1.5 mM MgCl2, 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 0.2 mM each deoxynucleoside trisphosphate, 0.2 µM each oligonucleotide, 1 U Platinum TaqDNA polymerase, and cDNA equivalent to 100 ng of total RNA. PCR was conducted with the following cycling program: 2 min at 95°C of initial denaturation, followed by 28–40 cycles of denaturation at 95°C for 30 s, annealing at 55–60°C (depending on oligonucleotide sequences) for 30 s, and extension at 72°C for 30 s. Appropriate controls were always performed in which either cDNA or reverse transcriptase was omitted. PCR products were size separated by electrophoresis in ethidium bromide-stained 2.5% agarose gels with bands visualized by ultraviolet transillumination.
Because qualitative RT-PCR showed that several TAARs were expressed, real-time RT-PCR was then performed to get a quantitative estimate of their relative expression. Real-time RT-PCR was conducted on a iQ5 Optical System (Bio-Rad) with the following cycle program: 30 s at 95°C, followed by 50 two-step amplification cycles consisting of 10 s denaturation at 95°C and 30 s annealing/extension at 60°. A final dissociation stage was run to generate a melting curve for verification of amplicon specificity and primer dimer formation. The reactions were performed in a total volume of 20 µl containing cDNA equivalent to 0.1 µg of total RNA, 0.4 µM each oligonucleotide, and 10 µl of iQ SYBR Green Supermix. All genes were amplified in triplicate with negative controls in which either cDNA or reverse transcriptase was omitted. Since we aimed at comparing the expression of different genes in a single RNA sample, we reported the 2–
Ct values obtained by taking the gene with the lowest Ct as a reference. GAPDH was also included in the analysis to compare TAAR expression to the expression of a constitutive housekeeping gene.
Oligonucleotide sequences for rat TAARs (TAAR1, TAAR2, TAAR3, TAAR4, TTAR5, TAAR6, TAAR7a, TAAR8a, and TAAR9) and for rat glyceraldehyde 3-phosphate dehydrogenase (GAPDH), which was used as a positive control, are shown in Table 1
. The sequences were designed on the basis of genes or coding sequences published in Gene Bank using Beacon Designer 4 software (Premier Biosoft International, Palo Alto, CA, USA). Oligonucleotide primers were chosen such that each set is specific for a given TAAR subtype. All TAARs but TAAR2 are encoded by a single exon, and TAAR2 primers were internal to the second exon. Primer specificity and quality were confirmed by sequencing PCR products obtained from rat ventricle genomic DNA: PCR conditions were the same as used for cDNA amplification, with 150 ng of template genomic DNA used per amplification.
|
Assay of [125I]T1AM binding
Perfused hearts were trimmed and homogenized on ice in 10 vol. of 50 mM Tris-HCl, 120 mM NaCl, 5 mM KCl, 250 mM sucrose, 12 µM leupeptin, 10 µM pepstatin A, and 100 µM PMSF by a Potter-Elvejheim homogenizer set at 800 rpm (15 passes, repeated twice). The homogenate was filtered through cheesecloth, and the resulting filtrate was cleared by centrifugation at 1,000 g for 10 min at 4°C; the resulting supernatant was centrifuged at 100,000 g for 30 min at 4°C to obtain a crude heart membrane preparation.
[125I]T1AM binding was determined in a buffer containing 50 mM Tris/HCl, 120 mM NaCl, 5 mM KCl, 50 µM ascorbic acid, 10 µM pargiline, 2 nM [125I]T1AM, and 0.6–1 mg/ml rat heart membrane protein and increasing concentrations of unlabeled T1AM, norepinephrine, or dopamine. After 10 min of incubation at 37°C, the binding reaction was stopped by incubating the preparation on ice for 1 min, followed by rapid filtration over GF/B filters (Whatman, Brentford, UK) presoaked in 0.3% polyethyleneimine. Filters were washed three times with 4 ml of ice-cold buffer (50 mM Tris-HCl, 120 mM NaCl, and 5 mM KCl). Radioactivity was counted on a gamma counter. Protein concentration was determined by Bio-Rad protein assay.
Cyclic AMP assay
Cyclic AMP (cAMP) assays were performed using a direct cAMP enzyme immunoassay kit (Sigma-Aldrich). Heart tissue was prepared according to the manufacturer’s instructions. Briefly, perfused hearts were frozen in liquid nitrogen, chopped, and homogenized in 10 vol. of 0.1 M HCl by a Potter-Elvejheim homogenizer set at 800 rpm (15 passes, repeated twice). The homogenates were centrifuged at 600 g for 15 min at room temperature, and the supernatant (postnuclear fraction) was stored at –80°C. Supernatants were diluted to the appropriate concentration with 0.1 M HCl provided in the kit.
Phosphotyrosine assay
At the end of each perfusion, ventricles were homogenized on ice in 5 vol. of ice-cold buffer containing 154 mM NaCl, 50 mM Tris-HCl, 12 µM leupeptin, 100 µM PMSF, 10 µM pepstatin A, 20 mM NaF, 1 mM vanadate, and 1 mM EDTA by 15 + 15 passes in a Potter-Elvejheim homogenizer set at 800 rpm. The standard subcellular fractions, i.e., the nuclear, mitochondrial, microsomal, and cytosolic fraction, were prepared as described elsewhere (11)
.
Western blotting was performed on 70–100 µg of protein prepared from each fraction. Samples were separated on a Nupage 4–12% Bis-Tris gradient gel (Invitrogen, San Diego, CA, USA). Proteins were electrically transferred to a nitrocellulose membrane according to the manufacturer’s instructions. After blocking with nonfat milk, the nitrocellulose membranes were incubated overnight at 4°C using primary 4G10 anti-phosphotyrosine antibodies (monoclonal, 1:1000) from Upstate (Charlottesvile, VA, USA). Membranes were washed and incubated for 120 min at room temperature with secondary antibody conjugated with horseradish peroxidase (HRP; 1:10,000) from Bio-Rad Laboratories (Milan, Italy). Immunoblots were visualized using a chemiluminescence reaction (Bio-Rad) and exposed to film. Quantitative comparison between different samples was performed by densitometric analysis.
Assay of endogenous thyronamimes
Hearts from rats were weighed and then homogenized in 10 ml of ice-cold 1 M HCl. The homogenate was centrifuged for 20 min at 14,000 rpm in a cold room, and the supernatant was transferred to falcon tubes and 10 µl of 10 µM deuterated T1AM (T1AM-d4) was added. The pH was adjusted to 10–11 with 40% KOH followed by the addition of 1 ml of 1 M CHES buffer. Ethyl acetate was used to extract the aqueous layer five times. Anhydrous sodium sulfate was added to the combined ethyl acetate layers, which were filtered and concentrated. The residue was redissolved by adding 31.5 µl of methanol and 38.5 µl of 0.1 M HCl, transferred to an Eppendorf tube, and stored at –80°C. T1AM and T1AM-d4 were detected using a triple-stage TSQ Quantum Discovery quadrupole mass spectrometer (Thermo Electron, San Jose, CA, USA). The ESI interface was operated using the following settings: sheath and aux gas flow rates, 45 and 20 respectively; source voltage, 3.0 kV; tube lens voltage, 180 V; capillary voltage, 35 V; and capillary temperature, 325°C. The transitions from m/z 356 
212, m/z 356 339, m/z 360 216, and m/z 360 343 were monitored. Scan event settings were Q2 collision gas, 1.0 mTorr; collision energy, 12; scan width, 1.2 m/z; scan time, 0.5 s; Q1 peak width, 0.7; and Q3 peak, width 0.90. An isocratic LC separation was utilized with a 200 x 2.1 mm, 5 µm Hypurity Advance C18 column (Thermo Hypersil, Waltham, MA, USA). The 0.4 ml/min mobile phase consisted of methanol:water (45:55) with 0.01% trifluoroacetic acid. Injection volumes were 10 µl in mobile phase with 0.1 M HCl. The signal to noise value for the lowest calibrator was 10:1. The intra-assay precision (CV) for calculated T1AM of buffer-extracted calibrators (over the range 5–350 pmol/sample) was 20% for the lowest calibrator and >15% for all other calibrators. Acceptable linearity was obtained for a linear regression of calibrator peak area ratios (T1AM/T1AM-d4) vs. pmol/sample T1AM, with typical correlation coefficients (r2) of >0.998. The average internal standard extraction efficiency from heart homogenates in buffer was 
20%.
Statistical analysis
Results are expressed as mean ± SE. Differences between groups were evaluated by one-way or two-way ANOVA, as appropriate, while individual group comparisons were performed using the Bonferroni post hoc test. When only two groups were compared, paired or unpaired t test was used, as appropriate. The threshold of statistical significance was set at P < 0.05. Radioligand binding data were analyzed by nonlinear fitting of competition experiments. GraphPad Prism version 4.1 for Windows (GraphPad Software, San Diego, CA, USA) was used for data processing and statistical analysis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. T1AM produced a dose-dependent decrease in cardiac output, aortic pressure, coronary flow, and heart rate. The IC50, defined as the concentration able to produce 50% reduction vs. the baseline value, averaged 27 µM for the effect on cardiac output and 37 µM for the effect on heart rate (Fig. 2
). The active form of the thyroid hormone, 3,5,3'-triiodothyronine (T3), when added to the perfusion buffer was totally ineffective at concentrations up to 40 µM. The deiodinated derivative of T1AM, thyronamine (T0AM), was much less effective than T1AM, producing 32% reduction of cardiac output at 56 µM concentration, with minor effects on heart rate and coronary flow. On the whole, we found that T1AM has both negative inotropic and chronotropic actions, the former occurring at lower concentrations than the latter. The decrease in coronary flow was proportional to the reduction in cardiac output, suggesting that it may represent the response to reduced energy demand. In paced hearts (250 bpm), contractile performance was also reduced by T1AM (e.g., in 2 experiments 54±5% decrease in cardiac output was observed 10 min after addition of 38 µM T1AM), and a negative inotropic response was produced by T0AM, in the absence of significant changes in heart rate.
|
|
As shown in Fig. 3
, the response to T1AM was extremely rapid. Pressure development and cardiac output declined after 30 s. The decrease in heart rate was out of proportion to the decrease in contractile performance, and its progression was substantially slower. At the highest T1AM concentration, these effects were preceded by a small transient increase in aortic pressure, which lasted only 15–20 s. The negative inotropic and chronotropic responses persisted for the whole duration of the perfusion (50 min) but were reversible after removal of T1AM. For instance, in one set of experiments, cardiac output averaged 54 ± 3% of the baseline after 50 min of perfusion with 25 µM T1AM and increased to 79 ± 3% of the baseline after a subsequent 30 min period of washout with standard buffer.
|
The direct contractile and electrophysiological effects of T1AM on cardiomyocytes were tested in patch-clamped cells by using the perforated patch technique. This technique allows for the simultaneous recording of action potential and cell shortening without dialysis of the cytoplasm. As shown in Fig. 4
, T1AM (50 µM) markedly prolonged the action potential elicited at 1Hz (Fig. 4A
, top traces); at the same time, cell shortening (an index of isotonic contractility) and its decay time were significantly reduced (Fig. 4A
, bottom traces).
|
Cardiac trace amine-associated receptors
Given the evidence that T1AM can stimulate the production of cAMP in cells heterologously expressing the cloned rat TAAR1 (1)
and knowing that 17 TAAR genes have been identified in the rat genome, it was of interest to establish the complement of TAAR subtypes expressed in rat heart. For this analysis, 40 cycles of RT-PCR were performed on total RNA prepared from adult rat ventricles as template with oligonucleotide primers specific for each subtype. DNA sequence analysis of the resulting products revealed that transcripts for TAAR1, TAAR2, TAAR3, TAAR4, and TAARR8a were detectable (Fig. 5
). TAAR9 also appeared to be expressed, but the signal was confounded by the presence of non-specific bands. TAAR7a was not detectable under the conditions we used (the bands observed in the TAAR7a lane did not correspond to the amplicon size), while we occasionally saw faint evidence of products corresponding to TAAR5 and TAAR6. To get a quantitative estimate of the relative expression of the different TAAR genes, real-time RT-PCR was performed (Table 2
). The expression of five different TAAR subtypes was confirmed, with a clear preponderance of TAAR8a mRNA, the content of which exceeded those of the other subtypes (namely, TAAR1, TAAR2, TAAR3, and TAAR4) by 35- to 116-fold.
|
|
To confirm the existence of specific binding sites for T1AM in cardiac membrane preparations, radioligand binding experiments were performed using custom-synthesized and labeled [125I]T1AM. Owing to the limited amount of ligand available, we only evaluated [125I]T1AM binding at nanomolar concentrations. Although no TAAR subtype-selective antagonist is currently available, competition binding experiments involving a panel of compounds provided evidence that [125I]T1AM binding is saturable and specific with an IC50 of 5 µM (Fig. 6
). [125I]T1AM was not displaced bynorepinephrine or by dopamine at concentrations of 
1 mM (data not shown).
|
Transduction pathways
In reconstituted cell models where TAAR1 has been heterologously expressed, T1AM can activate TAAR1 to produce cAMP (1)
. In our experimental model, cAMP concentrations were unchanged 20 s, 2 min, or 20 min after T1AM infusion (Fig. 7
). By contrast, a 2- to 3-fold increase was produced by isoprenaline.
|
The transduction of GPCR signaling often involves the activation of specific protein kinases. Therefore, other experiments were performed in an effort to evaluate the influence of different kinase inhibitors on the response to T1AM. The effects of T1AM or T0AM were not significantly modified by 10 nM H-89 (protein kinase A inhibitor), 2 µM chelerythrine (protein kinase C inhibitor), 1 µM lavendustin C (calcium-CaM-dependent kinase II inhibitor), 3 µM LY 294002 (phopshatidylinositol-3-kinase inhibitor), 13 µM SB 203580 (MAP kinase-2 inhibitor), and 37 µM PD 98059 (MAP kinase kinase inhibitor). Insulin (5 IU/l) and atropin (4.5 µM) were also ineffective. To investigate the role of Gi proteins, hearts were perfused for 10 min in the presence of 4 µg/L PTX and then challenged with 1.5 µM acetylcholine (Ach) or with T1AM: although the negative chronotropic response to Ach was remarkably blunted (14 vs. 63% decrease in heart rate) as expected (12)
, the effect of T1AM was not modified.
In contrast, a significant effect was produced by the tyrosine kinase inhibitor genistein (Fig. 8
, top). Although genistein (37 µM) did not modify cardiac output, if used alone, it increased dramatically the response to T1AM, such that after 30 min the heart was no longer able to support a significant cardiac output. In a few minutes severe bradycardia occurred as well. Similar findings, i.e., potentiation of the response to T1AM, were produced by the phospholipase C inhibitor U-73122 at a concentration (0.5 µM), which produced only a slight negative inotropic effect, if U-73122 was used alone (data not shown).
|
Opposite effects were observed with the tyrosine phosphatase inhibitor vanadate (50 µM), which allowed better preservation of inotropic state after infusion of T1AM (Fig. 8
, bottom). The effect of vanadate was blunted after 30 min, but a slight decrease in cardiac output was also observed after 30–50 min of perfusion with vanadate alone.
In another set of experiments, the concentration of calcium in the perfusion buffer was reduced from 1.5–0.75 mM. This effected a slight reduction in baseline cardiac output (50.0 vs. 64.4 ml/min), while leaving the heart rate unchanged. At low calcium concentration, the response to T1AM was potentiated. In particular, the negative inotropic effect of T1AM was remarkably increased, even after correction for the difference in baseline values (Fig. 9
).
|
Phosphotyrosine assay
Because experiments demonstrated that the response of the heart to T1AM was affected by tyrosine kinase inhibitors and tyrosine phosphatase inhibitors, it was of interest to determine whether the pattern of tyrosine phosphorylation was altered after exposure to T1AM. Antibodies raised against phosphotyrosine were used in Western blot experiments performed on the four standard subcellular fractions prepared from ventricular tissue: cytosolic, nuclear, mitochondrial, and microsomal. After 30 min of perfusion with T1AM, significant changes can be seen in the banding pattern of tyrosine phosphorylation, particularly in the microsomal and cytosolic fractions. As shown in Fig. 10
, a high molecular mass band (>100 KDa) present in the microsomal fraction disappeared or was remarkably attenuated, whereas a low molecular mass band (26 KDa) appeared. In addition, decreased tyrosine phosphorylation was observed in several protein bands present in the cytosolic fraction, especially in the region of the gel corresponding to species of 74 KDa. Densitometric analysis of three different preparations (not shown) confirmed that the changes induced by T1AM were statistically significant.
|
Thyronamine content of cardiac tissue
Previously, we reported T1AM to be present in several rodent tissues (1)
. Here we tried to determine the amount of endogenous thyronamines present in rat myocardial tissue using an HPLC tandem mass spectrometry method with synthetic standards. Quantitative analysis could be done with the endogenous T1AM signals using a synthetic deuterium labeled T1AM derivative as an internal standard. We evaluated three different groups of rat hearts and found that endogenous T1AM levels averaged 68 pmol/g wet wt and showed a wide range of variation from 1 to 210 pmol/g (see Fig. 11
for representative HPLC tracings).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. When heterologously expressed, rat, mouse, and human TAAR1s rapidly couple to the stimulation of cAMP production when exposed to T1AM. In vivo T1AM has been found to produce physiological effects in a matter of minutes that are opposite to those induced over time by thyroid hormone (1)
. In the present work, we report the characterization of the effects of T1AM in the isolated rat heart preparation, an experimental model system in which neurohumoral influences are eliminated, allowing an easy and direct evaluation of the effects of the drug on cardiac physiology and performance at a physiological work load.
Consistent with our previous findings (1)
, micromolar concentrations of T1AM decreased cardiac output and heart rate. We also observed that aortic pressure was reduced and that coronary flow decreased to a lower extent than cardiac output. These findings provide evidence for a negative inotropic and chronotropic effect. Notably, the reduction in heart rate cannot account for decreased cardiac output, because the former occurred at a lower concentration and was out of proportion to the latter. For instance, with 38 µM T1AM a 55% decrease in heart rate was matched by a 65% decrease in cardiac output, while with 1.5 µM Ach a 63% decrease in heart rate was matched by a 24% decrease in cardiac output. This interpretation is supported by the results obtained in paced hearts and by the response to the less effective compound T0AM, which decreased cardiac output, in the absence of significant effects on the heart rate.
The response to T1AM and T0AM developed within 30 s, and it was readily reversible, consistent with a non-genomic action and it being mediated by stimulation of a metabotropic GPCR. Notably, equimolar concentrations of T3, the active form of thyroid hormone, were totally ineffective over this time scale, when added to the perfusion buffer, and in general the effects of T1AM were opposite to those induced on a longer time scale by thyroid hormone.
The negative inotropic effect of T1AM was also apparent in isolated cardiomyocytes, suggesting that the latter are a direct target of this compound. Prolonged duration of the action potential might contribute to the negative chronotropic effect of T1AM, although this hypothesis should be validated by investigations performed in sinus node cells.
Because thyronamines have been reported to interact with heterologously expressed TAAR1 (1)
, a novel aminergic GPCR, our functional experiments suggest the existence of a novel aminergic system that is capable of modulating cardiac function. However, many different TAAR subtypes have been described previously. Nine different orthologous genes (i.e., genes generated through a speciation event) have been identified and designated as TAAR1 to TAAR9 (4)
. In some species, paralogue genes (i.e., a gene generated by duplication events within the lineage of the species) are present and they are distinguished by a letter suffix. On the whole, the rat genome includes 17 putatively functional genes (and 2 putative pseudogenes), and their cardiac expression has not been determined. Here we provide evidence for the expression of mRNA coding for at least five different TAAR subtypes in rat heart, with quantitative preponderance of TAAR8a. Unfortunately, reagents specific for each subtype of receptor protein are not available. However, radioligand binding experiments confirmed the presence of specific binding sites for T1AM.
The type(s) of G proteins coupled to every TAAR and the immediate downstream effectors have yet to be established, although there is widespread evidence that activated recombinant rat, mouse, chimp, and human TAAR1s heterologously expressed in HEK293 cells or Xenopus oocytes couple to Gs and stimulate cAMP production (1)
. In our model, we did not detect any change in tissue cAMP, and the functional effects produced by T1AM are opposite to those induced by agents that increase tissue cAMP, such as forskolin or ß-adrenergic stimulation. However, we cannot exclude that a localized increase in cAMP may occur, activating signaling pathway(s) able to reduce cardiac inotropic and chronotropic state. Alternatively, in a native environment (nonheterlogous cell), activation of TAAR1 may be coupled to some other second messenger pathway, or the receptor(s) mediating the effects of T1AM in the heart may not be TAAR1 but other TAAR subtypes. The latter hypothesis is consistent with the observed preferential expression of TAAR8a.
We further investigated the transduction pathway(s) that might be involved in mediating the cardiac effects of T1AM using a pharmacological approach. Inhibitors of protein kinase A, protein kinase C, calcium-CaM-dependent kinase II, phopshatidylinositol-3-kinase, MAP kinase-2, and MAP kinase kinase did not affect the response to T1AM, suggesting that these effectors do not play a major role. In contrast, the effect of T1AM was significantly blunted by vanadate, a tyrosine phosphatase inhibitor, while it was dramatically potentiated by genistein, a tyrosine kinase inhibitor. The interpretation of the vanadate experiments is complicated by the fact that vanadate inhibits P-type ATPases. This has been associated with direct inotropic effects, namely a negative inotropic effect at 5–10 µM concentration and a positive inotropic effect at higher concentration (20 µM; ref 13
). The former is attributed to inhibition of the Na+,K+-ATPase (digitalis-like effect), whereas the latter is probably due to inhibition of the sarcoplasmic reticulum Ca2+-ATPase that depletes the sarcoplasmic reticulum calcium pool. As expected, in our experimental model, the direct inotropic effect of 50 µM vanadate was a negative one, as shown by the time-dependent decrease in cardiac output, and therefore, it cannot account for the observed blunting of T1AM negative inotropic effect.
Based on these findings, it is our speculation that T1AM might induce the dephosphorylation of critical tyrosine residues, possibly by activating phosphotyrosine phosphatases and that this effect is accentuated if baseline tyrosine kinase activity is impaired by genistein. Alternatively, tyrosine kinases themselves might be the target of T1AM-dependent tyrosine phosphatases, since modulation of receptor tyrosine kinases by protein tyrosine phosphatases has been described in several tissues (14
, 15)
. Interestingly, U-73122, a phospholipase C inhibitor, also potentiated the response to T1AM. Mutual interactions between phospholipase C and receptor tyrosine kinases occur in several transduction pathways, and it might be the case that the pathway(s) involved in the response to T1AM is/are also targeted by phospholipase C.
Modulation of nonreceptor tyrosine phosphatases by G protein-coupled receptors has been reported in several cell types, namely pheocromocytoma cells (16)
, fibroblasts (17)
, endometrial and ovarian cancer cells (18
, 19)
, and glioma cells (20)
. In some of these systems, signal transduction is inhibited by PTX, suggesting the involvement of Gi proteins (16
, 19)
, while in others Gq proteins are probably involved (18)
. In our experimental model, the response to T1AM was not modified by pretreatment with PTX, which would argue against a major role for Gi proteins in mediating its effects on the heart. Further experiments will be necessary to test the potential role of Gq proteins and/or Rho family members.
To test the hypothesis that tyrosine phosphorylation state is affected by T1AM, we performed Western blot experiments using anti-phosphotyrosine antibodies that recognize a wide array of epitopes containing phosphotyrosine. Perfusion with T1AM was associated with decreased tyrosine phosphorylation of several proteins, i.e., microsomal fraction proteins migrating with high molecular mass (>100 KDa) and cytosolic proteins migrating at 74 kDa. Although this is a crude assay and it does not establish a cause and effect relationship, the results are consistent with the hypothesis that changes in the phosphorylation state of critical tyrosine residues may play an important role in the response of the heart to T1AM.
Further investigations will be necessary to establish the molecular identity and physiological function of the protein(s) present in these bands and the final molecular effectors of the response to T1AM. In the present work, we observed that the negative inotropic action was remarkably accentuated if calcium concentration was reduced. Interestingly, similar findings were reported with the calcium channel antagonist verapamil (21)
. In general, theoretical considerations suggest that the negative inotropic response to interventions that reduce calcium availability to contractile proteins should be more evident at low extracellular calcium concentration (22)
, and this prediction can be validated in the isolated working rat heart (Zucchi et al., unpublished observations). Therefore, it would be interesting to evaluate whether T1AM can affect calcium fluxes and whether the proteins the phosphorylation state of which is modified by T1AM are involved in the control of calcium homeostasis.
Competition binding experiments permitted the determination of an IC50 for T1AM that was on the order of 5 µM. This value should be a good estimate of the dissociation constant (KD) of T1AM, since [125I]T1AM concentration was three orders of magnitude lower. These results are in good agreement with the results of other experiments in which the IC50 for the negative inotropic effect averaged 28 µM.
Cardiac T1AM content averaged 68 pmol/g wet wt, which is about one order of magnitude lower than myocardial norepinephrine or Ach content but similar to epinephrine, dopamine, or adenosine content (23
24
25
26
27)
. By comparison, the myocardial content of bioactive enkephalins, which are thought to produce significant physiological and pathophysiological effects, is on the order of 20–30 pmol/g (28)
, i.e., it is two to three times lower than T1AM content.
It is difficult to establish the physiological relevance of endogenous T1AM unless its free concentration, subcellular distribution, and receptor binding properties are determined. Since myocardial density is close to 1.05, the average tissue concentration of T1AM is close to 70 nM. Intercellular space and ground substance account for 10% of total tissue volume, myofibrils for 35%, and mitochondria for another 25% (29
, 30)
. Therefore, if T1AM cannot cross the sarcolemma and/or is compartmented, its local concentration would increase by up to one order of magnitude. However, the occurrence and extent of T1AM binding to proteins or other subcellular structures are unknown. Interestingly, in the few assays that we performed, T1AM concentration varied over a wide range, and in individual hearts, it was remarkably higher or lower than the average value. Additional work is necessary to clarify these issues and in particular to determine the factors affecting endogenous T1AM levels. However, on the basis of our findings it seems likely that under certain physiological conditions T1AM concentrations may not be far from the range able to produce functionally relevant effects.
Cardiac T1AM content is 20- and 2-fold higher than T3 and thyroxine content, respectively (31
, 32)
. T1AM has been detected in blood (1)
, so circulating T1AM might access cardiac TAARs. However, the heart is known to contain aromatic amino acid decarboxylase (33
, 34)
, as well as type I and type III deiodinases (35)
. Therefore, T1AM might also be produced in situ from thyroid hormone and represent a novel branch of thyroid hormone signaling.
Notably, very few endogenous negative inotropic agents have been identified (36)
. The best known is adenosine, but it produces only minor effects on ventricular myocardium, unless the inotropic state is increased by adrenergic stimulation. Myocardial depression has been reported after exposure to cytokines, such as tumor necrosis factor 
and interleukin-6, while other endogenous compounds (platelet activating factor, arachidonic acid, and nitrogen monoxide) can either increase or decrease cardiac contractility depending on concentration and experimental models.
Although T1AM is a novel compound only recently discovered, a few studies with T0AM were performed over 30 yr ago. In the dog, intravenously administration of T0AM was seen to produce an increase in cardiac output and dP/dt, after a lag of 10 min (37)
. This effect was remarkably blunted by adrenergic blockade, suggesting that T0AM was acting by stimulating catecholamine release. It was also reported that T0AM appeared to have a central nervous system excitatory effect, so the experimental dogs were administered unusually high doses of general anesthetics and opioids for the whole duration of each study. Therefore, it is difficult to compare this in vivo investigation with our ex vivo denervated model. However if the anesthetized dog model was subjected to catecholamine depletion and/or adrenergic blockade, T0AM infusion produced an immediate, transient, negative inotropic effect. We suggest that the latter observation reflected the direct, TAAR-mediated, response to T0AM.
In conclusion, T1AM is a novel endogenous agent that produces negative inotropic and chronotropic effects in the isolated working rat heart. Several TAAR subtypes are expressed in the adult rat heart and specific binding sites for T1AM can be detected. The effect of T1AM is blunted by vanadate and potentiated by genistein, suggesting that TAAR-dependent changes in the phosphorylation state of critical tyrosine residues may be important. Evidence for reduced tyrosine phosphorylation after exposure to T1AM has been observed in cytosolic and microsomal proteins. It seems reasonable that this heretofore novel transduction pathway might be a target worthy of therapeutic exploitation. In addition, because T1AM can be detected in heart, endogenous T1AM levels could play a role in the physiological or pathophysiological modulation of cardiac function.
|
ACKNOWLEDGMENTS |
|---|
Received for publication October 6, 2006. Accepted for publication December 25, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-mediated activation of phosphotyrosine phosphatase. Endocrinology 142,2369-2380This article has been cited by other articles:
|
|
 |
|
  S. Piehl, T. Heberer, G. Balizs, T. S. Scanlan, R. Smits, B. Koksch, and J. Kohrle Thyronamines Are Isozyme-Specific Substrates of Deiodinases Endocrinology, June 1, 2008; 149(6): 3037 - 3045. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||
; n=7) or with buffer containing 21 µM 3,5,3' triiodothyronine (
; n=4), 19 µM T1AM (•; n=7), 25 µM T1AM (
; n=4), or 38 µM T1AM (
; n=4). Data are percentage of baseline values. Actual baseline values are given in the text. Bars are mean ± SE (bars are not shown when smaller than symbol dimensions). +P < 0.05, *P < 0.01 vs. control at the corresponding time point by two-way ANOVA followed by Bonferroni posthoc test.
; n=3), or their combination (
