Abstract
3-Iodothyronamine (T1AM) is an endogenous relative of thyroid hormone with profound metabolic effects. In different experimental models, T1AM increased blood glucose, and it is not clear whether this effect is entirely accounted by changes in insulin and/or glucagone secretion. Thus, in the present work, we investigated the uptake of T1AM by hepatocytes, which was compared with the uptake of thyroid hormones, and the effects of T1AM on hepatic glucose and ketone body production. Two different experimental models were used: HepG2 cells and perfused rat liver. Thyronines and thyronamines (T0AMs) were significantly taken up by hepatocytes. In HepG2 cells exposed to 1 μM T1AM, at the steady state, the cellular concentration of T1AM exceeded the medium concentration by six- to eightfold. Similar accumulation occurred with 3,5,3′-triiodothyronine and thyroxine. Liver experiments confirmed significant T1AM uptake. T1AM was partly catabolized and the major catabolites were 3-iodothyroacetic acid (TA1) (in HepG2 cells) and T0AM (in liver). In both preparations, infusion with 1 μM T1AM produced a significant increase in glucose production, if adequate gluconeogenetic substrates were provided. This effect was dampened at higher concentration (10 μM) or in the presence of the amine oxidase inhibitor iproniazid, while TA1 was ineffective, suggesting that T1AM may have a direct gluconeogenetic effect. Ketone body release was significantly increased in liver, while variable results were obtained in HepG2 cells incubated with gluconeogenetic substrates. These findings are consistent with the stimulation of fatty acid catabolism, and a shift of pyruvate toward gluconeogenesis. Notably, these effects are independent from hormonal changes and might have physiological and pathophysiological importance.
Introduction
3-Iodothyronamine (T1AM) is an endogenous relative of thyroid hormone, which is able to produce acute functional effects (Scanlan et al. 2004, Piehl et al. 2011). Recent advances in the quantitative analysis of T1AM using methods based on HPLC coupled to tandem mass spectrometry (MS/MS) have shown that T1AM can be detected in blood and tissues derived from rodents and humans, and that in several rat tissues, including heart, liver, kidney, skeletal muscle, and stomach, T1AM concentration is greater than 3,5,3′-triiodothyronine (T3) and thyroxine (T4) concentration (Saba et al. 2010, Galli et al. 2012). Although structural similarities between T4 and T1AM have led to the hypothesis that T1AM is produced in the peripheral tissues by T4 deiodination and decarboxylation, the biosynthetic origin of T1AM remains an open issue. Serum analysis of thyroid cancer patients treated with T4 (Hoefig et al. 2011) provided evidence for extrathyroidal T1AM production, and limited T1AM formation was observed in cardiomyocytes incubated with T3 (Saba et al. 2010). On the other hand, a recent study has revealed that T1AM is not an extrathyroidal metabolite of T4, and is produced by a process that requires the same biosynthetic factors necessary for T4 synthesis, namely the sodium–iodide symporter and thyroperoxidase (Hackenmueller et al. 2012).
The molecular target(s) of T1AM are largely unknown. In vitro studies provided evidence that T1AM can activate with high-affinity G protein-coupled receptors, including trace amine-associated receptor 1 (Scanlan et al. 2004, Zucchi et al. 2006) and possibly α2-adrenoceptors (Regard et al. 2007). In addition to these receptors, T1AM also interacts with plasma membrane and vesicular biogenic amine transporters (Snead et al. 2007) as well as different mitochondrial targets (Venditti et al. 2011, Cumero et al. 2012).
Administration of exogenous T1AM and other thyronamines (T0AMs) produced functional effects that showed a rapid onset and were often opposite to those induced by T3 (Liggett 2004, Weatherman 2007). In rodents, i.p. T1AM injection rapidly induced hypothermia, decreased cardiac function and decreased the respiratory quotient, suggesting a shift from primarily carbohydrate to predominantly lipid utilization (Scanlan et al. 2004, Chiellini et al. 2007, Braulke et al. 2008). Recent results from NMR-based metabolomics and breath studies have shown that chronic T1AM exposure induced a rapid increase in lipid mobilization, followed after a few days by increased protein breakdown (Haviland et al. 2013). T1AM-treated mice showed continued reduction in body weight, independent of food consumption, and after T1AM withdrawal they regained only 1.8% of the lost weight in the following 2 weeks. Intracerebroventricular (i.c.v.) injection of T1AM modified hormone secretion, food intake, and memory acquisition (Dhillo et al. 2009, Klieverik et al. 2009, Manni et al. 2012, 2013).
T1AM effects were not linearly related to the dosage and depended on the animal species and administration route (Dhillo et al. 2009, Klieverik et al. 2009, Manni et al. 2012). The response to T1AM may also be affected by its complex metabolism, which includes oxidative deamination (to 3-iodothyroacetic acid, TA1), deiodination (to T0AM), sulfation (to O-sulfonate-T1AM), acetylation (to N-acetyl-T1AM), and glucuronidation (to T1AM-glucuronide) (Wood et al. 2009, Hackenmueller & Scanlan 2012).
Among its metabolic effects, T1AM has been reported to produce hyperglycemia (Regard et al. 2007, Klieverik et al. 2009). Notably, hyperglycemia occurs after administration of relatively low doses of exogenous T1AM, producing changes in tissue concentration of about one order of magnitude (Manni et al. 2013). In addition a clinical investigation performed in a small series of patients revealed that serum T1AM concentration was significantly correlated with HbAlc, and significantly increased in a subgroup of diabetic patients (Galli et al. 2012). Therefore, the effects of T1AM on glucose metabolism might have physiological and pathophysiological relevance. The mechanism of this effect is largely unknown. As hyperglycemia was observed after i.c.v. injection, it was originally attributed to central regulation of endocrine function, namely stimulation of glucagon secretion and/or inhibition of insulin secretion (Klieverik et al. 2009). However, it was recently observed that even after i.c.v. injection, a significant increase in plasma T1AM was produced (Manni et al. 2012), and the possible occurrence of peripheral effects on glucose metabolism has been suggested. Because of the central role of liver in glucose homeostasis, the aim of the present work was to establish whether T1AM affects glucose metabolism in perfused liver and in a hepatocellular carcinoma cell line (HepG2 cells). An additional aim was to determine the uptake and metabolism of T1AM in the same preparations.
Materials and methods
Chemicals
T1AM, T0AM, TA1, TA0, and their deuterated derivatives used for HPLC–MS/MS were synthesized as described elsewhere (Hart et al. 2006, Miyakawa & Scanlan 2006, Wood et al 2009). Unless otherwise specified, all other reagents were from Sigma–Aldrich. Solvents for HPLC–MS/MS measurements were HPLC grade, and the other chemicals were reagent grade.
Cell culture and treatment
Human hepatocellular carcinoma cells (HepG2), obtained from American Type Culture Collection (Manassas, VA, USA), were cultured in DMEM supplemented with 10% (v/v) fetal bovine serum, 1 mM pyruvate, 100 U/ml penicillin, and 100 μg/ml streptomycin at 37 °C in a humidified atmosphere containing 5% CO2 and subcultured before confluence.
The experiments aimed at evaluating hormone uptake and metabolism were carried out as described previously, with minor modifications (Saba et al. 2010). Briefly, cells were seeded into 24-well plates (8.5×104 cells/well) and grown to 80% confluence. At the start of each experiment, the culture medium was removed and replaced with 0.5 ml of Krebs–Ringer medium (118 mM NaCl, 25 mM NaHCO3, 4.5 mM KCl, 1.2 mM KH2PO4, 1.2 mM MgSO4, 1.5 mM CaCl2, 1 g/l glucose, pH 7.4) and preincubated in a humidified atmosphere of 5% CO2 at 37 °C for 30 min. Incubation was initiated by quickly replacing the preincubation medium with 0.5 ml Krebs–Ringer buffer, containing one of the following compounds: T1AM, T3, T4 at 1 μM concentration. For each of the tested compound, a 1 mM stock solution in DMSO was previously prepared. The plate was returned to a humidified atmosphere of 5% CO2 at 37 °C, the medium was removed from each well at specific time points (5 min to 24 h), centrifuged (11 600 g for 2–3 min) to eliminate detached cells, and analyzed by HPLC–MS/MS analysis. The cell plates were then frozen for 24–48 h and lysed in 0.1 ml 0.1 M NaOH. After pH neutralization (0.01 ml 1.0 N HCl), the cell lysates were diluted with Krebs–Ringer to a final volume of 0.5 ml, centrifuged at 5000 g for 10 min, and used for HPLC–MS/MS analysis. As assessed in previous experiments, where cells were removed from the well by scraping in the presence of ice-cold PBS followed by centrifugation at 5000 g for 10 min, the packed cell volume was on the order of 0.02 ml. Therefore, we can assume that during cell lysates preparation (0.5 ml, final volume), the cellular content in the final sample was diluted by about 25-fold.
Similar experiments were carried out to examine the effect of amine oxidase inhibitor, iproniazid, on T1AM metabolism. The cells were preincubated for 30 min with 0.5 ml Krebs–Ringer buffer supplemented with iproniazid (10 mM stock solution in DMSO to a final concentration of 100 μM) before adding T1AM.
To assess glucose and ketone body release, HepG2 were seeded into six-well plates (5×105 cells/well) and grown to 80% of confluence with standard medium. Before treatment, the cells were washed twice with PBS and then exposed for 4 h to exogenous T1AM (0.1, 0.5, 1, 5, and 10 μM) or TA1 (0.5 and 1 μM) in 1 ml DMEM base, glucose- and phenol red-free, containing 100 U/ml penicillin, 100 μg/ml streptomycin, and 4 mM l-glutamine, supplemented with 2 mM sodium pyruvate and 20 mM sodium lactate (glucose production buffer; Yoon et al. 2001) at 37 °C in 5% CO2. In some experiments, the amine oxidase inhibitor iproniazide (100 μM) was also included. Control cells were incubated with DMEM containing DMSO (10–20 μl/well).
To assess free fatty acid (FFA) release, HepG2 were exposed to exogenous 1 μM T1AM for four or 24 h in DMEM base (glucose- and phenol red-free), containing 100 U/ml penicillin, 100 μg/ml streptomycin, and 4 mM glutamine, supplemented with 1 g/l glucose and 1 mM sodium pyruvate.
To induce steatosis, HepG2 were exposed to exogenous lipids as described (Di Nunzio et al. 2011, Yao et al. 2011). Cells were incubated with 1 μM T1AM in DMEM base (glucose- and phenol red-free) containing 100 U/ml penicillin, 100 μg/ml streptomycin, 4 mM glutamine, 1 g/l glucose, 1 mM sodium pyruvate, and supplemented with 1 mM of FFA mixture (2:1 ratio of oleate and palmitate) in 1% BSA for 24 h (stock solution of 30 mM FFA was conveniently diluted to 1 mM in culture medium). In each experimental protocol, control cells were incubated with supplemented DMEM-containing DMSO.
Cell culture medium was then collected and glucose, ketone bodies (acetoacetate and 3-hydroxybutyrate), and fatty acid levels were evaluated.
Rat liver perfusion
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 weight), fed a standard diet, were anesthetized with a mixture of ether and air and the livers were perfused in situ with glucose-free Krebs–Ringer buffer or glucose production buffer equilibrated with a mixture of O2 (95%) and CO2 (5%) at a constant flow rate of 10 ml/min. The fluid was introduced through a cannula inserted into the portal vein, while a second cannula inserted into the inferior cava vein was used to collect the effluent perfusate. After 10 min washout period and 10 min perfusion with Krebs–Ringer buffer alone, T1AM (1 μM or 50 nM) was added to the non-recirculating buffer, and the liver was perfused for another 40 min, followed by 10 min of perfusion with Krebs–Ringer buffer alone (Mario et al. 2009). In control experiments, liver was perfused for 60 min with Krebs–Ringer buffer alone. During the experiment, samples of the perfusion buffer were collected at 5 min intervals, and at the end of the experiments the liver was removed and frozen at −80 °C.
In the experiments to evaluate hormone up-take and metabolism, isolated rat liver was perfused with 50 nM or 1 μM T4, T3, or T1AM in Krebs–Ringer buffer, as described earlier, and tissue concentration was determined after 60 min. HPLC–MS/MS analysis that allowed the simultaneous detection of T3, T4, T1AM, and its putative metabolites, namely T0AM, TA0, and TA1, was carried out as described elsewhere (Saba et al. 2010).
Metabolite assays
Glucose concentration was assessed in 100 μl of medium or perfusate with a colorimetric glucose assay kit (Sigma–Aldrich). Ketone bodies were evaluated in the medium or perfusate by enzymatic methods as elsewhere described (Alberti & Hockaday 1972). Briefly, acetoacetate was determined in a reaction mix containing 0.05 mM K2HPO4 (pH 7), 0.02% NADH, and 0.25 U/ml 3-hydroxybutyrate dehydrogenase. The mixture was incubated for 45 min at 30 °C and absorbance was read at 340 nm to evaluate NADH reduction. 3-Hydroxybutyrate was measured in a reaction mix containing 0.03 M Tris–HCl (pH 8.5), 2% hydrazine, 0.06% EDTA, 0.02% NAD, and 0.25 U/ml 3-hydroxybutyrate dehydrogenase. The mixture was incubated for 45 min at 30 °C and absorbance was read at 340 nm to evaluate NADH production.
Fatty acid concentration was determined in the cell culture medium with a FFA quantification kit (BioVision Research Product, Malpitas, CA, USA) upon extraction. Fatty acids were extracted by adding 200 μl of dichloromethane to 850 μl of medium. The mixture was vortexed and spinned at 10 000 g for 5 min. The organic phase was collected and vacuum dried for 1 h to remove dichloromethane. Dried lipids were then dissolved in 50 μl of assay buffer provided by the kit and used for the assay.
Metabolite concentrations were referred to the total protein content (Bradford 1976) of whole-cell lysates (in HepG2 experiments) or to wet tissue weight (in liver experiments).
Statistical analysis
Results are expressed as the mean±s.e.m. Differences between groups were analyzed by ANOVA or, when only two groups were involved, by unpaired t-test. Regression analysis was performed by linear, exponential, or hyperbolic models, as detailed in the description of each experiment. The threshold of statistical significance was set at P<0.05. GraphPad Prism version 6.0 for Windows (GraphPad Software, San Diego, CA, USA) was used for data processing and statistical analysis.
Results
T1AM uptake and catabolism in HepG2 cells and in perfused rat liver
When cells were exposed to 1 μM T1AM, its concentration in the incubation medium decreased exponentially, while lysate T1AM increased, reaching a steady state after about 10 min (Fig. 1A). In these experiments, incubation medium volume was 500 μl, while the original volume of cell lysate was on the order of 20–25 μl, and the latter was diluted to a final volume of 500 μl to facilitate processing and assay. Steady-state concentrations were close to 500 and 150 nM in the incubation medium and in diluted cell lysate respectively (Fig. 1A). Taking into account the dilution factor, the actual steady-state cellular concentration can be estimated to average about 3–4 μM, exceeding medium concentration by over six- to eightfold. The overall recovery of T1AM was on the order of 65–70% (500 pmol were added and the final amounts detected in incubation medium and in cell lysate averaged ∼250 and 75 pmol respectively).
T1AM up-take and metabolism in HepG2 cells (subjected to 20–25 passages in vitro): (A) results of T1AM up-take in HepG2 cell during 6 h of incubation with 1 μM T1AM (mean±s.e.m. of nine biological replicates). (B) Catabolite production during incubation with 1 μM T1AM: representative results of T0AM and TA1 assay; (C) representative results of T1AM and TA1 assay in experiments carried out in the presence of 100 μM iproniazid. Assays were performed at different time in the incubation medium and in the cell lysate. During preparation, the lysate was diluted ∼20- to 25-fold to a final volume of 500 μl.
Citation: Journal of Endocrinology 221, 1; 10.1530/JOE-13-0311
T1AM adhesion to cell culture plates, as determined by carrying out the experiment in the absence of cells, was limited, accounting only for about 10% decrease in T1AM concentration. Additional experiments were also carried out at 4 °C, and the rate constant of the exponential decay decreased from 0.052±0.006/min to 0.025±0.004/min (P<0.01), suggesting that active biochemical process are involved in this phenomenon.
Medium and lysate were also assayed for T1AM catabolites. TA1 accumulated over time both in lysate and in medium, reaching a concentration of about 140 nM in both compartments after 240 min (due to lysate dilution, this corresponds to an estimated cellular concentration of 2.8–3.5 μM). T0AM was detected at very low concentration (<0.5 nM) (Fig. 1B), while TA0 was not revealed (data not shown). Overall catabolite production accounted for about 30% of T1AM administration.
Experiments were repeated in the presence of the amine oxidase inhibitor iproniazid, and in these conditions T1AM recovery averaged 99%, because after 240 min incubation medium and lysate concentrations averaged 700 and 300 nM, respectively, while TA1 production was almost abolished (Fig. 1C) and T0AM was not significantly modified (data not shown).
By comparison, similar experiments were performed by incubating HepG2 cells with T3 or T4. As observed with T1AM, lysate concentrations increased while medium concentrations decreased exponentially, until a steady state was obtained after about 120 min. At the steady state, medium concentrations averaged 651±73 and 413±91 nM for T3 and T4 respectively. The corresponding lysate concentrations were 135±59 and 343±137 nM respectively. The overall recovery was close to 80% for T3 and T4.
T1AM, T3, and T4 uptake were also investigated in perfused rat liver. After 40 min of perfusion with 50 nM or 1 μM T1AM, tissue concentration averaged 87.2±22.0 and 1013.8±312.5 pmol/g, respectively, vs a control value of 7.1±4.2 pmol/g. Among putative catabolites, only T0AM was detected in the liver homogenate at concentrations approximately one order of magnitude lower than T1AM. In this model, T1AM uptake was lower than T3 and T4 uptake, because after 40 min of perfusion with 50 nM T3 or T4 their tissue concentrations averaged 853.6 and 1531.2 pmol/g, respectively, vs control values of 3.9 and 15.5 pmol/g. If perfusate T3 and T4 concentrations were raised to 1 μM, tissue concentrations after 40 min of perfusion averaged 2.14 and 3.90 μmol/g respectively. In the latter experiments, the assay of T3 and T4 in the perfusion buffer confirmed that over 90% of infused hormones were taken up and stored in the tissue.
Notably, perfusion with T3 or T4 was not associated with significant increase in tissue T1AM, nor was perfusion with T1AM was not associated with any significant change in tissue T3 or T4 (data not shown).
Metabolite assays
In HepG2 cells cultured in glucose production buffer (Fig. 2A), incubation with 1 μM T1AM induced ∼25% increase in glucose production (P<0.01). In other experiments, the dose dependence of this effect was investigated (Fig. 2B): the estimated EC50 was 0.84 μM, but at 5 or 10 μM T1AM the stimulation of glucose production decreased, yielding a bell-shaped dose–response curve.
Glucose and ketone body (KB) production in HepG2 cell cultures that were incubated for 4 h in glucose production buffer as described in Materials and methods. Glucose or KB concentration was assayed in the medium. Results are expressed as mean±s.e.m. and are normalized to the total cell protein content determined in cell lysates. (A) Glucose concentration after incubation with 1 μM T1AM (n=9 per group). (B) A dose–response curve was obtained using 0.1, 0.5, 1, 5, 10 nmol/ml of T1AM, corresponding to 0.1, 0.5, 1, 5, and 10 μM concentrations (n=3 in each case). (C and D) KB production: (A and C) 1 μM T1AM administered to cells which underwent >30 in vitro passages (n=9), while in (B and D) a dose–response curve (n=3) was obtained in cells which underwent <20 in vitro passages. See text for further details. *P<0.05, **P<0.01 by ANOVA or unpaired t-test, as appropriate.
Citation: Journal of Endocrinology 221, 1; 10.1530/JOE-13-0311
More complex findings were obtained with regard to ketone body production. Baseline ketone body production and the response to T1AM apparently depended on the number of passages performed in vitro. Using cells which underwent >30 passages in vitro, treatment with 1 μM T1AM produced a slight (11%) but significant decrease in ketone body release (Fig. 2C), chiefly accounted for by decreased acetoacetate release (252.4±10.5 vs 286.8±7.2 nmol/mg protein, P<0.05), while no significant effect was detected using 0.1 or 10 μM T1AM (data not shown). However, when we took care to use cells that underwent no more than 20 passages, we observed that baseline ketone body production was higher (on the order of 630 nmol/mg proteins), and under these conditions T1AM produced a dose-dependent stimulation of ketone body release, with EC50 on the order of 1.12 μM (Fig. 2D).
In the presence of 100 μM iproniazid, the increase in glucose production induced by 1 μM T1AM was dampened (9%) and did not reach the threshold of statistical significance (5.99±0.17 vs 6.53±0.31 μg/mg protein in cell lysate, P=NS). We also tested the effect of TA1, the major T1AM catabolite, and, at 500 nM or 1 μM concentration, it did not affect either glucose or ketone body release. Notably, cellular uptake of TA1 was substantially lower than that observed with T1AM (15 vs 60%).
In the presence of exogenous glucose, T1AM did not produce any change in fatty acid production after 4 or 24 h of incubation (Fig. 3A and B), although at the later time point there was a slight increase that did not reach the threshold of statistical significance (P=0.088). Under conditions promoting steatosis, i.e. with supplementation of exogenous glucose and fatty acid mixture, after 24 h T1AM did not affect FFA concentration, while glucose concentration was increased (P<0.01), suggesting increased production and/or decreased consumption (Fig. 4A and B).
Free fatty acid (FFA) production in HepG2 cell cultures that were incubated for 4 h (A) or 24 h (B) in DMEM medium as described in Materials and methods. Results are expressed as mean±s.e.m. of nine biological replicates and are normalized to the total cell protein content determined in cell lysates. Differences between groups were not statistically significant (unpaired t-test).
Citation: Journal of Endocrinology 221, 1; 10.1530/JOE-13-0311
Glucose (A) and fatty acids (B) concentration in HepG2 cell cultures after incubation for 24 h in conditions promoting steatosis, i.e. DMEM medium containing 1 mM free fatty acids 1 mM and 1 g/l glucose. Results are expressed as mean±s.e.m. of nine biological replicates and are normalized to the total cell protein content determined in cell lysates. **P<0.01 by unpaired t-test.
Citation: Journal of Endocrinology 221, 1; 10.1530/JOE-13-0311
Liver perfusion with glucose production buffer in the presence of 1 μM T1AM (Fig. 5) also showed a significant increase in glucose production (Fig. 5A, P<0.05) that was associated with increased ketone body release (Fig. 5B, P<0.01), largely accounted for by increased 3-hydroxybutyrate release (246.0±21.04 vs 153.3±11.81 nmol/min per g, P<0.01).
Glucose (A) and ketone body (KB) (B) release by livers perfused with 1 μM T1AM in glucose production buffer. Glucose and KB concentrations were determined in the perfusate every 5 min for a total period of 40 min, and the average release rate was calculated, as described in Materials and methods. Results are expressed as mean±s.e.m. of the three experiments and are normalized to tissue wet weight. **P<0.01 by unpaired t-test.
Citation: Journal of Endocrinology 221, 1; 10.1530/JOE-13-0311
Infusion with Krebs–Ringer buffer did not produce any significant change in metabolite release (glucose: 0.24±0.17 vs 0.29±0.02 mg/min per g, P=NS; ketone bodies 40.1±4.9 vs 35.1±3.0 nmol/min per g, n=8–9). A few experiments (n=2) were also carried out in livers obtained from 4 h fasted rats, and a 26% increase in the release of ketone bodies was observed (231±22 vs 175±25 nmol/min per g), while glucose release was not affected (0.28±0.06 vs 0.28±0.05 mg/min per g).
Discussion
In this investigation, we observed that T1AM, such as T3 and T4, can be taken up and accumulated in hepatocytes. In HepG2 cell culture, these compounds were quickly absorbed because their presence was detected in cell lysate after a few minutes of infusion. At the steady state, estimated cellular concentration exceeded medium concentration by about six- to eightfold. These results are in agreement with previous observations performed in cardiomyocytes (Saba et al. 2010) and in FRTL-5 thyroid cells (Agretti et al. 2011), and they are consistent with the results obtained in perfused liver, where tissue concentration was higher than perfusate concentration. In this model, the ratio between tissue and perfusate concentration was lower than the lysate:medium ratio observed in cell cultures, but it should be considered that T1AM is likely to be taken up and catabolized by vascular and interstitial cells, and that it undergoes biliary excretion (Chiellini et al. 2012).
In HepG2 cells, T1AM recovery was on the order of 70%. Most of the balance was accounted for by oxidative deamination, yielding the thyroacetic derivative TA1 that was detectable in cell lysate and medium within a few minutes. Although D1 deiodinase has been reported to be expressed in HepG2 cells (Jakobs et al. 2002), T0AM production was minimal, probably because of the absence of D3 deiodinase and/or of essential cofactors for deiodination or decarboxylation, while it was greater in perfused liver, where it appears to be more relevant than oxidative deamination. We cannot, however, exclude that additional metabolic pathways may be active, particularly conjugation to sulfate and glucuronide (Hackenmueller & Scanlan 2012), because the corresponding derivatives could not be tested by the present method.
It should be stressed that about 50% of administered T1AM, T3, or T4, was detected in cell lysate after 2 h of infusion. As a consequence at the steady state the estimated cellular concentration of T3, T4, and T1AM was significantly higher than medium concentration. This suggests the existence of specific binding sites and/or transport pathways for T1AM and thyroid hormones.
The molecular mechanism by which T0AMs are taken up is still controversial. Eight potential transporters belonging to the solute carrier family have been tentatively identified (Ianculescu et al. 2009), although their affinity and sodium-independence did not seem to be consistent with the characteristics of T1AM uptake (Saba et al. 2010). Recently it has been reported that T1AM is largely bound to apolipoprotein B-100 (apo-B100), the protein component of LDLs (Roy et al. 2012). This could be a potential vehicle to transport T1AM and other T0AMs into cells which expressed the receptor for LDL, even though this is unlikely to contribute to our results, because in our experiments the standard growth medium supplemented with fetal bovine serum was replaced with other media, namely Krebs buffer or glucose production buffer, which do not contain lipoproteins. In any case, the kinetics of T1AM removal from the incubation medium was significantly lower at 4 °C, suggesting the presence of active transport mechanisms.
The primary aim of this work was to establish whether T1AM may directly stimulate hepatic glucose production. If adequate substrates for gluconeogenesis were provided, T1AM increased glucose production in two different experimental models, namely cultured HepG2 cells and in situ perfused rat liver. So consistent results were obtained in different experimental models, which also investigate different time scales, because the former was a subchronic and the latter an acute model. We conclude that modulation of hepatic metabolism may contribute to the hyperglycemic effect reported after administration of exogenous T1AM (Regard et al. 2007, Klieverik et al. 2009, Manni et al. 2012), which had previously been attributed only to modulation of insulin and/or glucagone secretion.
In HepG2 cells, the EC50 for the effect on gluconeogenesis was in the submicromolar range, and in perfused liver experiments a significant effect was observed with infusion of 1 μM T1AM, resulting in tissue concentration on the order of 1000 pmol/g. It is not easy to discuss the potential physiological relevance of these observations, as there is still some uncertainty about the physiological levels of T1AM. In humans, using an immunological assay (Hoefig et al. 2011), the plasma concentrations were reported on the order of 66 nM, while plasma T1AM concentration measured by a mass spectrometry-based assay was about 200-fold lower (Galli et al. 2012). As T1AM binds with high affinity to apo-B100 (Roy et al. 2012), it has been hypothesized that these different techniques estimate total and free T1AM respectively. However, T1AM is known to be concentrated in tissues, as discussed above, and we have previously reported that its liver content in rat in vivo was on the order of 90 pmol/g (Saba et al. 2010). If we rely on this value, then we may suggest that infusion of exogenous 1 μM T1AM raised liver concentration to a value that was one order of magnitude greater than the physiological concentration, after the depletion presumably occurring during prolonged perfusion with T1AM-free buffer (after over 60 min of perfusion ‘control’ values were close to 1 pmol/g). Further experiments will be necessary to clarify these important issues.
The effect of T1AM on gluconeogenesis may show a bell-shaped dose–response curve, because the response of HepG2 was reduced at the highest concentration tested (10 μM). This is not surprising, because other functional effects of T1AM, particularly on feeding and behavior, are also biphasic (Dhillo et al. 2009, Hettinger et al. 2010, Manni et al. 2012). Consistent with this interpretation, the response to 1 μM T1AM decreased in the presence of iproniazid, which inhibits its deamination to TA1, while we did not obtain any significant response after administration of exogenous TA1, whose cellular uptake was very low.
Another effect attributed to T1AM is the stimulation of lipid catabolism, with a shift from glucose to fatty acid as energy source (Braulke et al. 2008, Haviland et al. 2013). We observed a similar effect in the perfused liver model, because T1AM induced an increase in the release of ketone bodies. The results obtained in the HepG2 model were more complex, and in a subset of cells showing low values of baseline ketone body production T1AM reduced ketogenesis.
The different timing of these two experimental models might partly account for this discrepancy, because in the conscious mouse T1AM-induced stimulation of lipid catabolism was also time dependent (Haviland et al. 2013). Alternatively, it might be suggested that energy production by HepG2 cells, which are known to have limited ketogenic capability (Vilà-Brau et al. 2011), may be significantly dependent on pyruvate and amino acid catabolism, so that pyruvate shift toward gluconeogenesis would cause less acetyl-CoA to be available for ketogenesis. Notably, in HepG2 cells no significant effect was apparent on fatty acid production.
In conclusion, we observed that T1AM is actively accumulated in hepatocytes, where at concentrations in the micromolar range it is able to stimulate gluconeogenesis and to stimulate ketogenesis, provided that adequate energy substrates are available. These effects are independent from hormonal changes and might have physiological and pathophysiological importance. It would also been interesting to investigate whether T1AM may contribute to the metabolic effects usually attributed to thyroid hormone and/or to the metabolic abnormalities observed in hypothyroidism, a condition in which tissue and plasma T1AM levels have been reported to be decreased (Galli et al. 2012, Hackenmueller et al. 2012).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work was supported by a PRIN Italian research grant (Research Projects of National Interest) awarded to R Z in 2008.
Author contribution statement
S G and G C designed and carried out cell culture experiments and metabolic assays. S F performed the ex vivo experiments on liver. A S carried out mass spectrometry measurements. R Z designed and supervised the experimental work and wrote the manuscript.
References
Agretti P, De Marco G, Russo L, Saba A, Raffaelli A, Marchini M, Chiellini G, Grasso L, Pinchera A & Vitti P et al. 2011 3-Iodothyronamine metabolism and functional effects in FRTL5 thyroid cells. Journal of Molecular Endocrinology 47 23–32. (doi:10.1530/JME-10-0168)
Alberti KGMM & Hockaday TDR 1972 Rapid blood ketone body estimation in the diagnosis of diabetic ketoacidosis. BMJ 2 565–568. (doi:10.1136/bmj.2.5813.565)
Bradford MM 1976 A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein–dye binding. Analytical Biochemistry 72 248–254. (doi:10.1016/0003-2697(76)90527-3)
Braulke LJ, Klingenspor M, DeBarber A, Tobias SC, Grandy DK, Scanlan TS & Heldmaier G 2008 3-Iodothyronamine: a novel hormone controlling the balance between glucose and lipid utilisation. Journal of Comparative Physiology B 178 167–177. (doi:10.1007/s00360-007-0208-x)
Chiellini G, Frascarelli S, Ghelardoni S, Carnicelli V, Tobias SC, DeBarber A, Brogioni S, Ronca-Testoni S, Cerbai E & Grandy DK et al. 2007 Cardiac effects of 3-iodothyronamine: a new aminergic system modulating cardiac function. FASEB Journal 21 1597–1608. (doi:10.1096/fj.06-7474com)
Chiellini G, Erba P, Carnicelli V, Manfredi C, Frascarelli S, Ghelardoni S, Mariani G & Zucchi R 2012 Distribution of exogenous [125I]-3-iodothyronamine in mouse in vivo: relationship with trace amine-associated receptors. Journal of Endocrinology 213 223–230. (doi:10.1530/JOE-12-0055)
Cumero S, Fogolari F, Domenis R, Zucchi R, Mavelli I & Contessi S 2012 Mitochondrial F(0) F(1)-ATP synthase is a molecular target of 3-iodothyronamine, an endogenous metabolite of thyroid hormone. British Journal of Pharmacology 166 2331–2347. (doi:10.1111/j.1476-5381.2012.01958.x)
Dhillo WS, Bewick GA, White NE, Gardiner JV, Thompson EL, Bataveljic A, Murphy KG, Roy D, Patel NA & Scutt JN et al. 2009 The thyroid hormone derivative 3-iodothyronamine increases food intake in rodents. Diabetes, Obesity and Metabolism 11 251–260. (doi:10.1111/j.1463-1326.2008.00935.x)
Di Nunzio M, Valli V & Bordoni A 2011 Pro- and anti-oxidant effects of polyunsaturated fatty acid supplementation in HepG2 cells. Prostaglandins, Leukotrienes and Essential Fatty Acids 85 121–127. (doi:10.1016/j.plefa.2011.07.005)
Galli E, Marchini M, Saba A, Berti S, Tonacchera M, Vitti P, Scanlan TS, Iervasi G & Zucchi R 2012 Detection of 3-iodothyronamine in human patients: a preliminary study. Journal of Clinical Endocrinology and Metabolism 97 E69–E74. (doi:10.1210/jc.2011-1115)
Hackenmueller SA & Scanlan TS 2012 Identification and quantification of 3-iodothyronamine metabolites in mouse serum using liquid chromatography–tandem mass spectrometry. Journal of Chromatography 1256 89–97. (doi:10.1016/j.chroma.2012.07.052)
Hackenmueller SA, Marchini M, Saba A, Zucchi R & Scanlan TS 2012 Biosynthesis of 3-iodothyronamine (T1AM) is dependent on the sodium–iodide symporter and thyroperoxidase but does not involve extrathyroidal metabolism of T4. Endocrinology 153 5659–5667. (doi:10.1210/en.2012-1254)
Hart ME, Suchland KL, Miyakawa M, Bunzow JR, Grandy DK & Scanlan TS 2006 Trace amine-associated receptor agonists: synthesis and evaluation of thyronamines and related analogues. Medicinal Chemistry 49 1101–1112. (doi:10.1021/jm0505718)
Haviland JA, Reiland H, Butz DE, Tonelli M, Porter WP, Zucchi R, Scanlan TS, Chiellini G & Assadi-Porter FM 2013 NMR-based metabolomics and breath studies show lipid and protein catabolism during low dose chronic T1 AM treatment. Obesity 21 2538–2544. (doi:10.1002/oby.20391)
Hettinger BD, Schuff K, Marks D & Scanlan TS 2010 3-Iodothyronamine (T1AM) causes weight loss in mice via reduction in food consumption. 14th International Thyroid Congress, Paris, France. Abstract OC-141.
Hoefig CS, Köhrle J, Brabant G, Dixit K, Yap B, Strasburger CJ & Wu Z 2011 Evidence for extrathyroidal formation of 3-iodothyronamine in humans as provided by a novel monoclonal antibodybased chemiluminescent serum immunoassay. Journal of Clinical Endocrinology and Metabolism 96 1864–1872. (doi:10.1210/jc.2010-2680)
Ianculescu AG, Giacomini KM & Scanlan TS 2009 Identification and characterization of 3-iodothyronamine intracellular transport. Endocrinology 150 1991–1999. (doi:10.1210/en.2008-1339)
Jakobs TC, Mentrup B, Schmutzler C, Dreher I & Köhrle J 2002 Proinflammatory cytokines inhibit the expression and function of human type I 5′-deiodinase in HepG2 hepatocarcinoma cells. European Journal of Endocrinology 146 559–566. (doi:10.1530/eje.0.1460559)
Klieverik LP, Janssen SF, van Riel A, Foppen E, Bisschop PH, Serlie MJ, Boelen A, Ackermans MT, Sauerwein HP & Fliers E et al. 2009 Thyroid hormone modulates glucose production via a sympathetic pathway from the hypothalamic paraventricular nucleus to the liver. PNAS 106 5966–5971. (doi:10.1073/pnas.0805355106)
Liggett SB 2004 The two-timing thyroid. Nature Medicine 10 582–583. (doi:10.1038/nm0604-582)
Manni ME, De Siena G, Saba A, Marchini M, Dicembrini I, Bigagli E, Cinci L, Lodovici M, Chiellini G & Zucchi R et al. 2012 3-Iodothyronamine: a modulator of the hypothalamus–pancreas–thyroid axes in mice. British Journal of Pharmacology 166 650–658. (doi:10.1111/j.1476-5381.2011.01823.x)
Manni ME, De Siena G, Saba A, Marchini M, Landucci E, Gerace E, Zazzeri M, Musilli C, Pellegrini-Giampietro D & Matucci R et al. 2013 Pharmacological effects of 3-iodothyronamine (T1AM) in mice include facilitation of memory acquisition and retention and reduction of pain threshold. British Journal of Pharmacology 168 354–362. (doi:10.1111/j.1476-5381.2012.02137.x)
Mario EG, Leonardo ES, Bassoli BK, Cassolla P, Borba-Murad GR, Bazotte RB & de Souza HM 2009 Investigation of the acute effect of leptin on the inhibition of glycogen catabolism by insulin in rat liver perfused in situ. Pharmacological Reports 61 319–324. (doi:10.1016/S1734-1140(09)70038-6)
Miyakawa M & Scanlan TS 2006 Synthesis of [125I]-,[2H]-, and [3H]-labelled 3-iodothyronamine (T1AM). Synthetic Communications 36 891. (doi:10.1080/00397910500466074)
Piehl S, Hoefig CS, Scanlan TS & Köhrle J 2011 Thyronamines – past, present, and future. Endocrine Reviews 2 64–80. (doi:10.1210/er.2009-0040)
Regard JB, Kataoka H, Cano DA, Camerer E, Yin L, Zheng YW, Scanlan TS, Hebrok M & Coughlin SR 2007 Probing cell type-specific functions of Gi in vivo identifies GPCR regulators of insulin secretion. Journal of Clinical Investigation 117 4034–4043. (doi:10.1172/JCI32994)
Roy G, Placzek E & Scanlan TS 2012 ApoB-100-containing lipoproteins are major carriers of 3-iodothyronamine in circulation. Journal of Biological Chemistry 287 1790–1800. (doi:10.1074/jbc.M111.275552)
Saba A, Chiellini G, Frascarelli S, Marchini M, Ghelardoni S, Raffaelli A, Tonacchera M, Vitti P, Scanlan TS & Zucchi R 2010 Tissue distribution and cardiac metabolism of 3-iodothyronamine. Endocrinology 151 5063–5073. (doi:10.1210/en.2010-0491)
Scanlan TS, Suchland KL, Hart ME, Chiellini G, Huang Y, Kruzich PJ, Frascarelli S, Crossley DA, Bunzow JR & Ronca-Testoni S et al. 2004 3-Iodothyronamine is an endogenous and rapid-acting derivative of thyroid hormone. Nature Medicine 10 638–642. (doi:10.1038/nm1051)
Snead AN, Santos MS, Seal RP, Miyakawa M, Edwards RH & Scanlan TS 2007 Thyronamines inhibit plasma membrane and vesicular monoamine transport. ACS Chemical Biology 2 390–398. (doi:10.1021/cb700057b)
Venditti P, Napolitano G, Di Stefano L, Chiellini G, Zucchi R, Scanlan TS & Di Meo S 2011 Effects of the thyroid hormone derivatives 3-iodothyronamine and thyronamine on rat liver oxidative capacity. Molecular and Cellular Endocrinology 341 55–62. (doi:10.1016/j.mce.2011.05.013)
Vilà-Brau A, De Sousa-Coelho AL, Mayordomo C, Haro D & Marrero PF 2011 Human HMGCS2 regulates mitochondrial fatty acid oxidation and FGF21 expression in HepG2 cell line. Journal of Biological Chemistry 286 20423–20430. (doi:10.1074/jbc.M111.235044)
Weatherman RV 2007 A triple play for thyroid hormone. ACS Chemical Biology 2 377–379. (doi:10.1021/cb700104v)
Wood WJ, Geraci T, Nilsen A, DeBarber AE & Scanlan TS 2009 Iodothyronamines are oxidatively deaminated to iodothyroacetic acids in vivo. Chembiochem 10 361–365. (doi:10.1002/cbic.200800607)
Yao HR, Liu J, Plumeri D, Cao YB, He T, Lin L, Li Y, Jiang YY, Li J & Shang J 2011 Lipotoxicity in HepG2 cells triggered by free fatty acids. American Journal of Translational Research 3 284–291.
Yoon JC, Puigserver P, Chen G, Donovan J, Wu Z, Rhee J, Adelmant G, Stafford J, Kahn CR & Granner DK et al. 2001 Control of hepatic gluconeogenesis through the transcriptional coactivator PGC-1. Nature 413 131–138. (doi:10.1038/35093050)
Zucchi R, Chiellini G, Scanlan TS & Grandy DK 2006 Trace amine-associated receptors and their ligands. British Journal of Pharmacology 49 967–978. (doi:10.1038/sj.bjp.0706948)
