| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Institut für Experimentelle Endokrinologie und Endokrinologisches Forschungszentrum der Charité EnForCé (S.P., J.K.), Charité–Universitätsmedizin Berlin, Campus Virchowklinikum, D-13353 Berlin, Germany; Federal Institute for Risk Assessment (T.H., G.B.), D-12277 Berlin, Germany; Department of Physiology and Pharmacology (T.S.S.), Oregon Health and Science University, Portland, Oregon 97239; Department of Chemistry and Biochemistry–Organic Chemistry (R.S., B.K.), Free University of Berlin, D-14195 Berlin, Germany
Address all correspondence and requests for reprints to: Prof. Dr. Josef Köhrle, Institut für Experimentelle Endokrinologie und Endokrinologisches Forschungszentrum der Charité EnForCé, Charité–Universitätsmedizin Berlin, Augustenburger Platz 1, D-13353 Berlin, Germany. E-mail: josef.koehrle{at}charite.de.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) ARE a novel class of endogenous signaling compounds. In terms of structure, they differ from thyroid hormone L-thyroxine (T4) and deiodinated thyroid hormone derivatives only concerning the absence of the carboxylate group of the β-alanine side chain. Thyronamine nomenclature follows the rules applied for thyronines (TxAM with x indicating the number of iodine atoms per molecule). So far, only two representatives of thyronamines, namely 3-iodothyronamine (3-T1AM) and thyronamine (T0AM), have been detected in vivo in various species (1). Although their physiological roles still remain elusive, 3-T1AM and T0AM have exhibited short-term hypothermic, negative chronotropic, and negative inotropic effects that are opposite in direction to the actions of the classical active thyroid hormone T3 (1, 2, 3). Thus, thyronamines were suggested to be derivatives of thyroid hormones that might serve to fine-tune or even antagonize thyroid hormone effects.
|
So far, the three deiodinase isozymes (Dio1, Dio2, and Dio3) have been described to catalyze the sequential reductive removal of iodine from iodothyronines and various iodothyronine metabolites thus controlling the bioavailability of thyroid hormones (5). Although Dio1 exhibits both phenolic and tyrosyl ring deiodination activity (Fig. 1
), Dio2 and Dio3 are more specific with respect to the position of the iodine removed. Dio2 catalyzes only deiodinations of the phenolic ring, e.g. the conversion of the prohormone T4 to active T3, whereas Dio3 catalyzes only deiodinations of the tyrosyl ring, e.g. the conversion of T4 to inactive rT3 (5).
In this study, the complete panel of thyronamine deiodination reactions was investigated systematically. Because the various thyronamines differ only regarding the number or the position of the iodine atoms, their distinction by immunological methods has been hampered so far. Therefore, a novel liquid chromatography and tandem mass spectrometry (LC-MS/MS) method was developed for the simultaneous detection of all thyronamines in the same sample. The experiments revealed that all three deiodinases catalyze thyronamine deiodination reactions with each isozyme exhibiting a unique substrate specificity.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animals
Livers from two euthyroid, adult, male C57BL/6 wild-type mice were kindly provided by Dr. Ulrich Schweizer (Institute of Experimental Endocrinology, Charité, Berlin).
Cell culture
The human ECC-1 endometrium carcinoma cell line was kindly provided by Dr. Monique Kester (Department of Internal Medicine, Erasmus Medical Center, Rotterdam, The Netherlands). HepG2 and ECC-1 cells were propagated in DMEM-F12 medium (Invitrogen, Karlsruhe, Germany). HEK293 cells were cultured in DMEM (Biochrom, Berlin, Germany), and MSTO-211H cells were grown in RPMI 1640 medium (Invitrogen). All cell culture media were supplemented with 10% (vol/vol) fetal calf serum (Biochrom) and 100 nM sodium selenite for optimal expression of deiodinase selenoproteins (8). At confluence, HEK293 and HepG2 cells were directly harvested for deiodinase assays. By contrast, confluent MSTO-211H cells were serum starved overnight and then incubated with 10 µM forskolin and 700 nM MG-132 for 6 h to stimulate the endogenous Dio2 enzyme activity. Likewise, confluent and serum-starved ECC-1 cells were treated with 100 nM phorbol-12-myristate-13-acetate for 6 h to increase the endogenous Dio3 enzymatic activity.
LC-MS/MS detection of thyronamines
LC-MS/MS analyses were performed using an Agilent 1100 HPLC system (Agilent, Waldbronn, Germany) and an API 365 triple-quadrupole tandem mass spectrometer (Applied Biosystems, Darmstadt, Germany) equipped with TurboIonSpray interface. The detection was performed using positive electrospray ionization (ESI+) in the selected reaction monitoring (SRM) mode. Data processing was performed using Bio Analyst version 1.3.1 software (Applied Biosystems).
All compound-specific mass spectrometric working parameters were optimized by directly injecting thyronamine standard solutions (10 µg/ml) at a flow rate of 10 µl/min into the mass spectrometer and were summarized in Table 1. Because all thyronamines showed the loss of ammonia in their tandem mass spectra, the transition (M+H)+ 
(M+H-NH3)+ was used for their detection by MS/MS. Chromatographic separation of all thyronamines was achieved using a Synergi Polar-RP 80-Å column (150 x 2 mm; Phenomenex, Aschaffenburg, Germany) with a 0.3 ml/min gradient elution program (Fig. 2). The optimized elution parameters were mobile phase A (water-acetonitrile-acetic acid, 95:5:0.6) and mobile phase B (water-acetonitrile-acetic acid, 5:95:0.6): 0–1 min, 10% B; 1–15 min, 10–50% B; 15–20 min, 50% B; 20–24 min, 50–10% B; and 25–29 min, 10% B. The remaining mass spectrometric working parameters were source temperature, 300 C; dwell time, 90 msec; Q1 peak width, 0.7; and Q3 peak width, 0.7.
|
|
The limits of detection and lower limits of quantitation were calculated using a signal-to-noise ratio of at least 3:1 and 10:1, respectively. The limits of detection of the analyte concentrations in deiodinase reaction matrices were 100 pM for T0AM and mono- and diiodothyronamines and 1 nM for triiodothyronamines and T4AM. The lower limits of quantitation of the analyte concentrations in deiodinase reactions were 250 pM for T0AM and mono- and diiodothyronamines and 2.5 nM for triiodothyronamines and T4AM.
The assay stability was verified by determining the following parameters: intra- and interassay precision of retention times were less than 2.14 sec and less than 2.18 sec, respectively. Intra- and interassay precision analyte concentrations ranged from 5.0–8.3% and 5.6–11.0%, respectively. Thus, precision data were within the required limits of 15% coefficient of variation at all analyte concentrations studied (9, 10). Intra- and interassay bias of analyte concentration ranged from 3.3–10.3% and 6.1–12.4%. Thus, bias data were also within the acceptance range of ±15% of the nominal values at all analyte concentrations (9, 10).
LC-MS/MS-based deiodinase assays
Mouse liver membrane fractions were prepared as described before (11). To generate cell line-derived deiodinase preparations, confluent cells were washed twice with ice-cold 1x PBS at pH 7.4 and harvested by scraping into a homogenization buffer containing 250 mM sucrose, 20 mM HEPES, 1 mM EDTA, and 1 mM dithiothreitol (DTT). HepG2 and MSTO-211H cells were harvested into a homogenization buffer at pH 7.4, whereas ECC-1 cells were scraped into a homogenization buffer at pH 8.0 (Table 2). After sonication, the protein concentrations were determined using Bradford assay (12). Samples containing appropriate amounts of protein (Table 2) were prepared in the respective homogenization buffer and stored at –20 C until use.
|
). Thyronamines, rT3 (which served as a positive control substrate for Dio1 and Dio2), or T3 (which served as a positive control substrate for Dio3) were added to the deiodinase reactions at various final concentrations as listed in Table 2. Each substrate was assayed at all concentrations in at least three separate experiments. Within one such experiment, each substrate concentration was analyzed in three replicate reactions containing 1 mM 6n-propyl-2-thiouracil (PTU) and three replicate reactions devoid of PTU. As an exception to this, all reactions with Dio3-containing ECC-1 lysates were performed in the presence of 1 mM PTU to inhibit the weak Dio1 activity found in this cell line. The enzymatic reactions were stopped by adding 0.1 vol 100% (vol/vol) acetic acid. Subsequently, 4 pmol 3-T1AM-d4 was added to each vial to serve as internal standard. The reaction mixtures were incubated at 37 C for 60 min. Proteins were precipitated by adding 3 vol ice-cold acetone and incubating at –20 C for 15 min. After centrifugation at 14,000 x g and 4 C for 5 min, the supernatants were transferred into new Eppendorf tubes, acidified with 0.002 vol 30% (vol/vol) HCl, washed twice with 1 vol cyclohexane, and then subjected to three subsequent extraction cycles with 1.5 vol ethyl acetate. The organic layers were combined and evaporated to dryness at 45 C. The residues were redissolved in 30 µl H2O-methanol-acetic acid (90:10:1) and stored at –20 C until analysis of the deiodination products by LC-MS/MS.
The injection volumes were 10 µl with analytes dissolved in the mobile phase used for LC-MS/MS analysis.
The extraction efficiencies of thyronamines from deiodinase reaction matrices were routinely controlled in every run. Recovery rates were not significantly influenced by the type of cell line lysate, by the pH of the deiodinase reaction, or by PTU addition. Furthermore, in every experiment, calibration curves for each analyte were recorded taking the compound specific recovery rates into consideration.
Comparison of LC-MS/MS-based deiodinase assays and 125I– release assays
To validate the performance of the novel LC-MS/MS method in deiodinase assays, the apparent Km and Vmax values of selected iodothyronine deiodination reactions were determined by LC-MS/MS assays and compared with the values obtained from classical radioactive 125I– release assays. The following substrates were used: 50 nM to 4.4 µM rT3 for Dio1, 10–50 nM rT3 for Dio2, and 10–100 nM T3 for Dio3. All reactions were stopped after 30 min. The specific activities were expressed as moles 3,3'-T2 per milligram protein per minute and moles 125I– released per milligram protein per minute for LC-MS/MS and 125I– release experiments, respectively. The apparent Km and Vmax values were determined by fitting the enzyme kinetic data to the Michaelis-Menten equation by means of nonlinear regression using GraphPad Prism Software version 4.00 (GraphPad, San Diego, CA). The values reported for apparent Km and Vmax are from three separate experiments performed in triplicate and are presented as mean ± SD. For Dio1 and Dio2, the values obtained by LC-MS/MS-based deiodinase assays and 125I– release assays did not differ significantly as calculated by Wilcoxon test (data not shown). For Dio3 from ECC-1 lysates, the specific activities measured in LC-MS/MS-based deiodinase assays were in line with those obtained by a radiochemical, HPLC-based assay (13).
Apparent Km and Vmax of selected thyronamine deiodination reactions
The apparent Km and Vmax values of selected thyronamine deiodination reactions were measured by LC-MS/MS-based deiodinase assays using 10 nM–5 µM thyronamine substrate and an incubation time of 30 min. The specific activities were expressed as moles deiodinated product per milligram protein per minute. The values reported for apparent Km and Vmax were calculated as described above and are presented as mean ± SD from three separate experiments performed in triplicate.
Dio1 and Dio2 assays based on 125I-release (14)
Mouse liver membrane fractions and cell line-derived deiodinase preparations were generated as described for LC-MS/MS-based deiodinase assays (Table 2) except that different amounts of protein (micrograms per reaction) were used (HepG2, 50; mouse liver membrane fractions, 20; MSTO-211H, 150; ECC-1, 300; HEK293, 20–300).
Aliquots of cell line lysates and mouse liver membrane fractions were thawed on ice. Deiodination reactions were started by incubating the aliquots at 37 C in 1.5 vol potassium phosphate buffer at pH 6.8 containing 167 mM K2HPO4/KH2PO4, 1.67 µM Na2EDTA·H2O, 273 µM NaOH, 33 mM DTT, 500 cpm/µl [5'-125I]rT3, and various concentrations of unlabeled rT3 (Table. 2). Each sample was analyzed in three separate experiments. Within one such experiment, each sample was analyzed in three replicate reactions containing 1 mM PTU and three replicate reactions devoid of PTU.
Reactions were stopped by adding 0.5 vol 10% (wt/vol) BSA and 3 vol 10% (wt/vol) trichloroacetic acid. After centrifugation, the supernatants were analyzed for 125I– content using acetic acid-equilibrated Dowex 50W-X2 ion-exchange resins. The 125I– released in samples containing PTU was attributed to Dio2, whereas the difference in 125I– release between aliquots with and without PTU was used to calculate Dio1 activity.
Competition assays
125I– release assays were used to analyze the effect of thyronamines on the phenolic ring deiodination of rT3 by Dio1. Mouse liver membrane fractions were incubated for 15 min with 0.2–4.4 µM rT3 and 0.1–10.0 µM thyronamine at each rT3 concentration used.
The values reported for apparent Km and Vmax were determined as described above and are presented as mean ± SD from three separate experiments performed in triplicate. Linear regression lines from Eadie-Hofstee plots are used to display the effects of thyronamines on the Km (represented by slope) and Vmax (represented by y-intercept) of the phenolic ring deiodination of rT3 by Dio1 (see Fig. 4, left panel).
|
, right panel). The values reported for Ki are from three separate experiments performed in triplicate and are presented as mean ± SD.
Statistical analyses of competition assays
Statistical analyses were performed using GraphPad Prism Software version 4.00. The effect of thyronamines on the apparent Km and Vmax of the conversion of rT3 by Dio1 from mouse liver membrane fractions was analyzed using Friedman test followed by Dunn’s post test. P < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In HepG2 lysates, high specific Dio1 activity but no Dio2 and Dio3 activities were detected (Table 3). Accordingly, HepG2 lysates represented an isozyme-specific preparation of Dio1 and were used to study the ability of Dio1 to accept thyronamines as substrates.
|
). Thus, MSTO-211H lysates represented an isozyme-specific preparation of Dio2.
In ECC-1 lysates, Dio3 was detected at high specific activities (Table 3). Because also a minimal Dio1 activity was measured (specific activity of 23 ± 3 fmol 3,3'-T2/mg protein·min at a final rT3 concentration of 1 µM), 1 mM PTU was added to each reaction catalyzed by ECC-1 lysates to efficiently inhibit the Dio1 activity.
Lysates of HEK293 cells served as deiodinase-deficient negative control preparations because no deiodinase activity was measured (Table 3).
Dio1-containing HepG2 lysates deiodinate rT3AM and 3',5'-T2AM at the phenolic rings
In positive control experiments, HepG2 lysates were incubated as described in Table 2 using rT3 as a well-defined substrate of Dio1. rT3 was readily converted into its phenolic ring deiodination product 3,3'-T2 (Fig. 3A). This conversion was completely inhibited by 1 mM PTU. In negative control experiments using deiodinase-deficient HEK293 lysates or HepG2 lysates that had been heat-inactivated before, no deiodination of rT3 was observed (data not shown).
|
using various concentrations of thyronamines as substrates. Both rT3AM and 3',5'-T2AM were readily deiodinated at their phenolic rings yielding 3,3'-T2AM and 3'-T1AM, respectively, at all substrate concentrations tested (Fig. 3, B and C). Because both reactions were sensitive to PTU inhibition, they were unambiguously catalyzed by Dio1. No conversions of rT3AM and 3',5'-T2AM were observed in negative control experiments performed with HEK293 lysates or heat-inactivated HepG2 lysates (data not shown). All other thyronamines were not deiodinated by HepG2 cell lysates under these conditions and at the substrate concentrations tested.
Dio2-containing MSTO-211H lysates also deiodinate rT3AM and 3',5'-T2AM at the phenolic rings
In positive control experiments, rT3 was deiodinated at the phenolic ring by MSTO-211H lysates yielding 3,3'-T2 (data not shown). In contrast to HepG2 lysates, this deiodination was not sensitive to PTU, which is consistent with the known inhibition characteristics of Dio2. Of all thyronamines, only rT3AM and 3',5'-T2AM were converted by Dio2 from MSTO-211H lysates yielding 3,3'-T2AM and 3'-T1AM, respectively (Fig. 3, D and E).
Dio3-containing ECC-1 lysates deiodinate all thyronamines with iodine atoms in the tyrosyl ring
To inhibit the weak Dio1 enzymatic activity that had been found in ECC-1 cells, 1 mM PTU was added to all incubations. In positive control experiments, ECC-1 lysates were incubated with T3, which represents the preferred substrate of Dio3 (5). T3 was sequentially deiodinated into its tyrosyl ring deiodination products 3,3'-T2 and 3'-T1 (Fig. 3F). Because this reaction proceeded in the presence of PTU, it was clearly catalyzed by Dio3. All thyronamines with iodine atoms in the tyrosyl ring were deiodinated at their tyrosyl rings (Fig. 3, G–J, illustrating the respective deiodination reactions of T4AM, rT3AM, T3AM, and 3-T1AM). By contrast, those thyronamines without tyrosyl ring iodine, namely 3',5'-T2AM and 3'-T1AM, were not deiodinated (data not shown).
Apparent Km and Vmax of selected thyronamine deiodination reactions
To characterize the kinetic properties of thyronamine deiodinations, the apparent Km and Vmax of selected thyronamine deiodination reactions were measured and compared with that of the respective iodothyronine.
Accordingly, the phenolic ring deiodination of rT3AM by Dio1 from HepG2 and by Dio2 from MSTO-211H lysates was compared with that of rT3. With both deiodinase preparations, rT3AM was deiodinated at the phenolic ring at apparent Km values similar to those observed with rT3 but at lower Vmax values compared with rT3 (compare Table 3 with Table 4). Likewise, the tyrosyl ring deiodination of T3AM by Dio3 from ECC-1 lysates was characterized by a similar apparent Km but by a lower apparent Vmax when compared with T3 (compare Table 3 with Table 4).
|
Thyronamines identified as Dio1 substrates inhibit the Dio1-catalyzed 5'-deiodination of rT3
To substantiate the newly identified thyronamine deiodination reactions, we studied which thyronamines would interfere with an established iodothyronine deiodination reaction. Those thyronamines that had been identified as deiodinase substrates in the LC-MS/MS assays were expected to compete with a classical iodothyronine substrate in a deiodinase-mediated catalyzed reaction. By contrast, those thyronamines that were excluded as deiodinase substrates in the LC-MS/MS experiments were not expected to do so.
In the following studies, liver membrane fractions from male C57BL/6 wild-type mice were used as an exemplary deiodinase preparation for several reasons. First, high specific Dio1 activity but no Dio2 and Dio3 activities were detected in mouse liver membrane fractions using both classical 125I– release assays and LC-MS/MS-based deiodination assays (data not shown). Thus, mouse liver membrane fractions served as a rich and isozyme-specific source of Dio1. Second, mouse liver membrane fractions represented a Dio1 preparation of high physiological and ex vivo relevance. Finally, both phenolic and tyrosyl ring thyronamine deiodination reactions were identified for this enzyme preparation in preliminary LC-MS/MS studies. In line with HepG2 lysates, Dio1 from mouse liver membrane fractions catalyzed phenolic ring deiodinations of rT3AM and 3',5'-T2AM yielding 3,3'-T2AM and 3'-T1AM, respectively (data not shown). Furthermore, mouse liver membrane fractions deiodinated 3,3'-T2AM at the phenolic ring yielding 3-T1AM and catalyzed weak tyrosyl ring deiodinations of T3AM and 3,5-T2AM producing 3,3'-T2AM and 3-T1AM, respectively (data not shown). Because these reactions were all completely inhibited by the addition of 1 mM PTU, they were unambiguously catalyzed by Dio1.
To substantiate the identification of those thyronamine substrates of Dio1 in a functional assay, mouse liver membrane fractions were subjected to 125I– release assays using rT3 as substrate and increasing concentrations of the various thyronamines as inhibitors. rT3 was chosen because it represents the preferred substrate of Dio1 (5). In line with previous reports (17), rT3 was deiodinated at the phenolic ring by Dio1 from mouse liver membrane fractions following Michaelis-Menten kinetics with an apparent Km of 0.43 ± 0.023 µM and apparent Vmax of 142.7 ± 2.7 pmol 125I– released/mg protein·min (Table 5).
|
Both 3',5'-T2AM and 3,3'-T2AM led to a significant increase in Km as determined by Friedman test followed by Dunn’s post test (Fig. 4A for 3',5'-T2AM and Table 5). Moreover, 3',5'-T2AM and 3,3'-T2AM caused slight decreases in Vmax, which reached statistical significance. Therefore, 3',5'-T2AM and 3,3'-T2AM acted as mixed inhibitors, with competitive and noncompetitive elements. Dixon plots revealed Ki values of 2.6 ± 0.4 µM and 4.8 ± 0.1 µM for 3',5'-T2AM and 3,3'-T2AM, respectively (Table 5).
By contrast, rT3AM decreased the Vmax of rT3 conversion without significantly changing Km, indicating that rT3AM behaved as a noncompetitive inhibitor (Fig. 4B and Table 5). From the change in apparent Vmax for rT3, a Ki value of 1.3 ± 0.2 µM was calculated for rT3AM (Fig 4B and Table 5).
T3AM and 3,5-T2AM, which had been identified as weak tyrosyl ring substrates of Dio1 from mouse liver membrane fractions, also acted as noncompetitive inhibitors (Table 5). As expected from the LC-MS/MS studies, which had demonstrated weak tyrosyl ring deiodinations of T3AM and 3,5-T2AM, those substrates exhibited high Ki values of 34.0 ± 2.6 and 85.0 ± 7.7 µM, respectively (Table 5).
Thyronamines, which were not converted by Dio1 from mouse liver membrane fractions in the LC-MS/MS experiments, namely T4AM, 3'-T1AM, and 3-T1AM, failed to show any effect on the 5'-deiodination of rT3 by Dio1 up to a concentration of 10 µM (Table 5).
Taken together, the results obtained from the competition assays were consistent with the data obtained by the LC-MS/MS experiments and indicated selective deiodination reactions of the thyronamine substrates by Dio1.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
). This panel allows confining the biosynthetic pathways for 3-T1AM and T0AM. It suggests a synthesis of endogenous T0AM from 3-T1AM via tyrosyl ring deiodination by Dio3 rather than from 3'-T1AM via phenolic ring deiodination. By contrast, endogenous 3-T1AM is equally likely to be synthesized from 3,3'-T2AM via phenolic ring deiodination and from 3,5-T2AM via tyrosyl ring deiodination, provided that 3,5-T2AM or 3,3'-T2AM occurs in vivo. So far, only 3-T1AM and T0AM have been detected endogenously (1). The detection of thyronamines with higher iodine content in vivo might have been hampered by the fact that their mass spectrometric detection is technically more difficult. Alternatively or additionally, their levels in vivo could be very low due to putative rapid deiodination to 3-T1AM and T0AM.
The identified thyronamine substrates were functionally verified by inhibiting the Dio1-catalyzed 5'-deiodination of rT3. In these studies, mouse liver membrane fractions were chosen as an exemplary deiodinase source because both phenolic and tyrosyl ring thyronamine substrates had been identified for this enzyme preparation in LC-MS/MS experiments. The phenolic ring substrates rT3AM, 3',5'-T2AM and 3,3'-T2AM were potent inhibitors exhibiting Ki values in the lower micromolar range, i.e. comparable to the Km value of the phenolic ring deiodination of rT3. By contrast, the Ki values of the tyrosyl ring substrates were roughly one order of magnitude higher. These data were consistent with the LC-MS/MS experiments in which the phenolic and tyrosyl ring substrates were converted readily and weakly, respectively. So far, it is unclear why only some thyronamine substrates acted as mixed inhibitors with predominantly competitive elements (3',5'-T2AM and 3,3'-T2AM), whereas other thyronamines unexpectedly acted as noncompetitive inhibitors (rT3AM, T3AM, and 3,5-T2AM; Table 5). Because up to now, the molecular conformation of only one thyronamine, namely T3AM, has been resolved (21), no structure-function relationships for deiodinases and thyronamines (comparable to those of deiodinases and thyronines) are available to clarify this discrepancy. However, it can be excluded that the modes of inhibition exhibit any relationship solely to the iodine substitution pattern of the thyronamine substrates or to the position of the iodine removed.
In our study, we used two different Dio1 preparations, namely HepG2 lysates and mouse liver membrane fractions. Interestingly, their substrate specificities did not overlap completely. Although rT3AM and 3',5'-T2AM were deiodinated at the phenolic rings by both preparations, 3,3'-T2AM, T3AM, and 3,5-T2AM were exclusively deiodinated by mouse liver membrane fractions. This difference persisted even when the pH of the homogenization buffer and the potassium phosphate buffer were increased to 8.0 (refer to Table 2, data not shown). Therefore, our data might suggest that species-specific factors modulate the Dio1 substrate specificity toward thyronamines.
Besides supporting a role for deiodinases in thyronamine biosynthesis, our data also provide new insights into the structural requirements for deiodinase substrates in general. Deiodinases have long been known to convert not only iodothyronines but also sulfated iodothyronines, e.g. T4S and T3S (22), iodothyronine glucuronic acid esters, e.g. T4G (23), and iodothyronine acetic acid analogs, e.g. tetrac and triac (4) as well as N-acetylated iodothyronine derivatives (24, 25). In comparison with thyronines, these metabolites either carry additional negative charges or exhibit a negatively charged side chain. The identification of thyronamines as deiodinase substrates shows that deiodinases can indeed convert substrates with a positively charged side chain.
Our analyses were based on a newly established, robust, and sensitive analytical method applying LC-MS/MS with selected reaction monitoring for thyronamine detection. This method has several advantages over classical iodide release assays. It allows analyzing deiodinations of substrates without radioactive labeling. Product formation and substrate disappearance are monitored simultaneously and are both quantifiable. Moreover, sequential monodeiodinations are detectable. In contrast to RIAs, LC-MS/MS experiments are devoid of antibody cross-reactions. Compared with radioactive assays, this LC-MS/MS method is of course less sensitive and more time consuming but yields a more complete picture of deiodination cascades. Furthermore, we are not aware of any other HPLC method achieving complete baseline separation of all except one constitutional thyronamine and thyronine isomer. Thus, the LC-MS/MS method allows for backtracking the position of each iodine atom removed from the substrate, which to our knowledge has not been achieved by any other method so far.
Interestingly, sequential monodeiodination cascades were observed only in the case of the tyrosyl ring deiodinations of T3 and T3AM by Dio3-containing ECC-1 lysates (Fig. 3, F and I). In all other experiments, one-step deiodinations were monitored even if the product proved to be readily convertible by the respective deiodinase when used as a direct substrate. Considering that the sensitivity of the LC-MS/MS method used here was inversely proportional to the degree of iodination of thyronamines, the predominant absence of sequential monodeiodinations indicates a substrate preference of thyronamines with higher iodine content over those with lower iodine content for deiodinases.
In most instances, the thyronamine-deiodinating reactions are identical to the known deiodination reactions for the corresponding thyronines. For instance, in our study, rT3AM was readily converted by all three deiodinase isozymes, which corresponds to the deiodination reactions reported for rT3 (5).
Due to the recent discovery of thyronamine sulfation by hepatic, cardiac, and brain sulfotransferases (SULTs), the spectrum of deiodinase substrates could become even broader (26). T0AM and 3-T1AM, which occur in vivo, were found to be readily sulfated by human liver SULT1A3. Moreover, 3-T1AM was sulfated by homogenates of human brain and cardiac tissue, i.e. target tissues of thyronamine action. These SULT actions might serve to attenuate and thus regulate thyronamine action. It remains to be tested whether T0AMS, 3-T1AMS, and T3AMS occur in vivo, whether they can also be deiodinated and, if so, whether thyronamine sulfation has any effect on deiodination efficiency. According to literature data on iodothyronine sulfation, Dio1 would be the most likely candidate to catalyze deiodinations of sulfated thyronamines because deiodinations of iodothyronine sulfoconjugates by Dio2 and Dio3 are very limited (4).
In summary, we present here the systematic identification of thyronamines as efficient substrates for deiodinases by a novel LC-MS/MS-based method. Each deiodinase isozyme exhibited a unique substrate specificity toward thyronamines. The newly identified thyronamine substrates were functionally verified by inhibiting the 5'-deiodination of rT3 in a Dio1-catalyzed reaction in classical 125I– release assays. These data support a role for deiodinases in thyronamine biosynthesis. Moreover, by excluding some thyronamine deiodination reactions, the biosynthetic pathways of 3-T1AM and T0AM were confined.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: The authors of this manuscript have nothing to disclose.
First Published Online March 13, 2008
Abbreviations: Dio1, Deiodinase 1; DTT, dithiothreitol; LC-MS/MS, liquid chromatography and tandem mass spectrometry; PTU, 6n-propyl-2-thiouracil; SULT, sulfotransferase.
Received December 4, 2007.
Accepted for publication March 4, 2008.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. P Klieverik, E. Foppen, M. T Ackermans, M. J Serlie, H. P Sauerwein, T. S Scanlan, D. K Grandy, E. Fliers, and A. Kalsbeek Central effects of thyronamines on glucose metabolism in rats J. Endocrinol., June 1, 2009; 201(3): 377 - 386. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. S. Scanlan 3-Iodothyronamine (T1AM): A New Player on the Thyroid Endocrine Team? Endocrinology, March 1, 2009; 150(3): 1108 - 1111. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Gereben, A. M. Zavacki, S. Ribich, B. W. Kim, S. A. Huang, W. S. Simonides, A. Zeold, and A. C. Bianco Cellular and Molecular Basis of Deiodinase-Regulated Thyroid Hormone Signaling Endocr. Rev., December 1, 2008; 29(7): 898 - 938. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
), whereas the various thyronamine concentrations are presented as follows: 
, 0.1 µM; 
, 0.5 µM; 
, 1.0 µM; 
, 2.0 µM; 
, 3.0 µM 
, 5.0 µM; x, 7.0 µM; 
, 10.0 µM. The effect of thyronamines on the Km and Vmax of the phenolic ring deiodination of rT3 by Dio1 was analyzed using Friedman test followed by Dunn’s post test, and P < 0.05 was considered significant. In Dixon plots (right panels), the various rT3 concentrations are presented as follows: 
