Role of Sulfated Tyrosines of Thyroglobulin in Thyroid Hormonosynthesis
http://www.100md.com
《内分泌学杂志》
Institut National de la Santé et de la Recherche Médicale Unité 555 (N.V., O.C.), Faculté de Médecine, Université de la Méditerranée, 13385 Marseille, France
Northwestern University (M.-C.N.), Pulmonary Division & Critical Care Medicine, Chicago, Illinois 60611
The Scripps Research Institute (D.M.C.), Department of Molecular and Experimental Medicine, La Jolla, California 92037
Abstract
Our previous studies showed that sulfated tyrosines (Tyr-S) are involved in thyroid hormone synthesis and that Tyr5, the main hormonogenic site of thyroglobulin (Tg), is sulfated. In the present paper, we studied the role of Tyr-S in the formation and activity of the peroxidase-Tg complex. Results show that noniodinated 35SO3-Tg specifically binds (Kd = 1.758 μM) to immobilized lactoperoxidase (LPO) via Tyr-S linkage by using saturation binding and competition experiments. We found that NIFEY-S, a 15-amino acid peptide corresponding to the NH2-end sequence of Tg and containing the hormonogenic acceptor Tyr5-S, was a better competitor than cholecystokinin and Tyr-S. 35SO3-Tg, iodinated without peroxidase, bound to LPO with a Kd (1.668 μM) similar to that of noniodinated Tg, suggesting that 1) its binding occurs via Tyr-S linkage and 2) Tyr-S requires peroxidase to be iodinated, whereas nonsulfated Tyr does not. Iodination of NIFEY-S with [125I]iodide showed that Tyr5-S iodination increased with LPO concentration, whereas iodination of a nonsulfated peptide containing the donor Tyr130 was barely dependent on LPO concentration. Enzymatic hydrolysis of iodinated Tg or NIFEY-S showed that the amounts of sulfated iodotyrosines also depended on LPO amount. Sulfated iodotyrosines were detectable in the enzyme-substrate complex, suggesting they have a short life before the coupling reaction occurs. Our data suggest that after Tyr-S binding to peroxidase where it is iodinated, the sulfate group is removed, releasing an iodophenoxy anion available for coupling with an iodotyrosine donor.
Introduction
THE MAIN PARTNERS in thyroid hormone synthesis are: thyroglobulin (Tg), thyroperoxidase (TPO), the H2O2 generating system, and iodide. TSH, the pituitary hormone regulating thyroid function, is essential to control the level of expression of all the partners (1, 2, 3, 4), including the NaI symporter (5) and the apical iodide transporter(s) that transport iodide into the colloid (3, 6, 7). Thyroid hormone synthesis occurs only when, on one hand, Tg, iodide (7), and H2O2 (8) are secreted into the colloid and when, on the other hand, TPO is simultaneously present on the apical membrane of thyrocytes (1). In brief, the activation of TPO by H2O2 allows the oxidation of iodide and then the iodination of some tyrosines of Tg, leading to the formation of monoiodotyrosine (MIT) and diiodotyrosine (DIT). Several mechanisms have been proposed for the iodination reaction, including the various forms of a same iodide oxidation state that allow a two-electron reaction: the iodinium ion I+, the enzyme iodide complex [EOI]–, and the free hypoiodous acid (HOI) (for review see Refs.3 and 9). Then the coupling reaction between two iodotyrosines leads to the formation of thyroid hormones: T3 (DIT + MIT) and T4 (DIT + DIT). The catalytic action of TPO in the coupling reaction (10) involves the formation of a quinol ether intermediate containing two DIT. Removal of this intermediate from the peptide chain induces the formation of a dehydroalanine residue (11) on the one hand, and the formation of T4 linked to peptide chain on the other hand. Taurog et al. (12) showed that TPO catalyzed simultaneously the reactions of tyrosine iodination and coupling. Among the five hormonogenic sites identified in Tg molecule, tyrosine 5 (Tyr5) is the preferential T4 forming site (4) with tyrosine 130 (Tyr130) as donor site (13, 14).
The tyrosyl residues in Tg have different levels of reactivity to iodine. Iodination, controlled by the native structure of Tg, proceeds in a sequential order (15, 16, 17). At the hormonogenic site, the DIT acceptor and the DIT donor involved in the coupling reaction have different characteristics, such as their pKa (18). The peptide sequence surrounding each hormonogenic acceptor tyrosine in Tg (4) is similar to the known consensus sequence for tyrosine sulfation (19, 20, 21). Moreover, we previously showed that thyroid hormone synthesis was dependent on tyrosine sulfation (22), and more recently we showed that the major hormonogenic tyrosine (Tyr5) was sulfated (23). We hypothesize that sulfation could give the specific character for a tyrosine to become the DIT acceptor. Little information exists on the molecular interactions between Tg and TPO and their consequences in iodination and coupling reactions. A specific site for tyrosine binding has been described near the heme pocket on lactoperoxidase (LPO) (24, 25). TPO contains at least one specific active site for tyrosine that is independent of the iodide-binding site (26).
In this study, we propose to evaluate the role of sulfated tyrosines (Tyr-S) in the formation of the peroxidase-Tg complex as well as in the iodination of tyrosines and coupling reactions. Tyr-S are specifically involved in protein or peptide binding (27, 28), as in hormone-receptor interaction (29, 30), inflammatory leukocyte adhesion (31), hemostasis (31, 32), and chemokine signaling (31, 33). Given that Tyr5-S found in Tg (23) becomes the DIT acceptor involved in hormonosynthesis, we focused on the following questions. Does 35SO3-Tg specifically bind to peroxidase via Tyr-S linkage What is the involvement of the sulfate group in the iodination of tyrosine acceptor and in the coupling reaction
Our results show that 35SO3-Tg binds specifically to peroxidase through Tyr-S linkage and that the iodination level of sulfated Tyr5 depends on the peroxidase concentration. Moreover, when peroxidase concentration is increased, more sulfated iodotyrosine intermediates are found in the enzyme-substrate complex. Thus we propose that iodinated Tyr-S could remain bound to peroxidase until the structural conditions of coupling are reached in the peroxidase-Tg complex. The sulfate group of the DIT acceptor might then be released, leaving the phenoxy anion free for association with the DIT donor.
Materials and Methods
Cell culture and preparation of cellular porcine Tg
Thyroid epithelial cells were isolated from porcine glands, and primary cultures were made on porous filters coated with collagen (34) in the presence of TSH added from d 6 and without iodide (35). Cells were cultured without iodide to obtain noniodinated Tg. Cell culture conditions and [35S]sulfate-labeling were previously described (23). On d 18, apical media, in which Tg was essentially secreted, were collected and kept at –20 C until Tg purification. Nonlabeled Tg and [35S]sulfate-labeled Tg (35SO3-Tg) were purified from the apical media as described (23). Protein amount was evaluated either by Micro BCA protein assay (Pierce, Rockford, IL) or by measurement of absorbance at 280 nm (E1%1 cm = 10).
Preparation of Tyr-S and sulfated peptides
L-Tyrosine (Sigma Aldrich, St. Louis, MO) was sulfated according to the method of Reitz et al. (36). A peptide containing the NH2-end sequence of Tg, NIFEYQVDAQPLRP, named NIFEY, was synthesized (Sigma Genosys, Haverhill, UK). The Tyr5 on this peptide was sulfated according to Futaki et al. (37) with some modifications. In brief, NIFEY (15 μmol) was incubated with 1.5 mmol sulfur trioxide dimethylformamide complex (DMF-SO3; Fluka Chemie, Buchs, Switzerland) in a 1-ml dimethylformamide (Fluka)-pyridine (Prolabo; Merck Eurolab, Briare le Canal, France) mixture (4:1), for 24 h at room temperature. The sulfated peptide (NIFEY-S) was then purified on a Sephadex G-10 column (Amersham Biosciences, Orsay, France) equilibrated with 0.05 M ammonium bicarbonate (pH 8.2) and identified by the measure of absorbance at 262 and 280 nm. The ratio of A280 to A262 characteristic of each Tyr-S (32) was 0.5 for NIFEY-S and 0.75 for Tyr-S. Tyr-S and NIFEY-S were separated from the nonsulfated Tyr or NIFEY by cation-exchange chromatography (see below) on Dowex 50WX2 (H+). Cholecystokinin (CCK8, DYMGWMDF; Sigma Genosys) was used as nonsulfated peptide, and CCK8-S (Bachem Biochemica GmbH, Heidelberg, Germany) was also processed as described for NIFEY-S. The presence of a sulfate residue on the phenol function of tyrosine was characterized by nuclear magnetic resonance (NMR).
NMR analyses
NMR was performed by the Spectropole (Universités d’Aix-Marseille, Marseille, France). NMR spectra were recorded in D2O solutions at 300 K using a Bruker Avance DRX 500 spectrometer equipped with a Bruker CryoPlatform and a 5-mm cryo TXI probe. The temperature of the probe and preamplifier was 30 K. Chemical shifts were referenced to D2O: H = 4.79 ppm. 1H spectra were obtained with a standard zg sequence. Correlation spectroscopy spectra were obtained using gradient pulses for selection with a cosygp sequence.
Preparation of immobilized LPO
LPO (Sigma) was bound to Immobilon-P membrane by filtration in a 96-well multiscreen-IP opaque plate (Millipore Corporation, Bedford, MA) following the manufacturer’s instructions. Each well was treated 5 min with 100 μl 70% ethyl alcohol, filtered, and washed with 100 μl of ultrapure water, followed by 200 μl of 0.01 M sodium phosphate buffer (pH 7.4) containing 0.14 M sodium chloride (PBS). LPO (25 μg/100 μl PBS/well) was then filtered and washed by 200 μl PBS. Membranes were used immediately for binding assays. BSA (Sigma) used as negative control was also bound to membranes in the same conditions.
Tg binding assays
Saturation binding experiments.
Increasing concentrations of 35SO3-Tg or iodinated 35SO3-Tg were dissolved in 0.05 M phosphate buffer (pH 7.2), and 100 μl of each solution was added to an Immobilon-bound LPO- or BSA-containing well and incubated for 45 min at 37 C with gentle stirring. Media were then harvested, and membranes were washed three times with the same buffer. Then the 35SO3-radioactivity was counted in 200 μl/well of liquid scintillation cocktail Microscint 40 (Packard Biosciences, BV, Groningen, The Netherlands) using a TopCount NXT (Packard Instrument Co., Meriden, CT). Competition experiments were performed in the same conditions, but 35SO3-Tg (1.5 μM) was incubated with increasing concentrations of each compound individually: Tyr-S, CCK8-S, NIFEY-S, or Tyr, CCK8, NIFEY. Results and curves were given using GraphPad Prism software.
In vitro hormonosynthesis.
In vitro iodination and coupling experiments [40 atoms of iodine per mole of Tg, (38)] were performed with 35SO3-Tg or unlabeled Tg secreted by cells cultured without iodide. Tg (1.5 nmol) was incubated in 1 ml of 0.05 M sodium phosphate buffer, pH 7.4, containing 1.5 mg glucose, 100 nmol potassium iodide (KI), and 5 μg LPO. Carrier-free Na125I (0.25 μCi/ml, 0.5 μM; NEN Life Science Products, Zaventem, Belgium) was added to assays with unlabeled Tg, and the reaction was started by the addition of 2.5 μg glucose oxidase (Sigma) (38). The relative concentrations of these reagents were named 1KIGO-1LPO. Hormonosynthesis was carried out at 37 C for 30 min. The peptides NIFEY-S, NIFEY, and PY130 (AEGMEVYGTRQLGRP, a Tg-like peptide containing the hormonogenic Tyr130; Sigma Genosys) were iodinated (560 nmol/ml) in the same way but with variations in the concentrations of LPO, KI, and glucose-glucose oxidase (GGO) system, as indicated in Table 1. Hormonosynthesis was also performed with Immobilon-bound LPO in a 96-well multiscreen-IP opaque plate in the conditions of binding assays. The medium (100 μl), containing either Tg or Tg-peptides with [125I]iodide and the GGO system (1KIGO), was incubated 45 min at 37 C. Media were then harvested and frozen for further analysis. Membranes were washed three times with the same buffer and then cut, and [125I]iodide radioactivity was determined in a -counter (Cobra II; Packard Instrument Co.).
Gel electrophoresis
Aliquots of iodinated 35SO3-Tg or 125I-labeled Tg were analyzed by 5% SDS-PAGE (35). Gels were dried and then exposed to an imaging plate radioactive energy sensor (BAS-IP.MP 2040S; Fuji Photo Film Co., Ltd., Kanagawa, Japan) before scanning (Fujix BAS1000 IP reader; Fuji).
Enzymatic digestion after hormonosynthesis
125I-labeled Tg, iodinated 35SO3-Tg, and LPO-bound Tg or peptides on membranes were dissolved in a 0.05 M sodium phosphate buffer (pH 7.4), and the samples were digested for 48 h at 37 C by 20% (wt/wt) pronase (Roche Molecular Biochemicals, Mannheim, Germany). Samples were then subjected to digestion by 20% (wt/wt) leucine aminopeptidase, (cytosol type V from porcine kidney; Sigma) for an additional 48 h at 37 C. The iodoamino acids were then separated by cation-exchange chromatography or reverse-phase high performance liquid chromatography (RP-HPLC).
Cation-exchange chromatography
Dowex 50WX2 (H+ form) chromatography was performed as described (35, 39). The nonretained material, in 0.1 M formic acid (FA) fraction and in 0.1 M FA subsequent rinses, contained the majority (85–95%) of the [35S]sulfate radioactivity applied corresponding to sulfated amino acids, inorganic sulfate, sulfated carbohydrates (35), as well as free [125I]iodide. The column was then eluted (3 x 1 vol Dowex) successively with ethyl alcohol 25% and ethyl alcohol-ammonium 1N (EN) containing hydrophobic iodinated compounds. Fractions were collected, frozen, dried, and then analyzed by thin layer chromatography (TLC).
RP-HPLC
Separation by RP-HPLC of the different 125I-labeled or 35S-labeled samples was performed on silica C18 column (Uptisphere 5 μm ODSB UP50DB 10 M Interchim; Asnières, France). The column was eluted at a 0.5 ml/min flow rate (one fraction per minute) using a linear 60-min gradient of 0–60% acetonitrile in 20 mM ammonium acetate (pH 4). With this elution gradient, sulfated compounds were not retained and were recovered in the first minutes of elution, whereas the iodotyrosines (MIT, DIT, T3, T4) were retained and successively eluted. Fractions were collected, frozen, dried, and then analyzed by TLC.
Thin layer chromatography
Enzymatic hydrolysates or fractions from enzymatic hydrolysates purified on Dowex 50WX2 or on RP-HPLC were applied to silica-coated glass plates (10 x 10; Merck, Darmstadt, Germany) and chromatography was developed by the solvent system: N-butyl alcohol (Fluka Chemie)-glacial acetic acid (Carlo Erba Reagenti, Milan, Italy)-water (65:20:15) for 90 min. Labeled components were revealed with an imaging plate radioactive energy sensor.
Preparation of sulfated iodoamino acids
L-Tyrosine (Sigma) and L-tyrosine-sulfate were submitted to in vitro iodination as above with carrier-free Na125I (0.25 μCi/ml). The resulting 125I-labeled iodotyrosines were purified by cation-exchange chromatography (see above) and compounds of the two major elution fractions, 0.1 M FA and EN, were analyzed either directly on TLC or after separation by RP-HPLC (data not shown). Comparatively, 35SO3-Tyr isolated from 35SO3-Tg after alkaline hydrolysis (35) was iodinated as above, and the sulfated iodotyrosine derivatives were separated on RP-HPLC and analyzed by TLC. A spot at retention factor (Rf) of 0.44 was observed in the FA fraction of 125I-Tyr-S (Fig. 1A, lane 1), but not in the corresponding fraction of nonsulfated 125I-Tyr (Fig. 1A, lane 2). This compound was also present in the first hydrophilic fractions of RP-HPLC of 125I-Tyr-S (Fig. 1B, lanes 1 and 2) and of iodinated 35SO3-Tyr (Fig. 1C, lane 2). Unfortunately we did not obtain sufficient amounts to characterize this sulfated iodotyrosine (SIT) by NMR.
Results
Characterization of Tyr-S and NIFEY-S
In general, O-sulfation (ester bond) represents the dominant sulfation reaction (28) but SO3– is sometimes bound to the aromatic ring of tyrosine (sulfonation). NMR served to characterize the two types of bond. Before sulfation, Tyr showed two signals as doublets (6.77 and 7.07 ppm) corresponding to the 2 x 2 equivalent protons of the aromatic ring (Fig. 2A), whereas after sulfation (Fig. 2B), we observed a down-field shift of these two signals (7.18 and 7.23 ppm). Sulfonation (7.47 and 6.89 ppm determined by correlation spectroscopy experiment, Fig. 2B) was only weakly present (17%). Before sulfation (Fig. 2C), NIFEY showed four signals: three doublets (6.7, 7.0, 7.11 ppm) and one multiplet (7.19 ppm). The first two signals likely corresponded to the aromatic protons of Tyr, whereas the two other signals were due to phenylalanine, the other aromatic amino acid in NIFEY. After sulfation (Fig. 2D), the first two signals merged into a multiplet (7.1–7.22 ppm) with the other aromatic protons. Thus in each case, NMR showed that SO42– and the hydroxyl of tyrosine are linked by an ester bond.
Specific binding of noniodinated 35SO3-Tg to peroxidase
Increasing concentrations of noniodinated 35SO3-Tg were incubated with LPO or BSA linked to Immobilon-P membrane. After washings, the remaining radioactivity bound to LPO or BSA was counted. For each 35SO3-Tg concentration, the radioactivity associated to BSA (nonspecific binding) was subtracted from the radioactivity associated to LPO (total binding), giving a nearly hyperbolic curve corresponding to the specific dose-dependent binding of 35SO3-Tg to LPO (Fig. 3). The Scatchard representation gave the amount of 35SO3-Tg corresponding to the saturation binding: max = 768 ± 138 pmol Tg per milligram of LPO as well as the 35SO3-Tg equilibrium dissociation constant: Kd = 1.758 x 10–6 M.
Competition experiments by Tyr-S or Tyr-S-containing peptides (NIFEY-S, CCK8-S) were performed. Increasing amounts of each competitor were added to 35SO3-Tg solution (1.5 μM) and incubated with immobilized LPO or BSA. Fig. 4A displays competition characteristics. The concentration of competitor that shifted 50% of the specific 35SO3-Tg bound (IC50) was determined, and its equilibrium dissociation constant (Ki) was calculated using the Cheng and Prusoff equation (40): Ki = IC50/1+([L]/Kd), ([L]= concentration of free Tg). NIFEY-S was the best competitor (Ki = 1.95 x 10–4 M), followed by CCK8-S (Ki = 8.14 x 10–4 M) and Tyr-S (Ki = 19.1 x 10–4 M). Thus the peptide with a Tg-like sequence (NIFEY-S) was the best candidate to compete with 35SO3-Tg. When we used nonsulfated competitors, the shifting was always lower (Fig. 4B), suggesting that the two peptides and tyrosine also compete with Tg binding to LPO but with a lower efficacy, most likely for a specific Tyr site. Note that it was not possible to use higher concentrations of these compounds because of solubilization problems. All together, these data suggest that Tg was bound to peroxidase via Tyr-S linkage, and via Tyr linkage but more weakly.
Specific binding of iodinated 35SO3-Tg to LPO
Among the experiments that we performed to study the iodination of Tyr-S, we have obtained a Tg iodinated only by the GGO system, suggesting Tyr-S might be iodinated before binding to peroxidase. To determine whether this is true, 35SO3-Tg was iodinated in vitro with KI in conditions in which no coupling reaction could occur, i.e. under only H2O2 oxidation without LPO (see further in Fig. 7). The iodinated 35SO3-Tg was then incubated with immobilized LPO or BSA in conditions described above. As observed for noniodinated Tg, the specific binding of 35SO3-Tg to LPO tended toward a plateau (Fig. 5). The Scatchard representation gave the max = 1711 ± 60 pmol Tg per milligram of LPO, and the 35SO3-Tg equilibrium dissociation constant: Kd = 1.668 x 10–6 M. Higher amounts of Tg were necessary to reach binding saturation, suggesting a decrease in the number of Tyr-S available to bind LPO after iodination. However, the similar Kd value obtained for 35SO3-Tg noniodinated (1.758 x 10–6 M) and 35SO3-Tg iodinated via GGO system (1.668 x 10–6 M) suggested that iodinated Tg binds LPO via Tyr-S linkage, and consequently Tyr-S could not be iodinated by the GGO system only. It thus seems that peroxidase activity was necessary to iodinate Tyr-S.
Role of Tyr-S in the iodination and coupling reactions
To understand the process of Tyr-S binding in relationship with its iodination status, we studied the iodination and coupling reactions in different conditions. One of the criteria for the coupling reaction to occur would be the presence of tyrosines that are both sulfated and iodinated. Therefore, we looked for intermediate compounds such as MIT-sulfate or more importantly DIT-sulfate (DIT-S), because the acceptor iodotyrosine is always a DIT and we have previously shown that the main hormonogenic site contains a sulfated acceptor tyrosine (Tyr5-S) (23).
Iodination of Tg and coupling reactions with soluble LPO.
First we investigated the presence of SIT in the usual conditions of hormonosynthesis (1KIGO-1LPO). After iodination in vitro of 35SO3-Tg with KI and Tg with K125I, samples were submitted to enzymatic hydrolysis. Hydrolysates were analyzed using TLC after Dowex 50WX2 separation. By comparison with standard (Fig. 1), only very weak spots at Rf 0.44 (Fig. 6) were detectable with both 35S- and 125I-labeling, indicating that only traces of the corresponding SIT were present after Tg iodination.
Iodination of Tg and coupling reaction with immobilized LPO.
These experiments were performed in the conditions used to study specific Tyr-S binding. Immobilized LPO or BSA were incubated with Tg, [125I]iodide, and the GGO system (1KIGO), and electrophoresis of the media was then carried out in reducing conditions. After the coupling reaction in the presence of LPO, in addition to the specific band for 125I-Tg subunit, we observed other bands of higher molecular weight (Fig. 7A, lanes 3 and 4). In contrast, in the presence of BSA, only the band specific for 125I-Tg subunit could be detected (Fig. 7A, lanes 1 and 2). An experiment performed with 35SO3-Tg showed similar profiles with BSA (Fig. 7B, lanes 1 and 2), but after coupling with LPO (Fig. 7B, lanes 3 and 4), the Tg subunit signal was decreased, resulting in an increase in the higher molecular weight species. Media and membranes were separately treated with proteolytic enzymes. RP-HPLC analyses of media showed mainly MIT, but no thyroid hormones in the presence of BSA only (Fig. 8A), whereas MIT, DIT, T3, and T4 were observed in the presence of LPO (Fig. 8B). Hydrolysates from immobilized LPO-125I-Tg complex did not show T3 or T4 (Fig. 8C). TLC analysis of the first hydrophilic fractions of these last two RP-HPLC showed a spot at Rf 0.44 for media (Fig. 9A, lane 1) and for membranes (Fig. 9B, lane 2). This spot was specific for SIT because it was already observed in the same RP-HPLC fractions of standard (Fig. 1B). No iodoamino acids were present in the same portion of RP-HPLC of Tg iodinated without LPO (not shown), suggesting that Tyr-S was not iodinated with the GGO system. The 0.44 SIT was not detected with 35SO3-Tg, probably because the labeling was not sufficient to allow clear detection. SIT could be MIT-sulfate, DIT-S, or a nonfully hydrolyzed peptide released from LPO-Tg complex.
Role of Tyr-S: iodination of Tg peptides, NIFEY-S, NIFEY, and PY130
To better understand the role of Tyr-S in the iodination and coupling reactions, we iodinated NIFEY-S for comparison with either NIFEY or PY130, the peptide containing the nonsulfated tyrosine donor. NIFEY-S turned out to be poorly iodinated with the GGO system only (less than 2% of total iodide); on the other hand, using the same system, we found that the nonsulfated form of the same peptide was 3-fold more iodinated. When soluble LPO was added (1KIGO-1LPO), 1 nmol NIFEY-S incorporated 24 pmol [125I]iodine, whereas 1 nmol NIFEY or PY130 incorporated 15-fold more (Table 1). Different conditions of iodide, GGO system, and LPO concentrations were checked. Iodine incorporation into NIFEY-S was highly dependent on LPO concentration but not on iodide and H2O2 concentrations (Table 1). Comparatively, NIFEY and PY130 iodination was poorly dependent on iodide and H2O2 concentrations as well as on LPO concentration. Iodine incorporations into PY130 or NIFEY were similar (Table 1), showing that nonsulfated tyrosines had the same reactivity toward LPO, but that the reactivity of Tyr-S was different. Assays were also carried out with immobilized LPO (1KIGO-20LPO conditions). As in the soluble system, NIFEY-S was less iodinated than NIFEY or PY130 (49% vs. 95% or 96% of iodine incorporated into peptides), but the relative amount of iodide bound to membrane was higher for NIFEY-S than for NIFEY or PY130 (Table 2), suggesting that the major part of nonsulfated peptides were released from LPO after iodination. In the case of NIFEY-S, a relatively high proportion (22%, i.e. 60% of membrane-bound 125I-iodine) remained linked to membrane after proteolysis, showing that the iodinated NIFEY-S-LPO complex was only partly digested by proteolytic enzymes.
To study the coupling reaction between the Tyr acceptor (sulfated Tyr5 of NIFEY-S) and the Tyr donor (Tyr130 of PY130), NIFEY-S and PY130 were iodinated together. The weak concentrations of KI and glucose oxidase were limiting for iodination, whereas LPO concentration did not modify the incorporation of iodine (Table 1). In the 4KIGO-4LPO conditions only, the amount of iodine incorporated into the two peptides corresponded to the sum of the amounts incorporated into each peptide. With immobilized LPO, a low amount of iodinated peptides remained linked to LPO by comparison with NIFEY-S (Table 2), showing that, with PY130, either less NIFEY-S was bound to LPO or the coupling reaction occurred. However, using RP-HPLC analysis, only 1.5 ± 0.12 pmol T4 per nanomole of NIFEY-S was detected, showing that the coupling reaction occurred at low levels even with soluble LPO.
Samples from the first two conditions of iodination with LPO (Table 1) were hydrolyzed and analyzed by TLC. Spots corresponding to the compound at Rf 0.44 as well as a MIT and DIT migration front appeared clearly in NIFEY-S hydrolysates for high amounts of LPO (Fig. 10, lane 2), confirming that Tyr5-S iodination was dependent on LPO amount. The spot at Rf 0.44, which was not visible in assays with PY130 (Fig. 10, lanes 3 and 4), appeared also to be specific for Tyr-S. With immobilized LPO, 0.44 SIT was not visible in media (not shown) and was weakly detectable (Fig. 10, lane 7) in the membrane digest, probably because the NIFEY-S-LPO complex was not hydrolyzed by proteases and remained linked to membrane, as also observed in Table 2. With high concentrations of LPO, iodination of NIFEY-S with PY130 showed weak amounts of 0.44 SIT (Fig. 10, lanes 6 and 8), suggesting that when Tyr5-S and Tyr130 were present together, less Tyr5-S was bound to LPO.
Discussion
In previous studies, with either cultured thyrocytes or in vitro experiments for hormonosynthesis, we had shown that thyroid hormone synthesis decreased when tyrosine sulfation decreased (22) and most importantly that Tyr5, the major hormonogenic tyrosine, was sulfated (23). In this paper, using in vitro experimental procedures, we now report the function of Tyr-S in thyroid hormonosynthesis. We provide data indicating specific binding of Tg to LPO, which involves Tyr-S linkage, and the necessary presence of peroxidase to iodinate Tyr-S after binding to LPO. Weak amounts of SIT were detected in the peroxidase-Tg complex that represent short-lived intermediate compounds of hormone synthesis. These data, in addition to those obtained for the iodination of hormonogenic sulfated Tyr5 acceptor and Tyr130 donor, allow us to propose a mechanism of hormonosynthesis that completes the previous ones.
Using either cellular TPO (cultured thyrocytes reorganized into follicles) or LPO (in vitro experiments), we have previously obtained similar results for hormonosynthesis (22). These results agree with several reports that TPO and LPO have a similar efficiency in catalyzing iodination and coupling reactions (3, 9, 41) with the same preferential formation of DIT in Tg (42). Although some biochemical differences exist (41), particularly the production of HOI by LPO only (41, 43), we chose to perform these in vitro studies with LPO rather than TPO because native TPO is difficult to obtain in sufficient amounts and soluble recombinant TPO is known to be involved in iodination reaction (44) but have not been tested in coupling reaction.
We demonstrate that Tyr-S and the two peptides NIFEY-S and CCK8-S are good competitors for the binding of noniodinated 35SO3-Tg, showing the specificity of Tyr-S linkage. NIFEY-S was more efficient than CCK8-S, which was more efficient than Tyr-S, suggesting that the peptide sequence surrounding the Tyr5-S is also critical for the binding. Protein-protein interactions via the Tyr-S linkage (27, 28) have been described in hormone-receptor recognition (29, 30), inflammatory leukocyte adhesion (31), hemostasis (31, 32), and chemokine signaling (31, 33). However this is the first time that Tyr-S linkage was found to be involved in the formation and the catalytic reaction of a peroxidase-substrate complex. Tyr and nonsulfated peptides were also able to compete with Tg but with a lower efficacy, suggesting either that Tyr-S and Tyr could have different levels of affinity for the same site or, most likely, that LPO contains an additional specific site for nonsulfated tyrosine, as previously described (24, 25). The two-site hypothesis is also supported by 1) the differences in iodination of sulfated and nonsulfated Tg peptides (Table 1 and 2) and 2) the fact that the sulfate group confers to Tyr chemical properties probably requiring a specific structure in LPO. Indeed Tyr-S residue could be stabilized by forming conjugate ion pairs with the positive charges of amino acids located in the LPO sequence, such as the guanidino group of Arg residue (45). Thus two different tyrosine sites appear to be present on LPO: one for Tyr-S and one for Tyr.
When 35SO3-Tg was iodinated without LPO and then bound to LPO, the number of picomoles of Tg corresponding to the saturation of sites was 3-fold higher than with noniodinated 35SO3-Tg. A decrease in Tyr-S available for binding could be caused by modifications of Tg during its iodination because tyrosine iodination is known to result in alterations of the Tg structure and of its immunological and biochemical properties (46). That the Kd value of Tg-peroxidase interaction was not changed by the iodination of Tg (without LPO) before the binding, and that iodination of NIFEY-S was mainly dependent on the concentration of LPO, could indicate that Tg-bound Tyr-S was not iodinated in the absence of LPO. We assume that, in this case, the sulfate group might prevent the attachment of oxidized iodide to the aromatic ring of tyrosine, because the iodination of NIFEY-S was lower than that of the same nonsulfated peptide, and maybe due to a desulfation of NIFEY-S. Similar results were obtained by chemical iodination (47).
Then, to characterize Tyr5-S iodination, we looked for SIT that should be transiently present during the process. We were able to detect only traces of SIT (Rf 0.44) in enzymatically iodinated Tg. However, this compound was recovered in higher amounts in hydrolysates of immobilized LPO-Tg complex or of LPO-iodinated NIFEY-S, likely because the relative amount of LPO was higher. This SIT could be a nonfully hydrolyzed peptide from the Tyr-S-LPO complex. In immobilized LPO experiments with NIFEY-S, SIT was weakly present but more than 20% of iodine remained on the membrane fraction after proteolytic digestion. This finding supports the hypothesis that DIT-S at position 5 remains linked to LPO until the coupling reaction occurs, becoming the side chain undergoing the hormonogenic coupling reaction. On the other hand, Schepky et al. (47) showed that Tyr-S-containing peptide was more rapidly iodinated after the release of sulfate by acid hydrolysis. Based on these observations, we made the assumption that the binding of Tg to LPO by Tyr5-S linkage could modify the charges of the sulfate group and, consequently, the charges of the aromatic ring, and the Tyr5-S acceptor could become more reactive for a rapid iodination. Desulfation might occur when complete iodination is achieved just before the coupling reaction, DIT-S becoming a DIT (iodophenoxy anion) and thus making it available for the coupling reaction. This hypothesis is in agreement with pKa studies of iodotyrosines in hormonosynthesis, suggesting that the coupling reaction requires phenoxy anion formation (18). It is also in agreement with other results showing that iodination and coupling reactions occur simultaneously (12) and that iodide is essential for the coupling reaction (48, 49). Further studies on the NIFEY-S- or Tg-peroxidase complex will be needed to confirm this hypothesis.
The rate of iodination was dependent on LPO concentration only for the sulfated peptide, and SIT could be clearly observed after either NIFEY-S iodination with high concentration of LPO or Tg iodination with immobilized LPO. However, when NIFEY-S and PY130 were iodinated together with a high concentration of LPO, we did not observe the amount of SIT corresponding to NIFEY-S only. Moreover, immobilized LPO experiments showed that a low amount of iodinated NIFEY-S remained bound to membrane when PY130 was present. Consequently it appears that the two tyrosine sites may cooperate to equilibrate and control acceptor and donor iodination. This assumption is in line with previous reports on the different reactivities of tyrosyl residues toward iodine: iodination would proceed in a sequential order controlled by the native structure of Tg (15, 16, 17). In addition the coupling reaction with NIFEY-S and PY130 was very low, probably due to the small chance of having, in LPO, two DIT in close proximity. This finding is not surprising because the three-dimensional structure of Tg is necessary for the coupling reaction (50).
All these data allow us to propose a mechanism of thyroid hormonosynthesis that completes the previous ones. Taking into account that 1) a specific binding of Tg to LPO occurs via Tyr-S linkage, 2) LPO is necessary to iodinate Tyr-S, and 3) iodination of Tyr-S is dependent on the amount of LPO, Tyr5-S could first bind to peroxidase, allowing Tg to be in the right position to bind the Tyr donor. This agrees with the fact that TPO is able to bind Tyr before its activation by H2O2 (26) and also with the report of Xiao et al. (42) showing that preferential formation of DIT involves interaction between Tg and peroxidase. Thus the peroxidase could bind only the two types of tyrosine involved in hormonosynthesis to allow tyrosine iodination likely by [EOI]– complex (25) and to achieve the coupling reaction. Because a great part of NIFEY-S remained bound to immobilized LPO, iodinated Tyr5-S that becomes the DIT acceptor could remain bound to its site until the coupling occurs. On the other hand, the major part of PY130 containing the nonsulfated Tyr that becomes the DIT donor was released from peroxidase after iodination and could be waiting for the coupling reaction when involved in the Tg structure. The other nonsulfated tyrosines, not involved in hormone synthesis, could be iodinated by a free intermediate such as HOI, which is the active iodinating intermediate at physiological pH (41, 43) and probably corresponds to the diffusible iodinating compound in LPO-catalyzed iodination (51). The release of DIT donor from peroxidase as well as the possible action of HOI, particularly present with LPO (41), could explain that, in our experiments, soluble LPO led to a higher incorporation of iodide on NIFEY or PY130 than on NIFEY-S remaining partly bound to LPO.
Concerning Tg iodination in the absence of LPO, H2O2 appeared to be sufficient to oxidize iodide, but only MIT could be observed, suggesting a lower iodination efficiency. Therefore, in the absence of peroxidase, iodination could occur due to a Fenton-type reaction initiated by traces of transition-metal ions in the reaction, involving hydroxyl radical formation and subsequent HOI formation (52).
In conclusion, our results provide new information on the thyroid hormone synthesis process: the sulfate group, present on tyrosine acceptor in hormonogenic site of Tg, seems to be a necessary partner to bind Tg to peroxidase, likely stabilizing this interaction and allowing the formation of DIT acceptor and thus becoming a potential regulatory element of thyroid hormone synthesis. Further studies should be performed to confirm the binding of Tyr5-S with TPO, using immobilized TPO or more physiological conditions: membrane-bound TPO expressed at the apical surface of filter-cultured thyroid cell monolayers (53). Tyrosylprotein sulfotransferase (TPST)-deficient mice (54) should also be considered a good model serving as a negative control, although the phenotypes observed with TPST1 or TPST2 deficiency do not correspond to the characteristic phenotypes of severe hypothyroidism as early as at birth (55). To perform such a study maybe both TPST1 and TPST2 genes would need to be knocked out. This work should also be followed by structural studies of the Tg-peroxidase complex to locate the sites of Tyr-S and Tyr binding on LPO and TPO.
Acknowledgments
The authors thank Prof. P. Carayon for comments, Dr. C. Penel for reading the manuscript and helpful discussions, Dr. B. Mallet for advice about Fenton reaction and HPLC analyses, and Mrs. J. Barbaria for HPLC experiments. We also thank Dr. G. Herbette (Spectropole, Universités d’Aix-Marseille, Faculté de Saint Jérme) for help in NMR analysis and interpretation and Dr. P. H. Rolland for the gift of thyroids from control experimental pigs.
Footnotes
This study was supported by Institut National de la Santé et de la Recherche Médicale (Unité 555), Centre National de la Recherche Scientifique (SDI 401555), Institut Fédératif de Recherche 125, and the Université de la Méditerranée.
Abbreviations: CCK8, Cholecystokinin; DIT, diiodotyrosine; DIT-S, DIT-sulfate; EN, ethyl alcohol-ammonium 1N; FA, formic acid; GGO, glucose-glucose oxidase; HOI, hypoiodous acid; LPO, lactoperoxidase; MIT, monoiodotyrosine; NMR, nuclear magnetic resonance; Rf, retention factor; RP-HPLC, reverse phase high performance liquid chromatography; SIT, sulfated iodotyrosine; Tg, thyroglobulin; TLC, thin layer chromatography; TPO, thyroperoxidase; TPST, tyrosylprotein sulfotransferase; Tyr-S, sulfated tyrosine.
References
Ekholm R 1990 Biosynthesis of thyroid hormones. Int Rev Cytol 120:243–288
Vassart G, Dumont JE 1992 The thyrotropin receptor and the regulation of thyrocyte function and growth. Endocr Rev 13:596–611
Kopp P 2005 Thyroid hormone synthesis. In: Brawerman LE, Utiger RD, eds. Werner, Ingbar’s the thyroid: a fundamental, clinical text. 9th ed. Philadelphia: Lippincott Williams, Wilkins; 52–76
Arvan P, Di Jeso B 2005 Thyroglobulin structure, function, and biosynthesis. In: Brawerman LE, Utiger RD, eds. Werner, Ingbar’s the thyroid: a fundamental, clinical text. 9th ed. Philadelphia: Lippincott Williams, Wilkins; 77–95
Carrasco N 2005 Thyroid iodine transport. Brawerman LE, Utiger RD, eds. Werner, Ingbar’s the thyroid: a fundamental, clinical text. 9th ed. Philadelphia: Lippincott Williams, Wilkins; 37–52
Nilsson M, Bjorkman U, Ekholm R, Ericson LE 1990 Iodide transport in primary cultured thyroid follicle cells: evidence of a TSH-regulated channel mediating iodide efflux selectively across the apical domain of the plasma membrane. Eur J Cell Biol 52:270–281
Gruffat D, Gonzalvez S, Chambard M, Mauchamp J, Chabaud O 1991 Long-term iodination of thyroglobulin by porcine thyroid cells cultured in porous-bottomed culture chambers: regulation by thyrotrophin. J Endocrinol 128:51–61
Bjorkman U, Ekholm R 1988 Accelerated exocytosis and H2O2 generation in isolated thyroid follicles enhance protein iodination. Endocrinology 122:488–494
Nunez J, Pommier J 1982 Formation of thyroid hormones. Vitam Horm 39:175–229
Lamas L, Dorris ML, and Taurog A 1972 Evidence for a catalytic role for thyroid peroxidase in the conversion of diiodotyrosine to thyroxine. Endocrinology 90:1417–1426
Gavaret JM, Nunez J, Cahnmann HJ 1980 Formation of dehydroalanine residues during thyroid hormone synthesis in thyroglobulin. J Biol Chem 255:5281–5285
Taurog A, Dorris LM, Doerge RD 1996 Mechanism of simultaneous iodination and coupling catalyzed by thyroid peroxidase. Arch Biochem Biophys 330:24–32
Marriq C, Lejeune PJ, Venot N, Vinet L 1991 Hormone formation in the isolated fragment 1–171 of human thyroglobulin involves the couple tyrosine 5 and tyrosine 130. Mol Cell Endocrinol 81:155–164
Dunn AD, Corsi CM, Myers HE, Dunn JT 1998 Tyrosine 130 is an important outer ring donor for thyroxine formation in thyroglobulin. J Biol Chem 273:25223–25229
Edelhoch H, Lippoldt RE 1962 The properties of thyroglobulin IX. The molecular properties of iodinated thyroglobulin. J Biol Chem 237:2788–2794
Gavaret JM, Deme D, Nunez J, Salvatore G 1977 Sequential reactivity of tyrosyl residues of thyroglobulin upon iodination catalyzed by thyroid peroxidase. J Biol Chem 252:3281–3285
Lamas L, Anderson PC, Fox JW, Dunn JT 1989 Consensus sequences for early iodination and hormonogenesis in human thyroglobulin. J Biol Chem 264:13541–13545
De Viljder JJM, den Hartog MT 1998 Anionic iodotyrosine residues are required for iodothyronine synthesis. Eur J Endocrinol 138:227–231
Hortin G, Folz R, Gordon JI, Strauss AW 1986 Characterization of sites of tyrosine sulfation in proteins and criteria for predicting their occurrence. Biochem Biophys Res Commun 141:326–333
Rosenquist GL, Nicholas Jr HB 1993 Analysis of sequence requirements for protein tyrosine sulfation. Protein Sci 2:215–222
Bundgaard JR, Vuust J, Rehfeld JF 1997 New consensus features for tyrosine O-sulfation determined by mutational analysis. J Biol Chem 272:21700–21705
Nlend MC, Cauvi D, Venot N, Chabaud O 1999 Sulfated tyrosines of thyroglobulin are involved in thyroid hormone synthesis. Biochem Biophys Res Commun 262:193–197
Venot N, Nlend MC, Cauvi D, Chabaud O 2002 The hormonogenic tyrosine 5 of porcine thyroglobulin is sulfated. Biochem Biophys Res Commun 298:193–198
Casella L, Gullotti M, Poli S, Bonfa M, Ferrari RP, Marchesini A 1991 Spectroscopic and binding studies on the stereoselective interaction of tyrosine with horseradish peroxidase and lactoperoxidase. Biochem J 279:245–250
Sun W, Dunford BH 1993 Kinetics and mechanism of the peroxidase-catalyzed iodination of tyrosine. Biochemistry 32:1324–1331
Yamamoto K, Degroot LJ 1982 Affinity chromatography of thyroid peroxidase using tyrosine coupled to agarose. J Biochem (Tokyo) 91:775–782
Niehrs C, Beisswanger R, Huttner WB 1994 Protein tyrosine sulfation, 1993—an update. Chem Biol Interact 92:257–271
Strott CA 2002 Sulfonation and molecular action. Endocr Rev 23:703–732
Gigoux V, Escrieut C, Silvente-Poirot S, Maigret B, Gouilleux L, Fehrentz JA, Gully D, Moroder L, Vaysse N, Fourny D 1998 Met-195 of the cholecystokinin-A receptor interacts with the sulfated tyrosine of cholecystokinin and is crucial for receptor transition to high affinity state. J Biol Chem 273:14380–14386
Costagliola S, Panneels V, Bonomi M, Koch J, Many MC, Smits G, Vassart G 2002 Tyrosine sulfation is required for agonist recognition by glycoprotein hormone receptors. EMBO J 21:504–513
Kehoe JW, Bertozzi CR 2000 Tyrosine sulfation: a modulator of extracellular protein-protein interactions. Chem Biol 7:R57–R61
Ward CM, Andrews RK, Smith AI, Berndt MC 1996 Mocarhagin, a novel cobra venom metalloproteinase, cleaves the platelet von Willebrand factor receptor glycoprotein Ib. Identification of the sulfated tyrosine/anionic sequence Tyr-276-Glu-282 of glycoprotein Ib as a binding site for von Willebrand factor and -thrombin. Biochemistry 35:4929–4938
Farzan M, Chung S, Li W, Vasilieva N, Wright PL, Schnitzler CE, Marchione RJ, Gerard C, Gerard NP, Sodroski J, Choe H 2002 Tyrosine-sulfated peptides functionally reconstitute a CCR5 variant lacking a critical amino-terminal region. J Biol Chem 277:40397–40402
Chambard M, Mauchamp J, Chabaud O 1987 Synthesis and apical and basolateral secretion of thyroglobulin by thyroid cell monolayers on permeable substrate: modulation by TSH. J Cell Physiol 133:37–45
Nlend MC, Cauvi D, Venot N, Chabaud O 1999 Thyrotropin regulates tyrosine sulfation of thyroglobulin. Eur J Endocrinol 141:61–69
Reitz HC, Ferrel RE, Fraenkel-Conrat H, Olcott HS 1946 Action of sulfating agents on proteins and model substances. I. Concentrated sulfuric acid. J Am Chem Soc 68:1024–1031
Futaki S, Taike T, Yagami T, Ogawa T, Akita T, Kitagawa K 1990 Use of dimethylformamide-sulphur trioxide complex as a sulphating agent of tyrosine. J Chem Soc Perkin Trans 1:1739–1744
Gruffat D, Venot N, Marriq C, Chabaud O 1992 Thyroid hormone synthesis in thyroglobulin secreted by porcine thyroid cells cultured on porous bottom chambers. Endocrinology 131:2921–2927
Blode H, Heinrich T, Diringer H 1990 A quantitative assay for tyrosine sulfation and tyrosine phosphorylation in peptides. Biol Chem Hoppe-Seyler 371:145–151
Cheng Y, Prusoff WH 1973 Relationship between the inhibition constant (K1) and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction. Biochem Pharmacol 22:3099–3108
Magnusson RP, Taurog A, Dorris ML 1984 Mechanisms of thyroid peroxidase- and lactoperoxidase-catalyzed reactions involving iodide. J Biol Chem 259:13783–13790
Xiao S, Dorris ML, Rawitch AB, Taurog A 1996 Selectivity in tyrosyl iodination sites in human thyroglobulin. Arch Biochem Biophys 334:284–294
Dunford HB, Ralston IM 1983 On the mechanism of iodination of tyrosine. Biochem Biophys Res Commun 116:639–643
Gut P, Grennan Jones F, Sullivan A, Ziemnicka Smith S, Jaskolski D, Furmaniak J, Rees Smith B 2000 Recombinant human thyroid peroxidase produced in insect cells has similar properties to native human thyroid peroxidase. Thyroid 10:543–550
Yagami T, Kitagawa K, Aida C, Fujiwara H, Futaki S 2000 Stabilization of a tyrosine O-sulfate residue by a cationic functional group: formation of a conjugate acid-base pair. J Pept Res 56:239–249
Gardas A, Domek H 1993 Iodine induced alteration in immunological and biochemical properties of thyroglobulin. Acta Biochim Pol 40:237–240
Schepky AG, Schmidt AM, Schmidt T, Schulz-Knappe P, Forssmann WG 1994 Determination of sulfated peptides by differential iodination. Biol Chem Hoppe-Seyler 375:201–203
Dème D, Pommier J, Nunez J 1976 Kinetics of thyroglobulin iodination and of hormone synthesis catalyzed by thyroid peroxidase. Role of iodide in the coupling reaction. Eur J Biochem 70:435–440
Virion A, Dème D, Pommier J, Nunez J 1981 The role of iodide and of free diiodotyrosine in enzymatic and non-enzymatic thyroid hormone synthesis. Eur J Biochem 118:239–245
Dunn JT, Dunn AD 1999 The importance of thyroglobulin structure for thyroid hormone synthesis. Biochimie (Paris) 81:505–509
Huber RE, Edwards LA, Carne TJ 1989 Studies on the mechanism of the iodination of tyrosine by lactoperoxidase. J Biol Chem 264:1381–1386
Klebanoff SJ 1982 The iron-H2O2-iodide cytotoxic system. J Exp Med 156:1262–1267
Penel C, Gruffat D, Alquier C, Benoliel AM, Chabaud O 1998 Thyrotropin chronically regulates the pool of thyroperoxidase and its intracellular distribution: a quantitative confocal microscopy study. J Cell Physiol 174:160–169
Moore KL 2003 The biology and enzymology of protein tyrosine O-sulfation. J Biol Chem 278:24243–24246
Vulsma T, de Vijlder JJM 2005 Genetic defects causing hypothyroidism In: Brawerman LE, Utiger RD, eds. Werner, Ingbar’s the thyroid: a fundamental, clinical text. 9th ed. Philadelphia: Lippincott Williams, Wilkins; 714–730(Marie-Christine Nlend, Da)
Northwestern University (M.-C.N.), Pulmonary Division & Critical Care Medicine, Chicago, Illinois 60611
The Scripps Research Institute (D.M.C.), Department of Molecular and Experimental Medicine, La Jolla, California 92037
Abstract
Our previous studies showed that sulfated tyrosines (Tyr-S) are involved in thyroid hormone synthesis and that Tyr5, the main hormonogenic site of thyroglobulin (Tg), is sulfated. In the present paper, we studied the role of Tyr-S in the formation and activity of the peroxidase-Tg complex. Results show that noniodinated 35SO3-Tg specifically binds (Kd = 1.758 μM) to immobilized lactoperoxidase (LPO) via Tyr-S linkage by using saturation binding and competition experiments. We found that NIFEY-S, a 15-amino acid peptide corresponding to the NH2-end sequence of Tg and containing the hormonogenic acceptor Tyr5-S, was a better competitor than cholecystokinin and Tyr-S. 35SO3-Tg, iodinated without peroxidase, bound to LPO with a Kd (1.668 μM) similar to that of noniodinated Tg, suggesting that 1) its binding occurs via Tyr-S linkage and 2) Tyr-S requires peroxidase to be iodinated, whereas nonsulfated Tyr does not. Iodination of NIFEY-S with [125I]iodide showed that Tyr5-S iodination increased with LPO concentration, whereas iodination of a nonsulfated peptide containing the donor Tyr130 was barely dependent on LPO concentration. Enzymatic hydrolysis of iodinated Tg or NIFEY-S showed that the amounts of sulfated iodotyrosines also depended on LPO amount. Sulfated iodotyrosines were detectable in the enzyme-substrate complex, suggesting they have a short life before the coupling reaction occurs. Our data suggest that after Tyr-S binding to peroxidase where it is iodinated, the sulfate group is removed, releasing an iodophenoxy anion available for coupling with an iodotyrosine donor.
Introduction
THE MAIN PARTNERS in thyroid hormone synthesis are: thyroglobulin (Tg), thyroperoxidase (TPO), the H2O2 generating system, and iodide. TSH, the pituitary hormone regulating thyroid function, is essential to control the level of expression of all the partners (1, 2, 3, 4), including the NaI symporter (5) and the apical iodide transporter(s) that transport iodide into the colloid (3, 6, 7). Thyroid hormone synthesis occurs only when, on one hand, Tg, iodide (7), and H2O2 (8) are secreted into the colloid and when, on the other hand, TPO is simultaneously present on the apical membrane of thyrocytes (1). In brief, the activation of TPO by H2O2 allows the oxidation of iodide and then the iodination of some tyrosines of Tg, leading to the formation of monoiodotyrosine (MIT) and diiodotyrosine (DIT). Several mechanisms have been proposed for the iodination reaction, including the various forms of a same iodide oxidation state that allow a two-electron reaction: the iodinium ion I+, the enzyme iodide complex [EOI]–, and the free hypoiodous acid (HOI) (for review see Refs.3 and 9). Then the coupling reaction between two iodotyrosines leads to the formation of thyroid hormones: T3 (DIT + MIT) and T4 (DIT + DIT). The catalytic action of TPO in the coupling reaction (10) involves the formation of a quinol ether intermediate containing two DIT. Removal of this intermediate from the peptide chain induces the formation of a dehydroalanine residue (11) on the one hand, and the formation of T4 linked to peptide chain on the other hand. Taurog et al. (12) showed that TPO catalyzed simultaneously the reactions of tyrosine iodination and coupling. Among the five hormonogenic sites identified in Tg molecule, tyrosine 5 (Tyr5) is the preferential T4 forming site (4) with tyrosine 130 (Tyr130) as donor site (13, 14).
The tyrosyl residues in Tg have different levels of reactivity to iodine. Iodination, controlled by the native structure of Tg, proceeds in a sequential order (15, 16, 17). At the hormonogenic site, the DIT acceptor and the DIT donor involved in the coupling reaction have different characteristics, such as their pKa (18). The peptide sequence surrounding each hormonogenic acceptor tyrosine in Tg (4) is similar to the known consensus sequence for tyrosine sulfation (19, 20, 21). Moreover, we previously showed that thyroid hormone synthesis was dependent on tyrosine sulfation (22), and more recently we showed that the major hormonogenic tyrosine (Tyr5) was sulfated (23). We hypothesize that sulfation could give the specific character for a tyrosine to become the DIT acceptor. Little information exists on the molecular interactions between Tg and TPO and their consequences in iodination and coupling reactions. A specific site for tyrosine binding has been described near the heme pocket on lactoperoxidase (LPO) (24, 25). TPO contains at least one specific active site for tyrosine that is independent of the iodide-binding site (26).
In this study, we propose to evaluate the role of sulfated tyrosines (Tyr-S) in the formation of the peroxidase-Tg complex as well as in the iodination of tyrosines and coupling reactions. Tyr-S are specifically involved in protein or peptide binding (27, 28), as in hormone-receptor interaction (29, 30), inflammatory leukocyte adhesion (31), hemostasis (31, 32), and chemokine signaling (31, 33). Given that Tyr5-S found in Tg (23) becomes the DIT acceptor involved in hormonosynthesis, we focused on the following questions. Does 35SO3-Tg specifically bind to peroxidase via Tyr-S linkage What is the involvement of the sulfate group in the iodination of tyrosine acceptor and in the coupling reaction
Our results show that 35SO3-Tg binds specifically to peroxidase through Tyr-S linkage and that the iodination level of sulfated Tyr5 depends on the peroxidase concentration. Moreover, when peroxidase concentration is increased, more sulfated iodotyrosine intermediates are found in the enzyme-substrate complex. Thus we propose that iodinated Tyr-S could remain bound to peroxidase until the structural conditions of coupling are reached in the peroxidase-Tg complex. The sulfate group of the DIT acceptor might then be released, leaving the phenoxy anion free for association with the DIT donor.
Materials and Methods
Cell culture and preparation of cellular porcine Tg
Thyroid epithelial cells were isolated from porcine glands, and primary cultures were made on porous filters coated with collagen (34) in the presence of TSH added from d 6 and without iodide (35). Cells were cultured without iodide to obtain noniodinated Tg. Cell culture conditions and [35S]sulfate-labeling were previously described (23). On d 18, apical media, in which Tg was essentially secreted, were collected and kept at –20 C until Tg purification. Nonlabeled Tg and [35S]sulfate-labeled Tg (35SO3-Tg) were purified from the apical media as described (23). Protein amount was evaluated either by Micro BCA protein assay (Pierce, Rockford, IL) or by measurement of absorbance at 280 nm (E1%1 cm = 10).
Preparation of Tyr-S and sulfated peptides
L-Tyrosine (Sigma Aldrich, St. Louis, MO) was sulfated according to the method of Reitz et al. (36). A peptide containing the NH2-end sequence of Tg, NIFEYQVDAQPLRP, named NIFEY, was synthesized (Sigma Genosys, Haverhill, UK). The Tyr5 on this peptide was sulfated according to Futaki et al. (37) with some modifications. In brief, NIFEY (15 μmol) was incubated with 1.5 mmol sulfur trioxide dimethylformamide complex (DMF-SO3; Fluka Chemie, Buchs, Switzerland) in a 1-ml dimethylformamide (Fluka)-pyridine (Prolabo; Merck Eurolab, Briare le Canal, France) mixture (4:1), for 24 h at room temperature. The sulfated peptide (NIFEY-S) was then purified on a Sephadex G-10 column (Amersham Biosciences, Orsay, France) equilibrated with 0.05 M ammonium bicarbonate (pH 8.2) and identified by the measure of absorbance at 262 and 280 nm. The ratio of A280 to A262 characteristic of each Tyr-S (32) was 0.5 for NIFEY-S and 0.75 for Tyr-S. Tyr-S and NIFEY-S were separated from the nonsulfated Tyr or NIFEY by cation-exchange chromatography (see below) on Dowex 50WX2 (H+). Cholecystokinin (CCK8, DYMGWMDF; Sigma Genosys) was used as nonsulfated peptide, and CCK8-S (Bachem Biochemica GmbH, Heidelberg, Germany) was also processed as described for NIFEY-S. The presence of a sulfate residue on the phenol function of tyrosine was characterized by nuclear magnetic resonance (NMR).
NMR analyses
NMR was performed by the Spectropole (Universités d’Aix-Marseille, Marseille, France). NMR spectra were recorded in D2O solutions at 300 K using a Bruker Avance DRX 500 spectrometer equipped with a Bruker CryoPlatform and a 5-mm cryo TXI probe. The temperature of the probe and preamplifier was 30 K. Chemical shifts were referenced to D2O: H = 4.79 ppm. 1H spectra were obtained with a standard zg sequence. Correlation spectroscopy spectra were obtained using gradient pulses for selection with a cosygp sequence.
Preparation of immobilized LPO
LPO (Sigma) was bound to Immobilon-P membrane by filtration in a 96-well multiscreen-IP opaque plate (Millipore Corporation, Bedford, MA) following the manufacturer’s instructions. Each well was treated 5 min with 100 μl 70% ethyl alcohol, filtered, and washed with 100 μl of ultrapure water, followed by 200 μl of 0.01 M sodium phosphate buffer (pH 7.4) containing 0.14 M sodium chloride (PBS). LPO (25 μg/100 μl PBS/well) was then filtered and washed by 200 μl PBS. Membranes were used immediately for binding assays. BSA (Sigma) used as negative control was also bound to membranes in the same conditions.
Tg binding assays
Saturation binding experiments.
Increasing concentrations of 35SO3-Tg or iodinated 35SO3-Tg were dissolved in 0.05 M phosphate buffer (pH 7.2), and 100 μl of each solution was added to an Immobilon-bound LPO- or BSA-containing well and incubated for 45 min at 37 C with gentle stirring. Media were then harvested, and membranes were washed three times with the same buffer. Then the 35SO3-radioactivity was counted in 200 μl/well of liquid scintillation cocktail Microscint 40 (Packard Biosciences, BV, Groningen, The Netherlands) using a TopCount NXT (Packard Instrument Co., Meriden, CT). Competition experiments were performed in the same conditions, but 35SO3-Tg (1.5 μM) was incubated with increasing concentrations of each compound individually: Tyr-S, CCK8-S, NIFEY-S, or Tyr, CCK8, NIFEY. Results and curves were given using GraphPad Prism software.
In vitro hormonosynthesis.
In vitro iodination and coupling experiments [40 atoms of iodine per mole of Tg, (38)] were performed with 35SO3-Tg or unlabeled Tg secreted by cells cultured without iodide. Tg (1.5 nmol) was incubated in 1 ml of 0.05 M sodium phosphate buffer, pH 7.4, containing 1.5 mg glucose, 100 nmol potassium iodide (KI), and 5 μg LPO. Carrier-free Na125I (0.25 μCi/ml, 0.5 μM; NEN Life Science Products, Zaventem, Belgium) was added to assays with unlabeled Tg, and the reaction was started by the addition of 2.5 μg glucose oxidase (Sigma) (38). The relative concentrations of these reagents were named 1KIGO-1LPO. Hormonosynthesis was carried out at 37 C for 30 min. The peptides NIFEY-S, NIFEY, and PY130 (AEGMEVYGTRQLGRP, a Tg-like peptide containing the hormonogenic Tyr130; Sigma Genosys) were iodinated (560 nmol/ml) in the same way but with variations in the concentrations of LPO, KI, and glucose-glucose oxidase (GGO) system, as indicated in Table 1. Hormonosynthesis was also performed with Immobilon-bound LPO in a 96-well multiscreen-IP opaque plate in the conditions of binding assays. The medium (100 μl), containing either Tg or Tg-peptides with [125I]iodide and the GGO system (1KIGO), was incubated 45 min at 37 C. Media were then harvested and frozen for further analysis. Membranes were washed three times with the same buffer and then cut, and [125I]iodide radioactivity was determined in a -counter (Cobra II; Packard Instrument Co.).
Gel electrophoresis
Aliquots of iodinated 35SO3-Tg or 125I-labeled Tg were analyzed by 5% SDS-PAGE (35). Gels were dried and then exposed to an imaging plate radioactive energy sensor (BAS-IP.MP 2040S; Fuji Photo Film Co., Ltd., Kanagawa, Japan) before scanning (Fujix BAS1000 IP reader; Fuji).
Enzymatic digestion after hormonosynthesis
125I-labeled Tg, iodinated 35SO3-Tg, and LPO-bound Tg or peptides on membranes were dissolved in a 0.05 M sodium phosphate buffer (pH 7.4), and the samples were digested for 48 h at 37 C by 20% (wt/wt) pronase (Roche Molecular Biochemicals, Mannheim, Germany). Samples were then subjected to digestion by 20% (wt/wt) leucine aminopeptidase, (cytosol type V from porcine kidney; Sigma) for an additional 48 h at 37 C. The iodoamino acids were then separated by cation-exchange chromatography or reverse-phase high performance liquid chromatography (RP-HPLC).
Cation-exchange chromatography
Dowex 50WX2 (H+ form) chromatography was performed as described (35, 39). The nonretained material, in 0.1 M formic acid (FA) fraction and in 0.1 M FA subsequent rinses, contained the majority (85–95%) of the [35S]sulfate radioactivity applied corresponding to sulfated amino acids, inorganic sulfate, sulfated carbohydrates (35), as well as free [125I]iodide. The column was then eluted (3 x 1 vol Dowex) successively with ethyl alcohol 25% and ethyl alcohol-ammonium 1N (EN) containing hydrophobic iodinated compounds. Fractions were collected, frozen, dried, and then analyzed by thin layer chromatography (TLC).
RP-HPLC
Separation by RP-HPLC of the different 125I-labeled or 35S-labeled samples was performed on silica C18 column (Uptisphere 5 μm ODSB UP50DB 10 M Interchim; Asnières, France). The column was eluted at a 0.5 ml/min flow rate (one fraction per minute) using a linear 60-min gradient of 0–60% acetonitrile in 20 mM ammonium acetate (pH 4). With this elution gradient, sulfated compounds were not retained and were recovered in the first minutes of elution, whereas the iodotyrosines (MIT, DIT, T3, T4) were retained and successively eluted. Fractions were collected, frozen, dried, and then analyzed by TLC.
Thin layer chromatography
Enzymatic hydrolysates or fractions from enzymatic hydrolysates purified on Dowex 50WX2 or on RP-HPLC were applied to silica-coated glass plates (10 x 10; Merck, Darmstadt, Germany) and chromatography was developed by the solvent system: N-butyl alcohol (Fluka Chemie)-glacial acetic acid (Carlo Erba Reagenti, Milan, Italy)-water (65:20:15) for 90 min. Labeled components were revealed with an imaging plate radioactive energy sensor.
Preparation of sulfated iodoamino acids
L-Tyrosine (Sigma) and L-tyrosine-sulfate were submitted to in vitro iodination as above with carrier-free Na125I (0.25 μCi/ml). The resulting 125I-labeled iodotyrosines were purified by cation-exchange chromatography (see above) and compounds of the two major elution fractions, 0.1 M FA and EN, were analyzed either directly on TLC or after separation by RP-HPLC (data not shown). Comparatively, 35SO3-Tyr isolated from 35SO3-Tg after alkaline hydrolysis (35) was iodinated as above, and the sulfated iodotyrosine derivatives were separated on RP-HPLC and analyzed by TLC. A spot at retention factor (Rf) of 0.44 was observed in the FA fraction of 125I-Tyr-S (Fig. 1A, lane 1), but not in the corresponding fraction of nonsulfated 125I-Tyr (Fig. 1A, lane 2). This compound was also present in the first hydrophilic fractions of RP-HPLC of 125I-Tyr-S (Fig. 1B, lanes 1 and 2) and of iodinated 35SO3-Tyr (Fig. 1C, lane 2). Unfortunately we did not obtain sufficient amounts to characterize this sulfated iodotyrosine (SIT) by NMR.
Results
Characterization of Tyr-S and NIFEY-S
In general, O-sulfation (ester bond) represents the dominant sulfation reaction (28) but SO3– is sometimes bound to the aromatic ring of tyrosine (sulfonation). NMR served to characterize the two types of bond. Before sulfation, Tyr showed two signals as doublets (6.77 and 7.07 ppm) corresponding to the 2 x 2 equivalent protons of the aromatic ring (Fig. 2A), whereas after sulfation (Fig. 2B), we observed a down-field shift of these two signals (7.18 and 7.23 ppm). Sulfonation (7.47 and 6.89 ppm determined by correlation spectroscopy experiment, Fig. 2B) was only weakly present (17%). Before sulfation (Fig. 2C), NIFEY showed four signals: three doublets (6.7, 7.0, 7.11 ppm) and one multiplet (7.19 ppm). The first two signals likely corresponded to the aromatic protons of Tyr, whereas the two other signals were due to phenylalanine, the other aromatic amino acid in NIFEY. After sulfation (Fig. 2D), the first two signals merged into a multiplet (7.1–7.22 ppm) with the other aromatic protons. Thus in each case, NMR showed that SO42– and the hydroxyl of tyrosine are linked by an ester bond.
Specific binding of noniodinated 35SO3-Tg to peroxidase
Increasing concentrations of noniodinated 35SO3-Tg were incubated with LPO or BSA linked to Immobilon-P membrane. After washings, the remaining radioactivity bound to LPO or BSA was counted. For each 35SO3-Tg concentration, the radioactivity associated to BSA (nonspecific binding) was subtracted from the radioactivity associated to LPO (total binding), giving a nearly hyperbolic curve corresponding to the specific dose-dependent binding of 35SO3-Tg to LPO (Fig. 3). The Scatchard representation gave the amount of 35SO3-Tg corresponding to the saturation binding: max = 768 ± 138 pmol Tg per milligram of LPO as well as the 35SO3-Tg equilibrium dissociation constant: Kd = 1.758 x 10–6 M.
Competition experiments by Tyr-S or Tyr-S-containing peptides (NIFEY-S, CCK8-S) were performed. Increasing amounts of each competitor were added to 35SO3-Tg solution (1.5 μM) and incubated with immobilized LPO or BSA. Fig. 4A displays competition characteristics. The concentration of competitor that shifted 50% of the specific 35SO3-Tg bound (IC50) was determined, and its equilibrium dissociation constant (Ki) was calculated using the Cheng and Prusoff equation (40): Ki = IC50/1+([L]/Kd), ([L]= concentration of free Tg). NIFEY-S was the best competitor (Ki = 1.95 x 10–4 M), followed by CCK8-S (Ki = 8.14 x 10–4 M) and Tyr-S (Ki = 19.1 x 10–4 M). Thus the peptide with a Tg-like sequence (NIFEY-S) was the best candidate to compete with 35SO3-Tg. When we used nonsulfated competitors, the shifting was always lower (Fig. 4B), suggesting that the two peptides and tyrosine also compete with Tg binding to LPO but with a lower efficacy, most likely for a specific Tyr site. Note that it was not possible to use higher concentrations of these compounds because of solubilization problems. All together, these data suggest that Tg was bound to peroxidase via Tyr-S linkage, and via Tyr linkage but more weakly.
Specific binding of iodinated 35SO3-Tg to LPO
Among the experiments that we performed to study the iodination of Tyr-S, we have obtained a Tg iodinated only by the GGO system, suggesting Tyr-S might be iodinated before binding to peroxidase. To determine whether this is true, 35SO3-Tg was iodinated in vitro with KI in conditions in which no coupling reaction could occur, i.e. under only H2O2 oxidation without LPO (see further in Fig. 7). The iodinated 35SO3-Tg was then incubated with immobilized LPO or BSA in conditions described above. As observed for noniodinated Tg, the specific binding of 35SO3-Tg to LPO tended toward a plateau (Fig. 5). The Scatchard representation gave the max = 1711 ± 60 pmol Tg per milligram of LPO, and the 35SO3-Tg equilibrium dissociation constant: Kd = 1.668 x 10–6 M. Higher amounts of Tg were necessary to reach binding saturation, suggesting a decrease in the number of Tyr-S available to bind LPO after iodination. However, the similar Kd value obtained for 35SO3-Tg noniodinated (1.758 x 10–6 M) and 35SO3-Tg iodinated via GGO system (1.668 x 10–6 M) suggested that iodinated Tg binds LPO via Tyr-S linkage, and consequently Tyr-S could not be iodinated by the GGO system only. It thus seems that peroxidase activity was necessary to iodinate Tyr-S.
Role of Tyr-S in the iodination and coupling reactions
To understand the process of Tyr-S binding in relationship with its iodination status, we studied the iodination and coupling reactions in different conditions. One of the criteria for the coupling reaction to occur would be the presence of tyrosines that are both sulfated and iodinated. Therefore, we looked for intermediate compounds such as MIT-sulfate or more importantly DIT-sulfate (DIT-S), because the acceptor iodotyrosine is always a DIT and we have previously shown that the main hormonogenic site contains a sulfated acceptor tyrosine (Tyr5-S) (23).
Iodination of Tg and coupling reactions with soluble LPO.
First we investigated the presence of SIT in the usual conditions of hormonosynthesis (1KIGO-1LPO). After iodination in vitro of 35SO3-Tg with KI and Tg with K125I, samples were submitted to enzymatic hydrolysis. Hydrolysates were analyzed using TLC after Dowex 50WX2 separation. By comparison with standard (Fig. 1), only very weak spots at Rf 0.44 (Fig. 6) were detectable with both 35S- and 125I-labeling, indicating that only traces of the corresponding SIT were present after Tg iodination.
Iodination of Tg and coupling reaction with immobilized LPO.
These experiments were performed in the conditions used to study specific Tyr-S binding. Immobilized LPO or BSA were incubated with Tg, [125I]iodide, and the GGO system (1KIGO), and electrophoresis of the media was then carried out in reducing conditions. After the coupling reaction in the presence of LPO, in addition to the specific band for 125I-Tg subunit, we observed other bands of higher molecular weight (Fig. 7A, lanes 3 and 4). In contrast, in the presence of BSA, only the band specific for 125I-Tg subunit could be detected (Fig. 7A, lanes 1 and 2). An experiment performed with 35SO3-Tg showed similar profiles with BSA (Fig. 7B, lanes 1 and 2), but after coupling with LPO (Fig. 7B, lanes 3 and 4), the Tg subunit signal was decreased, resulting in an increase in the higher molecular weight species. Media and membranes were separately treated with proteolytic enzymes. RP-HPLC analyses of media showed mainly MIT, but no thyroid hormones in the presence of BSA only (Fig. 8A), whereas MIT, DIT, T3, and T4 were observed in the presence of LPO (Fig. 8B). Hydrolysates from immobilized LPO-125I-Tg complex did not show T3 or T4 (Fig. 8C). TLC analysis of the first hydrophilic fractions of these last two RP-HPLC showed a spot at Rf 0.44 for media (Fig. 9A, lane 1) and for membranes (Fig. 9B, lane 2). This spot was specific for SIT because it was already observed in the same RP-HPLC fractions of standard (Fig. 1B). No iodoamino acids were present in the same portion of RP-HPLC of Tg iodinated without LPO (not shown), suggesting that Tyr-S was not iodinated with the GGO system. The 0.44 SIT was not detected with 35SO3-Tg, probably because the labeling was not sufficient to allow clear detection. SIT could be MIT-sulfate, DIT-S, or a nonfully hydrolyzed peptide released from LPO-Tg complex.
Role of Tyr-S: iodination of Tg peptides, NIFEY-S, NIFEY, and PY130
To better understand the role of Tyr-S in the iodination and coupling reactions, we iodinated NIFEY-S for comparison with either NIFEY or PY130, the peptide containing the nonsulfated tyrosine donor. NIFEY-S turned out to be poorly iodinated with the GGO system only (less than 2% of total iodide); on the other hand, using the same system, we found that the nonsulfated form of the same peptide was 3-fold more iodinated. When soluble LPO was added (1KIGO-1LPO), 1 nmol NIFEY-S incorporated 24 pmol [125I]iodine, whereas 1 nmol NIFEY or PY130 incorporated 15-fold more (Table 1). Different conditions of iodide, GGO system, and LPO concentrations were checked. Iodine incorporation into NIFEY-S was highly dependent on LPO concentration but not on iodide and H2O2 concentrations (Table 1). Comparatively, NIFEY and PY130 iodination was poorly dependent on iodide and H2O2 concentrations as well as on LPO concentration. Iodine incorporations into PY130 or NIFEY were similar (Table 1), showing that nonsulfated tyrosines had the same reactivity toward LPO, but that the reactivity of Tyr-S was different. Assays were also carried out with immobilized LPO (1KIGO-20LPO conditions). As in the soluble system, NIFEY-S was less iodinated than NIFEY or PY130 (49% vs. 95% or 96% of iodine incorporated into peptides), but the relative amount of iodide bound to membrane was higher for NIFEY-S than for NIFEY or PY130 (Table 2), suggesting that the major part of nonsulfated peptides were released from LPO after iodination. In the case of NIFEY-S, a relatively high proportion (22%, i.e. 60% of membrane-bound 125I-iodine) remained linked to membrane after proteolysis, showing that the iodinated NIFEY-S-LPO complex was only partly digested by proteolytic enzymes.
To study the coupling reaction between the Tyr acceptor (sulfated Tyr5 of NIFEY-S) and the Tyr donor (Tyr130 of PY130), NIFEY-S and PY130 were iodinated together. The weak concentrations of KI and glucose oxidase were limiting for iodination, whereas LPO concentration did not modify the incorporation of iodine (Table 1). In the 4KIGO-4LPO conditions only, the amount of iodine incorporated into the two peptides corresponded to the sum of the amounts incorporated into each peptide. With immobilized LPO, a low amount of iodinated peptides remained linked to LPO by comparison with NIFEY-S (Table 2), showing that, with PY130, either less NIFEY-S was bound to LPO or the coupling reaction occurred. However, using RP-HPLC analysis, only 1.5 ± 0.12 pmol T4 per nanomole of NIFEY-S was detected, showing that the coupling reaction occurred at low levels even with soluble LPO.
Samples from the first two conditions of iodination with LPO (Table 1) were hydrolyzed and analyzed by TLC. Spots corresponding to the compound at Rf 0.44 as well as a MIT and DIT migration front appeared clearly in NIFEY-S hydrolysates for high amounts of LPO (Fig. 10, lane 2), confirming that Tyr5-S iodination was dependent on LPO amount. The spot at Rf 0.44, which was not visible in assays with PY130 (Fig. 10, lanes 3 and 4), appeared also to be specific for Tyr-S. With immobilized LPO, 0.44 SIT was not visible in media (not shown) and was weakly detectable (Fig. 10, lane 7) in the membrane digest, probably because the NIFEY-S-LPO complex was not hydrolyzed by proteases and remained linked to membrane, as also observed in Table 2. With high concentrations of LPO, iodination of NIFEY-S with PY130 showed weak amounts of 0.44 SIT (Fig. 10, lanes 6 and 8), suggesting that when Tyr5-S and Tyr130 were present together, less Tyr5-S was bound to LPO.
Discussion
In previous studies, with either cultured thyrocytes or in vitro experiments for hormonosynthesis, we had shown that thyroid hormone synthesis decreased when tyrosine sulfation decreased (22) and most importantly that Tyr5, the major hormonogenic tyrosine, was sulfated (23). In this paper, using in vitro experimental procedures, we now report the function of Tyr-S in thyroid hormonosynthesis. We provide data indicating specific binding of Tg to LPO, which involves Tyr-S linkage, and the necessary presence of peroxidase to iodinate Tyr-S after binding to LPO. Weak amounts of SIT were detected in the peroxidase-Tg complex that represent short-lived intermediate compounds of hormone synthesis. These data, in addition to those obtained for the iodination of hormonogenic sulfated Tyr5 acceptor and Tyr130 donor, allow us to propose a mechanism of hormonosynthesis that completes the previous ones.
Using either cellular TPO (cultured thyrocytes reorganized into follicles) or LPO (in vitro experiments), we have previously obtained similar results for hormonosynthesis (22). These results agree with several reports that TPO and LPO have a similar efficiency in catalyzing iodination and coupling reactions (3, 9, 41) with the same preferential formation of DIT in Tg (42). Although some biochemical differences exist (41), particularly the production of HOI by LPO only (41, 43), we chose to perform these in vitro studies with LPO rather than TPO because native TPO is difficult to obtain in sufficient amounts and soluble recombinant TPO is known to be involved in iodination reaction (44) but have not been tested in coupling reaction.
We demonstrate that Tyr-S and the two peptides NIFEY-S and CCK8-S are good competitors for the binding of noniodinated 35SO3-Tg, showing the specificity of Tyr-S linkage. NIFEY-S was more efficient than CCK8-S, which was more efficient than Tyr-S, suggesting that the peptide sequence surrounding the Tyr5-S is also critical for the binding. Protein-protein interactions via the Tyr-S linkage (27, 28) have been described in hormone-receptor recognition (29, 30), inflammatory leukocyte adhesion (31), hemostasis (31, 32), and chemokine signaling (31, 33). However this is the first time that Tyr-S linkage was found to be involved in the formation and the catalytic reaction of a peroxidase-substrate complex. Tyr and nonsulfated peptides were also able to compete with Tg but with a lower efficacy, suggesting either that Tyr-S and Tyr could have different levels of affinity for the same site or, most likely, that LPO contains an additional specific site for nonsulfated tyrosine, as previously described (24, 25). The two-site hypothesis is also supported by 1) the differences in iodination of sulfated and nonsulfated Tg peptides (Table 1 and 2) and 2) the fact that the sulfate group confers to Tyr chemical properties probably requiring a specific structure in LPO. Indeed Tyr-S residue could be stabilized by forming conjugate ion pairs with the positive charges of amino acids located in the LPO sequence, such as the guanidino group of Arg residue (45). Thus two different tyrosine sites appear to be present on LPO: one for Tyr-S and one for Tyr.
When 35SO3-Tg was iodinated without LPO and then bound to LPO, the number of picomoles of Tg corresponding to the saturation of sites was 3-fold higher than with noniodinated 35SO3-Tg. A decrease in Tyr-S available for binding could be caused by modifications of Tg during its iodination because tyrosine iodination is known to result in alterations of the Tg structure and of its immunological and biochemical properties (46). That the Kd value of Tg-peroxidase interaction was not changed by the iodination of Tg (without LPO) before the binding, and that iodination of NIFEY-S was mainly dependent on the concentration of LPO, could indicate that Tg-bound Tyr-S was not iodinated in the absence of LPO. We assume that, in this case, the sulfate group might prevent the attachment of oxidized iodide to the aromatic ring of tyrosine, because the iodination of NIFEY-S was lower than that of the same nonsulfated peptide, and maybe due to a desulfation of NIFEY-S. Similar results were obtained by chemical iodination (47).
Then, to characterize Tyr5-S iodination, we looked for SIT that should be transiently present during the process. We were able to detect only traces of SIT (Rf 0.44) in enzymatically iodinated Tg. However, this compound was recovered in higher amounts in hydrolysates of immobilized LPO-Tg complex or of LPO-iodinated NIFEY-S, likely because the relative amount of LPO was higher. This SIT could be a nonfully hydrolyzed peptide from the Tyr-S-LPO complex. In immobilized LPO experiments with NIFEY-S, SIT was weakly present but more than 20% of iodine remained on the membrane fraction after proteolytic digestion. This finding supports the hypothesis that DIT-S at position 5 remains linked to LPO until the coupling reaction occurs, becoming the side chain undergoing the hormonogenic coupling reaction. On the other hand, Schepky et al. (47) showed that Tyr-S-containing peptide was more rapidly iodinated after the release of sulfate by acid hydrolysis. Based on these observations, we made the assumption that the binding of Tg to LPO by Tyr5-S linkage could modify the charges of the sulfate group and, consequently, the charges of the aromatic ring, and the Tyr5-S acceptor could become more reactive for a rapid iodination. Desulfation might occur when complete iodination is achieved just before the coupling reaction, DIT-S becoming a DIT (iodophenoxy anion) and thus making it available for the coupling reaction. This hypothesis is in agreement with pKa studies of iodotyrosines in hormonosynthesis, suggesting that the coupling reaction requires phenoxy anion formation (18). It is also in agreement with other results showing that iodination and coupling reactions occur simultaneously (12) and that iodide is essential for the coupling reaction (48, 49). Further studies on the NIFEY-S- or Tg-peroxidase complex will be needed to confirm this hypothesis.
The rate of iodination was dependent on LPO concentration only for the sulfated peptide, and SIT could be clearly observed after either NIFEY-S iodination with high concentration of LPO or Tg iodination with immobilized LPO. However, when NIFEY-S and PY130 were iodinated together with a high concentration of LPO, we did not observe the amount of SIT corresponding to NIFEY-S only. Moreover, immobilized LPO experiments showed that a low amount of iodinated NIFEY-S remained bound to membrane when PY130 was present. Consequently it appears that the two tyrosine sites may cooperate to equilibrate and control acceptor and donor iodination. This assumption is in line with previous reports on the different reactivities of tyrosyl residues toward iodine: iodination would proceed in a sequential order controlled by the native structure of Tg (15, 16, 17). In addition the coupling reaction with NIFEY-S and PY130 was very low, probably due to the small chance of having, in LPO, two DIT in close proximity. This finding is not surprising because the three-dimensional structure of Tg is necessary for the coupling reaction (50).
All these data allow us to propose a mechanism of thyroid hormonosynthesis that completes the previous ones. Taking into account that 1) a specific binding of Tg to LPO occurs via Tyr-S linkage, 2) LPO is necessary to iodinate Tyr-S, and 3) iodination of Tyr-S is dependent on the amount of LPO, Tyr5-S could first bind to peroxidase, allowing Tg to be in the right position to bind the Tyr donor. This agrees with the fact that TPO is able to bind Tyr before its activation by H2O2 (26) and also with the report of Xiao et al. (42) showing that preferential formation of DIT involves interaction between Tg and peroxidase. Thus the peroxidase could bind only the two types of tyrosine involved in hormonosynthesis to allow tyrosine iodination likely by [EOI]– complex (25) and to achieve the coupling reaction. Because a great part of NIFEY-S remained bound to immobilized LPO, iodinated Tyr5-S that becomes the DIT acceptor could remain bound to its site until the coupling occurs. On the other hand, the major part of PY130 containing the nonsulfated Tyr that becomes the DIT donor was released from peroxidase after iodination and could be waiting for the coupling reaction when involved in the Tg structure. The other nonsulfated tyrosines, not involved in hormone synthesis, could be iodinated by a free intermediate such as HOI, which is the active iodinating intermediate at physiological pH (41, 43) and probably corresponds to the diffusible iodinating compound in LPO-catalyzed iodination (51). The release of DIT donor from peroxidase as well as the possible action of HOI, particularly present with LPO (41), could explain that, in our experiments, soluble LPO led to a higher incorporation of iodide on NIFEY or PY130 than on NIFEY-S remaining partly bound to LPO.
Concerning Tg iodination in the absence of LPO, H2O2 appeared to be sufficient to oxidize iodide, but only MIT could be observed, suggesting a lower iodination efficiency. Therefore, in the absence of peroxidase, iodination could occur due to a Fenton-type reaction initiated by traces of transition-metal ions in the reaction, involving hydroxyl radical formation and subsequent HOI formation (52).
In conclusion, our results provide new information on the thyroid hormone synthesis process: the sulfate group, present on tyrosine acceptor in hormonogenic site of Tg, seems to be a necessary partner to bind Tg to peroxidase, likely stabilizing this interaction and allowing the formation of DIT acceptor and thus becoming a potential regulatory element of thyroid hormone synthesis. Further studies should be performed to confirm the binding of Tyr5-S with TPO, using immobilized TPO or more physiological conditions: membrane-bound TPO expressed at the apical surface of filter-cultured thyroid cell monolayers (53). Tyrosylprotein sulfotransferase (TPST)-deficient mice (54) should also be considered a good model serving as a negative control, although the phenotypes observed with TPST1 or TPST2 deficiency do not correspond to the characteristic phenotypes of severe hypothyroidism as early as at birth (55). To perform such a study maybe both TPST1 and TPST2 genes would need to be knocked out. This work should also be followed by structural studies of the Tg-peroxidase complex to locate the sites of Tyr-S and Tyr binding on LPO and TPO.
Acknowledgments
The authors thank Prof. P. Carayon for comments, Dr. C. Penel for reading the manuscript and helpful discussions, Dr. B. Mallet for advice about Fenton reaction and HPLC analyses, and Mrs. J. Barbaria for HPLC experiments. We also thank Dr. G. Herbette (Spectropole, Universités d’Aix-Marseille, Faculté de Saint Jérme) for help in NMR analysis and interpretation and Dr. P. H. Rolland for the gift of thyroids from control experimental pigs.
Footnotes
This study was supported by Institut National de la Santé et de la Recherche Médicale (Unité 555), Centre National de la Recherche Scientifique (SDI 401555), Institut Fédératif de Recherche 125, and the Université de la Méditerranée.
Abbreviations: CCK8, Cholecystokinin; DIT, diiodotyrosine; DIT-S, DIT-sulfate; EN, ethyl alcohol-ammonium 1N; FA, formic acid; GGO, glucose-glucose oxidase; HOI, hypoiodous acid; LPO, lactoperoxidase; MIT, monoiodotyrosine; NMR, nuclear magnetic resonance; Rf, retention factor; RP-HPLC, reverse phase high performance liquid chromatography; SIT, sulfated iodotyrosine; Tg, thyroglobulin; TLC, thin layer chromatography; TPO, thyroperoxidase; TPST, tyrosylprotein sulfotransferase; Tyr-S, sulfated tyrosine.
References
Ekholm R 1990 Biosynthesis of thyroid hormones. Int Rev Cytol 120:243–288
Vassart G, Dumont JE 1992 The thyrotropin receptor and the regulation of thyrocyte function and growth. Endocr Rev 13:596–611
Kopp P 2005 Thyroid hormone synthesis. In: Brawerman LE, Utiger RD, eds. Werner, Ingbar’s the thyroid: a fundamental, clinical text. 9th ed. Philadelphia: Lippincott Williams, Wilkins; 52–76
Arvan P, Di Jeso B 2005 Thyroglobulin structure, function, and biosynthesis. In: Brawerman LE, Utiger RD, eds. Werner, Ingbar’s the thyroid: a fundamental, clinical text. 9th ed. Philadelphia: Lippincott Williams, Wilkins; 77–95
Carrasco N 2005 Thyroid iodine transport. Brawerman LE, Utiger RD, eds. Werner, Ingbar’s the thyroid: a fundamental, clinical text. 9th ed. Philadelphia: Lippincott Williams, Wilkins; 37–52
Nilsson M, Bjorkman U, Ekholm R, Ericson LE 1990 Iodide transport in primary cultured thyroid follicle cells: evidence of a TSH-regulated channel mediating iodide efflux selectively across the apical domain of the plasma membrane. Eur J Cell Biol 52:270–281
Gruffat D, Gonzalvez S, Chambard M, Mauchamp J, Chabaud O 1991 Long-term iodination of thyroglobulin by porcine thyroid cells cultured in porous-bottomed culture chambers: regulation by thyrotrophin. J Endocrinol 128:51–61
Bjorkman U, Ekholm R 1988 Accelerated exocytosis and H2O2 generation in isolated thyroid follicles enhance protein iodination. Endocrinology 122:488–494
Nunez J, Pommier J 1982 Formation of thyroid hormones. Vitam Horm 39:175–229
Lamas L, Dorris ML, and Taurog A 1972 Evidence for a catalytic role for thyroid peroxidase in the conversion of diiodotyrosine to thyroxine. Endocrinology 90:1417–1426
Gavaret JM, Nunez J, Cahnmann HJ 1980 Formation of dehydroalanine residues during thyroid hormone synthesis in thyroglobulin. J Biol Chem 255:5281–5285
Taurog A, Dorris LM, Doerge RD 1996 Mechanism of simultaneous iodination and coupling catalyzed by thyroid peroxidase. Arch Biochem Biophys 330:24–32
Marriq C, Lejeune PJ, Venot N, Vinet L 1991 Hormone formation in the isolated fragment 1–171 of human thyroglobulin involves the couple tyrosine 5 and tyrosine 130. Mol Cell Endocrinol 81:155–164
Dunn AD, Corsi CM, Myers HE, Dunn JT 1998 Tyrosine 130 is an important outer ring donor for thyroxine formation in thyroglobulin. J Biol Chem 273:25223–25229
Edelhoch H, Lippoldt RE 1962 The properties of thyroglobulin IX. The molecular properties of iodinated thyroglobulin. J Biol Chem 237:2788–2794
Gavaret JM, Deme D, Nunez J, Salvatore G 1977 Sequential reactivity of tyrosyl residues of thyroglobulin upon iodination catalyzed by thyroid peroxidase. J Biol Chem 252:3281–3285
Lamas L, Anderson PC, Fox JW, Dunn JT 1989 Consensus sequences for early iodination and hormonogenesis in human thyroglobulin. J Biol Chem 264:13541–13545
De Viljder JJM, den Hartog MT 1998 Anionic iodotyrosine residues are required for iodothyronine synthesis. Eur J Endocrinol 138:227–231
Hortin G, Folz R, Gordon JI, Strauss AW 1986 Characterization of sites of tyrosine sulfation in proteins and criteria for predicting their occurrence. Biochem Biophys Res Commun 141:326–333
Rosenquist GL, Nicholas Jr HB 1993 Analysis of sequence requirements for protein tyrosine sulfation. Protein Sci 2:215–222
Bundgaard JR, Vuust J, Rehfeld JF 1997 New consensus features for tyrosine O-sulfation determined by mutational analysis. J Biol Chem 272:21700–21705
Nlend MC, Cauvi D, Venot N, Chabaud O 1999 Sulfated tyrosines of thyroglobulin are involved in thyroid hormone synthesis. Biochem Biophys Res Commun 262:193–197
Venot N, Nlend MC, Cauvi D, Chabaud O 2002 The hormonogenic tyrosine 5 of porcine thyroglobulin is sulfated. Biochem Biophys Res Commun 298:193–198
Casella L, Gullotti M, Poli S, Bonfa M, Ferrari RP, Marchesini A 1991 Spectroscopic and binding studies on the stereoselective interaction of tyrosine with horseradish peroxidase and lactoperoxidase. Biochem J 279:245–250
Sun W, Dunford BH 1993 Kinetics and mechanism of the peroxidase-catalyzed iodination of tyrosine. Biochemistry 32:1324–1331
Yamamoto K, Degroot LJ 1982 Affinity chromatography of thyroid peroxidase using tyrosine coupled to agarose. J Biochem (Tokyo) 91:775–782
Niehrs C, Beisswanger R, Huttner WB 1994 Protein tyrosine sulfation, 1993—an update. Chem Biol Interact 92:257–271
Strott CA 2002 Sulfonation and molecular action. Endocr Rev 23:703–732
Gigoux V, Escrieut C, Silvente-Poirot S, Maigret B, Gouilleux L, Fehrentz JA, Gully D, Moroder L, Vaysse N, Fourny D 1998 Met-195 of the cholecystokinin-A receptor interacts with the sulfated tyrosine of cholecystokinin and is crucial for receptor transition to high affinity state. J Biol Chem 273:14380–14386
Costagliola S, Panneels V, Bonomi M, Koch J, Many MC, Smits G, Vassart G 2002 Tyrosine sulfation is required for agonist recognition by glycoprotein hormone receptors. EMBO J 21:504–513
Kehoe JW, Bertozzi CR 2000 Tyrosine sulfation: a modulator of extracellular protein-protein interactions. Chem Biol 7:R57–R61
Ward CM, Andrews RK, Smith AI, Berndt MC 1996 Mocarhagin, a novel cobra venom metalloproteinase, cleaves the platelet von Willebrand factor receptor glycoprotein Ib. Identification of the sulfated tyrosine/anionic sequence Tyr-276-Glu-282 of glycoprotein Ib as a binding site for von Willebrand factor and -thrombin. Biochemistry 35:4929–4938
Farzan M, Chung S, Li W, Vasilieva N, Wright PL, Schnitzler CE, Marchione RJ, Gerard C, Gerard NP, Sodroski J, Choe H 2002 Tyrosine-sulfated peptides functionally reconstitute a CCR5 variant lacking a critical amino-terminal region. J Biol Chem 277:40397–40402
Chambard M, Mauchamp J, Chabaud O 1987 Synthesis and apical and basolateral secretion of thyroglobulin by thyroid cell monolayers on permeable substrate: modulation by TSH. J Cell Physiol 133:37–45
Nlend MC, Cauvi D, Venot N, Chabaud O 1999 Thyrotropin regulates tyrosine sulfation of thyroglobulin. Eur J Endocrinol 141:61–69
Reitz HC, Ferrel RE, Fraenkel-Conrat H, Olcott HS 1946 Action of sulfating agents on proteins and model substances. I. Concentrated sulfuric acid. J Am Chem Soc 68:1024–1031
Futaki S, Taike T, Yagami T, Ogawa T, Akita T, Kitagawa K 1990 Use of dimethylformamide-sulphur trioxide complex as a sulphating agent of tyrosine. J Chem Soc Perkin Trans 1:1739–1744
Gruffat D, Venot N, Marriq C, Chabaud O 1992 Thyroid hormone synthesis in thyroglobulin secreted by porcine thyroid cells cultured on porous bottom chambers. Endocrinology 131:2921–2927
Blode H, Heinrich T, Diringer H 1990 A quantitative assay for tyrosine sulfation and tyrosine phosphorylation in peptides. Biol Chem Hoppe-Seyler 371:145–151
Cheng Y, Prusoff WH 1973 Relationship between the inhibition constant (K1) and the concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic reaction. Biochem Pharmacol 22:3099–3108
Magnusson RP, Taurog A, Dorris ML 1984 Mechanisms of thyroid peroxidase- and lactoperoxidase-catalyzed reactions involving iodide. J Biol Chem 259:13783–13790
Xiao S, Dorris ML, Rawitch AB, Taurog A 1996 Selectivity in tyrosyl iodination sites in human thyroglobulin. Arch Biochem Biophys 334:284–294
Dunford HB, Ralston IM 1983 On the mechanism of iodination of tyrosine. Biochem Biophys Res Commun 116:639–643
Gut P, Grennan Jones F, Sullivan A, Ziemnicka Smith S, Jaskolski D, Furmaniak J, Rees Smith B 2000 Recombinant human thyroid peroxidase produced in insect cells has similar properties to native human thyroid peroxidase. Thyroid 10:543–550
Yagami T, Kitagawa K, Aida C, Fujiwara H, Futaki S 2000 Stabilization of a tyrosine O-sulfate residue by a cationic functional group: formation of a conjugate acid-base pair. J Pept Res 56:239–249
Gardas A, Domek H 1993 Iodine induced alteration in immunological and biochemical properties of thyroglobulin. Acta Biochim Pol 40:237–240
Schepky AG, Schmidt AM, Schmidt T, Schulz-Knappe P, Forssmann WG 1994 Determination of sulfated peptides by differential iodination. Biol Chem Hoppe-Seyler 375:201–203
Dème D, Pommier J, Nunez J 1976 Kinetics of thyroglobulin iodination and of hormone synthesis catalyzed by thyroid peroxidase. Role of iodide in the coupling reaction. Eur J Biochem 70:435–440
Virion A, Dème D, Pommier J, Nunez J 1981 The role of iodide and of free diiodotyrosine in enzymatic and non-enzymatic thyroid hormone synthesis. Eur J Biochem 118:239–245
Dunn JT, Dunn AD 1999 The importance of thyroglobulin structure for thyroid hormone synthesis. Biochimie (Paris) 81:505–509
Huber RE, Edwards LA, Carne TJ 1989 Studies on the mechanism of the iodination of tyrosine by lactoperoxidase. J Biol Chem 264:1381–1386
Klebanoff SJ 1982 The iron-H2O2-iodide cytotoxic system. J Exp Med 156:1262–1267
Penel C, Gruffat D, Alquier C, Benoliel AM, Chabaud O 1998 Thyrotropin chronically regulates the pool of thyroperoxidase and its intracellular distribution: a quantitative confocal microscopy study. J Cell Physiol 174:160–169
Moore KL 2003 The biology and enzymology of protein tyrosine O-sulfation. J Biol Chem 278:24243–24246
Vulsma T, de Vijlder JJM 2005 Genetic defects causing hypothyroidism In: Brawerman LE, Utiger RD, eds. Werner, Ingbar’s the thyroid: a fundamental, clinical text. 9th ed. Philadelphia: Lippincott Williams, Wilkins; 714–730(Marie-Christine Nlend, Da)