| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute of Interdisciplinary Research (J.V.S., C.M., J.E.D., C.E.), University of Brussels, School of Medicine, 1070 Brussels, Belgium; and Institut Bordet (D.D., P.L.), University of Brussels, 1000 Brussels, Belgium
Address all correspondence and requests for reprints to: Christophe Erneux, Institute of Interdisciplinary Research, Campus Erasme Building C, 808 Route de Lennik, 1070 Brussels, Belgium. E-mail: cerneux{at}ulb.ac.be.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Objective: We investigated the formation of inositol phosphates by HPLC techniques in human thyroid slices to separate the inositol phosphate isomers.
Results: Ins(1,4,5)P3, inositol 1,3,4-trisphosphate, and inositol 1,3,4,5-tetrakisphosphate were increased after TSH stimulation. The effect of TSH in human thyroid cells was reproduced by recombinant TSH and prevented by antibodies blocking the TSH receptor. Thyroid-stimulating antibodies at concentrations eliciting a cAMP response equivalent to TSH failed to stimulate inositol phosphate generation.
Conclusions: TSH, but not thyroid-stimulating antibodies, activates both cAMP and the phospholipase C cascade in human thyroid as now demonstrated by an increase in Ins(1,4,5)P3 and its inositol phosphate metabolites. Therefore, this effect cannot be extrapolated to the FRTL-5 cell line. The apparent discrepancy may be due to a difference between species (human vs. rat) or to the loss of the fresh tissue properties in a cell line. The dual effect of TSH in human cells, through cAMP on secretion of thyroid hormones and through the diacylglycerol, Ins(1,4,5)P3 Ca2+ pathway on thyroid hormone synthesis, implies the possible separation of these effects in thyroid disease.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the thyroid cell, most effects of TSH are mediated by its activation of adenylate cyclase and cAMP generation (4, 5). However, in 1987 Laurent et al. (6) showed that, as suggested by Freinkel’s indirect data (1), TSH stimulates the generation of Ins(1, 4, 5)P3 and other inositol phosphates in human thyroid slices. This effect required a 10 times higher hormone concentration than needed for enhancing cAMP accumulation (6, 7). Parallel results were obtained in pig thyroid slices (8, 9). However, no similar effects were obtained in dog (10, 11, 12) and rat thyroid slices (13). In FRTL-5 cells, the concentrations of TSH used to stimulate PtdIns(4, 5)P2 hydrolysis were 1000 times higher than those required to enhance maximally cAMP (14, 15).
The difference between dog and human thyroid fitted nicely with physiological data on thyroid hormone synthesis. Iodide oxidation and binding to thyroglobulin, as well as the oxidative coupling of iodotyrosines into iodothyronines are catalyzed by thyroperoxidase using H2O2 as oxidizing reagent. H2O2, which is limiting these processes (16), is generated by thyroid oxidase (17, 18). It is remarkable that, in human cells, in which TSH stimulates both the cAMP and the Ins(1, 4, 5)P3 cascades, H2O2 generation is only stimulated by the second messengers, Ca2+ and diacylglycerol of the second cascade, whereas in dog thyroid cells, in which TSH only activates the cAMP cascade (13, 16, 19, 20), H2O2 generation is also activated by cAMP. In support of these findings, it was later found that in CHO cells transfected with the human TSH receptor, TSH again activated both cascades, the PtdIns(4, 5)P2 cascade requiring 10 times higher concentrations of TSH compared with the cAMP cascade (21). With this background, Singh et al. (22) remarked that the work of Laurent et al. did not make use of HPLC, i.e. did not use later developed and fully proved methodology. Using HPLC and FRTL-5 cells as a model, they showed that TSH did not activate the Ins(1, 4, 5)P3 formation in these cells and questioned whether the TSH effect occurs in human thyroid cells (22). As stated by Singh et al.: "for the moment the burden of the proof therefore passes to workers who wish to continue to promote the view that activated TSH receptor activates PLC catalyzed PtdIns(4, 5)P2 hydrolysis" (22).
In this study, we investigated the formation of inositol phosphates by HPLC techniques in human thyroid slices. We show that Ins(1, 4, 5)P3, inositol 1,3,4-trisphosphate [Ins(1, 3, 4)P3], and inositol 1,3,4,5-tetrakisphosphate (InsP4) levels were increased by TSH but not by thyroid-stimulating antibodies (TSAbs). These effects were reproduced in CHO cells transfected with the human TSH receptor. These effects were also observed using recombinant TSH, and abolished by TSH receptor-blocking antibodies.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Bovine TSH was from Sigma-Aldrich (Bornem, Belgium). Bacterially produced human TSH (Thyrogen) used in vivo in human patients was obtained from Genzyme Europe (Naarden, The Netherlands). Partisphere SAX column, 25 cm x 4.6 mm was from Whatmann VWR (Leuven, Belgium). [3H]Inositol was from Amersham, Biosciences Benelux (Rosendaal, The Netherlands).
Thyroid tissue
Human thyroid tissue was obtained from nine patients undergoing partial or total thyroidectomy for resection of solitary cold nodules or multinodular goiters. Serum TSH was normal. Only the healthy, normal-looking, nonnodular tissue verified by pathological examination was used within 30 min of surgical removal. The protocol had been approved by the Ethics Committee of the Hospital, and the patients routinely gave their informed consent for use of pathological material. The tissue was sliced at room temperature with a Stadie-Riggs microtome (Arthur Thomas, Philadelphia, PA). Thyroid slices (80–100 mg wet weight tissue) were incubated at 37 C in 2 ml Krebs-Ringer bicarbonate buffer, pH 7.4, supplemented with glucose 8 mM and 0.5 g/liter BSA, and under a gas phase of 95% O2-5% CO2.
CHO-transfected cells
We used a CHO-K1 cell line stably expressing the human TSH receptor referred to as JP19 cells. Details of construction of the human TSH receptor expression vector and transfection, have been previously reported (23). For comparison, cells transfected with an empty vector (JP02) were used.
Inositol phosphates measurements
In tissues. For HPLC, the thyroid slices were preincubated 4 h in the presence of 180 µCi/ml [3H]inositol. The slices were then lightly dried on paper and transferred to fresh unlabeled incubation medium to which 10 mM LiCl was added after 15 min. After a further 5 min, the test agents were added for the test incubation (see figure legends). The incubation was stopped by rapid immersion of the slices in 3% ice-cold perchloric acid and 125 µM phytic acid. After homogenization, followed by 30 min on ice, and centrifugation (15,000 x g, 10 min), the supernatant was collected. The pellet was washed with 1% ice cold perchloric acid and recentrifuged. The combined supernatants were neutralized by KOH (1.53 M) in the presence of HEPES buffer (0.75 M). After 30 min on ice, the samples were centrifuged and the supernatants were carefully aspirated (1.2 ml); 0.97 ml was analyzed by HPLC on Partisphere SAX column, 25 cm x 4.6 mm as reported before (24). Radioactivity was detected with an on-line detector from Raytest (Straubenhardt, Germany). The data were normalized with respect to the total radioactivity present in the lipid fraction, i.e. total PI. The identity of each peak was verified using labeled standards; [3H]Ins(1, 3, 4, 5)P4 and [3H]Ins(1, 3, 4)P3 were prepared using recombinant enzymes (25, 26).
For Dowex columns, the protocol of incubation was the same except that the amount of [3H]inositol was 40 µCi/ml. The principles of the protocol, after incubation, are the same as for HPLC, but the last supernatant was brought in between pH 8 and 9 with sodium tetraborate (5 mM)-EDTA (0.5 mM) and eluted in a stepwise fashion through an anion exchange column of AG1-X8 resin (formate form, 200–400 mesh; Bio-Rad, Watford, UK) (27). After homogenization of the thyroid slices and centrifugation, the pellet was dissolved in 1 M NaOH and counted as total PI. The results are expressed as counts per minute per 100 mg tissue.
In cells. For HPLC, 1.5 x 106 JP19 cells were seeded in 6-cm diameter dishes with 60 µCi/ml [3H]inositol. After labeling for 24 h, the cells were rinsed twice with Krebs-Ringer HEPES buffer (KRH) and preincubated 30 min in KRH buffer supplemented with 10 mM LiCl. The medium was then replaced by KRH + LiCl medium containing the test agent for 15 min. The cells were treated exactly as the tissue except that they did not need homogenization.
For Dowex columns, 0.5 x 106 JP19 cells were seeded in 3.5-cm diameter dishes with 60 µCi/ml [3H]inositol. The next day, the cells were incubated as those for HPLC measurements and treated as described above for tissue (see description, In tissues). The cell debris in the bottom of the dishes was dissolved in 1 M NaOH and counted as total PI. The results are expressed as counts per minute per dish.
cAMP measurements in tissue
The thyroid slices were preincubated for 1 h in 2 ml Krebs-Ringer bicarbonate buffer medium and transferred to fresh medium supplemented with 25 µM rolipram and the test agents under study. One hour later, the slices were dropped in 1 ml boiling water (5 min), homogenized, and centrifuged, and the supernatant was dried in a speed-Vac concentrator. cAMP was measured as reported in Ref.28 .
TSH receptor-blocking antibodies
Two mouse monoclonal TSH receptor-blocking antibodies have been obtained in our laboratory by Dr. S. Costagliola (IRIBHM, University of Brussels, Brussels, Belgium) (29, 30). They were shown to block the TSH binding to its receptor and are devoid of TSAb activity. A mixture of these two antibodies, each at 10 µg/ml, was added at the beginning of the incubation.
TSAbs
Two sera with high TSAb activity (nos. 1 and 2), from patients suffering from Graves’ disease, were obtained from the department of clinical biology from the local hospital.
Presentation of the results
Results are presented as one representative experiment of a minimum of two or more according to the tissue availability. When the results of different experiments were pooled, means, SD of the mean, and Student’s t values were calculated from the logarithms of the individual data. This procedure has been shown to normalize the distribution of metabolic variables in the thyroid (31). Results are expressed as antilogarithms of the means and of the mean minus or plus the SD of the means. When a representative experiment is shown, the data are presented as means of duplicates ± range. In the case of HPLC profiles of single injections, a table describing the numerical values of the chromatograms is presented.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In both human thyroid slices and CHO cells transfected with the human TSH receptor, TSH added for 15 min stimulated the formation of [3H]Ins(1, 4, 5)P3 as determined by anion exchange HPLC (Fig. 1
, A and B). In transfected cells, maximal production of [3H]Ins(1, 4, 5)P3 was reached after 15 min of TSH stimulation as determined previously (data not shown). The lack of sufficient material did not allow us to do complete kinetics in human thyroid slices followed by HPLC analysis, but such kinetics using Dowex columns have been published before (6). In the typical experiment shown in Fig. 1A, the increase in [3H]Ins(1, 4, 5)P3 was about 2-fold (Table 1). At the same time, TSH provoked the accumulation of a large amount of [3H]Ins(1, 3, 4)P3, which was nondetectable in nonstimulated cells. [3H]InsP2 and InsP1 levels were also increased in the presence of TSH (Fig. 1A and Table 1). [3H]Ins(1, 3, 4, 5)P4, which was nondetectable in the absence of TSH, could be measured in the presence of TSH in human thyroid slices (Fig. 1A).
|
|
As expected, the effect of TSH on [3H]Ins(1, 4, 5)P3 was not reproduced by forskolin, a direct activator of adenylate cyclase. ATP, by activating ATP receptors, present in human thyroid (7) and Gq, did increase the formation of [3H]inositol trisphosphate. This was shown both on Dowex columns and by HPLC (Tables 1
and 2 and Fig. 2). The HPLC data of this last experiment show that [3H]Ins(1, 4, 5)P3 was not stimulated in the presence of TSH presumably as it was rapidly converted into Ins(1, 3, 4)P3 via its phosphorylation by an active Ins(1, 4, 5)P3 3-kinase. As shown in this and in all our experiments, [3H]Ins(1, 3, 4)P3 was undetectable at basal level, but was always produced in the presence of TSH. Therefore, the overall production of both isomers was increased as shown also in the data obtained on Dowex columns (Table 2). Moreover in the data of Fig. 2, [3H]InsP2 levels were increased from 39,376 cpm to 252,938 cpm in the presence of TSH (6-fold), and InsP1 were increased from 249,342 cpm to 4,296,957 cpm (17-fold, see Table 1). These values are very much comparable to the data obtained on Dowex columns (Table 2).
|
|
The results obtained so far indicated that TSH stimulated [3H]Ins(1, 4, 5)P3 formation in human thyroid slices, that this effect was reproduced in CHO cells transfected with the human TSH receptor, and that it was not mimicked by forskolin. We wanted to test whether this effect could result from a contamination of the commercial TSH preparation, and therefore, have used purified recombinant TSH produced in bacteria. As shown in Fig. 3, recombinant TSH at 3 and 10 mU/ml stimulated [3H]Ins(1, 4, 5)P3 formation in CHO cells transfected with the human TSH receptor (an effect which could also be seen with the TSH preparation from Sigma-Aldrich). No effect could be seen with untransfected CHO cells. More generally, recombinant TSH stimulated inositol phosphates derived from the metabolism of [3H]Ins(1, 4, 5)P3: [3H]InsP1, InsP2, Ins(1, 3, 4)P3, and InsP4. Both [3H]Ins(1, 3, 4)P3 and InsP4 are undetectable in unstimulated cells.
|
shows the levels of [3H]inositol phosphates in human thyroid slices that had been stimulated by 3 mU/ml recombinant TSH in the presence and absence of TSH receptor-blocking antibodies. Recombinant TSH stimulated [3H]Ins(1, 4, 5)P3 formation, and this effect was prevented by the receptor-blocking antibodies. This further confirms that the increase in Ins(1, 4, 5)P3 formation was generated after an interaction between TSH and its receptor.
|
In previous work, using the Dowex method, TSAb, unlike TSH did not activate PLC in human thyroid slices (32). Figure 5A shows an HPLC profile of human thyroid slices simulated by two potent TSAbs. Although the stimulating antibodies had an effect of the same magnitude as TSH on cAMP accumulation (Fig. 5B), they did not stimulate [3H]Ins(1, 4, 5)P3 formation in contrast to TSH (Fig. 5A). In our experiments, nine thyroid tissues tested showed a stimulation of inositol phosphates generation by TSH.
|
After 1 h of incubation with TSH or ATP, the labeling of PI was increased in human thyroid slices. Results, obtained in seven different thyroids for TSH and four of these thyroids for ATP, are shown in Table 3. Forskolin 10 µM, a potent activator of the cAMP cascade but not of the PI cascade, was without effect on two thyroids (ratio forskolin over control was 0.98 and 1.01, respectively) in which TSH and ATP increased the labeling of PI.
|
). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
With regard to the generation of Ins(1, 4, 5)P3, the negative results of Singh et al. (22) on the action of TSH in FRTL-5 cells raised an interesting question; however, the study was questionable on several aspects. 1) It did not include measurements of the metabolite InsP1. 2) InsP2 was not enhanced in the experiments. 3) It concerned a thyroid cell line, the FRTL-5 cells, which is very different from the in vivo models (33) and in which only huge concentrations of an unpurified hormone had been reported to have an effect. At such concentrations, any effect could be due to a contaminant. 4) The results of Singh et al. (22) led to the suggestion of a totally new mechanism of inositol polyphosphate formation without activation of a PLC. The concept of an action of TSH on the PLC cascade in human thyroid cells remained questionable until proven by more recently developed methodology.
Therefore, we looked for the presence and eventual increase in Ins(1, 4, 5)P3 levels in human thyroid slices exposed to TSH, by HPLC techniques. Human thyroid slices represent, for short-term experiments, by far the best model of the thyroid in vivo; they conserve the tissue structure and they respond to TSH by increasing protein iodination and thyroid hormone secretion, i.e. the two most important early effects of the hormone (4, 5). In our hands and in this model, TSH enhances the generation of Ins(1, 4, 5)P3 and Ins(1, 3, 4)P3 and produces marked increases in InsP1 and InsP2 isomers. The effect we report in this study was obtained with much lower concentrations than those used in FRTL-5 cells (100 times less) and just above the level of those eliciting cAMP accumulation (6). It was obtained with impure commercial bovine TSH from Sigma-Aldrich as in this other work, but also with purified recombinant human TSH for use as injection in human patients. The effect was reproduced in CHO cells overexpressing the human TSH receptor. It was prevented by antibodies blocking the TSH receptor. The effect is fully compatible with the known in vivo effect of TSH on thyroglobulin iodination and with the fact that these effects are not reproduced by forskolin, i.e. not mediated by cAMP, the other arm of the TSH action.
Several characteristics of this TSH action are worth mentioning. 1) The effect is species specific, and further work will be necessary to establish in various species whether the absence of an Ins(1, 4, 5)P3 response parallels the presence of a stimulation of H2O2 generation by cAMP. 2) A similar effect is observed in CHO cells expressing the human and the dog TSH receptors but not in dog thyroid slices. 3) The effect is not reproduced by forskolin and, therefore, is not a consequence of the activation of the cAMP cascade. 4) The measurable effect of TSH on Ins(1, 4, 5)P3 levels in human thyroid slices is transient with a maximal of 6-fold stimulation (Fig. 4). Due to the limited amounts of biological material (i.e. normal human thyroid) we could not perform complete kinetics but have preferred to repeat the observations with thyroids from different patients (Figs. 1A, 2, 4, and 5). The variability we observe in Ins(1, 4, 5)P3 stimulation has been observed before on cAMP measurements in response to TSH (32). We propose that Ins(1, 4, 5)P3 is rapidly phosphorylated by an Ins(1, 4, 5)P3 3-kinase to InsP4, dephosphorylated by type I inositol 5-phosphatase to Ins(1, 3, 4)P3, and then to InsP2 or InsP1 isomers that accumulate due to the inhibition of inositol bis and monophosphatases by lithium (34, 35). This could explain why, in some experiments, the increase in Ins(1, 4, 5)P3 levels were minor (at the time point studied) but other inositol phosphates, InsP2 and InsP1, accumulated. It suggests a very fast but steady turnover of Ins(1, 4, 5)P3, previously observed in dog thyroid slices stimulated by carbamylcholine (26). When compared with a much longer rise in the accumulation of intracellular Ca2+ levels (7), this suggests that the small remaining Ins(1, 4, 5)P3 is compartmentalized. 5) The very fast turnover of Ins(1, 4, 5)P3 and InsP4 in thyroid cells contrasts with the situation in mouse thymocytes in which Ins(1, 3, 4, 5)P4 accumulates in response to concanavalin A, suggesting a functional role in T cells (24). This further suggests that the couple of Ins(1, 4, 5)P3-kinase and Ins(1, 4, 5)P3/Ins(1, 3, 4, 5)P4 5 phosphatase may function as an efficient switch sustaining or shortening the time life of InsP4 depending on the cell type. 6) As already described by cruder methodology, TSAbs, at concentrations eliciting a cAMP response similar to 10 mU/ml TSH, did not activate inositol phosphate generation. Therefore, the effect in TSH receptor expressing CHO cells is an artifact of this system presumably due to the much higher TSH receptor concentration in the membranes of these cells (36).
Two other effects of TSH may be discussed: the stimulation of PI synthesis and the release of Ins. In fact, the first was suggested by the initial experiments on PI turnover in various thyroids (2, 4, 37, 38). Increased synthesis of PI in the presence of TSH or ATP has been previously documented more extensively. In human thyroid slices, TSH activates the synthesis of phosphatidic acid, cytidine monophosphate phosphatidic acid, and PI (39). Similar effects have been demonstrated with [32P] as a marker in calf thyroid slices (40). Increased incorporation of inositol into PIs concomitant with an increased hydrolysis and inositol phosphate generation can only be explained by an increased synthesis. More extensive but similar results and conclusions have been presented for dog and sheep thyroid slices stimulated by TSH (37) and for other cell types and agonists (41). As in human thyroid slices, the effect on porcine thyroid cells was not mediated by cAMP (38). Such increased synthesis distinct from the PLC cycle is probably widespread as judging from the literature on phospholipid or PI turnover. It has been emphasized in the action of trophic hormones (2, 42). PI transfer protein has been shown to dictate the rate of inositol trisphosphate production by promoting the synthesis of PtdIns(4, 5)P2 (43). Resynthesis of PI in the phosphoinositide cycle has been shown to be regulated by the availability of either substrate or product. The obvious hypothesis to explain it could be a positive feedback between hydrolysis of PI bisphosphate and resynthesis through PI (44).
The decreased [3H]inositol content of our cells might be explained by two phenomena: either the depletion due to increased PI resynthesis or an increased efflux of Ins. The first is demonstrated by the increased resynthesis of the PI despite the loss of [3H]inositol in the inositol phosphate products due to stimulated PLC activity. An increased efflux has been demonstrated in response to TSH in dog thyroid slices (45).
In summary, in humans, TSH activates at least two signaling cascades through its receptor: the cAMP pathway and the Ins(1, 4, 5)P3 Ca2+ pathway. As shown by previous work (46), the first is responsible for the stimulation of iodide transport and thyroid hormone secretion, differentiation, and growth, the second for thyroid hormone synthesis. The activation of the two pathways is dissociated in autonomous adenomas, and we show again here that they are dissociated in Graves disease (32, 47). The functional consequences of this dissociation on the iodinating capacity of the tissue in these diseases are now investigated in vivo by use of the perchlorate-induced iodide discharge test (Corvilain, B., unpublished data).
|
Footnotes |
|---|
First Published Online December 29, 2005
Abbreviations: Ins(1 3 4 )P3, Inositol 1,3,4-trisphosphate; Ins(1 4 5 )P3, inositol 1,4,5-trisphosphate; InsP4, inositol 1,3,4,5-tetrakisphosphate; KRH, Krebs-Ringer HEPES buffer; PI, phosphatidylinositol; PLC, phospholipase C; PtdIns(4 5 )P2, phosphatidylinositol 4,5-bisphosphate; TSAb, thyroid-stimulating antibody.
Received June 14, 2005.
Accepted December 19, 2005.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. Metaye, P. Levillain, J.-L. Kraimps, and R. Perdrisot Immunohistochemical detection, regulation and antiproliferative function of G-protein-coupled receptor kinase 2 in thyroid carcinomas J. Endocrinol., July 1, 2008; 198(1): 101 - 110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. T. Antunes, A. Gagnon, M. L. Langille, and A. Sorisky Thyroid-Stimulating Hormone Induces Interleukin-6 Release from Human Adipocytes through Activation of the Nuclear Factor-{kappa}B Pathway Endocrinology, June 1, 2008; 149(6): 3062 - 3066. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Grasberger, J. Van Sande, A. Hag-Dahood Mahameed, Y. Tenenbaum-Rakover, and S. Refetoff A Familial Thyrotropin (TSH) Receptor Mutation Provides in Vivo Evidence that the Inositol Phosphates/Ca2+ Cascade Mediates TSH Action on Thyroid Hormone Synthesis J. Clin. Endocrinol. Metab., July 1, 2007; 92(7): 2816 - 2820. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
