- Email: [email protected]
Thyroxine secretion by isolated hog thyroid cells: A cyclic AMP independent pathway
Molecular and Cellular Endocrinology, 9 (1977) 33-43 0 Elsevier/North-Holland Scientific Publishers, Ltd.
THYROXINE
SECRETION BY ISOLATED HOG THYROID CELLS: A CYCLIC
AMP INDEPENDENT PATHWAY
B. ROUSSET, Y. MUNARI, A. ROSTAGNAT Groupe de Physiopathologie France Received
24 March
Endocrinienne,
1977; accepted
12 July
and R. MORNEX
H6pital de 1 ‘Antiquaille, 69321 Lyon Cedex I,
1977
The release of 13tI-labeled thyroxine (T4) from isolated hog thyroid cells was increased 1.5-2-fold by thyrotropin (TSH). Dibutyryl cyclic AMP failed to reproduce this TSH action. In this in vitro system another cell activity, T4 synthesis, was stimulated in an essentially identical fashion by TSH and dibutyryl cyclic AMP (time course of action, dose-response relationship). 3-Isobutyl-1-methylxanthine (IBMX), 0.5 mM, did not alter the basal [ 1311]T4 release whereas it enhanced the [ 13tI]T4 synthesis. TSH, 60 mu/ml, increased the intracellular cyclic AMP concentration 3-4-fold. Chlorpromazine (5 X 10e4 M) abolished the TSH stimulation of cyclic AMP accumulation but did not alter the TSH-induced increase in [ t311]T4 secretion. It is concluded that the TSH action on [ 1311]T4 secretion by isolated thyroid cells is not mediated by the adenylate cyclase-cyclic AMP system. Keywords:
isolated
thyroid
cells; thyroxine
secretion;TSH
action;
cyclic AMP pathway.
It is well known that most of the effects of TSH upon thyroidal metabolism in intact tissue, e.g. lobes or slices (Pastan and Macchia, 1967; Ahn and Rosenberg, 1970; Onaya and Solomon, 1969;Williams and Wolff, 1971;Van Sande et al., 1975), or in dispersed thyroid cells (Wilson et al., 1968; Lissitzky et al., 1971; Rousset et al., 1976a) may be reduplicated by cyclic AMP or dibutyryl cyclic AMP. As sug gested by many authors, a TSH-like action of dibutyryl cyclic AMP would provide a first indication for a role of the adenylate cyclase-cyclic AMP system in the mediation of TSH action. On the contrary, the failure of dibutyryl cyclic AMP to mimic a given effect of TSH could suggest that this effect of TSH is not mediated by cyclic AMP. Similarly, the mimicry or potentiation by methylxanthine of a TSH effect further suggests that intracellular cyclic AMP could account for this hormonal stimulation. The best evidence that a given effect of TSH is not mediated by cyclic AMP would be that TSH retains its stimulatory action on the metabolic pathway in question when the TSH-induced increase in intracellular cyclic AMP concentration is inhibited. In a previous study (Rousset et al., 1976b) we have reported that isolated hog 33
34
B. Rousset et al.
thyroid cells are capable of secreting neosynthetized thyroxine. This process was stimulated by TSH. In the present report we have investigated the role of the adenylate cyclase-cyclic AMP system in the TSH action. Attempts were made to assess whether cyclic AMP or its acylated derivatives or IBMX (an inhibitor of phosphodiesterases) could increase the thyroxine secretion by dispersed thyroid cells. We then examined whether TSH retained its stimulatory action on T, secretion when the TSH stimulation of cyclic AMP generation was blocked by chlorpromazine. Both indirect (augmentation of intracellular cyclic AMP concentration by addition of exogenous cyclic AMP, monobutyryl cyclic AMP, dibutyryl cyclic AMP or by inhibition of cyclic AMP degradation) and direct (inhibition of intracellular cyclic AMP generation) experiments indicate that the TSH action on T4 secretion by dispersed thyroid cells is not mediated by the adenylate cyclase-cyclic AMP system.
METHODS AND MATERIALS
Cell incubation Isolated hog thyroid cells were obtained by the discontinuous trypsinization procedure as previously described (Rousset et al., 1976b). Aliquots of lo-50 ~1 of freshly isolated cells (packed cell volume) were incubated in 3 ml of Earle’s balanced salt solution, pH 7.4. Incubations were carried out in 30 ml plastic vials with constant shaking (60 cycles/mm) at 37°C under air. Incubations were performed in triplicate.
Procedure to assess T, secretion Thyroid cells were preincubated with 2-5 ,uCi Na 1311 (carrier-free) for 3 h. The final iodide concentration in the medium was that of Earle’s medium (5 X lops M). After the time of labeling, cells were pelleted by centrifugation at 80-1OOg for 7 min, washed in cold Earle’s medium and incubated in the presence of 2 mM methimazole for 3 or 4 h. When present, porcine TSH, cyclic AMP, monobutyryl or dibutyryl cyclic AMP and IBMX were added at the beginning or 1 h after the start of the incubation period. At the end of the incubation period, cell suspensions were centrifuged (2OOg, 10 min) and the supernatant medium was pipeted off and stored at 4°C for subsequent assays. Free [ r3rI ] T 4 in the medium was extracted by anionexchange chromatography as previously described (Rousset et al., 1976b). This experimental procedure enabled us to assess the release of preformed [13’I]T4 under the condition of blocked T4 synthesis (Rousset et al., 1976b).
Procedure for the measurement of T4 synthesis Thyroid cells were incubated with 2-5 /&iNa1311 and 5 X lo-’ M stable iodide, with or without TSH, sodium butyrate, monobutyryl cyclic AMP, dibutyryl cyclic
TSH action on thyroid secretion
35
AMP and dibutyryl cyclic GMP for 6 h. At the end of the incubation period, thyroid cells were analyzed for their [ 13*I]T4 content as described previously (Rousset et al., 1976b). The procedure included a pronase treatment and the extraction chromatography. Medium-free [ 1311]T4 was meaof [ r311]T4 by anion-exchange sured as described above. Total synthetized [ 1311]Tq was calculated as the sum of intracellular [ 13’I]T4 plus medium [13’I]T4. Cyclic AMP assay
Thyroid cells were preincubated in Earle’s medium for 15 min in the presence of 0.5 mM IBMX. Then TSH was added and the cells were incubated for 15 min. The cells were then transferred to a 5 ml tube, pelleted by centrifugation (1 min at SOO-lOOOg), suspended in 0.1 ml 5 mM IBMX in Earle’s medium, and immediately frozen at -80°C. Cyclic AMP was extracted with 0.4 ml methanol-ethanol mixture (1 : 3) and measured by the method of Orgiazzi et al. (1974). Materials
Porcine thyrotropin was purchased from Organon (Paris, France), cyclic AMP, N6-monobutyryl cyclic AMP, N6 02dibutyryl cyclic AMP, N2 02-dibutyryl cyclic GMP and methimazole from Sigma (St. Louis, MO., U.S.A.), 3-isobutyl-lmethylxanthine was obtained from Aldrich Europe (Beerse, Belgium) and chlorpromazine hydrochloride from Specia (Paris, France). ‘311-Labeled iodide was provided by CEA (Saclay, France).
RESULTS AND DISCUSSION The time course of basal or TSH-stimulated [ 1311]T4 release from 3 h prelabeled thyroid cells is depicted in fig. 1A. A TSH stimulation was observed within 30 min of incubation. Fig. 1B illustrates the effect of varying concentrations of TSH. The release of [ 1311]T4 by prelabeled dispersed thyroid cells was stimulated by TSH in the concentration range 2-60 mu/ml. In a 3 h incubation period the maximal TSHinduced increase in [ r311]T4 release was 173 + 5% of control (mean + SEM, 10 experiments). Neither dibutyryl cyclic AMP (3 mM), monobutyryl cyclic AMP (3 mM) nor cyclic AMP (10 mM) altered the [r311]T4 release by dispersed thyroid cells (fig. 2). The presence of dibutyryl cyclic AMP in the incubation medium did not prevent the TSH-induced increase in [ 1311]T4 release (fig. 2). Therefore, it seemed likely that the lack of effect of dibutyryl cyclic AMP was not due to an alteration of the secretory process. In our cell preparations the effects of TSH on iodide uptake (Rousset et al., 1976a) and on [13’I]T4 synthesis (fig. 3) were mimicked by dibutyryl cyclic AMP. Similar results had been previously reported by Wilson et al. (i968). The time course of action of TSH (60 mu/ml) and dibutyryl cyclic AMP (3 mM) on [1311]T4
B. Rousset
36
et al.
11
l,,.
30
90 INCUBATION
180 TIME
(min.)
0
,
2.4
12 TSH
60
300
(mu/ml)
Fig. 1. A (left): Time course of basal OI TSH-stimulated [ 1311]T4 release by dispersed thyroid cells. 3 h pr:labeled cells were incubated in the presence of 2 mM methimazole with OI without TSH (60 mu/ml). B (right): Plot of the medium [ ’ 31I] T4 release by dispersed thyroid cells as a function of the TSH concentration. 3 h prelabeled cells were incubated in the presence of 2 mM methimazole with or without TSH for 3 h. Each point and vertical line represents the mean i SEM of triplicates.
synthesis were very similar (fig. 3A). The active concentration range of TSH and dibutyryl cyclic AMP on [ 1311]T4 synthesis was respectively 2-60 mu/ml and 0.1-3 mM (fig. 3B). It should be mentioned that TSH stimulated [r311] T4 synthesis and [ 1311]T4 release in the same concentration range. The stimulatory effect of dibutyryl cyclic AMP on [ 1311]T4 synthesis was specific since sodium butyrate or dibutyryl cyclic GMP at the same concentration (2 mM) were devoid of any effect (fig. 4). Under our experimental conditions the slight inhibitory effect of dibutyryl cyclic GMP was never statistically significant. Monobutyryl cyclic AMP was less potent than dibutyryl cyclic AMP in increasing [ 1311]T4 synthesis in dispersed thyroid cells. Since dibutyryl cyclic AMP exerted a TSH-like action on some aspects of iodine metabolism in our cell preparations, but not on T4 secretion, and since it did not alter this secretion process, we would suggest that TSH does not act through the cyclic AMP pathway to increase the T4 release by dispersed thyroid cells. Our data fulfilled the conditions stated by Dumont (1971) for a refutation of the Sutherland model. However, dibutyryl cyclic AMP may not act like intracellular cyclic AMP on some target enzyme systems. This has been suggested by Bidey et al. (1976). We tried to confirm the data obtained with dibutyryl cyclic AMP using an inhibitor of phosphodiesterase activity, IBMX, to increase intracellular cyclic AMP con-
37
TSH mtion on thyroid secretion
15 EXPERIMENT
EXPERIMENT
1
2
10
L1
5
*1
NONE
TSH
MSC
DBC
NONE
TSH
DBC
DBC
CAMP
T&H
Fig. 2. Lack of effect of cyclic AMP, monobutyryl and dibutyryl cyclic AMP on [ 1J11]T4 release by dispersed thyroid cells. Thyroid cells were preincubated with r311-labeled iodide for 3 h, then washed and incubated in a labeled iodide free medium, with 2 mM methimazole, for 3 h. Cyclic AMP (10 mM), monobutyryl cyclic AMP (3 mM), dibutyryl cyclic AMP (3 mM) and TSH (60 mu/ml) were added at the start of the incubation period. Monobutyryl cyclic AMP and dibutyryl cyclic AMP are abbreviated to MBC and DBC respectively. Each bar and vertical line represents the mean f SEM of triplicates.
centration. IBMX (0.5 mM) increased the [ r311]T4 synthesis (fig. 5). The lack of additivity of the effect of IBMX to that of a maximal concentration of TSH (60 mu/ml) would suggest that TSH and IBMX act through the same mechanism to increase T4 synthesis. In contrast, IBMX failed to increase the basal [ 13tI] T4 release. IBMX did not impair the secretory process since it did not modify the TSH response. These data indicate that intracellular cyclic AMP reproduced the TSH action on [1311]T4 synthesis but did not mimic the TSH action on [ 1311]T4 release. In order to test directly the involvement of cyclic AMP in the TSH stimulation of T4 secretion by dispersed thyroid cells, we tried to find an experimental design which enabled us to inhibit the TSH-induced increase in cyclic AMP without altering the TSH stimulation of T4 secretion. In the presence of 0.5 mM IBMX, TSH increased intracellular cyclic AMP concentration in a dose-related manner (fig. 6). Chlorpromazine, a membrane stabilizer (Onaya and Solomon, 1969; Williams and Wolff, 1971), was known to abolish the TSH-stimulated rise in cyclic AMP in mouse thyroid lobes (Williams, 1972). In our system, 0.5 mM chlorpromazine completely inhibited the increase in intracellular cyclic AMP concentration induced by 60 mu/ml of TSH (fig. 7). Chlorpromazine did not interfere in the cyclic AMP
38
B. Rousset et al.
assay (fig. 8). In contrast, 0.5 mM chlorpromazine did not alter the TSH-induced increase in [r311]T4 release (fig. 7). These data strengthen the conclusion drawn from the lack of effect of dibutyryl cyclic AMP and IBMX. It is thus reasonable to conclude that the TSH stimulation of T4 secretion by dispersed thyroid cells is not mediated by the adenylate cyclase-cyclic AMP system. The mechanism of TSH action on T4 secretion by dispersed thyroid cells appears different from that involved in follicular organized thyroid tissue. Indeed, TSH acts through the adenylate cyclase-cyclic AMP system to stimulate the hormonal secretion from intact thyroid tissue in vivo or in vitro (lobes or slices) (Williams, 1972). The difference in the mechanism of TSH action on hormone secretion between the two systems seems to be linked to some difference in the mechanism of hormone secretion. In vivo or in intact thyroid tissue in vitro, the first event in the secretion process is the endocytosis of thyroglobulin. This step, which is TSH-dependent via cyclic AMP (Williams, 1972), is not involved in the T4 secretion by dispersed thy-
3
6
4.5
Dibutyryl cyclic
1
6
3 INCUBATION
Fig. 3A. Time course of [ 1311]T4 dibutyryl
AMP
synthesis
TIME
by dispersed
thyroid
cyclic AMP. Values are the mean +SEM of triplicate
cells incubated
incubations.
with TSH OI
TSH action on thyroid secretion
La’-
0.1
0’3
39
1
i
DIBUTYRYL
i CYCLIC
AMP
(mM)
Fig. 3B. Effect of varying the concentrations of TSH or dibutyryl cyclic AMP on [ 1311]T4 synthesis. Thyroid cells were incubated with 1311-labeled iodide for 6 h. Values are the mean iSEM of triplicate incubations.
roid cells (Rousset et al., 1976~). The T4 secretion by dispersed thyroid cells differs on another point from the hormone secretion by follicular organized thyroid tissue since the former occurs from neosynthetized Ta-containing thyroglobulin (Rousset et al., 1976b) and the latter from the bulk of thyroglobulin stored in the follicular lumen (Dumont, 1971). If the processes subsequent to endocytosis,i.e. thyroglobulin digestion and hormone release, are similar in intact thyroid tissue and in dispersed thyroid cells, they might be cyclic AMP independent pathways. In the thyroid, another metabolic pathway seems to be independent of cyclic AMP metabolism: phospholipogenesis. Using either isolated pig thyroid cells (Scott et al., 1970) or pig thyroid slices (Scott et al., 1970; Jacquemin and Haye, 1970), attempts to reduplicate the effect of TSH on the synthesis of phosphatidylinositol by dibutyryl cyclic AMP have been unsuccessful. Concerning these and our own observations, what remains to be determined is the nature of the intracellular intermediate(s), if any, involved in the TSH action.
____-_---L-ka ____---__--
rtrol
butyrate 2mM
r
IPdZbzP
Fig. 4. Specificity of action of cyclic AMP derivatives on roid cells. Thyroid cells were incubated with ‘311-labeled line represents the mean + SEM of triplicates.
-
I
[ 1311]T4 synthesis iodide
by dispersed thyfor 6 h. Each bar and vertical
P
II TSH
00
++
00
++
I BMX
o+
o+
o+
o+
Fig. 5. Comparison between the effects of 3-isobutyl-l-methylxanthine (IBMX) on [ 1311]T4 synthesis (left panel) and on [ 1311]T4 release (right panel). Thyroid cells were incubated with 1311-labeled iodide for 6 h to measure [ 1 311]T4 synthesis. 3 h prelabeled cells were incubated for 3 h in the presence of 2 mM methimazole to assess [ 1311]T4 release. Each bar and vertical line represents the mean k SEM of triplicates.
6 60 TSH (mu / ml) Fig. 6. Effect of various concentrations of TSH on the intracellular cyclic AMP accumulation. Thyroid cells were preincubated with 0.5 mM IBMX for 15 min, then incubated for 15 min with I ‘SH. 15
CPZ(0.5mM) TSH (60mU/ml)O
0
0
+
+
+
0
0
Fig. 7. Effect of chlorpromazine (CPZ) on cyclic AMP accumulation (left panel) and on [ 13 ‘I]T4 secretion (right panel) by dispersed thyroid cells. For cyclic AMP assay, thyroid cells were preincubated for 15 min with chlorpromazine and 0.5 mM IBMX and then incubated for 15 min with or without TSH. To assess T4 secretion, 3 h prelabeled cells were preincubated for 1 h with chlorpromazine and 0.5 mM IBMX and then incubated for 3 h with or without TSH. Each bar and vertical line represents the mean f SEM of duplicates or triplicates in one typical experiment.
B. Rousset et al.
ol8MX
$25
0.5
0.775 c ,c,ik”
5
10 cell
10’3M
10
AM2i25[pmole5s] e%,t
[$j
I
,
20
130
Fig. 8. Standard curves of cyclic AMP assay. Comparison of the inhibition pattern of the binding of 3H-labeled cyclic AMP to hog thyroid cytosol by nonradioactive cyclic AMP and cell extract. Lack of effect of 3-isobutyl-I-methyIxan~ine (IBMX, IO-3 M) and chlorpromazine (CPZ, 1O-3 M) on the cyclic AMP assay.
ACKNOWLEDGEMENTS The authors wish to express their gratitude to C. Poncet for her technical assistance, to 1. Orgiazzi for the help he provided in the cyclic AMP assay and to B. Rigaud for the typing of the manuscript. This work was supported by a grant from INSERM, No. 74.1.132.4.
REFERENCES Ahn, C.S. and Rosenberg, I.N. (1970) Endocrinology 86,396-405. Bidey, S.P., Marsden, P., McKerron, C.C. and Anderson, J. (1976) Biochem. Biophys. Res. Commun. 70;418-424. Dumont, J.E. (1971) Vitam. Horm. 29,287-412. Jacquemin, Cand Haye, B. (1970) Bull. Sac. Chim. Biol. 52,153-165. Lissitzky, S., Fayet, G., Giraud, A. Verrier, B. and Torresani, J. (1971) Eur. J. Biochem. 24, 88-99. Onaya, T. and Solomon, D.H. (1969) Endocrinology 85,1010-1017. Orgiazzi, _I.,Chopra, I.J., Williams, DE. and Solomon, D.H. (1974) Anal. Lett. 7,609-619. Pastan, I. and Macchia, V. (1967) J. Biol. Chem. 242,5757-5761. Rousset, B.,Poncet, C. and Mornex, R. (1976a) Ann. Endocrinol. (Paris) 37, 109-110. Rousset, B., Poncet, C. and Mornex, R. (1976b) Biochim. Biophys. Acta 437,543-561.
TSH action on thyroid secretion Rousset, B., Poncet, C. and Mornex, R. (1976c) Acta Endocrinol., Suppl. 204,44. Scott, T.W., Freinkel, N., Klein, J.H. and Nitzan, M. (1970) Endocrinology 87,854-863. Van Sande, J., Grenier, G., Willems, C. and Dumont, J.E. (1975) Endocrinology 96,781-786. Williams, J.A. and Wolff, J. (1970) Endocrinology 88,206-217. Williams, J.A. (1972) Endocrinology 91, 1411-1417. Wilson, B., Raghupathy, E., Tonoue, T. and Tong, W. (1968) Endocrinology 83,877-884.
43
THYROXINE
SECRETION BY ISOLATED HOG THYROID CELLS: A CYCLIC
AMP INDEPENDENT PATHWAY
B. ROUSSET, Y. MUNARI, A. ROSTAGNAT Groupe de Physiopathologie France Received
24 March
Endocrinienne,
1977; accepted
12 July
and R. MORNEX
H6pital de 1 ‘Antiquaille, 69321 Lyon Cedex I,
1977
The release of 13tI-labeled thyroxine (T4) from isolated hog thyroid cells was increased 1.5-2-fold by thyrotropin (TSH). Dibutyryl cyclic AMP failed to reproduce this TSH action. In this in vitro system another cell activity, T4 synthesis, was stimulated in an essentially identical fashion by TSH and dibutyryl cyclic AMP (time course of action, dose-response relationship). 3-Isobutyl-1-methylxanthine (IBMX), 0.5 mM, did not alter the basal [ 1311]T4 release whereas it enhanced the [ 13tI]T4 synthesis. TSH, 60 mu/ml, increased the intracellular cyclic AMP concentration 3-4-fold. Chlorpromazine (5 X 10e4 M) abolished the TSH stimulation of cyclic AMP accumulation but did not alter the TSH-induced increase in [ t311]T4 secretion. It is concluded that the TSH action on [ 1311]T4 secretion by isolated thyroid cells is not mediated by the adenylate cyclase-cyclic AMP system. Keywords:
isolated
thyroid
cells; thyroxine
secretion;TSH
action;
cyclic AMP pathway.
It is well known that most of the effects of TSH upon thyroidal metabolism in intact tissue, e.g. lobes or slices (Pastan and Macchia, 1967; Ahn and Rosenberg, 1970; Onaya and Solomon, 1969;Williams and Wolff, 1971;Van Sande et al., 1975), or in dispersed thyroid cells (Wilson et al., 1968; Lissitzky et al., 1971; Rousset et al., 1976a) may be reduplicated by cyclic AMP or dibutyryl cyclic AMP. As sug gested by many authors, a TSH-like action of dibutyryl cyclic AMP would provide a first indication for a role of the adenylate cyclase-cyclic AMP system in the mediation of TSH action. On the contrary, the failure of dibutyryl cyclic AMP to mimic a given effect of TSH could suggest that this effect of TSH is not mediated by cyclic AMP. Similarly, the mimicry or potentiation by methylxanthine of a TSH effect further suggests that intracellular cyclic AMP could account for this hormonal stimulation. The best evidence that a given effect of TSH is not mediated by cyclic AMP would be that TSH retains its stimulatory action on the metabolic pathway in question when the TSH-induced increase in intracellular cyclic AMP concentration is inhibited. In a previous study (Rousset et al., 1976b) we have reported that isolated hog 33
34
B. Rousset et al.
thyroid cells are capable of secreting neosynthetized thyroxine. This process was stimulated by TSH. In the present report we have investigated the role of the adenylate cyclase-cyclic AMP system in the TSH action. Attempts were made to assess whether cyclic AMP or its acylated derivatives or IBMX (an inhibitor of phosphodiesterases) could increase the thyroxine secretion by dispersed thyroid cells. We then examined whether TSH retained its stimulatory action on T, secretion when the TSH stimulation of cyclic AMP generation was blocked by chlorpromazine. Both indirect (augmentation of intracellular cyclic AMP concentration by addition of exogenous cyclic AMP, monobutyryl cyclic AMP, dibutyryl cyclic AMP or by inhibition of cyclic AMP degradation) and direct (inhibition of intracellular cyclic AMP generation) experiments indicate that the TSH action on T4 secretion by dispersed thyroid cells is not mediated by the adenylate cyclase-cyclic AMP system.
METHODS AND MATERIALS
Cell incubation Isolated hog thyroid cells were obtained by the discontinuous trypsinization procedure as previously described (Rousset et al., 1976b). Aliquots of lo-50 ~1 of freshly isolated cells (packed cell volume) were incubated in 3 ml of Earle’s balanced salt solution, pH 7.4. Incubations were carried out in 30 ml plastic vials with constant shaking (60 cycles/mm) at 37°C under air. Incubations were performed in triplicate.
Procedure to assess T, secretion Thyroid cells were preincubated with 2-5 ,uCi Na 1311 (carrier-free) for 3 h. The final iodide concentration in the medium was that of Earle’s medium (5 X lops M). After the time of labeling, cells were pelleted by centrifugation at 80-1OOg for 7 min, washed in cold Earle’s medium and incubated in the presence of 2 mM methimazole for 3 or 4 h. When present, porcine TSH, cyclic AMP, monobutyryl or dibutyryl cyclic AMP and IBMX were added at the beginning or 1 h after the start of the incubation period. At the end of the incubation period, cell suspensions were centrifuged (2OOg, 10 min) and the supernatant medium was pipeted off and stored at 4°C for subsequent assays. Free [ r3rI ] T 4 in the medium was extracted by anionexchange chromatography as previously described (Rousset et al., 1976b). This experimental procedure enabled us to assess the release of preformed [13’I]T4 under the condition of blocked T4 synthesis (Rousset et al., 1976b).
Procedure for the measurement of T4 synthesis Thyroid cells were incubated with 2-5 /&iNa1311 and 5 X lo-’ M stable iodide, with or without TSH, sodium butyrate, monobutyryl cyclic AMP, dibutyryl cyclic
TSH action on thyroid secretion
35
AMP and dibutyryl cyclic GMP for 6 h. At the end of the incubation period, thyroid cells were analyzed for their [ 13*I]T4 content as described previously (Rousset et al., 1976b). The procedure included a pronase treatment and the extraction chromatography. Medium-free [ 1311]T4 was meaof [ r311]T4 by anion-exchange sured as described above. Total synthetized [ 1311]Tq was calculated as the sum of intracellular [ 13’I]T4 plus medium [13’I]T4. Cyclic AMP assay
Thyroid cells were preincubated in Earle’s medium for 15 min in the presence of 0.5 mM IBMX. Then TSH was added and the cells were incubated for 15 min. The cells were then transferred to a 5 ml tube, pelleted by centrifugation (1 min at SOO-lOOOg), suspended in 0.1 ml 5 mM IBMX in Earle’s medium, and immediately frozen at -80°C. Cyclic AMP was extracted with 0.4 ml methanol-ethanol mixture (1 : 3) and measured by the method of Orgiazzi et al. (1974). Materials
Porcine thyrotropin was purchased from Organon (Paris, France), cyclic AMP, N6-monobutyryl cyclic AMP, N6 02dibutyryl cyclic AMP, N2 02-dibutyryl cyclic GMP and methimazole from Sigma (St. Louis, MO., U.S.A.), 3-isobutyl-lmethylxanthine was obtained from Aldrich Europe (Beerse, Belgium) and chlorpromazine hydrochloride from Specia (Paris, France). ‘311-Labeled iodide was provided by CEA (Saclay, France).
RESULTS AND DISCUSSION The time course of basal or TSH-stimulated [ 1311]T4 release from 3 h prelabeled thyroid cells is depicted in fig. 1A. A TSH stimulation was observed within 30 min of incubation. Fig. 1B illustrates the effect of varying concentrations of TSH. The release of [ 1311]T4 by prelabeled dispersed thyroid cells was stimulated by TSH in the concentration range 2-60 mu/ml. In a 3 h incubation period the maximal TSHinduced increase in [ r311]T4 release was 173 + 5% of control (mean + SEM, 10 experiments). Neither dibutyryl cyclic AMP (3 mM), monobutyryl cyclic AMP (3 mM) nor cyclic AMP (10 mM) altered the [r311]T4 release by dispersed thyroid cells (fig. 2). The presence of dibutyryl cyclic AMP in the incubation medium did not prevent the TSH-induced increase in [ 1311]T4 release (fig. 2). Therefore, it seemed likely that the lack of effect of dibutyryl cyclic AMP was not due to an alteration of the secretory process. In our cell preparations the effects of TSH on iodide uptake (Rousset et al., 1976a) and on [13’I]T4 synthesis (fig. 3) were mimicked by dibutyryl cyclic AMP. Similar results had been previously reported by Wilson et al. (i968). The time course of action of TSH (60 mu/ml) and dibutyryl cyclic AMP (3 mM) on [1311]T4
B. Rousset
36
et al.
11
l,,.
30
90 INCUBATION
180 TIME
(min.)
0
,
2.4
12 TSH
60
300
(mu/ml)
Fig. 1. A (left): Time course of basal OI TSH-stimulated [ 1311]T4 release by dispersed thyroid cells. 3 h pr:labeled cells were incubated in the presence of 2 mM methimazole with OI without TSH (60 mu/ml). B (right): Plot of the medium [ ’ 31I] T4 release by dispersed thyroid cells as a function of the TSH concentration. 3 h prelabeled cells were incubated in the presence of 2 mM methimazole with or without TSH for 3 h. Each point and vertical line represents the mean i SEM of triplicates.
synthesis were very similar (fig. 3A). The active concentration range of TSH and dibutyryl cyclic AMP on [ 1311]T4 synthesis was respectively 2-60 mu/ml and 0.1-3 mM (fig. 3B). It should be mentioned that TSH stimulated [r311] T4 synthesis and [ 1311]T4 release in the same concentration range. The stimulatory effect of dibutyryl cyclic AMP on [ 1311]T4 synthesis was specific since sodium butyrate or dibutyryl cyclic GMP at the same concentration (2 mM) were devoid of any effect (fig. 4). Under our experimental conditions the slight inhibitory effect of dibutyryl cyclic GMP was never statistically significant. Monobutyryl cyclic AMP was less potent than dibutyryl cyclic AMP in increasing [ 1311]T4 synthesis in dispersed thyroid cells. Since dibutyryl cyclic AMP exerted a TSH-like action on some aspects of iodine metabolism in our cell preparations, but not on T4 secretion, and since it did not alter this secretion process, we would suggest that TSH does not act through the cyclic AMP pathway to increase the T4 release by dispersed thyroid cells. Our data fulfilled the conditions stated by Dumont (1971) for a refutation of the Sutherland model. However, dibutyryl cyclic AMP may not act like intracellular cyclic AMP on some target enzyme systems. This has been suggested by Bidey et al. (1976). We tried to confirm the data obtained with dibutyryl cyclic AMP using an inhibitor of phosphodiesterase activity, IBMX, to increase intracellular cyclic AMP con-
37
TSH mtion on thyroid secretion
15 EXPERIMENT
EXPERIMENT
1
2
10
L1
5
*1
NONE
TSH
MSC
DBC
NONE
TSH
DBC
DBC
CAMP
T&H
Fig. 2. Lack of effect of cyclic AMP, monobutyryl and dibutyryl cyclic AMP on [ 1J11]T4 release by dispersed thyroid cells. Thyroid cells were preincubated with r311-labeled iodide for 3 h, then washed and incubated in a labeled iodide free medium, with 2 mM methimazole, for 3 h. Cyclic AMP (10 mM), monobutyryl cyclic AMP (3 mM), dibutyryl cyclic AMP (3 mM) and TSH (60 mu/ml) were added at the start of the incubation period. Monobutyryl cyclic AMP and dibutyryl cyclic AMP are abbreviated to MBC and DBC respectively. Each bar and vertical line represents the mean f SEM of triplicates.
centration. IBMX (0.5 mM) increased the [ r311]T4 synthesis (fig. 5). The lack of additivity of the effect of IBMX to that of a maximal concentration of TSH (60 mu/ml) would suggest that TSH and IBMX act through the same mechanism to increase T4 synthesis. In contrast, IBMX failed to increase the basal [ 13tI] T4 release. IBMX did not impair the secretory process since it did not modify the TSH response. These data indicate that intracellular cyclic AMP reproduced the TSH action on [1311]T4 synthesis but did not mimic the TSH action on [ 1311]T4 release. In order to test directly the involvement of cyclic AMP in the TSH stimulation of T4 secretion by dispersed thyroid cells, we tried to find an experimental design which enabled us to inhibit the TSH-induced increase in cyclic AMP without altering the TSH stimulation of T4 secretion. In the presence of 0.5 mM IBMX, TSH increased intracellular cyclic AMP concentration in a dose-related manner (fig. 6). Chlorpromazine, a membrane stabilizer (Onaya and Solomon, 1969; Williams and Wolff, 1971), was known to abolish the TSH-stimulated rise in cyclic AMP in mouse thyroid lobes (Williams, 1972). In our system, 0.5 mM chlorpromazine completely inhibited the increase in intracellular cyclic AMP concentration induced by 60 mu/ml of TSH (fig. 7). Chlorpromazine did not interfere in the cyclic AMP
38
B. Rousset et al.
assay (fig. 8). In contrast, 0.5 mM chlorpromazine did not alter the TSH-induced increase in [r311]T4 release (fig. 7). These data strengthen the conclusion drawn from the lack of effect of dibutyryl cyclic AMP and IBMX. It is thus reasonable to conclude that the TSH stimulation of T4 secretion by dispersed thyroid cells is not mediated by the adenylate cyclase-cyclic AMP system. The mechanism of TSH action on T4 secretion by dispersed thyroid cells appears different from that involved in follicular organized thyroid tissue. Indeed, TSH acts through the adenylate cyclase-cyclic AMP system to stimulate the hormonal secretion from intact thyroid tissue in vivo or in vitro (lobes or slices) (Williams, 1972). The difference in the mechanism of TSH action on hormone secretion between the two systems seems to be linked to some difference in the mechanism of hormone secretion. In vivo or in intact thyroid tissue in vitro, the first event in the secretion process is the endocytosis of thyroglobulin. This step, which is TSH-dependent via cyclic AMP (Williams, 1972), is not involved in the T4 secretion by dispersed thy-
3
6
4.5
Dibutyryl cyclic
1
6
3 INCUBATION
Fig. 3A. Time course of [ 1311]T4 dibutyryl
AMP
synthesis
TIME
by dispersed
thyroid
cyclic AMP. Values are the mean +SEM of triplicate
cells incubated
incubations.
with TSH OI
TSH action on thyroid secretion
La’-
0.1
0’3
39
1
i
DIBUTYRYL
i CYCLIC
AMP
(mM)
Fig. 3B. Effect of varying the concentrations of TSH or dibutyryl cyclic AMP on [ 1311]T4 synthesis. Thyroid cells were incubated with 1311-labeled iodide for 6 h. Values are the mean iSEM of triplicate incubations.
roid cells (Rousset et al., 1976~). The T4 secretion by dispersed thyroid cells differs on another point from the hormone secretion by follicular organized thyroid tissue since the former occurs from neosynthetized Ta-containing thyroglobulin (Rousset et al., 1976b) and the latter from the bulk of thyroglobulin stored in the follicular lumen (Dumont, 1971). If the processes subsequent to endocytosis,i.e. thyroglobulin digestion and hormone release, are similar in intact thyroid tissue and in dispersed thyroid cells, they might be cyclic AMP independent pathways. In the thyroid, another metabolic pathway seems to be independent of cyclic AMP metabolism: phospholipogenesis. Using either isolated pig thyroid cells (Scott et al., 1970) or pig thyroid slices (Scott et al., 1970; Jacquemin and Haye, 1970), attempts to reduplicate the effect of TSH on the synthesis of phosphatidylinositol by dibutyryl cyclic AMP have been unsuccessful. Concerning these and our own observations, what remains to be determined is the nature of the intracellular intermediate(s), if any, involved in the TSH action.
____-_---L-ka ____---__--
rtrol
butyrate 2mM
r
IPdZbzP
Fig. 4. Specificity of action of cyclic AMP derivatives on roid cells. Thyroid cells were incubated with ‘311-labeled line represents the mean + SEM of triplicates.
-
I
[ 1311]T4 synthesis iodide
by dispersed thyfor 6 h. Each bar and vertical
P
II TSH
00
++
00
++
I BMX
o+
o+
o+
o+
Fig. 5. Comparison between the effects of 3-isobutyl-l-methylxanthine (IBMX) on [ 1311]T4 synthesis (left panel) and on [ 1311]T4 release (right panel). Thyroid cells were incubated with 1311-labeled iodide for 6 h to measure [ 1 311]T4 synthesis. 3 h prelabeled cells were incubated for 3 h in the presence of 2 mM methimazole to assess [ 1311]T4 release. Each bar and vertical line represents the mean k SEM of triplicates.
6 60 TSH (mu / ml) Fig. 6. Effect of various concentrations of TSH on the intracellular cyclic AMP accumulation. Thyroid cells were preincubated with 0.5 mM IBMX for 15 min, then incubated for 15 min with I ‘SH. 15
CPZ(0.5mM) TSH (60mU/ml)O
0
0
+
+
+
0
0
Fig. 7. Effect of chlorpromazine (CPZ) on cyclic AMP accumulation (left panel) and on [ 13 ‘I]T4 secretion (right panel) by dispersed thyroid cells. For cyclic AMP assay, thyroid cells were preincubated for 15 min with chlorpromazine and 0.5 mM IBMX and then incubated for 15 min with or without TSH. To assess T4 secretion, 3 h prelabeled cells were preincubated for 1 h with chlorpromazine and 0.5 mM IBMX and then incubated for 3 h with or without TSH. Each bar and vertical line represents the mean f SEM of duplicates or triplicates in one typical experiment.
B. Rousset et al.
ol8MX
$25
0.5
0.775 c ,c,ik”
5
10 cell
10’3M
10
AM2i25[pmole5s] e%,t
[$j
I
,
20
130
Fig. 8. Standard curves of cyclic AMP assay. Comparison of the inhibition pattern of the binding of 3H-labeled cyclic AMP to hog thyroid cytosol by nonradioactive cyclic AMP and cell extract. Lack of effect of 3-isobutyl-I-methyIxan~ine (IBMX, IO-3 M) and chlorpromazine (CPZ, 1O-3 M) on the cyclic AMP assay.
ACKNOWLEDGEMENTS The authors wish to express their gratitude to C. Poncet for her technical assistance, to 1. Orgiazzi for the help he provided in the cyclic AMP assay and to B. Rigaud for the typing of the manuscript. This work was supported by a grant from INSERM, No. 74.1.132.4.
REFERENCES Ahn, C.S. and Rosenberg, I.N. (1970) Endocrinology 86,396-405. Bidey, S.P., Marsden, P., McKerron, C.C. and Anderson, J. (1976) Biochem. Biophys. Res. Commun. 70;418-424. Dumont, J.E. (1971) Vitam. Horm. 29,287-412. Jacquemin, Cand Haye, B. (1970) Bull. Sac. Chim. Biol. 52,153-165. Lissitzky, S., Fayet, G., Giraud, A. Verrier, B. and Torresani, J. (1971) Eur. J. Biochem. 24, 88-99. Onaya, T. and Solomon, D.H. (1969) Endocrinology 85,1010-1017. Orgiazzi, _I.,Chopra, I.J., Williams, DE. and Solomon, D.H. (1974) Anal. Lett. 7,609-619. Pastan, I. and Macchia, V. (1967) J. Biol. Chem. 242,5757-5761. Rousset, B.,Poncet, C. and Mornex, R. (1976a) Ann. Endocrinol. (Paris) 37, 109-110. Rousset, B., Poncet, C. and Mornex, R. (1976b) Biochim. Biophys. Acta 437,543-561.
TSH action on thyroid secretion Rousset, B., Poncet, C. and Mornex, R. (1976c) Acta Endocrinol., Suppl. 204,44. Scott, T.W., Freinkel, N., Klein, J.H. and Nitzan, M. (1970) Endocrinology 87,854-863. Van Sande, J., Grenier, G., Willems, C. and Dumont, J.E. (1975) Endocrinology 96,781-786. Williams, J.A. and Wolff, J. (1970) Endocrinology 88,206-217. Williams, J.A. (1972) Endocrinology 91, 1411-1417. Wilson, B., Raghupathy, E., Tonoue, T. and Tong, W. (1968) Endocrinology 83,877-884.
43