Binding of Thyrotropin-Releasing Hormone to Plasma Membranes

Proc. Nat. Acad. Sci. USA Vol. 69, No. 1, pp. 283-287, January 1972

Binding of Thyrotropin-Releasing Hormone to Plasma Membranes of Bovine Anterior Pituitary Gland (hormone receptor/adenylate cyclase/equilibrium constant/[3Hfthyrotropin)

FERNAND LABRIE, NICHOLAS BARDEN, GUY POIRIER, AND ANDRE DE LEAN Laboratory of Molecular Endocrinology, Faculty of Medicine, Laval University, Quebec 10, Canada. Communicated by Alexander Rich, November 1, 1971

ABSTRACT An assay for the binding of ['Hlthyro- erties of hormone binding to its receptors in the adenohy- tropin-releasing hormone ([3HJTRH) is described. Plasma pophyseal plasma membrane. membranes isolated from bovine anterior pituitary gland bind about 600 femtomoles of this hormone per mg of MATERIALS AND METHODS protein, as compared to 15 femtomoles per mg of protein in the total adenohypophyseal homogenate (40-fold Thyrotropin-releasing hormone (L- [2,3-3H ]-Proline, 40 Ci/mol) purification). The equilibrium constant of membrane was purchased from New- England Nuclear. Radiochemical receptor-[3H]TRH binding at 00C is 4.3 X 107 L- M-1, or a purity, determined by paper chromatography on Whatman half-maximal binding of this hormone at 23 nM. The 1 water 25:4:10 was more binding is time-dependent; addition of unlabeled hor- no. in n-butyl alcohol-acetic acid- mone induces dissociation of the receptor-[3H]TRH com- than 98% at the end of the reported experiments. Biological plex with a half-life of 14 min. The binding of TRH is not activity of [3H]TRH was determined before delivery by bio- altered by 10 uM melanocyte-stimulating hormone- assay (23), and was found to be identical to that of a synthetic release inhibiting hormone, lysine-vasopressin, adreno- standard of TRH and, hence, active. corticotropin, growth hormone, prolactin, luteinizing completely biologically hormone, insulin, glucagon, L-thyroxine, or L-triiodo- Synthetic TRH was supplied by Dr. W. F. White, Abbott thyronine. K+ and Mg + + increase formation of the recep- Laboratories, Chicago, and by Dr. R. Guillemin, Salk Insti- tor-TRH complex at optimal concentrations of 5-25 mM tute, La Jolla. Synthetic melanocyte-stimulating hormone- and 0.5-2.5 mM, respectively, with inhibition at higher release inhibiting hormone was a gift from Drs. A. V. Schally concentrations. Ca++ inhibits binding of TRH at all concentrations tested. and J. A. Kastin, New Orleans. Porcine adrenocorticotropin was a gift from Dr. A. F. Dixon, Cambridge, England, Bovine growth hormone (NIH-GH-B16) and prolactin The secretion of thyrotropin by the anterior pituitary gland is (NIH-P-B3) were supplied by the National Pituitary Agency, controlled by a neurohormone synthesized in the hypotha- Endocrinology Study Section, NIH. Bovine LH was supplied lamic area and carried to its adenohypophyseal site of action by by Dr. C. Courte, Laval University, Quebec. Bovine insulin a portal system (1-3). A major contribution in the field of (PJ-4609) and crystalline porcine glucagon (GLF-599A) neuroendocrinology has been the recent elucidation of the were gifts from Dr. M. A. Root, The Lilly Research Labora- structure of this neurohormone (thyrotropin-releasing hor- tories, Indianapolis. Synthetic lysine-vasopressin was pur- rnone, TRH) as L-pyroglutamyl-ihistidyl-rproline amide chased from Sigma. Esters of cellulose membranes (HAWP- (4, 5). The availability of the first synthetic hypothalamic re- HA 0.45 um) were obtained from the Millipore Co. leasing hormone in a pure form opens new possibilities for the study of the mechanism of action of this peptide in the anterior Isolation of Adenohypophyseal Plasma Membranes. Anterior pituitary gland. pituitaries from adult cattle were collected in local slaughter- There is strong evidence of a role for adenylate cyclase as houses and rapidly brought to the laboratory in ice-cold a mediator of the action of the hypothalamic releasing Krebs-Ringer bicarbonate buffer (24) containing 11 mM hormones (6-11). We have found (unpublished data) that D-glucose. Plasma membranes were isolated by our modifica- adenylate cyclase, the enzyme catalyzing the formation of tion (9) of Neville's technique (25). The fraction of a 3% cyclic AMP, is associated with the plasma membranes in homogenate of bovine anterior pituitary tissue in 1 mM anterior pituitary tissue. The possibility of a specific binding NaHCO3r5 mM mercaptoethanol that sediments between of [3HJTRH to the adenohypophyseal plasma membranes as 1220 and 2000 X g is fractionated by isopycnic centrifugation their primary action was somewhat strengthened by studies (25) in a stepwise sucrose gradient of densities 1.14-1.22. with radio-iodinated adrenocorticotropin (12, 13), insulin As evidenced by electron microscopy and assays of enzyme (14-18), angiotensin (19), and glucagon (20-22), that indicate markers, the material sedimenting at the interfaces 1.14- 1.16 that the first step in the action of these polypeptide hormones and 1.16-1.18 consists of pure plasma membranes. Protein is their interaction with specific recognition sites on the plasma concentration was determined (26) with bovine serum albumin membrane of target cells. This paper describes an assay for as standard. measurement of [3H]TRH binding and the pertinent prop- Binding of [3H]TRH to Adenohypophyseal Plasma Mem- branes. In the standard assay, 50 jig of plasma-membrane Abbreviation: TRH, thyrotropin-releasing hormone. protein and 25 nM (27,500 cpm) [3HJTRH are incubated, in 283 Downloaded by guest on September 26, 2021 284 Biochemistry: Labrie et al. Proc. Nat. Acad. Sci. USA 69 (1972)

a final volume of 55 1l, in buffer F (20 mM sodium phosphate 20, buffer (pH 7.35)-7.5 mM KCl-2 mM MgCl2), at 00C for 40 min. The reaction is stopped by the addition of 1.0 ml of ice-cold buffer F, and the medium is immediately filtered with z 16 E gentle suction through a Millipore filter, with three successive 0 washes of the adsorbed material with 2 ml of the ice-cold buffer. After the filter is dried at 600C for 30 min, 10 ml of toluene-based scintillation fluid is added and the radioactivity is measured in a Packard Liquid Scintillation Spec- trometer. Unless otherwise stated, all assays are done in triplicate. z he RESULTS s Time Course of [8H]TRH-Receptor Association and Disso- ciation. Fig. 1 shows that half-maximal binding of [8H]TRH to the adenohypophyseal plasma membranes is obtained after 5 min of incubation at 00C, a plateau being reached between 20 and 30 min. No appreciable change in binding occurs when 0 2 4 6 8 10 12 14 16 incubation is extended to 150 min (data not shown). When Minutes excess unlabeled TRH (0.1 mM) is added after formation of FIG. 2. Time course of dissociation of bound ['HJTRH from the [3H]TRH-receptor complex has reached equilibrium, the adenohypophyseal plasma membranes at 00C. 25 nM [3HJTRH complex dissociates, with an initial half-life of 14 4- 1 min and plasma membranes (1 mg of protein/ml) were incubated for (Fig. 2). 40 min at 00C before the addition of 0.1 mM unlabeled TRH and Plasma Membrane Concentration. With 80 nM [3H]TRH, measurement of the remaining bound [3H]TRH at the indicated the binding is proportional to membrane concentration up times after addition of the unlabeled hormone. to 1.6 mg/ml (Fig. 3). A concentration of 1 mg of protein per increase in TRH binding between 2.5 and 5.0 mM K+, a ml was used in the standard binding assay. At saturating plateau being observed between 5 and 25 mM, with a sub- concentrations of [3H]TRH, about 40-fold purification of sequent decrease reaching values below the control at the number of TRH binding sites is usually observed in the 100 mM K+ (data not shown). When NaCl is added to the plasma membrane fraction as compared to the total adeno- standard incubation medium, there is no effect of Na++ ions hypophyseal homogenate. up to a concentration of 120 mM. Maximal stimulation by Effect of Potassium, Magnesium, Manganese, Calcium, and Mg++ is observed at concentrations of 0.5-2.5 mM, with Chelating Agents on [8H]TRH Binding. There is a sharp subsequent inhibition of binding at higher concentrations (Fig. 4). When the binding assay is performed in the presence of 1 mM Mg++, Mn++ gives a pattern similar to that of Mg++, while Ca++ is inhibitory at all concentrations (Fig. 4). Equilibrium Constant of TRH-Receptor Interaction. There is 30 a small amount of [8H]TRH that binds unspecifically to the

a, 0

*-20 0 E 80 _

a,n1 o 60 z S 40

o40 C

20e 0 L 0 10 20 30 40 50 60 MINUTES OF INCUBATION 0

FIG. 1. Time course of binding of [3H]TRH to adenohypo- 0 20 40 60 80 physeal plasma membranes. 25 nM [3H]TRH (27,500 cpm) and ,ag -membrane protein anterior pituitary plasma membranes (1 mg of protein/ml) were incubated at 00C. The bound fraction was separated by adsorp- FIG. 3. Effect of protein concentration of the anterior pituitary tion on Millipore membranes. plasma membrane fraction on [3H]TRH binding. Downloaded by guest on September 26, 2021 Proc. Nat. Acad. Sci. USA 69 (1972) Plasma-Membrane Binding of Thyrotropin-Releasing Hormone 285

50 In view of the well-known inhibitory feedback effect of thyroid hormone on TRH action in the anterior pituitary gland (29, 30), it is interesting to note that thyroxine or triiodothyronine have no influence on the binding of TRH to plasma membranes (Table 1), thus providing evidence for an action of thyroid hormone at a site subsequent to the binding of TRH to its receptor. DISCUSSION Despite its central importance in the direct control of thyrotropin secretion, and its indirect effect on the secretion of the many hormones influenced by the concentration of cir- culating thyroid hormone (31, 32), little is understood about the mechanism of action of TRH in molecular terms. The availability of synthetic TRH labeled at high specific activity in its proline residue makes it possible to study the interaction of this very important molecule with its receptors in the anterior pituitary gland. Such a study avoids limitations inherent in the use of radiolabeled derivatives of peptide hormones. A high degree of specificity of the TRH receptor is sug- gested by the absence of competition for TRH binding by two 8 12 peptide hormones of hypothalamic origin, and by six other Concentration (mM) peptide hormones. The interaction of TRH and its receptor follows standard bimolecular reaction kinetics, and the FIG. 4. Effect of Mg +, Mn++,and Ca++on ['H]TRH binding to adenohypophyseal plasma membranes. 1 mM Mg++ reaction is reversible. Unlabeled TRH displaces the labeled was presented during incubation with Mn++ or Ca++. *@, hormone in proportion to its relative concentration; virtually Mg++**,Mn +;**,Ca ++ complete inhibition of [3H]TRH binding was observed in the presence of 10 gM unlabeled TRH. However, when [3HJTRH and receptor have reached equilibrium, excess membrane (lower curve, Fig. 5A). This nonspecific component unlabeled TRH usually fails to displace 20-40% of the is related linearly to the concentration of labeled TRH, and bound radioactivity after 1 hr of incubation. The remaining accounts for about 0.05% of the total radioactivity filtered bound radioactivity is somewhat higher than the value through the membrane and for about 5% of the radioactivity expected from the rate of dissociation measured in the first bound in the presence of membranes. This nonspecific 20 min. Under comparable conditions, only 10-20% of glu- adsorption was subtracted from the total radioactivity cagon was displaced from liver membranes at 00C in the bound to the filter. presence of 1 mM EDTA, and negligible dissociation could be With increasing concentrations of [3H]TRH, the hormone measured in the absence of the chelating agent (21). About binding to plasma membranes follows the equation y = 10% of bound '25I-labeled insulin was not displaced by STRHU/[U+(1/KTRH) 1, where STRH is the number of TRH- excess unlabeled hormone (16). There is no satisfactory binding sites, U is the unbound TRH, and KTRH is the explanation for these results. equilibrium constant of the TRH-receptor interaction. The concentration of TRH giving half-maximal binding to KTRH was measured as 4.3 X 107 L - M 1 and STRH as the receptor (23 nM) is in the range of concentrations found 0.5 nanomoles * L-1 or 600 femtomoles per mg of membrane efficient in eliciting the best-known effect of this neuro- protein (Fig. 5A and 5B). hormone, TSH release by the anterior pituitary gland incubated in vitro (33). It is most likely that the concentration Specificity of [3H]TRH Binding. Increasing concentrations of TRH in the portal blood supplying the TSH-secreting of unlabeled TRH lead, by dilution, to a progressive decrease cells of the anterior pituitary gland is in this range of con- of binding of the labeled hormone. Virtually complete inhibi- centrations. tion of binding is observed after the addition of 10 uM Since calcium is required for the TRH-induced release of unlabeled TRH. 10 uM lysine-vasopressin and melanocyte- thyrotropin (34), it is noteworthy that addition of calcium stimulating hormone-release inhibiting hormone, two oligo- depresses binding of the labeled hormone at all concentrations peptides of hypothalamic origin, have no effect on [3H]TRH studied. It should be mentioned that physiological con- binding (Table 1). Similarly, no competition is observed centrations of Ca++ only inhibit [8H]TRH binding about 5%. when various peptide hormones (porcine adrenocorticotropin, These findings eliminate the TRH-receptor interaction as a bovine luteinizing hormone, bovine growth hormone, bovine possible site for the permissive action of Ca++ on thyrotropin prolactin, bovine insulin, and porcine glucagon) are added at release. Likewise, the binding of ACTH to the adrenal 10 JAM, well above the physiological plasma concentration receptor does not require Ca++ (13) and high concentrations (Table 1). Coupled with the dilution experiment with un- of Ca++ depress hormone binding. This inhibitory effect of labeled TRH, these data indicate that the site of binding of Ca++ might be secondary to competition of Ca++ with the TREI is highly specific, and validate the use of the described Mg++ sites on the hormone receptor, or to some direct assay for investigation of the specific binding of TRH. interaction of Ca++ with the receptor or the hormone. Downloaded by guest on September 26, 2021 286 Biochemistry: Labrie et al. Proc. Nat. Acad. Sci. USA 69 (1972) TABLE 1. Effect of unlabeled TRH, lysine-vasopressin, 60 A melanocyte-stimulating hormone-release inhibiting hormone, adrenocorticotropin, growth hormone, prolactin, luteinizing hormone, insulin, glucagon, thyroxin, and triiodothyronine on E 50 / the binding of 25 nM [3H]TRH to adenohypophyseal IW plasma membranes*

I- 40 Femtomoles of ['H]TRH Added hormone bound/mg membrane protein P-

E 30- Control 368 ± 14

n TRH 8 MSH-release inhibiting hormone 334 ± 6 co 20 Lysine-vasopressin 344 i 12 w Adrenocorticotropin 316 i 22 tr Growth hormone 360 i 22 Prolactin 350 i 20 Insulin 366 + 6 10 Glucagon 352 +t 8 ± 0.025 0.25 0.5 1.0 Control 370 20 L-thyroxine (5 376 i 38 [3H] TRH ( x let M) pig/ml) L-thyroxine (0.5 pg/ml) 394 + 18 L-triiodothyronine (0.2 ,ug/ml) 364 + 22 B L-triiodothyronine (0.02 ,ug/ml) 348 + 16

*Incubations were performed as described under "Methods", 8.0 except for the presence of indicated hormones. The molecular weights of growth hormone and prolactin were taker, respectively, as 20,846 and 23,500 (27, 28). .

- 6.0 This research was supported by Grant MA-3525 from the Medical Research Council of Canada. F. L. and N. B. are, respectively, Scholar and postdoctoral Fellow of the Medical Research C- Council of Canada. The technical assistance of A. Petitclerc and N. Lemay is gratefully acknowledged. .4e 4.0 1. Burgus, R. & Guillemin, R. (1970) Annu. Rev. Biochem. 39, 499-526. 2. McCann, S. M. & Porter, J. C. (1969) Physiol. Rev. 49, 240-284. 20 3. Mitnick, M. & Reichlin, S. (1971) Science 172, 1241-1243. 4. Boler, J., Enzman, F., Folkers, K., Bowers, C. Y. & Schally, A. V. (1969) Biochem. Biophys. Res. Commun. 37, 705-710. 5. Burgus, R., Dunn, T. F., Desiderio, D. & Guillemin, R. (1969) C. R. H. Acad. Sci. 269, 1870-1873. 6. Labrie, F., Beraud, G., Gauthier, M. & Lemay, A. (1971) YTRH J. Biol. Chem. 246, 1902-1908. (nM) 7. Labrie, F., Lemaire, S. & Courte, C., J. Biol. Chem., in FIG. 5. A. Effect of increasing concentrations of [3H]TRH on press. 8. Lemaire, S., Pelletier, G. & Labrie, F., J. Biol. Chem., in the binding of the hormone to anterior-pituitary plasma mem- press. branes. *-*, unspecific adsorption to the filter; 0-0, binding in 9. Labrie, F., Poirier, G., Lemaire, S., Pelletier, G. & Boucher, the presence of 1 mg/ml of plasma membrane protein, correction R., J. Biol. Chem., in press. being made for nonspecific adsorption. 10. Wilber, J., Peake, G. T. & Utiger, R. (1968) Endocrinology B. Double-reciprocal plot of data obtained in an experiment 84, 758-760. performed as described in Fig. 5A. KTRH, from the slope of this 11. Fleischer, H., Donald, R. A. & Butcher, R. W. (1969) Amer. line, is 4.3 X 107 L-M-1. J. Physiol. 217, 1287-1291. 12. Lefkowitz, R. J., Roth, J., Pricer, W. & Pastan, I. (1970) Proc. Nat. Acad. Sci. USA 65, 745-752. There are very interesting recent data on the interaction 13. Lefkowitz, R. J., Roth, J. & Pastan, I. (1970) Nature 228, of ACTH (12, 13), insulin (14-18), and glucagon (20-22) 864-866. 14. Cuatrecasas, P. (1971) Proc. Nat. Acad. Sci. USA 68, 1264- with receptors in their target cells. Among the striking 1268. similarities of the characteristics of binding of these relatively 15. House, P. D. R. & Weidemann, M. J. (1970) Biochem. large peptides with those of the tripeptide TRH, there are Biophys. Res. Commun. 41, 541-548. the high equilibrium constant and specificity and, possibly 16. Freychet, P., Roth, J. & Neville, D. M., Jr. (1971) Proc. 1833-1837. more important, the location of the receptor in the plasma Nat. Acad. Sci. USA 68, 17. Freychet, P., Roth, J. & Neville, D. M., Jr. (1971) Biochem. membrane, which is also the main site of the adenylate cyclase Biophys. Res. Commun. 43, 400-408. in mammalian tissues (35-36), including the anterior pituitary 18. Freychet, P., Roth, J. & Neville, D. M., Jr. (1971) J. Clin. gland. Invest. 50, 34a. Downloaded by guest on September 26, 2021 Proc. Nat. Acad. Sci. USA 69 (1972) Plasnma-Membrane Binding of Thyrotropin-Releasing Hormone 287

19. Lin, S. Y. & Goodfriend, T. L. (1970) Amer. J. Physiol. 29. Vale, W., Burgus, R. & Guillemin, R. (1968) Neuroendocrin- 218 (5) 1319-1328. ology 3, 34-46. 20. Tomasi, V., Koretz, S., Ray, T. K., Dunnick, J. & Marinetti, 30. Bowers, C. Y., Lee, K. L. & Schally, A. V. (1968) Endo- G. V. (1970) Biochem. Biophys. Acta 211, 31-42. crinology 82, 75-82. 21. Rodbell, M., Krans, H. M. J., Pohl, S. L. & Birnbaumer, L. (1971) J. Biol. Chem. 246, 1861-1871. 31. Labrie, F., Pelletier, G., Labrie, R., Ho-Kim, M. A., 22. Rodbell, M., Krans, H. M. J., Pohl, S. L. & Birnbaumer, L. Delgado, A., MacIntosh, B. & Fortier, C. (1968) Ann. (1971) J. Biol. Chem. 246, 1872-1876. Endocrinol., Paris, 29, 29-43. 23. Burgus, R., Dunn, T. F., Desiderio, D., Ward, D. N., Wale, 32. Labrie, F., Pelletier, G., Raynaud, J. P., Ducommun, P. & W. & Guillemin, R. (1970) Nature 226, 321-325. Fortier, C. (1969) in Metabolisme Pgriphirique et Transport 24. Cohen, P. P. (1957) in Monometric Techniques, ed. Umbreit, Humoral des Hormones Thyroidiennes et Stgroides (Masson & W. W., Burris, R. H. & Stauffer, J. F. (Burgess Publishing Cie, Paris), p. 115. Co., Minneapolis), p. 149. 33. Bowers, C. Y., Weil, A., Chang, J. K., Sievertsson, H., 25. Neville, D. M., Jr. (1960) J. Biophys. Biochem. Cytol. 8,413- Enzmann, F. & Folkers, K. (1970) Biochem. Biophys. Res. 422. Commun. 40, 683-691. 26. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, 34. Vale, W., Burgus, R. & Guillemin, R. (1967) Experientia 23 R. J. (1951) J. Biol. Chem. 193, 265-275. 853-855. 27. Fellows, R. E. & Rogol, A. D. (1969) J. Biol. Chem. 244, 35. Sutherland, E. W., Rall, T. W. & Menon, T. (1962) J. Biol. 1567-1575. Chem., 237, 1220-1227. 28. Cheever, E. V. & Lewis, J. J. (1969) Endocrinology 85, 465- 36. Robison, G. A., Butcher, R. W. & Sutherland, E. W. (1969) 473. Annu. Rev. Biochem. 37, 149-174. Downloaded by guest on September 26, 2021