studies of thyroid-stimulating hormone binding to bovine thyroid plasma membranes

Studies of Thyroid-stimulating Hormone Binding to Bovine Thyroid Plasma Membranes

Masanobu Kotani, Toshitsugu Kariya, and James B. Field

13’I-TSH prepared by the lactoperoxidate

method was used to study the binding of hormone to bovine thyroid plasma membranes. Specific binding was obtained using as little as 0.12 mu/ml 1311-TSH. Half-maximal binding occurred with 17.1 + 3.5 mu/ml and saturation a+ approximately 40 mu/ml. Scatchard plot analysis revealed two classes of binding sites, with association constants of 1 .l f 0.06 x lO’M_‘and 1.4 x 10’1M~ for the high- and low-affinity sites, respectively. Binding of 1311-TSH was linearly related to the amount of thyroid plasma membrane protein. Other polypeptide hormones and prostaglandin E, did not inhibit specific

TSH binding. Identical results were obtained using two TSH preparations of different biologic specific activity 12.5 mU/ ml unlabeled TSH decreased ’ ‘I-TSH binding 50%, and 156 mu/ml caused com-

P lete inhibition. After equilibrium of 311-TSH binding was established, max-

imal displacement was achieved by 120 min using about 300 mu/ml TSH. How- ever, only about one-half of the “‘I-TSH was displaced. Although GTP potentiated the stimulation of adenylate cyclase by TSH, it inhibited binding of 1311-TSH. Bind- ing of TSH correlated very well with activation of adenylate cyclase.

T HE INITIAL ACTION of thyroid-stimulating hormone (TSH) involves binding to receptors on the plasma membranetm6 and subsequent genera-

tion of increased cyclic 3’,5’-adenosine monophosphate (CAMP) in the cell.7m9 The augmented CAMP probably mediates the effects of TSH by dissociating protein kinase into its receptor and catalytic subunits.‘0-‘5 Previous reports of modification of stimulation of adenylate cyclase by TSH have not discriminated between binding of the hormone or activation of the enzyme.8~9J6J7

Although binding of TSH fo receptors on thyroid cells and plasma membranes has been reported, 2-6 the results have not always been consistent. Amir et al. reported that binding of ‘H-TSH did not correlate with biologic activity,* while Manley et al. obtained identical binding based on biologic potency.3 Dis-

agreement also exists as to whether there are one2.4-6 or two3 orders of binding sites. Differences of time and temperature of TSH binding have also been re-

ported.2-4.6 The present studies examined the binding of ‘3’I-TSH to thyroid plasma

membranes and correlated binding to stimulation of adenylate cyclase activity in such preparations.

From the Clinical Research Unit and the Department of Medicine, Universitv of Pirtsburgh School of Medicine. Pittsburgh, Pa.

Receivedfor publication November 12. 1974. Supported by USPHS. NIH Grant AM-06865. Reprint requesis should be addressed to James B. Field, M.D., Clinical Research Unit. 3304

Presbyterian-Universify Hospital. 230 Lothrop Street, Pittsburgh. Pa. 15261. Abbreviations used in this paper are: TSH. rh_vroid-stimulating hormone; CAMP, cvclic 3’S’-

adenosine monophosphare: BSA, bovine serum albumin; ACTH, adrenocorticotrophin: PGE,. prosraglandin E,

0 1975 bv Grune & Stratton, Inc.

Metabolism, Vol. 24, No. 8 (August), 1975 959

960

MATERIALS

Iodination of Bovine TSH

TSH was iodinated by the lactoperoxidase In . . .

KOTANI, KARIYA, AND FIELD

AND METHODS

method of Thorell and Johansson ‘* and Miyachi

et al.” Carrier free Na “‘I (l-3 mCi) was added to IO ~1 of 0.1 M potassium phosphate

buffer, pH 7.4, at 22’C. Lactoperoxidase (10 ~1, 100 rg/ml) in 0.1 M sodium acetate buffer, pH

5.6, was added, followed by 5 ~1 of bovine TSH (30 U/mg/ml). Five microliters of 0.88 mM

HzOz initiated the reaction. Hz02 (5 ~1) was added after 10 min and the reaction stopped at 20

min with 0.5 ml of 0.1 M potassium phosphate buffer. 13’I-TSH was separated from Na 13’1 by

Sephadex G-25 column chromatography (I x I5 cm). One-milliliter fractions were collected using

0.1 M potassium phosphate buffer containing 1% bovine serum albumin (BSA). 13’1-TSH eluted

as a sharp peak in fraction 5, while 13’1 was present in fractions 9-14. The specific activity of “‘I-TSH ranged from 70 to 160 pCi/pg. 13’1-TSH was further purified using Sephadex G-100

column chromatography (1 x 20 cm). One-milliliter fractions were eluted with 0.1 M potassium

phosphate buffer containing 1% BSA. They were stored at - 20” C until used.

Chromatography of 13’1-TSH (8 PCi, 5 x IO6 cpm) and unlabeled TSH (2 U, 30 U/mg) was

compared using a Sephadex G-100 column and eluted as outlined above. Radioactivity of the

l-ml fractions was determined and biologic activity assayed based on CAMP generation in dog-

thyroid slices.*‘**’ Radioactivity and biologic activity were closely associated in fractions 6-9 (Fig.

1). Fractions 7, 8, and 9 were satisfactory for binding of 13’1-TSH to plasma membranes.

Biologic activity of TSH treated with lactoperoxidase and H202 was assessed. Ten microliters of

TSH (500 pg, 1500 mu), IO ~1 of 0.1 M potassium phosphate buffer, and IO ~1 of lactoperoxidase

(IO mg/ml) in 0.1 M sodium acetate buffer were mixed. Addition of 5 rl of 88 mM H202 initiated the reaction. After IO min. 5 pl of Hz02 was again added. The reaction was terminated IO min

later by addition of 0.5 ml 0.1 M potassium phosphate buffer. An identical amount of TSH was

treated in a similar fashion except that lactoperoxidase and H202 were not added. After ap-

propriate dilutions, the reaction mixtures were assayed for biologic activity. No loss of biologic

activity resulted from treatment of TSH with lactoperoxidase and H202 (unpublished observa-

tions).

Measurement of ‘:“I-TSH Binding to Plasma Membranes

The buffer for binding contained 0.02 or 0.08 M Tris-HCI (pH 7.5) 3.5 x 10m3 M MgSOd.

I x 10m3 M EDTA, IO-* M theophylline, 2.1% BSA, 6 x 10m5 M ATP, 4 x 10m3 M cyclic AMP,

TUBE NUMBER

Fig. 1. Comparison of radioactivity and biologic activity of ‘3’I-TSH and unlabeled TSH after

column chromatography. TSH (2 U) and 13’I-TSH (8 pCi and 5 x 10’ cpm) were applied to a

Sephadex G-lOOcolumn and eluted with 0.1 M potassium phosphate buffer containing 1% BSA.

One-milliliter fractions were collected, and radioactivity and biologic activity determined.

THYROID-STIMULATING HORMONE BINDING 961

and an ATP regeneration system consisting of 2 x IO-‘M phosphoenol pyruvate and 250 pg per

ml pyruvate kinase. The final incubation volume included 100 ~1 of buffer, 15 ~1 of plasma mem-

branes (20-50 rg of protein), 20 ~1 of ‘3’I-TSH (9 x 104-20 x IO4 cpm, 90 rU-400 rU), and 25

~1 of0.08 M Tris-HCI buffer. Unlabeled TSH, when present, was added in 25 ~1 of 0.08 M Tris-

HCI buffer. Incubation was at 22°C for 120 min unless stated otherwise. ‘3’I-TSH bound to

membranes was separated from free hormone by the method of Amir et aL2 Tris-HCl-2.5% BSA

buffer (0.4 ml) was added to the incubation mixture which then was quantitatively applied to a

Millipore filter. Filters were presoaked with 10% BSA for 2 hr and washed with 1.5 ml of 0.02 or

0.08 M Tris-HCI buffer containing 2.5’4 BSA. After vacuum filtration, the filter was washed twice

with 1.5 ml of Tris-HCI-2.54/ BSA buffer. Filters were transferred to counting tubes and radio-

activity determined in a Packard Auto-gamma scintillation counter.

Nonspecific binding is defined as the 13’I-TSH which was not displaced by addition of 100~500

mU of unlabeled TSH at the onset of the incubation. Specific binding was corrected for nonspecific

binding. ‘3’I-TSH not bound to membranes but absorbed on the filter was determined using the

complete incubation mixture but without plasma membranes. Such binding of 13’I-TSH to filters

was less than 0.3% of the total radioactivity added to the incubation mixture and less than 25”,;

of the 13’ I-TSH bound by membranes. This was also subtracted in calculating specific binding.

The microcentrifugation method** gave identical results (unpublished observations).

Measurement of Dissociation of Bound 13’I-TSH

‘3’I-TSH was bound to plasma membranes as described above in a volume of 0.145 ml at 22°C

for 120 min. Unlabeled TSH was then added and the incubation continued. At the appropriate

times, free and membrane-bound hormone were separated.

Preparation of Bovine Thyroid Plasma Membranes

Bovine thyroid, rat liver, and rat kidney plasma membranes were prepared by the method of

Yamashita and Field.’ They were stored at -20°C for as long as 3 mo without loss of binding

capacity or adenylate cyclase activity. Adenylate cyclase activity was determined as previously

described.2’ Protein was measured by the method of Lowry et a1.23

Dr. J. G. Pierce, Department of Biological Chemistry. University of California School of

Medicine, Los Angeles, Calif., generously provided TSH (30 U/mg). TSH (NIH-B6, 2.5 U/mg)

and prolactin (NIH-PS-10) were supplied by the National Institutes of Arthritis, Metabolism,

and Digestive Diseases. Rat 13’1-TSH, prepared by the chloramine-T method, was donated by

Dr. P. Reed Larsen, Department of Medicine, University of Pittsburgh School of Medicine. Rat

TSH and bovine TSH had identical biologic activity (unpublished observations). Lacto-

peroxidase, grade B, was purchased from Calbiochem Corporation. Union Carbide Company

provided Na 13’1 in 0.05 N NaOH solution. Si4C ATP (27 mCi/mM) and 3H CAMP (24.1

Ci/mM) were purchased from Schwartz Bioresearch. GTP was from Sigma Chemical Company,

Dr. John Pike, Upjohn Company, kindly provided prostaglandin E’ (PGE’ ). Porcine insulin was

a gift from the Lilly Research Laboratories, and ACTH was obtained from Parke-Davis Company.

RESULTS

Binding of 13’1-TSH prepared by the lactoperoxidase method indicated that this preparation was better than chloramine-T labeled hormone (Table 1). Non-

specific binding of the former preparation varied from 6’%-21% of the total binding, compared to approximately 70% for the chloramine-T labeled hormone. The data in Fig. 2 demonstrate specific binding of 13’I-TSH using as little

as 0.12 mu/ml TSH and increasing in a relatively linear fashion with increasing amounts of hormone. Using 25 rg of membrane protein, binding was not sat- urated with 4.6 mU/ml of ‘3’I-TSH. Using a constant amount of 13’1-TSH (1.3 x 105cpm, 234 PU), binding was linearly related to the concentration of plasma membranes up to 0.5 mg protein per ml (Fig. 3). Approximately 500,” of bound 13’ I-TSH was displaced using 5 pg/ml TSH (12.5 mu/ml), and al-

962 KOTANI, KARIYA, AND FIELD

Table 1. Comparison of Binding to Thyroid Plasma Membranes of ‘N I-TSH Prepared by Chloramine-T and lactoperoxidase Methods

Type of Binding

(cpm boundfmg membrane protein)

Method of Preparation of 13’ I-TSH

Loctoperoxidase

Chloramine-T Before Fractionation Fraction 8

Total binding 3 1,000 45,500 50,900

Nonspecific binding 2 1,500 10,700 8,100

Specific binding 9,500 34,800 42,800

Thyroid plasma membranes (50 pg of protein) were incubated at 22°C for 2 hr in 0.16 ml containing

1 x lo6 cpm of each labeled hormone preparation. The loctoperoxidose preparation was tested before

and after purification on Sephadex G-100 as described in Materials and Methods. Nonspecific binding

was measured in the presence of 100 mU unlabeled TSH per tube. The specific radioactivities were 21

pCi/pg for the chloramine-T and 160 pCi/pg for the loctoperoxidose preparations, respectively. The

concentration of TSH used with chloromine-T was 2 U/mg while that with lactoperoxidase was 30 U/mg.

The results ore the average of closely agreeing duplicate determinations.

/

.

’ “Iv , , , , Fig. 2. Binding of ‘3’I-TSH to thyroid plasma membranes as a function of hormone

0 , 2 3 4 5 concentration. The plasma membranes (25

13’1 TSH CONCENTRATION pg of protein) were incubated with the appropriate amounts of 13’ I-TSH in 0.16 ml at

( mu/ml 1 22’C for 120 min.

60

I

Fig. 3. Binding of 13’ I-TSH to thyroid plasma membranes as o function of membrane concentration. The incubation was at 22°C for 120 min in 0.16 ml containing 234 PU 13’1-TSH (1.3 x lo5 cpm).

.li’Le 0.4 0 0.2

MEMBRANE CONCENTRATION (mg/ml)


Fig. 4. Effect of unlabeled TSH and other hormones on displacement of

specific 13’ I-TSH binding to thyroid

plasma membranes. The incubation

medium contained ?ji6 pg of membran;

protein and 91 &I I-TSH (1.5 x 10

cpm). Incubation was at 22°C for 120

min. The unlabeled hormones were

added at the beginning of the incuba-

tion.

60

g 40

I

r 20

A ACTH

q Proloctin

1 Insulin

0 Prostoglandln E,

OJ 0 100 200 300

HORMONE ADDED (yg/ml)

most complete displacement occurred with 62.5 pg/ml (156 mu/ml) (Fig. 4). Similar results were obtained with both 2.5 U/mg and 30 U/mg TSH when the

data were expressed based on biologic activity (unpublished observations). The

bound TSH was not displaced by large amounts of other polypeptide hormones or by PGE,. These concentrations of PGEI were sufficient to increase adenylate

cyclase activity in thyroid plasma membranes”~25 and mimic other effects of

TSH on thyroid slices. 21*26 “‘I-TSH did not bind specifically to rat liver or kid-

ney plasma membranes (Table 2). 13’1-TSH binding based on biologic activity was identical using two TSH preparations of widely different purity (Fig. 5).

A representative Scatchard plot analysis 27 of TSH binding demonstrated two classes of binding sites (Fig. 6). The assumption was made that pure TSH con-

tains 50 U of biologic activity per mg and has a molecular weight of 28,000.

Since the unitage of pure TSH is unknown, the association constants are at best only an approximation. The average association constant, K,, of the high-

affinity and low-capacity site was 1.1 f 0.06 x lo* M-’ (five different plasma

membrane preparations). The association constant of the low-affinity and high- capacity site was 1.4 x lo7 M-‘. After 120 min incubation with 13’1-TSH, excess unlabeled TSH displaced the bound hormone (Fig. 7). Maximum displacement occurred by 120 min of incubation, although about 507; of the 13’1-TSH still remained bound. Displacement of previously bound 13’I-TSH was achieved with 15 mu/ml unlabeled TSH and near maximal effects with about

Table 2. Comparison of 13’ I-TSH Binding to Plasma Membranes

Prepared From Beef Thyroid, Rat Liver, and Rat Kidney

131 I-TSH Binding pU/mg Membrane Protein

Source of Membranes Experiment 1 Experiment 2

Beef thyroid 246 43

Rat liver 3 1

Rat kidney 19 0

Plasma membranes (thyroid, 25 pg. experiments 1 and 2; liver, 33 ~9, experiment 1 and 66 ~9, experi-

ment 2; kidney, 26 pg, experiment 1 and 52 pg. experiment 2) were incubated for 120 min at 22°C. The

incubation mixture contained 81 PU and 97 MU I31 I-TSH in experiments 1 and 2, respectively.

1 1

x

.

x . x NIH-TSH

/-.

x l Pierce -TSH

0 1 0 5 IO 15

TSH (mu/ml)

KOTANI, KARIYA, AND FIELD

Fig. 5. Comparison of TSH propara- tions of different purity on “’ I-TSH binding to thyroid plasma membranes. The incubation medium contained 192 pU 13’I-TSH and appropriate amounts of unlabeled TSH (either NIH-B6, 2.5 U/mg or Pierce TSH, 30 U/mg). The incubation was for 120 min at 22°C.

Fig. 6. Scatchard plot of the specific binding of TSH to thyroid plasma membranes. BOUND TSH (mU/ml)

Fig. 7. Time course of displacement of 13’ I-TSH bound to thyroid plasma membranes by addition of excess unlabeled TSH. lncobation of 27 vg of plasma membrane protein with 214 rtJ 13’ I-TSH for 120 min at 22°C in 0.145 ml was followed

. by addition of 400 mu/ml unlabeled TSH and incubation for an additional appropriate time. The specific bind-

. ing to thyroid plasma membranes 60 120 I.90 at the end of the initial 120 min

MINUTES incubation represents 100% binding.


. Fig. 8. Displacement of lO0.k

‘3’I-TSH bound to thyroid plasma membranes by addition of increasing amounts of unlabeled 80- TSH. Incubation of 25 fig of .* plasma membrane protein with 4 \

97 FlJ 13’1-TSH for 120 min at $ \ \

22°C in 0.145 ml was followed ap 60. .\., by addition of the appropriate ~\ amount of unlabeled TSH and \LV

__ incubation for an additional 120 min. The specific binding to thyroid plasma membranes at the end of the initial 120 min incubation is taken as 100% binding.

40s

0: 0 loo 200 300 400 500 600 700

TSH ImU/mll

300 mu/ml (Fig, 8). Even with 625 mu/ml unlabeled TSH, approximately

407; of the bound 13’I-TSH could not be displaced. Binding of 1311-TSH was dependent on both temperature and time (Fig. 9).

Very little specific binding occurred at 4°C. At 22°C binding began at 1 min and increased gradually to near maximum by 120 min. Initially, binding was more

rapid at 37°C but the peak was significantly less than at 22°C. By 240 min at 37”C, binding was actually decreased. The optimal pH for binding was 7.3. At pH 6.0 binding was approximately 70% of that at pH 7.3. The Tris concentra-

tion influenced binding. It was maximal at 0.08 M Tris and reduced by 40”,/, and 6O”,b at 0.2 M and 0.3 M Tris, respectively.

Although GTP potentiates the TSH stimulation of adenylate cyclase,24 it does

not increase binding of ‘3’I-TSH to the plasma membranes (Fig. 10). With

10m3 M GTP, binding of TSH was reduced approximately 507;. Figure 11 com-

pares binding of TSH and its stimulation of adenylate cyclase activity. A very

good correlation exists between the concentrations of hormone producing both effects. The half-maximal binding of TSH occurred with 17.1 =I= 3.5 mu/ml (mean f SEM of four experiments) and half-maximal activation of adenylate

40- - 22°C

*_______* 37” c

t--4 4’C

Fig. 9. Effect of temperature and :iTe on binding of

..A.---- ,‘_..A I-TSH to thyroid

..I/ plasma membranes. .-’ Incubation medium

contained 50 fig of . plasma membrane

15 15 30 45 60 240 protein and 234 FU MINUTES ‘3’I-TSH.


2 OLL ,o 0 to-8 10-7 10-G 10-5 10-4

’ .Jo 10-3 5

GTP (Ml Y

Fig. 10. Effects of GTP on stimulation of adenylate cyclase activity by TSH and binding of 13’1-TSH to thyroid plasma membranes. For measurement of binding of l3 I-TSH, the incubation medium contained 22 Ag of plasma membrane protein and 192 AU 13’1-TSH. Incubation was at 22°C for 120 min. Assay of adenylate cyclase activity was done using 50 Ag of plasma membrane protein in a volume of 0.11 ml. Incubation was for 10 min at 37°C. The TSH concentration was 91 mu/ml.

I 0 . . . . . . .._....- - ._.... - 1600

-G __--

_____ 0

. 5 1500- % - 1400

yg % 1200

% IOOO-

e 3

- 1000

2

600 0 500 - 5 8 -600

H 400

0 20 40 60 60 100 IO

TSH (mU/ml)

Fig. 11. Correlation of binding of TSH and activation of adanylate cyclase in thyroid plasma membranes. In the binding studies, 27 pug of plasma membrane protein was incubated with 137 plJ ‘3’I-TSH for 120 min at 22-C. Adenylate cyclase activity was measured using 50 pg plasma membmne protein incubated in 0.11 ml for 10 min at 37°C. The buffer also contained 5 x 10e5 M GTP for the adenylate cyclase assay.


cyclase activity with 13.7 mu/ml (average of two experiments). Maximal ef-

fects on both parameters were obtained with about 40 mU /ml TSH.

DISCUSSION

The present study demonstrating specific binding of ‘3’I-TSH to thyroid

plasma membranes confirms and extends previous reports.2-6 Although

13’1-TSH prepared by the chloramine-T method has been used to measure TSH

binding’*’ our results indicate that the lactoperoxidase method is better (Table 1). The discrepancy between the two methods does not reflect species

differences since rat and bovine TSH had equal biologic activity when the results were expressed on the basis of stated biologic potency (unpublished ob-

servations). Amir et al. reported that ‘2SI-TSH prepared using very low concentrations of chloramine-T still retained 30x-40% of its biologic activity,

while 707; was bound by excess thyroid plasma membranes.40 In contrast, TSH treated with lactoperoxidase and Hz02 retained full biologic potency6 (un-

published observations). Over 800/o of hormone labeled by either method re-

acted with excess antibody, 3~6,40 but in view of the difference in biologic activity,

this might not be a good criterion to use. ‘H-TSH, which retained over 80% of its biologic activity,* also has been used for binding studies, but its specific radioactivity is less than iodinated preparations.4’

In our experiments specific binding represented 79%-94x of the total 13’I-TSH bound to plasma membranes, confirming similar results using isolated

thyroid cells4 and slices.’ Decreased specific binding of 13’1-TSH prepared with chloramine-T (Table 1) is consistent with the diminished biologic activity of

that preparation. However, in bovine thyroid slices, specific binding of

“‘I-TSH prepared with lactoperoxidase represented only about 33% of the total binding (unpublished observations). Half-maximal and maximal inhibi-

tion of “‘I-TSH binding to membranes by 12.5 mu/ml and 156 mu/ml TSH, respectively, is similar to what has been reported previously.3*4 This latter amount is significantly less than the 1.8 U/ml which Amir et al. reported was

necessary for complete inhibition of binding of ‘251-TSH prepared with

chloramine T.QO Binding of hormone to thyroid plasma membranes was specific for TSH

(Fig. 4), in confirmation of similar results previously published.2*4 Our observa-

tion that up to 300 pg/ml PGE’ did not inhibit binding of “‘I-TSH indicates separate receptor sites for the two substances. Moore and Wolff reached the same conclusion based on studies of ‘H PGEl binding to thyroid plasma mem-

branes.*” This amount of PGE’ exceeds that required for biologic effects on thyroid plasma membranes. 25 The absence of significant binding of ‘3’I-TSH to liver and kidney plasma membranes provides additional support for its specificity (Table 2).

In contrast to the observations of Amir et al.,* binding of 13’1-TSH was equivalent, based on biologic activity, regardless of the purity of the TSH preparation (Fig. 5). Our results confirm those of Manley et al. who also used TSH preparations varying tenfold in their biologic specific activity.3 If the biologic effects of TSH reflect initial binding to receptors on the plasma membrane, it is difficult to understand the discrepancy reported by Amir et al.* They


found decreasing binding of ‘H TSH per unit biologic activity as the purity of the TSH increased.’

The present results concerning effects of time and temperature on ‘3’I-TSH binding also differ from those of Amir et al.* They originally reported that maximum binding of ‘H-TSH was immediate and inversely dependent upon temperature. More recently they found that ‘3’I-TSH binding was half-maximal by lo-15 min and maximal at 2 hr.40 They found maximum binding was similar

at 0°C 15°C and 22°C but was decreased 45% at 37°C. Lissitzky et al. found

binding of TSH at 0°C was only 25% of that present at 35”C.4 The apparent decrease of binding at 37°C compared to 22°C (Fig. 9) could reflect more rapid

degradation of the receptors at the higher temperature. Lissitzky et al. also ob-

served that binding of TSH was time dependent.4 Such rapid binding of TSH is compatible with the rapid onset of effects of the hormone on thyroid plasma membranes,’ thyroid homogenates,” and intact thyroid cells.29

Controversy exists concerning the number of types of binding sites and their

association constants. The present results indicating two different classes of

receptor sites is in accord with that of Manley et al.,3 Moore and W01ff,~i and

Amir et a1.,40 but is in contrast to several others2,4-6 which indicated only a

single class of receptors. Some of this discrepancy may reflect the different types of thyroid preparations which have been used and the failure to use high concentrations of TSH. The association constant of the high-affinity, low-capacity

site was 1.1 f 0.06 x lo* M-‘, a value very similar to that obtained by Man-

ley et al.’ and Moore and W01ff.~’ It is two orders of magnitude less than the value of 2.6 x 10” M-l reported by Smith and Hall’ and one order of magni-

tude less than that obtained by Lissitzky et a1.,4 and Verrier et a1.6 Some of the differences might relate to the different incubation conditions, TSH prepara-

tions, and thyroid preparations which were used. Obviously the purity of the TSH preparation has a significant effect on the calculation of the association

constant. Since the specific biologic activity of pure TSH is not known, data of

binding experiments should be reported based on biologic activity. Some previous reports presented the data based on either molarity or weight of

TSH.2.4,6 Similar considerations are probably important in the reported differences for the association constant of the low-affinity, high-capacity site. Our value of 1.4 x 10’ M-l is very similar to that of 2.7 x lo7 M-’ observed by

Manley et al.,3 but is an order of magnitude less than that reported by Moore ’ and Wolff (0.7-1.0 x lo6 M- ). 41 Although the association constant for the

high-affinity, low-capacity site exceeds the physiologic concentrations of TSH,30

it is similar to that required for most of its in vitro effects.29 The explanation for this disparity is not clear.

‘311-TSH bound to thyroid plasma membranes could not be completely displaced by addition of a large excess of unlabeled hormone (Fig. 8). Verrier et al. reported a rapid component to the dissociation of 12’I-TSH from membranes, but even at the end of 50 min about 30% of the hormone was still bound.6 About 60% of I25 I-TSH was dissociated rapidly from cultured thyroid cells with a half-time of 3 min, while the remainder dissociated with a half-time of 30 min. Using 3H-TSH, Amir et al. observed that previously bound hormone could be


completely displaced from plasma membranes by unlabeled TSH within 3 min.2

However, only 40% of “‘1-TSH bound to guinea pig thyroid slices was dis-

placed over 1 hr and 99.5”,: over 30 hr when excess unlabeled TSH was added.3

This slower dissociation with this preparation is compatible with the slower

binding which the same authors reported. The apparent retention of a signiti- cant amount of I31 I-TSH to plasma membranes in our experiments is consistent

with the demonstration of persistent binding of the hormone to beef thyroid

slices when generation of CAMP was measured. 42 This suggests that TSH is not

rapidly metabolized by thyroid tissue, in contrast to results with insulin,

glucagon and hepatic plasma membranes.3’,32 In our system the pH optimum for binding was 7.3. Moore and Wolff re-

ported maximal binding at pH 5.5,4’ while Amir et al. found it at pH 6.0.40

Nonetheless, the latter investigators did their studies at pH 7.4 where binding

was only 257; of maximum.

Potentiation by GTP of stimulation of adenylate cyclase by TSH in membranes (Fig. 10) confirms earlier results.24,33*34 Such potentiation of hormone

stimulation of adenylate cyclase was first observed by Rodbell et al. using

glucagon and liver plasma membranes,35 and has also been noted in other tis-

sues.36,37 The stimulating effect of GTP was associated with a reduction of 13’1-TSH binding to plasma membranes (Fig. 10). Rodbell et al.,” using gluca-

gon and liver plasma membranes, and Soloff and Swartz,3g using oxytocin and a

particulate fraction obtained from mammary gland, reported similar findings. In contrast, Moore and Wolff reported that GTP had no effect on TSH binding

to thyroid plasma membranes despite potentiation of the stimulation of adenylate cyclase.4’ Because of the inverse relationship between effects of GTP

on hormone binding and activation of adenylate cyclase, Rodbell et al. postu- lated that dissociation of hormone from its receptor sites was more important

in initiation of its action.38

The correlation between binding of 13’ I-TSH and hormone activation of adenylate cyclase in thyroid plasma membranes provides further support for the biologic validity of our system. Since the biologic actions of TSH have been

attributed to activation of the adenylate cyclase-CAMP system,2’ binding of hormone should be intimately related to stimulation of adenylate cyclase. Verrier et al. also reported good correlation between binding of TSH and acti-

vation of adenylate cyclase.6 Half-maximum activation of adenylate cyclase required 1.8 nM TSH (approximately 2 mu/ml, assuming a TSH concentration

of 37 U/mg and a molecular weight of 28,000). Although Amir et al. found similar good correlation between binding of TSH and activation of adenylate

cyclase using between 1 and 8 nM TSH/ml, this relationship did not hold when more than 200 pg/ml plasma membrane protein was used in the assay.’ The concentrations of TSH which Amir et al.* used are much higher than those we required for either activation of adenylate cyclase or binding of TSH to plasma membranes. Based on a biologic activity of 6.8 U/mg, 1 nM/ml of TSH would be 196 mu/ml. Our studies were done using the same buffer system for both binding of ‘3’I-TSH and activation of adenylate cyclase so that the two activi- ties could be compared more directly. We found (unpublished observations),


as did Verrier et a1.6 and Moore and W~lff,~l that TSH binding was significantly decreased using the adenylate cyclase buffer compared to this buffer without ATP.

ACKNOWLEDGMENT

The authors are grateful to Mary Matthews and Barbara Sheehan for excellent help in the prep-

aration of the manuscript.

REFERENCES

I. Pastan I, Roth J, Macchia V: Binding of

hormone to tissue: The first step in polypeptide

hormone action. Proc Nat Acad Sci USA 56:

1802, 1966

2. Amir SM, Carraway TF Jr, Kohn LD:

The binding of thyrotropin to isolated bovine

thyroid plasma membranes. J Biol Chem 248:

4092, 1973

3. Manley SW, Bourke JR, Hawker RW: Re-

versible binding of labelled and non-labelled

thyrotrophin by intact thyroid tissue in vitro. J

Endocrinol55:555, 1972

4. Lissitzky S, Fayet G, Verrier B, Hennen

G, Jaquet P: Thyroid-stimulating hormone

binding to cultured thyroid cells. FEBS Let-

ters 29:20, 1973

5. Smith BR, Hall R: Binding of thyroid

stimulators to thyroid membranes. FEBS Let-

ters 42:301, 1974

6. Verrier B, Fayet G, Lissitzky S: Thyrotro-

pin-binding properties of isolated thyroid cells

and their purified plasma membranes: Relation

of thyrotropin-specific binding to adenylate-

cyclase activation. Em J Biochem 42:355, 1974

7. Yamashita K, Field JB: Preparation of

thyroid plasma membranes containing a TSH-

responsive adenyl cyclase. Biochem Biophys

Res Commun 40:171, 1970

8. Wolff J, Jones AB: The purification of

bovine thyroid plasma membranes and the

properties of membrane-bound adenyl cyclase.

J Biol Chem 24613939, 1971

9. Yamashita K, Field JB: Effects of long-

acting thyroid stimulation on thyrotropin stim-

ulation of adenyl cyclase activity in thyroid

plasma membranes. J Clin Invest 51:463, 1972 10. Yamashita K, Field JB: Cyclic AMP-

stimulated protein kinase prepared from bovine

thyroid glands. Metabolism 21: 150, 1972

11. Spaulding SW, Burrow GN: Several

adenosine 3’,5’-monophosphate-dependent pro-

tein kinases in the thyroid. Endocrinology 91:

1343, 1972

12. Corbin JD, Brostom CO, Alexander RL,

Krebs EC: Adenosine 3’,5’-monophosphate-

dependent protein kinase from adipose tissue.

J Biol Chem 247:3736, 1972

13. Gill GN, Garren LD: A cyclic 3’,5’-

adenosine monophosphate dependent protein

kinase from the adrenal cortex: Comparison

with a cyclic AMP binding protein. Biochem

Biophys Res Commun 39:335, 1970

14. Tao M, Salas ML, Lipmann F: Mech-

anism of activation by adenosine 3’,5’-cyclic

monophosphate of a protein phosphokinase

from rabbit reticulocytes. Proc Natl Acad Sci

USA 671408, 1970

IS. Rubin CS, Erlichman J, Rosen OM:

Molecular forms and subunit composition of a

cyclic adenosine 3’,5’-monophosphate-depen-

dent protein kinase purified from bovine heart

muscle. J Biol Chem 247:36, 1972

16. Haye B, Jacquemen Cl: Interaction de

la thyreostimuline avec ses recepteurs cellu-

laires: Effet de la phospholipase c sur la

fixation et L’activite biologique. FEBS Letters

18:47. 1971

17. Yamashita K, Field JB: The role of phos-

pholipids in TSH stimulation of adenylate cy-

clase in thyroid plasma membranes. Biochim

Biophys Acta 304:686, 1973

18. Thorell JI, Johansson BG: Enzymatic

iodination of polypeptides with “‘1 to high

specific activity. Biochim Biophys Acta 251:363,

1971

19. Miyachi Y, Vaitukaitis JL, Neischlag E,

Lipsett MB: Enzymatic radioiodination of

gonadotropins. J Clin Endocrinol Metab 34:

23, 1972

20. Bloom G, Field JB: Effects of subunits

on TSH and LH on TSH stimulation of cyclic

AMP accumulation in dog thyroid slices.

Endocrinology 92:642, 1973

21. Zor U, Kaneko T, Lowe IP, Bloom G,

Field JB: Effect of thyroid-stimulating hor-

mone and prostaglandins on thyroid adenyl

cyclase activation and cyclic adenosine 3’,5’-

monophosphate. J Biol Chem 244:5189, 1969

22. Rodbell M, Krans HMJ, Pohl SL,

Birnbaumer L: The glucagon-sensitive adenyl

cyclase system in plasma membranes of rat liver. III. Binding of glucagon method of as-

say and specificity. J Biol Chem 246:1861, 1971

23. Lowry OH, Rosenbrough NJ, Farr AL,


Randall RJ: Protein measurement with the folin

phenol reagent. J Biol Chem 193:265, 1951

24. Wolff J, Cook GH: Activation of thyroid

membrane adenylate cyclase by purine nucleo-

tides. J Biol Chem 248:350, 1973

25. Mashiter K, Mashiter G, Field JB: Ef-

fects of prostaglandin Et, ethanol and TSH on

the adenylate cyclase activity of beef thyroid

plasma membranes and cyclic AMP content of

dog thyroid slices. Endocrinology 94:370, 1974

26. Kaneko T, Zor U, Field JB: Thyroid

stimulating hormone and prostaglandin El

stimulation of cyclic 3’,5’-adenosine monophos-

phate in thyroid slices. Science 163: 1062. 1969

27. Scatchard G: The attractions of proteins

for small molecules and ions. Ann N Y Acad

Sci 51:660, 1949

28. Moore WV, Wolff J: Binding of prosta-

glandin Et to beef thyroid membranes. J Biol

Chem 248:5705, 1973

29. Dumont JE: The action of thyrotropin

on thyroid metabolism. Vitam Horm 29:287.

1971

30. Ode11 WD, Wilber JF, Utiger RD:

Studies of thyrotropin physiology by means of

radioimmunoassay. Recent Prog Horm Res 23:

41. 1967

3 I. Pohl SL, Krans MJ, Birnbaumer L,

Rodbell M: Inactivation of glucagon by plasma

membranes of rat liver. J Biol Chem 247:2295,

1972

32. Freychet P, Kahn R, Roth J, Neville

DM Jr: Insulin interactions with liver plasma

membranes. Independence of binding of the

hormone and its degradation. J Biol Chem 247:

3953, 1972

33. Sato S, Yamada T, Furihata R.

Makiuchi M: Effect of guanyl nucleotides on

the stimulation of adenyl cyclase activity in

human thyroid plasma membranes by thyroid-

stimulating hormone and prostaglandin Ez,

Biochim Biophys Acta 332:166, 1974

34. Burke G: Comparative effects of purine

nucleotides on thyrotropin and prostaglandin

Et responsive adenylate cyclase in thyroid

plasma membrane. Prostaglandins 3:537, 1973

35. Rodbell M, Birnbaumer L, Pohl SL,

Krans MJ: The glucagon-sensitive adenyl

cyclase system in plasma membranes of rat

liver. V. An obligatory role of guanyl nucleo-

tides in glucagon action. J Biol Chem 246: 1877,

1971

36. Krishna G, Harwood JP. Barber AJ,

Jamieson GA: Requirement for guanosine

triphosphate in the prostaglandin activation

of adenylate cyclase of platelet membranes.

J Biol Chem 247:2253, 1972

37. Bilezikian JP, Aurbach CD: The effects

of nucleotides on the expression of fi-adrenergic

adenylate cyclase activity in membranes from

turkey erythrocytes. J Biol Chem 249: 157. 1974

38. Rodbell M, Krans HMJ, Pohl SL, Birn-

baumer L: The glucagon-sensitive adenyl

cyclase system in plasma membranes of rat

liver. IV. Effects of guanyl nucleotides on bind-

ing of ‘ZSI-glucagon. J Biol Chem 246~1872,

1971

39. Soloff MS, Swartz TL: Characterization

of a proposed oxytocin receptor in rat mam-

mary gland. J Biol Chem 248:6471, 1973

40. Amir SD, Goldfine ID, lngbar SH: Pro-

gram of the Fifty-sixth Annual Meeting of the

Endocrine Society. 1974, p A94

41. Moore W, Wolff J: Program of the Fifty-

sixth Annual Meeting of the Endocrine Society,

1974, p A93

42. DeRubertis FR, Field JB: Program of

the Fifty-sixth Annual Meeting of the Endo-

crine Society, 1974, p A93

studies of thyroid-stimulating hormone binding to bovine thyroid plasma membranes

Documents