Properties of the interaction between bovine thyrotropin and bovine thyroid plasma membranes.

Studies have been conducted to characterize further the interaction between 125I-labeled bovine thyrotropin (TSH) and bovine thyroid plasma membranes. Sequential subcellular fractionation of thyroid homogenates yielded preparations of progressively greater specific binding activity, highest activity being found in fractions previously shown to contain predominately plasma membranes (Amir, S. M., Carraway, T.F., Kohn, L.D., and Winand, R.J. (1973) J. Biol. Chem. 248, 4092-4100). Although binding of 125I-TSH by plasma membranes was greatest at pH 6.0, studies were conducted at pH 7.45 as well as pH 6.0, and results obtained differed quantitatively, but not qualitatively. Binding was maximal at 0 degrees, 15 degrees, and 22 degrees and steady state values remained unchanged for at least 22 hours. At 37 degrees, binding was decreased by 40% at 1 hour; the loss was even greater (65%) at 50 degrees. A similar loss of binding was evident when membranes were preincubated without TSH at 37 degrees or higher and were then incubated with 125I-TSH at 0 degrees. Lineweaver-Burk analysis indicated that preincubation resulted in loss of receptor sites without change in affinity of residual receptors. Addition of Ca2+ (1 to 10 mM) to the preincubation medium prevented the effect of preincubation at 37 degrees by preserving the number of receptor sites without altering their affinity. Under similar conditions, Na+ and K+ were without protective effect. Membranes bound 45Ca2+ in a specific and saturable manner. Scatchard plots indicated a dissociatiion constant (Kd) of 9 X 10(-5) M and a capacity (n) of 54 nmol/mg of membrane protein. 45Ca2+ was also displaced from membranes by Mg2+ and Mn2+. Ca2+ had a biphasic effect on binding; low concentrations (1 to 10 muM) added to the incubation mixture stimulated binding, while higher concentrations (0.1 mM) caused inhibition. Mg2+ and Mn2+, at comparable concentrations, were also inhibitory, Na+ and K+ less so. In the case of Ca2+, both the stimulatory and inhibitory concentrations were lower than those required to achieve saturation of Ca2+-binding sites. Proteolytic enzymes (trypsin, alpha-chymotrypsin, and pronase) sharply reduced binding of 125I-TSH, owing to a decrease in receptor sites. Phospholipases A and C enhanced binding of TSH, while neuraminidase and beta-galactosidase were without measurable effect.


Properties of the Interaction between Bovine Thyrotropin
and Bovine Thyroid Plasma Membranes* (Received for publication, December 11, 1975) SYED M. AMIR, IRA D. GOLDFINE Corp. All other chemicals and reagents were obtained from commercial sources and were of the highest grade of purity available.

Methods
Binding of "SCalcium-Bovine thyroid membranes (78 pg of protein) were suspended in 20 mM Tris-Cl with 0.5% albumin, pH 7.45. "CaCl, (20 ng, 0.09 j&i) and varying concentrations of 'OCaCl, were added and the mixture (650 ul) was allowed to incubate at 5" for 20 hours. Nonspecific binding of '%a'+ was assessed in the presence of '%a*+ at a concentration of 390 pg/ml and was subtracted from total binding to obtain values for specific binding of ?a2+.
Bound and free Ca*+ were separated by centrifugation of the reaction mixture at 2,500 x g for 10 min. The supernatant was removed and the pellet resuspended in water, without prior washing. The radioactivity was then measured in the presence of Aquasol (New England Nuclear), using a Packard scintillation counter.
Purification of Z'SH-Partially purified bovine TSH (2.5 i.u./mg, and less than 0.01 x NIH-LH-Sl units/mg) was further purified by sequential column chromatography using CM52, DE52, and Sephadex G-100 resins, as previously described (8). As evaluated in the McKenzie assay system (9), freshly purified preparations of bovine TSH contained 10 to 15 i.u./mg. Purified preparations were used either fresh or were lyophilized and stored until needed at -60" for periods of up to 2 months.

Protein
Concentration-Measurements of protein concentration were performed by the method of Lowry et al. (ll), using crystalline bovine serum albumin as the standard. Membrane protein was determined following solubilization of appropriate aliquots in 1 N NaOH for 2 to 3 min in a boiling water bath.

Iodination
of TSH-Purified TSH, prepared either in our laboratory or supplied by Dr. Pierce, was radioiodinated by means of a technique, which employs stoichiometric amounts of chloramine T and no reductant (7), to a specific activity of 75 to 150 pCi/pg (1 to 2 atoms of iodine/molecule of TSH). Aliquots were either used directly or were stored frozen at -60". No significant loss in binding activity was noted for periods of up to 4 weeks, and binding of the two types of TSH preparation used for radioiodination did not differ appreciably.
For each experiment, the results shown are the mean of those obtained from triplicate vessels. In addition, all experiments were carried out at least twice, and only those findings that were consistent in these experiments are presented.

Preparation
of Thyroid Plasma Membranes-The method described by Wolff and Jones (10) was employed to obtain a membrane-rich pellet from homogenates of thyroid glands obtained from freshly killed cattle. At this stage, the method was altered to incorporate the system of discontinuous sucrose gradient centrifugation employed earlier (8). Briefly, the pellet was suspended in 1 rnM NaHCO, and was applied to a discontinuous sucrose gradient consisting of 45% (17 ml), 40% (17 ml), and 30% (17 ml) of sucrose (densities 1.20, 1.18, and 1.13, respectively) in 20 mM Tris-Cl, pH 7.45. Centrifugation was carried out in a Beckman model L5-65 and SW 25.2 rotor at 22,000 rpm for 2 hours. Material from the 30 to 40% and 40 to 45% interfaces was collected, diluted lo-fold with 20 mM Tris-Cl, pH 7.45, and sedimented by centrifugation.
The resulting membrane preparations were either used fresh, or were divided into small aliquots and were stored at -60" for future use. Immediately prior to use, membranes were thawed at 4" and diluted with appropriate buffer. Clumping was avoided by gently dispersing membranes with a Pasteur pipette. Membranes prepared in this manner were shown to retain responsiveness of the adenylate cyclase system to both TSH and F-.   (12).
conducted at pH 6.0.

Effect of Incubation
Temperature-Early experiments were conducted to ascertain the effect of incubation temperature on the binding of '2JI-TSH in the standard binding assay. When studied after 2.5 hours at pH 7.45, binding of TSH was greatest at 0, 15, and 22", no difference between these three temperatures being observed. In experiments carried out at 0" in the presence of increasing membrane concentrations, greater than 15% of labeled TSH was bound at pH 7.45, and as much as 70% at pH 6.0. At 37", however, binding was reduced by 40%, and at 50" by 65%. Similar effects of incubation temperature were observed at pH 6.0.
In view ef these findings, the time-course of binding of TSH at 0" and 37" was studied ( Fig. 1). At both temperatures, binding of TSH at pH 7.45 was maximum within a very short interval; i.e. within the lo-min period required for processing of samples for counting. The fraction of added TSH bound was similar at the two temperatures.
At O", the fraction of labeled TSH bound remained essentially unchanged during the subsequent 22 hours. At 37", in contrast, a progressive decrease in binding occurred up to approximately 2 hours, with no appreciable further decrease up to 5 hours. In other experiments using lower TSH concentrations, the highest binding at 0" was noted after 1.5 to 2 hours. A similar inverse relationship between the concentration of TSH employed and the time required for maximum binding has been noted in other reports (13). Accordingly, all subsequent incubations of membranes with TSH were conducted at 0" and were allowed to proceed for 2.5 hours.
In view of the fact that calcium is required for the action of many hormones (17) and is stimulatory to ligand-receptor interactions in certain systems (12), further studies of the effects of calcium were conducted. Preincubation of membranes with EDTA (4.3 mM) for 1 hour at 0" or 37" and pH 7.45, followed by careful washing of the membranes with incubation medium free of EDTA, did not affect the subsequent binding of 'Y-TSH when assessed under the standard conditions. However, in accord with previous reports (15), direct addition of very low concentrations of calcium (10 to 30 pM) to the standard incubation medium resulted in a significant (approximately 20%) increase in '"SI-TSH uptake (Fig. 3). Within the same experiments, progressively higher concentrations of calcium were associated with progressive inhibition of binding to values below those found in the absence of calcium. were assessed by enriching the basic incubation medium at pH 7.45 with concentrations of the chloride salts of Na+, K+, Caz+, Mg2+, and Mn2+, ranging between 0.1 and 100 mM. As reported previously (6,8,(14)(15)(16), all proved to be inhibitory (Fig. 2). The inhibitory potencies of Caz+, Mg'+, and MnZ+ were very similar to one another, as were the potencies of Na+ and K+. A strikingly greater inhibitory effect was ob-  "The enzyme-substrate ratio was 1:15 for trypsin, chymotrypsin, and pronase. Incubation was at pH 7.5 for 15 min at 37". In control experiments, trypsin inhibitor at twice the concentration of trypsin was added prior to the addition of the enzyme. *The enzyme-substrate ratio was 1:200 for phospholipase A and 1:6 for phospholipase C and D. The incubation with phospholipase A and C was at pH 7.5 at 30" for 10 min. Phospholipase D was incubated at pH 5.6 and 37" for 15 min. The medium in all cases contained 1.8 rnM CaCl, and the reaction was stopped by addition of EDTA (2.8 mM). 'The medium contained an enzyme-substrate ratio of 1:18 for neuraminidase and fl-galactosidase. Incubation was at pH 5.0 for neuraminidase and 7.5 for /%galactosidase and was continued for 15 min at 37". dIncubation was first performed with neuraminidase, the enzyme removed and the membrane reincubated with ,!I-galactosidase as noted in Footnote c. 'Incubation of membranes with DNase and RNase was carried out at an enzyme-substrate ratio of 1:9 at pH 7.5 and 37" for 15 min. In case of DNase, the medium also contained 3 rnM MgCl,.
phospholipase D had no effect on binding at pH 7.45. Regardless of whether binding studies were conducted at pH 7.45 or pH 6.0, preincubation of membranes with P-galactosidase or neuraminidase at their respective pH optima (7.5 and 5.0) had no effect on binding of '*"I-TSH. A similar lack of effect was evident when membranes were preincubated with DNase.
As has been noted previously (22)