Mast cell binding of neurotensin. I. Iodination of neurotensin and characterization of the interaction of neurotensin with mast cell receptor sites.

Neurotensin was iodinated at equimolar concentrations of peptide, iodide, and chloramine-T, producing a labeled peptide with a specific activity of 1000 to 2000 Ci/mmol. Rat mast cells specifically and reversibly bound 1.27 pmol of neurotensin/10(6) cells with a reversible affinity, KD, of 154 nM. Optimum specific binding occurred betwen pH 6.8 and 7.2 under hypotonic conditions and dropped sharply as buffer concentration increased beyond 10 mM. The divalent cations Ca2+ and Mg2+ prevented binding with 50% inhibition at 1.5 and 4 mM, respectively. Binding was strongly and equally inhibited by the sodium and potassium salts of chloride, bromide, and iodide, and to a lesser degree by LiCl. Maximum binding of 125I-neurotensin occurred within 10 min at 0 degrees, and within 1.5 to 2 min binding was reduced to half-maximum in the presence of excess unlabeled neurotensin or upon 20-fold dilution in buffer. Both CaCl2 and NaCl were able to dissociate 60% of the total bound neurotensin: half the label bound was removed in 4 to 6 min. EDTA inhibited the binding only at high concentrations and no requirement was found for sulfhydryl groups, ATP, or a glycoprotein in the binding of neurotensin.

Neurotensin was iodinated at equimolar concentrations of peptide, iodide, and chloramine-T, producing a labeled peptide with a specific activity of 1000 to 2000 Ci/mmol. Rat mast cells specifically and reversibly bound 1.27 pmol of neurotensin/lO" cells with a reversible affinity, K,, of 154 nu. Optimum specific binding occurred between pH 6.8 and 7.2 under hypotonic conditions and dropped sharply as buffer concentration increased beyond 10 mu. The divalent cations Ca'+ and Mg'+ prevented binding with 50% inhibition at 1.5 and 4 mn, respectively.
Binding was strongly and equally inhibited by the sodium and potassium salts of chloride, bromide, and iodide, and to a lesser degree by LiCl. Maximum binding of 'ZSI-neurotensin occurred within 10 min at o", and within 1.5 to 2 min binding was reduced to half-maximum in the presence of excess unlabeled neurotensin or upon 20-fold dilution in buffer. Both CaCl, and NaCl were able to dissociate 60% of the total bound neurotensin: half the label bound was removed in 4 to 6 min. EDTA inhibited the binding only at high concentrations and no requirement was found for sulfhydryl groups, ATP, or a glycoprotein in the binding of neurotensin.
Neurotensin, a tridecapeptide (pGlu-Leu-Tyr-Glu-Asn-Lys-Pro-Arg-Arg-Pro-Tyr-Ile-Leu-OH), has been isolated from the hypothalamus (1, 2) and immunoreactive neurotensin has been detected in other distinct areas of rat brain (3-5) as well as in the gastrointestinal tract (6-8). A direct central nervous system effect of neurotensin is seen in its ability to produce hypothermia in cold-exposed (4") rats following intracisternal administration (9)(10)(11) Na'""I, and CT (24). To 10 ~1 of 0.4 M sodium phosphate, pH 6.75, 3.9 ~1 of neurotensin and 1 mCi of Na""I were added and the reaction was initiated by the rapid mixing of 10 ~1 of CT (0.13 pg) in 0.4 M sodium phosphate buffer, pH 6.75. After 30 s at room temperature, 10 ~1 of sodium metabisultite (0.35 /*g; 4 molar excess) was added to terminate the reaction. Iodinated neurotensin was isolated from Sephadex LH-20 columns (10 x 0.7 cm with sintered polyethylene discs'), pre-equilibrated with 5 to 10 column volumes BAW (10:2:11 and eluted by the same at a flow rate of 9 to 10 ml/h. Isolation of Mast Cells -Mast cells were isolated from both the peritoneal and thoracic cavities of 20 to 40 guillotined and exsanguinated rats according to method of Johnson and Moran (19). The mast cells were washed thrice, in isotonic saline and once in 0.32 M sucrose. They were stored at 4" in 0.32 M sucrose containing 0.02% NaN,, at concentrations of 0.2 to 1.2 x 10" cells/ml (and used without further purification'). The proportion of mast cells ranged from 30 to 50% as determined by direct counting using methylene blue and eosin-hemotoxylin stains for mast cells and non-mast cells, respectively.
Radioreceptor Assay-The binding assays, carried out in triplicate, contained approximately 20 to 120 fmol of '"sI-neurotensin (neutralized to pH 6.5 to 7) of Peak I (Fig. l), 10 mM Tris/acetate buffer, pH 7.0, 1 mg of bovine serum albumin, 80 mM sucrose, and 1 to 6 x lo" mast cells in a total volume of 100 ~1. After 15 min at O", the peptideicell mixture was rapidly filtered using presoaked (2% albumin) glass fiber filters and washed once with 0.5 ml of cold 2% albumin within 3 to 4 s. The specific binding is the difference in counts per min between tub& containing varying amounts of unlabeled neurotensin and those containing 6 nmol of unlabeled neurotensin. The range of total binding was 2.25 to 13.64% (average, 6.68 f 2.45%, n = 211. The filter background, in the absence of mast cells, ranged from 0.45 to 3.79% (average, 2.09 ? 0.6%, n = 18). lz51-Neurotensin which had been stored for 24 h or longer had backgrounds of 5.14 -t 0.1%. Binding was a linear function of the cell concentration.

Zodination
of Neurotensin -Consistently active 1251-neurotensin molecules were prepared with freshly solubilized neurotensin because frozen solutions of neurotensin yielded an iodinated peptide which gave binding data that were irreproducible. When a lo-to 20-fold molar excess of CT was used for iodination, the resulting peptide exhibited very low or no specific binding. CT solutions could either be prepared fresh or stored frozen at loo-fold concentrated stock solutions in the dark without loss of oxidizing activity.
The iodinated peptide was isolated within 8 to 12 min after application to the column (Peak I, Fig. 1) and kept on ice thereafter. Each column fractionation of the iodination reaction consistently yielded three peaks, the proportions of which differed from one iodination to another. Iodinated neurotensin eluted ahead of noniodinated or "H-neurotensin by approximately 1 to 1.5 ml. Storage of some preparations of 12"1neurotensin at temperatures from 4" to -90" either under acidic (BAW) or neutral conditions resulted in loss of binding activity (whereas for other preparations, the activity was good for 3 weeks). Iodinated neurotensin was routinely prepared before every binding study. The specific activity of IzaIz Glass wool strongly absorbs 'z51-neurotensin. " The major cellular contaminant was erythrocytes which do not exhibit any specific binding for neurotensin; furthermore, a lymphocyte cell line (lM-9, kindly supplied by Jesse Roth) was similarly inactive.
The cells were stable with respect to neurotensin binding upon storage for several days at 4" in the presence of NaN,. neurotensin obtained was 1000 to 2000 Ci/mmol. The labeled peptide was analyzed for homogeneity by thin layer chromatography and paper electrophoresis. Thin layer chromatography of '251-neurotensin revealed the following: (a) a single major radioactive spot from Peak I was evident which migrated ahead of neurotensin in BAW and BPyA (Fig. 2). More than one spot was usually, but not always, found in 2BA; the significance of this is not known. (b) Peak II was qualitatively similar to the first in terms of the radiolabeled peptide. (c) The largest radioactive peak (III) migrated near the solvent front, ahead of Na'Y.
Paper electrophoresis confirmed that most of the label in Peaks I and II is bound to a molecule that remains at the origin while the negatively charged iodide migrates anodally (Fig. 3). The electrophoresis pattern for Peak III was superimposable on that of NalZ"I, and thus represents an unknown iodine-containing molecule that also appears to be present in the commercial Na""I solution. The amount of free nonpeptide-bound radioiodide in the lZ51-neurotensin peaks averaged 5.4%.
The specific binding of ""I-neurotensin was constant through the peak tubes of Peak I and decreased thereafter. The fraction with the highest radioactivity in Peak I was used in all the binding experiments. pH Optimum and Buffer Concentration -Using several buffers over a pH range from 4 to 8.8, a sharp optimum in the specific binding of 1251-neurotensin to mast cells was observed around neutrality, with good agreement at the overlapping pH values (Fig. 4A). A similar optimum pH was found for the binding of neurotensin to rat brain synaptosomes (25) and release of histamine by vasoactive peptides from the mast cells (20). Varying the concentration of Tris/acetate buffer, pH 7.0, from 10 to 150 mM, gave no true maximum: specific binding decreased precipitously with increasing buffer concentration beyond 10 mM with approximately 10 to 15% binding left at isotonicity (Fig. 4B). of the ionic milieu on the binding of neurotensin, the effects of the major extracellular, as well as intracellular cations were investigated separately. Both CaCl, and MgCl* inhibit the specific binding of neurotensin ( Fig. 5A): CaCl, decreased the binding by 50% at 1.5 mM, comparable to the effect of CaClz on the binding of neurotensin to synaptosomes (25). At 20 mM CaCl,, approximately 30% of the rzSI-neurotensin remained bound. MgCI, also prevented the interaction of lz51neurotensin with its receptor site, although it was slightly less effective than CaCl,.
The monovalent cations, Na+ and K+, strongly interfered with the binding of neurotensin (Fig. 6A) and exhibited the same degree of inhibition as chloride, bromide, or iodide salts. Although LiCl also served as an antagonist for binding of neurotensin to mast cells, the extent of its interference was less than that of NaCl or KC1 (Fig. 6A). Association and Dissociation of Neurotensin from Mast Cells-The time to reach maximum specific binding of *W neurotensin to mast cells at 0" was 10 min (Fig. 7A) and the binding remained constant for at least 60 min. Although the specific binding found at 22" (room temperature) was comparable to that at O", ice bath temperatures were used to ensure minimum degradation by proteolytic enzymes that might be released from damaged cells. Compared to 0", incubations at 37" reduced by one-third the amount of specifically bound lz51neurotensin.
Bacitracin (2.8 pg), which inhibits the degradation of TRF and LRF in vitro (26), had no effect upon the reduction of bound '251-neurotensin at 37". In fact, at both 22 and 37", the addition of bacitracin suppressed binding by as much as 40%.
After reaching the maximum level of binding (i.e. steady state conditions), 50% of the labeled neurotensin could be dissociated from its receptor binding site in approximately 1.5 min by the addition of excess neurotensin (6 nmol) and 100% dissociated by 10 min (Fig. 7B ). Upon dilution in a 20-fold excess of buffer (containing bovine albumin), which reduced the ratio of labeled neurotensin bound to mast cells, dissociation of bound 1z51-neurotensin was evident after 2 min; the maximum amount displaced in 10 min, however, was only 70% of the total bound. These dissociation effects are comparable to those seen with synaptosomes, but are smaller for mast cells by factors of 4 to 6 (25).
The addition of CaCl, to the '251-neurotensin.mast cell TIME (min) cells with 83.3 fmol of '""I-neurotensin in the standard assay), one set of triplicate tubes was diluted 20-fold in 10 rnM Tris/acetate, pH 7.0, containing 1 mg of albumin (0) and another set received 10 pg of neurotensin (NT) in 1 ~1 (6 nmol) (0). At specified times, from 30 s to 10 min, the samples were filtered and processed as given under "Experimental Procedures." complex at steady state conditions brought about its dissociation: 56% of the bound neurotensin was displaced at the end of 10 min, although 50% dissociated in approximately 5 min (Fig. 5B). NaCl also caused the dissociation of bound "jIneurotensin from mast cells (Fig. 6B), an effect similar to that observed with CaCly: 66% of the bound neurotensin was displaced in 3 to 4 min.

Influence
of Other Factors on Binding-Appreciable inhibition of binding was seen at a concentration of EDTA greater than 10 mM and none at 2 mM or less (Fig. 5A ).
Iodoacetamide, N-ethylmaleimide, and dithiothreitol at concentrations up to 5 mM had no effect on the binding of neurotensin to mast cells. From these results, it appears that there is no obvious interaction of the peptide with membrane sulfhydryl groups. The requirement for a glycoprotein in the binding of a variety of peptide hormones, including neuroten-sin4 to synaptosomes has not been detected (27), and none was found to be involved here in the interaction with mast cells: the binding data were iden..ical in the presence or absence of 20 pg of concanavalin A. ATP (3 mM) did not affect the binding of 'Z51-neurotensin to mast cells, whereas the binding to synaptosomes was reduced." Treatment with 0.1% Triton X-100 and centrifugation through a 5 to 25% (wl w) sucrose gradient did not yield a soluble binding component from mast cells as determined by our assay method.

Constant of Neurotensin
for Mast Cells -The affinity of 9-neurotensin for the mast cell receptor site was determined in the presence of increasing concentrations of unlabeled neurotensin under the optimal conditions observed for binding (Fig. 8).
In the calculation of the data, it was assumed that there were two types of binding sites; one representing specific binding and the other unsaturatable, nonspecific binding.