Purification of the type II insulin-like growth factor receptor from rat placenta.

The membrane receptor for insulin-like growth factor II (IGF II) has been purified to near homogeneity from rat placenta by chromatography of crude plasma membranes solubilized in Triton X-100 on agarose-immobilized IGF II. Elution of the IGF II receptor from the matrix at pH 5.0 in the presence of 1.5 M NaCl resulted in a receptor purification of 1100-fold from isolated plasma membranes, or 340-fold from the Triton extract with an average yield of about 50% in five separate purifications. Analysis of 125I-IGF II binding to the solubilized receptor in the Triton extract and in purified form by the method of Scatchard demonstrated no change in receptor affinity (Kd = 0.72 nM). Sodium dodecyl sulfate electrophoresis of the purified receptor showed one major band at Mr = 250,000 with only minor contamination. Affinity labeling of the receptor in isolated placenta membranes and in purified form using 125I-IGF II and the cross-linking agent disuccinimidyl suberate resulted in covalent labeling of only the Mr = 250,000 band. Such labeling was abolished by unlabeled IGF II but was unaffected by insulin, consistent with the previously reported specificity of IGF II receptor (Massague, J., and Czech, M.P. (1982) J. Biol. Chem. 257, 5038-5045). These results establish a one step affinity method for the purification of the type II IGF receptor that is rapid and highly efficient.

The membrane receptor for insulin-like growth factor I1 (IGF 11) has been purified to near homogeneity from rat placenta by chromatography of crude plasma membranes solubilized in Triton X-100 on agaroseimmobilized IGF 11. Elution of the IGF I1 receptor from the matrix at pH 5.0 in the presence of 1.5 M NaCl resulted in a receptor purification of 1100-fold from isolated plasma membranes, or 340-fold from the Triton extract with an average yield of about 50% in five separate purifications. Analysis of "'I-IGF I1 binding to the solubilized receptor in the Triton extract and in purified form by the method of Scatchard demonstrated no change in receptor affinity ( IGF 11' is a small polypeptide which closely resembles insulin in sequence, structure, and biological effects (1)(2)(3). The potency of IGF I1 in mediating short term metabolic effects in adipose tissue is 60-fold less than that of insulin, while its capacity to stimulate cell multiplication, DNA synthesis, and incorporation of sulfate into cartilage is 50-100 times greater (3). A receptor for IGF 11, distinct from the insulin and insulin-like growth factor I receptors, has been demonstrated in a variety of tissues (4) and has been shown by affinity labeling to be a single polypeptide of M , = 250,000 ~~ * Supported by Grant AM 30648 from the National Institutes of Health and by a grant from the Kroc Foundation. The costs of publication of this article were defrayed in part by the payment of Page charges. This article must therefore be hereby marked " d u e rtisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. (5-7). This receptor, also denoted as the type I1 IGF receptor, has high affinity for IGF 11, moderate affinity for IGF I, and no affinity for insulin (7). Recent studies on H-35 rat hepatoma cells suggest, but do not prove, that the IGF I1 receptor may mediate the growth stimulatory effects of IGF I1 on thest cells (8).
In rat adipocytes and H-35 rat hepatoma cells, binding of IGF I1 to the IGF I1 receptor is rapidly increased by physiological concentrations of insulin, apparently due to an insulininduced increase in IGF I1 receptor affinity (3,(8)(9)(10)(11). In the course of studies investigating the molecular mechanism of this receptor-receptor interaction, it was necessary to develop a rapid method to obtain microgram quantities of purified receptor. We report here the rapid, one step purification of the IGF I1 receptor from rat placenta to near homogeneity using affinity chromatography on IGF 11-agarose.

MATERIALS AND METHODS
Preparation of IGF I1 and IGF II-Agarose-IGF I1 was purified from the conditioned media of BRL-3A rat liver cells as previously described (11,12). Purified hormone was iodinated by the chloramine T method (13) to a specific activity of 100 Ci/g. IGF I was a generous gift of Professor R. E. Humbel (Biochemisches Institut der Universitat, Zurich, Switzerland). IGF I1 was covalently coupled to CNBr-activated Sepharose 4B (Pharmacia) according to directions supplied by the manufacturer.
One mg of hormone containing tracer lz5I-IGF I1 was incubated with 1 ml of activated gel for 16 h a t 4 "c in 0.1 M NaHC03, pH 8.0,0.5 M NaC1. Unreacted sites were quenched by incubation with 0.2 M glycine, pH 8.3, for 2 h at 24 "C. Coupled gel was extensively washed on a sintered glass filter with alternating cycles of 0.1 M sodium acetate, pH 4.0, 0.5 M NaCl and 0.1 M sodium bicarbonate, pH 8.0, 0.5 M NaCI, and was stored in 50 mM Hepes, 0.05% sodium azide. After use, the gel was recycled using the same washing procedure. It could be successfully reused without change in performance a t least five times.
Hormone Binding-IGF I1 binding to membranes was carried out as previously described (11). Triton extracts were incubated in 200 p1 of Krebs-Ringer phosphate buffer, pH 7.4, containing 10 mg/ml of bovine serum albumin and 0.1% (w/v) Triton X-100, T -I G F I1 (1 nM), and unlabeled IGF I1 for 30 min at 24 "C. Binding was terminated by the addition of 0.55 ml of ice-cold 0.1 M sodium phosphate, pH 7.4, containing 1 mg/ml of bovine y globulin (Sigma), and 0.5 ml of ice-cold 25% (w/w) polyethylene glycol. The mixture was vortexed, allowed to stand on ice for 15 min, and filtered under vacuum on an Amicon Microporous Filter, 0.45 micron size, which had been presoaked in Krebs-Ringer phosphate buffer containing 10 mg/ml of bovine serum albumin. The assay tube was washed with 2.5 ml of icecold 0.1 M Tris-C1, pH 7.4, containing 8% (w/w) polyethylene glycol, which was also applied to the filter. Filters were counted in a y counter to quantitate binding. Nonspecific binding, determined at a 250-fold excess of unlabeled IGF 11, was subtracted and represented 10-20% of maximum binding. Binding was analyzed by the method of Scatchard (14).
Preparation of Membranes-Rat placentas were obtained from Sprague-Dawley rats (Taconic Farms) in the seventeenth day of pregnancy. The tissue (50 g) was immediately homogenized (2 X 30 s) in 250 ml of ice-cold 10 mM sodium phosphate, pH 7.4, 1 mM EDTA, 0.25 M sucrose (PES) containing 1 mM PMSF using a Brinkmann Polytron. All subsequent steps were carried out at 4 "C. The homogenate was filtered through 2 layers of cheesecloth and centrifuged at 600 X g for 10 min. The pellets were discarded and the supernatant centrifuged at 3,000 X g for 10 min. The supernatant was then centrifuged at 30,000 X g for 40 min. The pellets were resuspended in 10 mM Tris CI, pH 7.4, 1 mM EDTA, 1 mM PMSF, and centrifuged at 30,000 X g for 30 min. The pellets were resuspended in a small volume of the same buffer and stored at -26 "C. They were 8539 by guest on March 24, 2020 http://www.jbc.org/ Downloaded from stable for at least a month with little or no change in Iz5I-IGF 11 binding activity.
Purification of the ZGF ZZ Receptor-Membranes (20 mg) were resuspended in 4 ml of 50 mM Hepes, pH 7.4, 1% (w/v) Triton X-100, containing 1 mM PMSF, 10 pg/ml of leupeptin, and 20 pg/ml of aprotinin and incubated with end-over-end mixing for 1 h at 4 "C. The mixture was centrifuged at 100,000 X g for 1 h. The supernatant was made 0.5 M in NaCl, then incubated with 0.8 ml of IGF 11-agarose which had been washed with 50 mM Hepes, 1% (w/v) Triton X-100. The suspension was incubated on a Fisher Hematology mixer for 1 h at 4 "C, then poured into a column (0.7 X 2.0 cm) and the flowthrough collected. The column was washed with 10 ml of 10 mM Hepes, pH 7.4, 0.5 M NaCl, 0.5% (w/v) Triton X-100 containing 1 mM PMSF, 10 pg/ml of leupeptin and 20 pg/ml of aprotinin. The receptor was eluted with 4 ml of 10 mM sodium acetate, pH 5.0, 1.5 M NaCl, 0.2% (w/v) Triton X-100 containing 1 mM PMSF, 10 pg/ml of leupeptin, and 20 pg/ml of aprotinin. Fractions (0.5 ml) were collected directly in 0.5 ml of 0.1 M sodium phosphate, pH 7.4, for immediate neutralization. After assay, fractions containing active receptor were pooled, made 20% (v/v) in glycerol, and stored at -26 "C.
Protein Assay-Membrane protein concentration was determined by the method of Lowry (15), and in Triton extracts by the method of Markwell et al. (16). Protein concentration of the purified receptor was determined by amino acid analysis using precolumn derivatization with o-phthaldehyde in the presence of 0-mercaptoethanol followed by reverse phase high pressure liquid chromatography by a modification of the method of Jones et al. (17).
Gel Ekctrophoresis and Autoradiography-Samples were boiled in the presence of Laemmli sample buffer (18) containing 1% sodium dodecyl sulfate and 50 mM dithiothreitol, then subjected to sodium dodecyl sulfate electrophoresis on 5% polyacrylamide gels or 5 to 12% gradient polyacrylamide gels according to the method of Laemmli (18). Gels were silver stained according to a method supplied by Bio-Rad or stained with Coomassie blue, dried, and subjected to autoradiography using Kodak X-OMAT film with enhancing screen (19).
Affinity Labeling-Affinity labeling of the receptor in membranes using DSS was carried out as previously described (11). Membranes

RESULTS
Rat placenta was chosen as a source of IGF I1 receptor for our studies because affinity labeling studies carried out as previously described (7) demonstrated that this tissue is rich in IGF I1 receptors but lacks IGF I receptors (data not shown).
Because IGF I receptors in some tissues have moderately high affinity for IGF I1 as well as IGF I (7), such receptors might be expected to copurify with the IGF I1 receptor on IGF IIagarose. When crude plasma membranes prepared from 17day pregnant rats were solubilized with 1% (w/v) Triton X-100 or 1% (w/v) octylglucoside, an increase in receptor binding activity of approximately %fold is observed. This may be due to the presence of receptor in inside-out vesicles which become available for binding in the presence of detergent. Increasing detergent concentration to 2% did not improve solubilization of activity. Triton X-100 was used for solubili- zation in preference to octylglucoside as the latter gave unacceptably high nonspecific binding levels in the polyethylene glycol assay. Table I summarizes the purification of the IGF I1 receptor from the Triton X-100 extract prepared from 20 mg of placenta plasma membrane. Essentially complete purification is attained with one passage of the extract over immobilized IGF 11-agarose. In five separate purifications, an average of 68% of the binding activity in the extract was retained on the column. This figure was not increased by reducing the amount of extract applied to the column or increasing adsorption time. The presence of 0.5 M NaCl during adsorption to the affinity matrix did not diminish the amount of receptor bound to the affinity matrix. These high salt conditions, as well as washing the column with 12 volumes of buffer containing 0.5 M NaC1, minimized nonspecific adsorption of a number of low molecular weight species which otherwise eluted with the purified receptor.
Elution of the receptor from the immobilized IGF I1 was achieved with 10 mM sodium acetate, pH 5.0, containing 1.5 M NaCl and 0.2% Triton X-100. Attempts to elute IGF I1 binding activity at pH 7.0 in the presence of 2 M NaCl, at pH 5.0 in the absence of 1.5 M NaCl, or at pH 9.0 or above in the presence or absence of 2 M NaCl were unsuccessful. Similarly, elution at pH 4.0 in the presence of 2 M NaCl resulted in a much lower yield of binding activity, presumably due to denaturation of the receptor at the lower pH. In order to minimize such acid-induced denaturation, eluting fractions were neutralized immediately by direct collection in phosphate buffer at neutral pH. Elution was normally complete in two bed volumes, so that further concentration of the pooled receptor was not necessary. Purified receptor was stored at -26 "C in the presence of 20% glycerol. Under these conditions, receptor binding activity was stable for months. Fig. 1 shows analysis by the method of Scatchard (14) of lZ5I-IGF I1 binding to the IGF I1 receptor in Triton X-100 extracts of plasma membrane and in the purified receptor preparation. No significant difference in affinity is seen with an average Kd = 0.72 nM. No evidence of curvilinearity of the Scatchard plots was observed. Assuming 1 mol of '"I-IGF 11 Binding of '2sI-IGF I1 was performed as described under "Materials and Methods" using 1 nM tracer and analyzed according to the method of Scatchard (14). The imets show binding saturation curves. A, Triton X-100 extract of crude rat placenta plasma membranes (15 pg/assay); R, purified IGF I1 receptor (40 ng/assay). All points are corrected for nonspecific binding, determined at a 250-fold excess of unlabeled '2sI-IGF 11.
bound per mol of receptor, one can calculate from Scatchard analysis that the affinity-purified fraction in the experiment presented in Table I should contain 7 pg of active receptor. Amino acid analyses indicated that 13 pg of total protein was present. Thus, in this preparation, the IGF I1 receptor attained at least 55% of theoretical purity by Scatchard analysis, indicating that it is close to homogeneity. Electrophoretic analysis of the purified receptor (Fig. 2) indicated that the actual purity is much higher (see below). Overall purification of 1100-fold from membranes or 340-fold from the Triton X-100 extract was achieved with a final yield (relative to the Triton extract) of 37%. In five separate experiments, the yield averaged 48% with a range of 31-63%. The receptor could be recycled over IGF 11-agarose for an additional one or two elutions. However, this gave no further purification with a substantial decrease in overall yield. Fig. 2 shows the sodium dodecyl sulfate-gel electrophoresis pattern obtained for crude plasma membranes, Triton extracts, and purified receptor when electrophoresed in the presence of dithiothreitol. The purified receptor preparation contained a major species a t M, = 250,000 as well as a minor contaminant at M, = 67,000 which is probably bovine serum albumin (lane C). When the purified receptor was examined on a 5-12% gradient gel (lane D), no other contaminant bands of molecular weight 25,000 or higher were observed. Fig. 3 shows autoradiographs of affinity labeled IGF I1 receptor in rat placenta plasma membranes and in the purified preparation, using "'I-IGF I1 (5 nM) and disuccinimidyl suberate (0.2 mM). In both cases, a single band at M, = 250,000 was observed, which co-migrated with the major silver-stained species in the purified preparations. No lower molecular weight bands were seen in the purified preparation even on overexposure of the autoradiograph (Fig. 3, lanes G-I). Electrophoresis of the affinity labeled receptor on a 5-12% gradient polyacrylamide gel to examine lower molecular weight proteins showed that no other band, including the M, = 67,000 Rat placenta plasma membranes and purified IGF I1 receptor were affinity labeled with '2sI-IGF I1 (5 nM) and disuccinimidyl suberate (0.2 mM) as described under "Materials and Methods" and electrophoresed on 5% polyacrylamide gels in the presence of dithiothreitol by the method of Laemmli (18). Autoradiographs of the dried gel are shown. Lanes A-C, plasma membranes (200 pg) affinity labeled with lZsII-IGF I1 in the presence of no unlabeled hormone ( A ) , unlabeled IGF I1 (600 nM) ( R ) , unlabeled insulin ( 3 p~) (C). D-F, purified IGF I1 receptor affinity labeled with '2sII-IGF I1 in the presence of no unlabeled hormone ( D ) , unlabeled IGF I1 (600 nM) ( E ) , and unlabeled insulin (3 p~) ( F ) . Lanes (3-1, same as 11-F but autoradiograph exposed for four days rather than one. Arrows show the location of molecular weight standards: bovine serum albumin (66,2001, phosphorylase b (92,500), p-galactosidase (116.250). and myosin (200,000).
band, was specifically labeled by "'I-IGF I1 (data not shown). This demonstrated the absence of proteolytic fragments of the IGF I1 receptor and lack of contamination with the IGF I receptor, whose major affinity labeled subunit migrates a t M, = 130,000 under reducing conditions. Labeling of the M, = 250,000 band was completely abolished by the presence of 600 nM unlabeled IGF I1 but was unaffected by the presence of 3 p~ unlabeled insulin. This is consistent with the results of binding assays in which binding of 12"I-IGF I1 (1 nM) to the purified receptor was abolished by the addition of 500 nM unlabeled IGF I1 but was unaffected by the presence of 5 p~ insulin or 5 p~ proinsulin. The addition of 10 nM unlabeled IGF I inhibited binding of I2'I-IGF I1 (

DISCUSSION
We report here the purification of the IGF I1 receptor to near homogeneity by a single step procedure of affinity chromatography on IGF 11-agarose. Such immobilized ligand chromatography is the only feasible means of purifying the receptor in the native state since it represents only 0.1% of the protein in crude plasma membranes. The technique has been widely applied to other growth factor receptors including insulin (20), epidermal growth factor (21), and nerve growth factor (22). Preliminary steps such as ion exchange or lectin chromatography prior to the affinity column proved unnecessary in our present studies, increasing the speed and yield of the purification.
The conditions reported here for eluting the IGF I1 receptor from the immobilized IGF I1 (pH 5.0 and high salt) resulted in elution of the receptor in small volume and high yield ( Table I). The conditions also avoided the use of denaturing agents such as urea, which has been used for elution of the insulin receptor from insulin-agarose (20). Dissociation of the ligand-receptor complex below pH 5.5 has been reported for several ligands including epidermal growth factor (23), insulin (24), lysosomal enzymes (25), and asialoglycoproteins (26). This may represent the physiological mechanism for removal of internalized ligand, as suggested by recent findings that endocytotic vesicles containing receptors for a2-macroglobulin (27) and transferrin (28) are rapidly acidified to pH 5.0 before fusion with lysosomes.
The affinity of the purified IGF I1 receptor from rat placenta (Kd = 0.72 nM, Fig. 1) is somewhat higher than that reported for the high affinity form of this receptor previously observed in plasma membranes and low density microsomes of control and insulin-treated adipocytes (Kd = 3.8 nM) (11). If the low affinity form of the IGF I1 receptor observed in intact adipocytes in the absence of insulin (11) is present in rat placenta, it does not survive homogenization and preparation of membranes. Affinity labeling of the purified receptor shows that the M , = 250,000 species, the only band observed upon silver staining except for a minor contaminant of bovine serum albumin (Fig. 2), is the only band covalently labeled by '"I-IGF I1 and the cross-linking agent DSS (Fig. 3). The fact that unlabeled insulin does not compete for binding or affinity labeling of the receptor, even a t a concentration of 3 p M , is consistent with previous finding that the IGF I1 receptor has no affinity for insulin (7). IGF I, however, shows significant affinity for this receptor in competition binding assays, again consistent with previous results (7). The purification reported here is sufficiently rapid and quantitative to permit studies on possible physiological COValent modifications of the IGF I1 receptor and in particular to investigate the molecular mechanism of its rapid modulation by insulin. The method should be applicable to purification of the receptor from many other tissues. In addition, the method can easily be scaled up to obtain large amounts of purified material suitable for antibody production and protein chemistry studies.