Somatomedin-C Stimulates the Phosphorylation of the ,&Subunit of Its Own Receptor*

Phosphorylation of the somatomedin-C receptor was investigated both in intact IM-9 cells and in IM-9 cells that had been solubilized with Triton X-100. Intact IM-9 cells were incubated with [32P]H3P04 for 1 h and for an additional 5 min in the absence or presence of insulin or somatomedin-C. The cells were then solubilized and subjected to wheat germ agglutinin Sepha- rose chromatography. The extent of phosphorylation of insulin and somatomedin-C receptors was assessed by immunoprecipitating the wheat germ agglutinin Sepharose eluates with monoclonal antibodies specific for each receptor and analyzing the immunoprecipitates by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The @-subunits of both receptors were phosphorylated in the absence of hormone, and the extent of phosphorylation of each receptor was enhanced by both hormones. However, each hormone was more potent than the other in enhancing phosphorylation of its own receptor. The &subunit of the somatomedin-C receptor was also phosphorylated when solubilized IM-9 cells that had been purified on wheat germ agglutinin Sepharose were incubated

Phosphorylation of the somatomedin-C receptor was investigated both in intact IM-9 cells and in IM-9 cells that had been solubilized with Triton X-100. Intact IM-9 cells were incubated with [32P]H3P04 for 1 h and for an additional 5 min in the absence or presence of insulin or somatomedin-C. The cells were then solubilized and subjected to wheat germ agglutinin Sepharose chromatography. The extent of phosphorylation of insulin and somatomedin-C receptors was assessed by immunoprecipitating the wheat germ agglutinin Sepharose eluates with monoclonal antibodies specific for each receptor and analyzing the immunoprecipitates by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The @-subunits of both receptors were phosphorylated in the absence of hormone, and the extent of phosphorylation of each receptor was enhanced by both hormones. However, each hormone was more potent than the other in enhancing phosphorylation of its own receptor.
The &subunit of the somatomedin-C receptor was also phosphorylated when solubilized IM-9 cells that had been purified on wheat germ agglutinin Sepharose were incubated with [-p3'P]ATP. In this soluble preparation, phosphorylation occurred on tyrosyl residues and was enhanced by concentrations of somatomedin-C in the range of 2.5 to 250 ng/ml, which is consistent with its receptor affinity. Tyrosyl phosphorylation of the somatomedin-C receptor also occurred when highly purified receptor, prepared by wheat germ agglutinin Sepharose affinity chromatography followed by immunoprecipitation, was incubated with [-p3'P]ATP. This indicates that the responsible tyrosyl kinase activity is intrinsic to the receptor or tightly associated with it.
Several viral oncogene products and their cellular homologues are tyrosine specific protein kinases which also serve as substrates for tyrosine phosphorylation (for example, see Refs. 1-5). In cells transformed by the Rous sarcoma virus, transformation results from the action of t h e oncogene product and is correlated with its protein kinase activity (6, 7). It is, therefore, interesting that receptors for several growth * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. factors including epidermal growth factor (8-131, insulin (14-22), and platelet-derived growth factor (23,241 contain tyrosyl residues which are phosphorylated, and that phosphorylation of these residues is enhanced by binding of the respective growth factor. In the case of epidermal growth factor and insulin, the evidence indicates that the receptor is itself a tyrosine-specific protein kinase, which is capable not only of autophosphorylation but also of phosphorylating exogenous proteins (10,11,(18)(19)(20)(21). In the present studies, we investigate phosphorylation of the somatomedin-C receptor.

MATERIALS AND METHODS
Somatomedin-C was purified from Cohn Fraction IV ( 2 5 ) . nIR-1 is a monoclonal antibody to the insulin receptor. aIR-2 and nIR-3 are monoclonal antibodies to the somatomedin-C receptor. The production and properties of these antibodies have been described (26)(27)(28). IM-9 (human lymphoblastoid) cells were grown in RPMI 1640 containing 10% bovine serum (Sterile Systems).
Immunopreczpitation of Solubilized Receptor-Immunoprecipitations were carried out with nIR-l (final concentration of 19 pg/ml), aIR-2 (final dilution of ascites fluid 1:420), or aIR-3 (final concentration 11 pg/ml). Normal mouse serum was included in all tubes at a final dilution of 1:400. Control tubes contained only normal mouse serum without monoclonal antibodies. Solubilized receptor was incubated with the indicated antibody in 50 mM Tris.HC1, pH 7.7, containing 0.2% Triton X-100 and the protease inhibitors bacitracin (1 mg/ml) and phenylmethylsulfonyl fluoride (20 pg/ml). In studies in which phosphorylated receptor was immunoprecipitated, the following were added to inhibit phosphatases: 5 mM EDTA, 10 mM sodium pyrophosphate, 10 rnM NaF, and 0.1 mM sodium vanadate. After 6 h, anti-mouse immunoglobulin (Cappel) was added at a final dilution of 1:18 and the incubation was continued for an additional 18 h. Four ml of Tris buffer containing 0.2% Triton X-100 was added, the tubes were centrifuged at 3,000 X g and the immunoprecipitate washed with an additional 4 ml of Tris buffer.
SDS'-Polyacylamide Gel Electrophoresis and Phosphoamino Acid Determination-The immunoprecipitates were washed with an additional 4 ml of H,O, lyophilized, and subjected to SDS-polyacrylamide gel electrophoresis with the Laemmli buffer system. After staining, the gels were either dried and subjected to autoradiography and densitometry, or the labeled bands were excised, electrophoretically eluted, and after hydrolysis in 6 N HCl, two-dimensional phosphoamino acid analysis was performed by electrophoresis (glacial acetic acid, 88% formic acid, H20, 78:25:897, pH 1.9) and ascending chromatography (isobutyric acid, 0.5 M NH,OH, 5:3) (13). T h e cells were washed rapidly and solubilized with a mixture designed to inhibit both phosphatases and kinases, and a glycoprotein fraction was isolated by wheat germ agglutinin affinity chromatography (see Fig. 1 legend). The labeled glycoproteins were then immunoprecipitated. A labeled band with a molecular weight of approximately 90,000, which corresponds to the P-subunit of the insulin receptor, was immunoprecipitated by (uIR-1, the antibody to the insulin receptor (Fig. lA, lanes 2, 6, and IO). After a 1-h incuhation at 37 "C, 5-ml aliquots were incubated for an additional 5 min at 3'7 "C with no additions (lanes 1-4). 0.5 pg/ml of insulin (lancs 5-0), or 0.5 pg/ml of somatomedin-C (lanes 9-12). The cells were then washed twice with 45 ml of phosphate-buffered saline containing 10 mM sodium pyrophosphate, 4 mM EDTA, 10 mM NaF, and 0.1 mM sodium vanadate, and solubilized in 50 mM Tris.HCI, pH '7.7, containing 1 5 Triton X-100, 1 mg/ml of bacitracin, 200 pg/ ml of phenylmethylsulfonyl fluoride, 10 mM sodium pyrophosphate, 4 mM EDTA, 10 mM sodium fluoride, and 0.1 mM sodium vanadate. The soluhilized cells were centrifuged at 100,000 X g for 1 h and the supernatants applied to a 0.5-ml wheat germ agglutinin-Sepharose column. The columns were washed and eluted with 1.5 ml of 0.5 M N-acetylglucosamine in 50 mM Tris.HC1, pH 7.7, containing 0.1% Triton X-100 and 1 mg/ml of bacitracin. The eluate was then immunoprecipitated with normal mouse serum (lanes 1 , 5 , and 9), uIR-1 (lanes 2.6, and I O ) , trlR -2 (lanes 3 , 7, and 1 1 ) , or trIR-3 (lanes 4, 8, and 12) as described under "Materials and Methods." The immunoprecipitates were then subjected to SDS-polyacrylamide gel electrophoresis on a 6.5% gel. Shown is an autoradiogram of the dried gel. Standard proteins are "C-acetylated albumin and phosphorylase. R, IM-9 cells were labeled with [:"P]HIPO,, as described above, and incuhated for 5 min at 37 "C with no additions (lanes I -3 ) . 0.5 pg/ml of insulin (lanes 4-6), or 20 pg/ml of insulin (lanes 7-9). The cells were then washed, solubilized, and purified on a wheat germ agglutinin-Sepharose column as described in A. The wheat germ agglutinin-Sepharose eluates were immunoprecipitated with normal mouse serum (lanes I , 4 , and 7 ) . crIR-1 (lanes 2, .5, and R ) , or crIR-X (lanes 3, 6, and 9), as described under "Materials and the extent of labeling is increased approximately 4-fold by 0.5 pg/ml of insulin and slightly less than 2-fold by 0.5 pg/ml of somatomedin-C. (Although other bands appear slightly darker in the lanes 2, 6, and 10 in which nIR-1 immunoprecipitates were run, we doubt that these bands correspond to components of the insulin receptor since they were also precipitated in significant amounts by normal mouse serum. The extent of phosphorylation of these bands was not enhanced by insulin or somatomedin-C.) d R -2 and nIR-3, antibodies to the somatomedin-C receptor, specifically immunoprecipitated a labeled hand (Fig. lA, lanes 3 , 4, 7, 8, 11, and 12) which was somewhat broader and had a somewhat slower mobility (apparent M, = 92,000-98,000) than the @-subunit of the insulin receptor and in some cases appeared as a doublet. This is characteristic of the @-subunit of the somatomedin-C receptor in IM-9 cells (27,28). Labeling of this protein was enhanced approximately 2-fold by 0.5 pg/ml of somatomedin-C, but only little, if at all, by 0.5 pg/ml of insulin. In similar studies (Fig. l R ) , higher concentrations of insulin (20 pglml), however, are quite effective a t stimulating phosphorylation of this protein. (With these high concentrations of insulin, there was also enhanced labeling of a very high molecular weight band which is seen at the very top of the separating gel (lanes 7-9). The significance of this band is not clear; however, it is clear that immunoprecipitation of this band is not specific since it is equally dark with normal mouse serum or either anti-receptor antibody.) Phosphorylation of Solubilized Receptor-IM-9 cells were solubilized with Triton X-100 and somatomedin-C receptors partially purified on a wheat germ agglutinin Sepharose column. The eluted receptor was incubated with somatomedin-C and then with [y"'P]ATP (Fig. 2). The &subunit of the somatomedin-C receptor was phosphorylated in the absence of somatomedin-C, hut phosphorylation was enhanced by as little as 2.5 ng/ml of somatomedin-C and maximal enhancement occurred with 250 ng/ml. Phosphoamino acid analysis indicated that phosphorylation occurred on tyrosyl residues both in the basal state and after somatomedin-C stimulation (Fig. 3).

Receptor Phosphorylation in Intact
Solubilized somatomedin-C receptor was partially purified on a wheat germ agglutinin Sepharose column and then immunoprecipitated with trIR-3. The 52.50 per microfiche or $1.00 per set of photocopy. Full sized photocopy is also included in the microfilm edition of the Journal that is available from Waverly Press. the antibody, itself, stimulated phosphorylation, or that somatomedin-C could not further increase phosphorylation under these circumstances.

DISCUSSION
The interpretation of the results presented in this study depend in part upon the specificity of tuIR-2 and tuIR-3. This has been evaluated extensively (27,28). With the concentrations of antibody used in this study, both nIR-2 and nIR-3 immunoprecipitate less than 2% of the amount of insulin receptor immunoprecipitated by nIR-1 (27). Furthermore, when IM-9 cells that were covalently affinity labeled with '"'I-insulin were solubilized, no detectable labeled receptor bands were immunoprecipitated by nIR-2 or nIR-3, while a very prominent labeled receptor band was immunoprecipitated by (VIR-1 (28). Therefore, in the present studies, the 92,000-98,000 molecular weight phosphotyrosine containing polypeptide immunoprecipitated by nIR-2 and tuIR-3 cannot be the 8-subunit of the insulin receptor.
In addition to its immunochemical specificity, there are other reasons for believing that this polypeptide is the @subunit of the somatomedin-C receptor. Its phosphorylation is enhanced by ng/ml concentrations of somatomedin-C and pg/ml concentrations of insulin, which is consistent with the known affinities of these hormones for the somatomedin-C receptor. Furthermore, it has a somewhat broader pattern and a slightly slower mobility than the @-subunit of the insulin receptor, properties characteristic of the @-subunit of the somatomedin-C receptor in IM-9 cells. In some autoradiograms, this broad band appears to be a doublet. We had previously suggested that the lower component of this doublet might be the @-subunit of the insulin receptor (27). This now seems unlikely, because the phosphorylation of both components of this band are enhanced to a similar extent by various concentrations of insulin and somatomedin-C, whereas if the lower component was the @-subunit of the insulin receptor, one would expect a differential effect on the extent of phosphorylation.
Because of the great similarity between receptors for insulin and somatomedin-C, it may be anticipated by analogy with the insulin receptor that the somatomedin-C receptor is a protein kinase and that phosphorylation of its @-subunit results from endogenous protein kinase activity. The retention of protein kinase activity by a highly purified receptor preparation (i.e. after wheat germ agglutinin-Sepharose chromatography and immunoprecipitation) is consistent with this; however, further studies will be required to conclusively demonstrate that the receptor, itself, is a protein kinase.
The present studies indicate that when solubilized somatomedin-C receptor is phosphorylated, the phosphorylation occurs on tyrosyl residues. This is true also for the solubilized insulin receptor; however, when the insulin receptor is phosphorylated in intact cells, a considerable amount of phosphoserine and phosphothreonine is present in addition to phosphotyrosine (15). Further studies will be required to determine if in intact cells this more complex pattern of phosphorylation also occurs with somatomedin-C receptors.
The results presented in Fig. 1 illustrate that high concentrations of insulin stimulate the phosphorylation of the somatomedin-C receptor, and high concentrations of somatomedin-C stimulate phosphorylation of the insulin receptor. As previously discussed, the most likely explanation for this is the known cross-reactivity of each receptor for the other's hormone; however, another possibility, which cannot be excluded, is that each receptor may be a substrate for the other. Activation of one receptor could then trigger the phosphoryl-