Hormone-dependent phosphorylation of ribosomal and plasma membrane proteins in mouse mammary gland in vitro.

Abstract Mouse mammary cells in organ culture incorporate 32Pi into the serine and threonine residues of 19 specific plasma membrane proteins and 8 specific ribosomal proteins as resolved by polyacrylamide electrophoresis. These proteins are also phosphorylated in cell-free reactions by tightly bound protein kinases which are endogenous to these organelles and not activated by cyclic adenosine 3',5'-monophosphate (cyclic AMP), and by a purified cytosol protein kinase which is activated by cyclic AMP. The radioactivity patterns of phosphorylated plasma membrane proteins are nearly identical after reactions with the endogenous or cytosol protein kinases or after phosphorylation in intact cultured cells. Certain differences in the radioactivity profiles of 32P-labeled ribosomal proteins are observed after phosphorylation by the ribosomal or cytosol protein kinases as compared to the 32P-proteins of ribosomes labeled in cultured cells. Insulin stimulates the rate of phosphorylation of specific plasma membrane and ribosomal proteins in undifferentiated mammary stem cells with maximal stimulation observed after 16 hours of culture. Prolactin synergizes with insulin to stimulate the rate of phosphorylation of all 19 membrane proteins and four ribosomal proteins in cells which differentiate in vitro after treatment with insulin and hydrocortisone. Stimulation of phosphorylation of these specific proteins occurs subsequent to the rapid induction by prolactin of the catalytic and regulatory subunits of the cyclic AMP-dependent protein kinase of the cytosol, as previously described. The results suggest that the phosphorylation of multiple specific proteins in these organelles represents the propagation of the initial prolactin stimulus at the cell surface to various functionally distinct compartments of the mammary epithelial cell.

These proteins are also phosphorylated in cell-free reactions by tightly bound protein kinases which are endogenous to these organelles and not activated by cyclic adenosine 3',5'-monophosphate (cyclic AMP), and by a purified cytosol protein kinase which is activated by cyclic AMP.
The radioactivity patterns of phosphorylated plasma membrane proteins are nearly identical after reactions with the endogenous or cytosol protein kinases or after phosphorylation in intact cultured cells.
Certain differences in the radioactivity profiles of 32P-labeled ribosomal proteins are observed after phosphorylation by the ribosomal or cytosol protein kinases as compared to the 32P-proteins of ribosomes labeled in cultured cells.
Insulin stimulates the rate of phosphorylation of specific plasma membrane and ribosomal proteins in undifferentiated mammary stem cells with maximal stimulation observed after 16 hours of culture.
Prolactin synergizes with insulin to stimulate the rate of phosphorylation of all 19 membrane proteins and four ribosomal proteins in cells which differentiate in vitro after treatment with insulin and hydrocortisone.
Stimulation of phosphorylation of these specific proteins occurs subsequent to the rapid induction by prolactin of the catalytic and regulatory subunits of the cyclic AMP-dependent protein kinase of the cytosol, as previously described.
The results suggest that the phosphorylation of multiple specific proteins in these organelles represents the propagation of the initial prolactin stimulus at the cell surface to various functionally distinct compartments of the mammary epithelial cell.
l'revious studies from our laboratory have characterized the enzymatic properties of t-70 forms of protein kinase (I and II) *  which are hormonally induced in mouse mammary gland (I, 2). Protein kinase II requires cyclic adenosine 3', 5'.monophosphate for maximal activity, while kinase I is not stimulated by cyclic nucleotides.
The hormone-dependent differentiation of mammary gIand in viva as well as in vitro is associated with marked increases in the specific activities of both forms of this cytosol protein kinase. It was shown that prolactin acts synergistically with insulin to induce the catalytic activity of protein kinase as well as a cyclic AMP'-binding protein in cultured mammary cells (2). To elucidate the functional roles of the hormonally induced protein kinases, it was essential to evaluate further the physiological substrates of both forms of this enzyme.
The present studies demonstrate that several specific protein components of highly purified plasma membranes and ribosomes may serve as substrates of the purified protein kinases, and the phosphorylation of these proteins in cultured mammary epithelial cells is under hormonal regulation.
The results implicate the induction and activation of protein kinase and the phosphorylation of specific proteins in various mammary cell orga.nelles in a sequence of biochemicai events which serves to propagate the prolactin stimulus to various functionally distinct compartments of the mammary cell.
Calf thymus histones, pancreatic RNase (five times crystallized), and crude collagenase were purchased from Worthington. cGMP, cUMP, and cCMP were products of Boehringer Mannheim, and [Y-~~P]-ATP (31 to 34 Ci per mmole) and inorganic [3'P]phosphate (carrier-free) were from International Chemical and Nuclear Corporation.
Cycloheximide was a product of Nutritional Biochemicals, density gradient grade sucrose (ribonuclease-free) was from Mann, and actinomycin D was obtained from AIerck. DEAE-cellulose and calcium phosphate gel, ATP, and bovine serum albumin were purchased from Sigma. Hydroxylapatite (Bio-Gel HT) was from Bio-Rad and ultrapure gua.nidine HCl was a product of Heico. Isolation of Mammary cAXP-binding Protein-Lactating mouse mammary gland was homogenized (1:8, w/v) in 5 m&r sodium glycerophosphate-HCl buffer (pH 6.5) containing 0.15 M KC1 and 0.2 mnr dithiothreitol.
The homogenate was centrifuged at 110,000 x g for 60 min, and the sediment was discarded (Step I). Solid ammonium sulfate was added to the above supernate (31.5 g per 100 ml) with stirring.
After 30 min the precipitate was collected by centrifugation at 16,000 x g for 10 min and the supernatant fluid was discarded. The residue was dissolved in 10 mn~ Tris-HCI buffer (pH 7.5) containing 10% glycerol and 6 m&r 2-mercsptoethanol (Buffer A) and dialyzed against 100 volumes of the same buffer with two changes of the buffer.
Calcium phosphate gel suspension in Buffer A was added to the above dialyzed protein preparation (30 mg per ml) with stirring.
After 15 min the gel was sedimented by centrifugation at 3,000 x g for 5 min and the supernate was discarded.
The sedimented gel was washed three times with 50 ml of Buffer A. Protein was finally eluted from the gel by extracting four times with 50 ml of 0.2 M potassium phosphate buffer (pH 8.1) containing 107; glycerol and 6 mM 2-mercaptoethanol.
To the combined extracts of the calcium phosphate gel, solid ammonium sulfate was added (31.5 g per 100 ml). After 30 min the suspension was centrifuged at 16,000 x g for 10 min and the supernate was discarded.
The pellet was dissolved in Buffer A and dialyzed extensively against the same buffer (Step 2). Solid guanidine HCl was added to the above solution with stirring to make the final concentration of guanidine 6 M. After 72 hours the solution was dialyzed against 30 volumes of the same buffer with six changes of the buffer. The dialyzed turbid solution was centrifuged at 100,000 x g for 20 min and the pellet was discarded (Step 3). The resulting supernate was applied to an hydroxylapatite column (1.5 x 3.5 cm) previously equilibrated with Buffer A. The column was washed with 5 ml of the equilibrating buffer prior to further elution with a linear gradient of potassium phosphate (0 to 400 mm) in a total volume of 250 ml of the buffer. The flow rate was 1.3 ml per hour and the volume in each fraction was 1.5 ml (Step 4). All steps were carried out at O-4".
Isolation of Mammary Cell Membranes-Minced tissue, explants from organ culture, or isolated epithelial cells were stirred in deionized water, the pH was adjusted to 7.0 with sodium bicarbonate (1:8, w/v) for 10 min, and the suspension was centrifuged at 1,000 X g for 10 min. The pellet was homogenized in 0.25 M sucrose containing 0.1 M KCI, 0.01 M Tris-HCl buffer (pH 7.4), and 0.003 M MgClz in a glass homogenizer wit'h eight passes of a loosely fitting glass pestle driven by motor at 60 rpm. The homogenate was filtered through four layers of cheese cloth and then centrifuged at 3,000 x g for 10 min, and the pellet was washed with the same buffer. Cell membranes were isolated from the pellet by discontinuous sucrose density gradient centifugation (2). The membrane preparation recovered at the interface between layers of 1.08 M and 1.80 M sucrose was diluted with 3 volumes of water and the membranes were finally resedimented by centrifugation at 27,000 x g for 30 min. For the assay of protein kinase activity in membranes, the sediment was suspended in 0.35 M sucrose containing 10 mM sodium glycerophosphate-HCI buffer (pH 6.5) and the suspension was dialyzed against the same buffer.
Isolation of Jfammary Ribosomes-Tissues or isolated epithelial cells were homogenized in 0.25 nt sucrose containing 0.1 M KCl, 0.01 M Tris-HCl (pH 7.4), and 0.003 M filgC12 in an all glass homogenizer, and the homogenate was centrifuged at 17,000 X g for 10 min. The resulting supcrnate was centrifuged at 110,000 x g for 60 min to sediment the microsomes, and the crude ribosomes were isolated from the microsomal pellet by extraction with sodium deoxycholate (3). For studies involving the assay of endogenous protein kinase activity, the crude ribosomes were further purified by treatments with 0.05 M MgClz and 0.01 ELI MgC14 and by repeated extractions with 0.5 M NH&I (3); they were finally dialyzed against 0.35 1\r sucrose containing IO mM sodium glycerophosphate-HCl buffer (pH 6.5).
Organ Culture ~Vethods-The abdominal, thoracic, and inguinal mammary glands were removed with aseptic technique from C3H/HeJ mice at the midpoint of their first pregnancy, and explants were prepared and cultured in Medium 199 (hIicrobiological Associates) as previously described (4, 5). Crysst.alline bovine insulin (Lilly), hydrocortisone, and ovine prolactin (Endocrinology Study Section, Xational Institutes of Health) were each present in the medium at a concentration of 5 pg per ml. The culture medium was replenished after each 48-hour period.
Culture of Isolated Mammary Epithelial Cells-Explants prepared fresh or removed from organ cultures were incubated in plastic vials at 37" for 60 min with collagenase (0.8 mg per ml) prepared in hfediurn 199 containing 0.1% serum albumin and penicillin and streptomycin (50 units each per ml) (6). The epithelial cells were separated from fat cells by centrifugation at 1000 x g for 10 min at O-4" and washed four times wit,h Medium 199. The suspension of the cells in Medium 199 was finally cultured in plastic Petri dishes for specified periods at 37".
Assay of Protein K&use-The activity of protein kinase in the purified cell membranes and ribosomes was measured by a slight modification of the method described earlier (2). The endogenous protein of these preparations served as the substrate for the endogenous protein kinase. The standard assay medium contained 10 pmoles of sodium glycerophosphate-HCl (pH 6.5), 1 nmole of [y-32P]ATP containing 2 to 5 X lo5 cpm, 2 pmoles of sodium fluoride, 0.4 pmole of theophylline, 0.06 pmole of ethylene glycol his@-aminoethyl ether)-N, N'-tetraacetic acid, 2 pmoles of cobalt chloride, and cell membranes or ribosomes with or without 200 pmoles of CAMP in a total volume of 0.2 ml. The incubation was carried out at 30" for 30 min and the reaction was stopped by the addition of 5 ml of cold 6% trichloroacetic acid. Bovine casein (2.5 mg), nonisotopic ATP (1.25 pmoles), and disodium phosphate (10 pmoles) 7209 were then added. After 30 min the suspension was heated at 90" for 20 min to hydrolyze nucleic acids. The precipitate was collected on Whatman GF/C glass fiber discs and washed with 25 ml of 6% trichloroacetic acid. To remove lipid, the discs were washed successively with 25 ml each of ethanol, ethanolether (3:1, v/v), and ether. The filter discs were finally counted in toluene scintillation liquid as described above. A unit of enzyme activity was defined as the amount of enzyme which catalyzed the transfer of 1 pmole of 32P from [-@*P]ATP to the recovered protein during 30 min under the standard assay conditions. c.UIP-independent protein kinase (I) and CAMP-dependent protein kinase (II) were isolated from lactating mouse mammary gland (1). The protein kinase activities of these enzymes and of CAMP-binding protein preparations were evaluated using calf thymus histones (1 mg) as the substrate in the above stmtdard assay medium.
After termination of the reaction by the addition of 10% trichloroacetic acid, the precipitate was assayed for radioactivity (2).
To an aliquot of these suspensions (1.0 ml), 6 ml of 6% trichloroacetic acid, bovine casein (2.5 mg), nonisotopic ATP (1.25 pmoles), and inorganic phosphate (10 pmoles) were added, and the mixture was allowed to stand at 0" for at least 30 min. Nucleic acids and lipids were extracted from precipitates by the procedures described above prior to counting radioactivity in toluene scintillation liquid.
Aliquots of the homogenates of the isolated epithelial cells, suspensions of 32P-membranes and -ribosomes were used for protein estimation, and the results were corrected for small losses of proteins during the isolation procedure.

DEAE-Cellulose
Chromatography of Protein Kinasez-Isolat.ed mammary protein kineses (I and II), fresh or pretreated with CAMP, were chromatographed on a column of DEAEcellulose by a slight modification of the procedure reported earlier (2). Fresh preparations of these enzymes were applied to a DEAE-cellulose column (0.9 x 24 cm) previously equilibrated with 5 m&f potassium phosphate buffer (pll 7.0) containing 2 mM EDTA.
The column was washed with 18 ml of the equilibrating buffer prior to elut,ion with a linear gradient of potassium phosphate (5 to 400 mM) in a total volume of 300 ml of the buffer. The flow rate was 8 ml per hour and the volume of each fraction was 3 ml. Enzyme preparations, preincubated with 0.5 pM CAMP, were also chromatographed by the above procedure except that all of the buffer solutions also contained 0.5 PM c~kkW.
Protein kinase and CAMP-binding protein preparations (0.2 ml), dialyzed previously against the above buffer, were layered over 4.5 ml of the sucrose gradient.
After centrifugation at 46,000 rpm for 21 hours at 5" in an SW 50L rotor, the tube bottom was punctured and IO-drop fractions were collected and assayed for protein kinase and CAMP-binding activities.
Preparation oj 32P-Labeled Jfembrane and Ribosomal Proteins--Mammary explants were exposed to media containing [32P]orthophosphatc (75 &i per ml) for 4 hours, and 32P-labeled epithelial cells were isolated by digestion of the explants with collagenase using 200 mg of lactating mammary gland as the carrier.
3*P-Labeled membranes were isolated from the homogenate of these cells without the addition of any carrier membrane. For the isolation of labeled ribosomes, highly purified mammary ribosomes (0.5 mg of protein) were used as the carrier. Membrane and ribosomal proteins were also labeled with 32P in cell-free preparations by the endogenous protein kinase activities of the membranes and ribosomes, respectively, and by exogenous protein kinase II. The reaction mixture contained 40 pmoles of sodium glycerophosphate-HCl (pH 6.5), 5 nmoles of [Y-~~P]ATI' containing 3.2 x 10" cpm, 12 pmoles of sodium fluoride, 2 pmoles of theophylline, 0.24 pmole of ethylene glycol bis(S-aminoethyl ether)-N,N'-tetraacetic acid, 8 pmoles of cobalt chloride, 1 nrnole of CAMP, and cell mem branes (1.6 mg of protein) or ribosomes (1.6 mg of protein) with or without 75 units of mammary protein kinase II in a total volume of 0.85 ml. The reaction was carried out at 30" for 60 min. The phosphorylation of membrane protein was stopped by the addition of 6 ml of cold 0.05 M sodium acetate buffer (pH 6.0) and the 32P-labeled membrane protein was sedimented by centrifugation at 27,000 X g for 30 min. The phosphorylation of ribosomal protein was arrested by the addition of 6 ml of cold 0.35 M sucrose containing 0.035 M KHCOB, 0.025 M KCl, 0.004 M MgC&, and 0.02 M KpHPOh (pH 7.4) (3) and the labeled ribosomes were isolated by centrifugation at 110.000 X u for 60 min.
Phosphorylation of llfembrane and Ribosomal Proteins in Intact Cells in Vitro-Explants or isolated mammary epithelial cells were exposed to media containing [32P]orthophosphate (40 to 50 FCi per ml) for 4 hours.
"Chase" experiments were performed by rinsing the isotopically labeled explants with Medium 199 and incubating them on nonisotopic medium for the indicated period.
After labeling with [32P]orthophosphate the explants were weighed and mixed with carrier tissue (200 mg of minced lactating mouse mammary gland), and digested with collagenase to separate the epithelial cells from the fat cells by the procedure described above. The 32P-labeled epithelial cells were washed four times with 0.25 M sucrose containing 0.1 M KCl, 0.01 M Tris-HCl (pH 7.4), and 0.003 nz MgCIZ prior to the isolation of 32P-labeled cell membranes and ribosomes by the methods described above. Purified mouse mammary membrane (10 mg of protein) and ribosomes (5 mg of protein) were added as carrier to the homogenates of these cells. The isolated 32P-membranes and -ribosomes were each suspended in water (1.2 ml) by gentle homogenization.
Ribosomal protein was extracted from the labeled ribosomes with 67% acetic acid in the presence of 0.1 M magnesium chloride (8), the acetic acid extract of the protein was dialyzed against 50 volumes of 77, acetic acid with two changes of the acid, and the dialysate was finally lyophilized (9). For the isolation of membrane protein, 32P-labeled membranes were digested at 37" for 60 min with RNase (75 pg) and DNase (70 pg) in 0.1 M sodium acetate buffer (pH 6.0) containing 5 mM magnesium chloride in a total volume of 1.0 ml. The reaction was stopped by the addition of 7 ml of 0.1 M sodium acetate buffer (pH 6.0) and the membranes were recovered by centrifugation at 27,000 x g for 30 min. To the membrane pellet 8 ml of 570 trichloroacetic acid were added, and the suspension was centrifuged at 1,000 x g for 10 min. For the removal of lipid, the precipitate was extracted with ethanol, ethanol-ether (3 : 1, v/v), and finally with ether. Polyacrylamide Gel Electrophoresis of Ribosomal Protein-3VLabeled ribosomal proteins prepared by the above procedure were subjected to polyacrylamide gel electrophoresis by a slight modification of the method of Leboy et al. (10). The lower gel (60 x 6 mm) contained 7.5% acrylamide (Eastman) and 0.150/, methylenebisacrylamide (Eastman) and the upper gel (10 x 6 mm) contained 2.5% acrylamide and 0.075SE methylenebisacrylamide. The lyophilized 32P-labeled ribosomal protein was suspended in 7 M urea containing 0.1 M 2-mercaptoethanol and the mixture was incubated at 37" for 2 hours (9). The protein in suspension was solubilized by the addition of glacial acetic acid (final concentration, 10%). Both the polyacrylamide gels and protein samples were 6 M in urea, and 150 to 200 pg of protein were applied to each gel for electrophoretic analysis.
Electrophoresis was carried out at 4" with a current of 3 ma per gel, and the time of electrophoresis, approximately 333 hours, was judged by the migration of Pyronine Y to the tip of the gel cathode from the upper reservoir.
After the electrophoresis, the gels were stained in 1% Amido black in 70/, acetic acid for 30 min, destained in 70/, acetic acid, and scanned for absorbance at 600 nm in a Gilford model 2400-S spectrophotometer equipped with a Gilford linear transport apparatus.
The gels were then sectioned with a manual slicing device (ll), and the thickness of each gel section was 1 mm. The gel sections were incubated in 1 ml of Soluene 100 (Packard) at 60" for 3 hours prior to incubation at room temperature overnight, and the samples were counted in toluene scintillation liquid as described above. The relative electrophoretic mobility (RF) of the protein bands was calculated from the ratio of the distance migrated by the protein band to the distance migrated by the tracking dye. blue in 25ol, isopropyl alcohol containing 10% acetic acid (12). The gels were destained first in 0.0025(;; Coomassie blue in lo<,& isopropyl alcohol containing lo?;, acetic acid for 4 to 5 hours and finally in loo/o acetic acid (12). The gels were scanned for absorbance at 550 nm and then sectioned with a manual slicing device for the assay of radioactivity in t,he gel sections by the procedure described above.
Other Analytical Procedures-3i1-'-Labeled membrane and ribosomal proteins were hydrolyzed in 2 K HCl in a boiling water bath for 16 hours (13). HCl was removed under vacuum, and the hydrolysate was subjected to paper electrophoresis on Whatman No. 1 paper strips (5 x 26 cm) using 8Cb formic acid as the electrophoresis buffer. Electrophoresis was con ducted at 4" for 10 hours at a constant voltage of 400 volts, the current flow being 2.8 to 3.0 ma per filter paper strip.
Ninhydrin-staining spots were cut out and counted in toluene scintillation liquid. The protein concentration of membrane and ribosome preparations and homogenates of the isolated mammary epithelial cells was estimated by the method of Lowry et al. (14). The protein content of the other samples was measured by the micro-biuret method (15). In both methods bovine serum albumin was used as the standard protein.

Polyacrylamide
Gel Electrophoresis of ddembrane Protein-Properties of cAXP'-binding Protein j'rom Mouse Jf ammary 3VLabeled membrane protein prepared by the above procedure Gland-Mammary CAMP-binding protein was separated l)arwas subjected to polyacrylamide gel electrophoresis by a mod-tially from the protein kinase activity by hydrosylapatite ification of the method of Fairbanks et al. (12) 32P-Labeled mem-cedure are shown in Table I. Treatment with 6 nz guanidine brane protein was suspended in 0.01 M Tris-HCI (pH 8.0) con-HCI selectively removed approximat,ely 94%; of the protein taining 0.04 M dithiothreitol, 0.002 M EDTA, 6 fir urea, 0.4TQ kinase activity. The CAMP-binding protein was purified to sodium dodecyl sulfate, O.l$& Triton X-100, and 0.2y0 sodium approximately 209fold as compared to the whole tissue homogdeoxycholate, and the mixture was incubated at 37" for 2 hours. enate and the preparat,ion was approximately 9954 free of the The suspension was centrifuged at 500 x g for 5 min to sedi-initial protein kinase activity per unit of activity of the CAMPment the small amount of insoluble material, and the supernate binding protein.
The binding protein showed a higher degree (30 to 40 pg of protein) containing lOcl, sucrose was used for of specificity for the binding of CAMP as compared to the the electrophoretic separation of proteins. The buffer for both binding of other cyclic nucleotides (cGMP, cUMP, and cCMP). of the reservoirs was 0.04 bf Tris-0.02 M sodium acetate-O.002 Isokinetic sucrose gradient centrifugation showed that CAMP-M EDTA-acetic acid buffer (pH 7.4) containing l%, sodium binding protein sediments as a single peak (2.4 S). c,4MPdodecyl sulfate.
Electrophoresis toward the anode was carried binding protein markedly inhibited the activity of protein out at 30" for approximately 75 min with a current of 8 ma per kinase I, and the addition of CAMP completely restored the gel using Pyronine Y as the tracking dye. After the elec-original activity of the protein kinase. However, the CAMPtrophoresis, the gels were stained overnight in 0.05% Coomassie binding protein had no significant effect on the activity of CAMP-activated protein kinase II. The other characteristics of CAMP-binding protein were almost identical with those reported earlier (2) using a less purified preparation of CAMPbinding protein. Fig. 1 shows the DEAE-cellulose chromatography profiles of protein kinase II before and after treatment with 0.5 pM CAMP.
Untreated protein kinase II showed the presence of a small a.rnount of protein kinase I as a contaminant.
Pretreatment of kinase II with CAMP caused it to be eluted in the form of protein kinase I. Analysis of the dialyzed fractions showed that t,he protein kinase activity recovered in the peak corresponding to kinase I was not activated by CAMP, and there was no detectable protein kinase peak corresponding to kinase II. There was, however, no change in the DEAEcellulose chromatography pattern of protein kinase I after preincubation with cAhlP. The product of the interaction of protein kinase I with CAMP-binding protein was analyzed by isokinetic sucrose density gradient centrifugation (Fig. 2). Purified protein kinases I and II sedimented in the sucrose gradient as 2.6 S and 3.9 S peaks. Preincubation of protein kinase I with CAMPbinding protein caused the enzymatic activity to have a sedimentation pattern which was nearly identical with that of native protein kinase II and caused the enzymatic activity to be dependent upon CAMP. Such interaetion with protein kinase I caused the 2.4 S CAMP-binding protein to sediment as two peaks (2.4 S and 3.9 S). Treatment of protein kinase II with CAMP-binding protein did not change its sedimentation pattern on the sucrose gradient. Uoth the cell membranes and the ribosomes contained protein kinase activity which was not stimulated significantly by CAMP, and this protein kinase activity was lost completely when the membranes and ribosomes were heated at 100" for 3 min. Treatment of the membranes with cetyltrimethylammonium bromide enhanced the membrane-associated protein kinase activity by apprositnately 5-fold, whereas no such latent protein kinase activity was detected in the ribosomes.
There was no appreciable extraction of the membrane-and ribosome-bound protein kinase activity by treatments with 0.2% Triton X-100, 0.2% sodium deoxycholate, 0.2% cetyltrimethylammonium bromide, 0.1% phosphatidylethanolamine, 0.9 M potassium chloride, 6 M guanidine HCl, 1.5 M 2-mercaptoethanol, or 5 M urea. Isolated cell membranes (cetyltrimethylammonium bromidetreated) and ribosomes contained approximately 1% and 0.4%, respectively, of the total protein kinase activity of lactating mammary tissue when calf thymus histones were used as the substrate for phosphorylation. Fig. 3 shows the rate of phosphorylation at varying concentrations of membrane and ribosomal protein substrates by exogenous cytosol protein kinase II. Protein kinase II phosphorylated ribosomal proteins at a faster rate (approximately a-fold greater) than membrane proteins, and the rate of phosphorylation of mammary histones by protein kinase II was approximately six times greater than that of ribosomal proteins.
The apparent K, value for membrane protein (1.05 mg per ml) was slightly greater than that of ribosomal protein (0.8 mg per ml) for protein kinase II. Plasma membrane proteins were phosphorylated with 32P in cell-free reactions with endogenous protein kinase of membrane or with cytosol protein kinase II or by incubating lactating The purified membrane proteins were then fractionated by polyacrylamide gel electrophoresis (Fig. 4). Thirty protein bands were detected in the electrophoretogram of the membrane proteins, and 19 of these bands were labeled with 32P. Four proteins with RF values of 0.58, 0.61, 0.65, and 0.68 represent the major 32P-labeled phosphoproteins. of phosphorylation of membrane proteins by membrane-bound protein kinase and by cytosol protein kinase II was nearly identical with that obtained with 32Pproteins derived from plasma membranes of intact mammary cells incubated with 32p. Fig. 5 shows the electrophoretogram radioactivity profiles of 32P-labeled ribosomal proteins which were phosphorylated by endogenous protein kinase of purified ribosomes or by reacting ribosomes wit'h cytosol protein kinase II, or by incubating lactational mammary gland with [32P]orthophosphate in organ culture.
Eighteen protein bands were detected in the electrophoretogram of ribosomal proteins and at least eight of these proteins are phosphorylated in intact mammary tissue in vitro. Addition of cytosol protein kinase II to suspensions of purified ribosomes augmented the phosphorylation of four ribosomal proteins (RF = 0.28, 0.40, 0.45, and 0.73) which were also phosphorylated by the endogenous protein kinase activity of the ribosomes.
Cytosol protein kinase II also phosphorylated three other ribosomal proteins (RF = 0.50, 0.60, and 0.67) which were either not phosphorylated or phosphorylated to a small extent by the endogenous protein kinase of ribosomes. The patterns of ribosomal 32P-protein formed by reacting the endogenous protein kinase or the cytosol protein kinase IL with ribosomes were quite different from that obtained from intact lactating mammary tissue cultured ilz vitro with 32Pi.
Two ribosomal proteins (RF = 0.33 and 0.37) which were most heavily phosphorylated in the whole lactating tissue were not phosphorylated significantly by the endogenous protein kinase of isolated ribosomes or by the cytosol protein kinase II in cellfree systems. The ribosomal proteins having RF values of 0.60 and 0.67 were phosphorylated to a small extent in the lactating mammary gland, whereas these proteins were phosphorylated to the great,est degree in purified ribosomes incubated with protein kinase II.
Effect of Hormones poration of 3*Pi into membrane and ribosomal proteins of mammary epithelial cells in mammary explants cultured in vitro. Insulin caused an increase of approximately 160% and 120y0 in the rate of incorporation of "Pi into membrane and ribosomal proteins, respectively, of mammary epithelial cells. The maximal effect of insulin on the phosphorylation of membrane and ribosomal proteins was observed after 16 hours of incubation. The degree of stimulation of membrane and ribosomal protein phosphorylation was slightly lower in explants incubated with insulin and hydrocortisone as compared with those incubated with insulin alone. Results similar to those shown in Fig. 6 were also observed with epithelial cells isolated from the cultured explants.
It has been previously shown that prolactin induces specific milk proteins only in alveolar cells formed in vitro and previously treated with insulin and hydrocortisone (16). The period of cell division required to form these cells is essentially complete after 72 hours of incubation (17). Tissues previously incubated on medium containing insulin and hydrocortisone for 72 hours were then exposed to the same medium or to the medium containing insulin, hydrocortisone, and prolactin for the specified periods.
The explants after exposure to "Pi (45 &i per ml) were digested with collagenase, and 3Wlabeled membrane and ribosomal proteins were obtained from the isolated epithelial cells.
Each point represents the amount of isotopic precursor incorporated during the preceding 4-hour period.
O-O, membrane 32P-protein, insulin and hydrocortisone; A---A, ribosomal 32P-protein, insulin and hydrocortisone. action of prolactin for the phosphorylation of membrane protein was maximal after 8 hours, and the half-maximal effect was observed after approximately 4 hours. The maximal effect of prolactin on the phosphorylation of ribosomal protein was observed after 16 hours, and the half-maximal effect was noted after approximately 8 hours. Similar results were obtained when epithelial cells, isolated from explants previously incubated first with insulin and hydrocortisone for 72 hours and then with prolactin for the specified periods, were exposed to medium containing 32Pi. Table II shows the effect of insulin and hydrocortisone for the prolactin-mediated stimulation of the phosphorylation of mammary membrane and ribosomal proteins.
Prolactin stimulated the rate of incorporation of 32Pi into these proteins in cells which were previously treated with hydrocortisone, and prolactin required the continued presence of insulin for its stimulatory action on the phosphorylation of membrane and ribosomal proteins. Results similar to those in Table II were also observed with isolated epithelial cells derived from the 32P-labeled mammary explants.
In order to distinguish the responses of the daughter cells formed in vitro from those of the epithelial cells which did not divide, colchicine was added to the culture to arrest dividing cells in mitosis.
As shown by the results listed in Table III, prolactin did not stimulate the incorporation of 32Pi into membrane and ribosomal proteins after colchicine treatment, in-7214 dicating that prolactin stimulates the phosphorylation of these tion were similar to those observed with whok membrane and proteins in cells formed in uitra. Addition of actinomycin 11 ribosomal proteins. This result indicates that the hormonal or cycloheximide to the culture along with prolactin prevented effects relate to the phosphorylation of the polypeptide chains the stimulation of phosphorylation of membrane and ribosornal and not to the phosphorylation of some protein-associated proteins, a result consistent w&h the conclusion that these molecules. Serine was phosphorylated at more than twice the prolactin-mediated increases in protein phosphorylation require rate of the threonine residues in both the membrane and riboconcomitant synthesis of RNA and protein.
somal proteins. The V-labeled membrane and ribosomal proteins derived from each hormone system were subjected to partial acid hydrolysis, and the phosphorylated amino acids were separated by paper electrophoresis.
As shown in Table IV prolactin stimulated the phosphorylation of serine and threonine residues of membrane and ribosomal proteins and the degrees of stimula- ,Vammary explants previously incubated first with insulin and hydrocortisone for 0 to 72 hours and then with or without prolactin for 72 to 85 hours were labeled with 32Pi in vitro, and the purified cell membrane and ribosomal proteins were subjected to polyacrylamide gel electrophoresis. It is clear from the eiectrophoretogram radioactivity profiles of V-labeled membrane protein (Fig. 8) that at least 19 proteins of the membrane are phosphorylated when the mammary explants are incubated in insulin and hydro'cortisone, and the addition of prolactin caused a general stimulation of the phosphorylation of these proteins of the membrane. Fig. 9  Estirnation of the phosphorylation of casein and membrane protein in mammary explants incubated in different hormone systems disclosed that less than 1% of the observed radioactivity of the membrane protein may be due to contamination with [V]casein. This conclusion is further strengthened by the marked differences in the kinetics of prolactiu-induced phosphorylation of cell membrane protein (Fig.  7) and of casein (18). 7216 phorylation by the endogenous protein kinase or by the cytosol kinase as compared to the pattern of phosphorylation by intact mammary cells in organ culture (Fig. 5). These differences may relate to differences in relative substrate availability in the intact cells as compared to the broken cell preparations.
In the present studies, it has been possible to compare the patterns of protein phosphorylation of membrane and ribosomal proteins by purified enzyme preparations with the patterns of phosphorylation observed in cultured cells responding to specific hormonal signals.
Previous studies demonstrated that prolactin and insulin act synergistically to induce rapidly the cytosol protein kinase in mammary epithelial cells in organ culture (2). The present studies demonstrate that this induction is associated with a marked increase in the rates of phosphorylation of specific membrane and ribosomal proteins in the cultured mammary cells formed in vitro (Fig. 7). Roth enzyme induction and the phosphorylation of specific plasma membrane and ribosomal proteins occur selectively in the mammary epithelial cells formed during the period of incubation in vitro as shown by the experiments with colchicine (Table III).
The induction of protein kinase required only the actions of insulin and prolactin on the cultured cells (2). A stimulation of the rate of phosphorylation of the membrane and ribosomal proteins required in addition a preliminary period of treatment of the cells with hydrocortisone (Table II).
Experiments with inhibitors of RNA and protein synthesis indicated that the prolactin-mediated stimulation of membrane and ribosomal protein phosphorylation is dependent upon the concomitant synthesis of RNA and protein.
It has been previously shown that the specific activity of the ATP pool of these cells does not change appreciably after stimulation with these hormones (13), indicating that the increased rate of incorporation of 3"P into specific proteins represents an increase in the net rate of phosphorylation of these proteins rather than a change in the availability of isotopically labeled precursor. Since 110 turnover of the 32P-labeled protein was detected during the R-hour chase periods, the net rate of incorporation of isotope into the phosphorylated proteins may be taken to represent the rate of phosphorylation of these substrates. Studies on the time course of the hormonally induced phosphorylation of membrane and ribosomal proteins (Fig. 7) demonstrated that the maximal rates of phosphorylation were obtained following the induction of maximal levels of cytosol protein kinases (2). This result is consistent with the concept that the protein kinase may be rate-limiting for protein phosphorylation, and that the phosphorylation of these proteins occurs as a consequence of the induction of protein kinase by prolactin. Characterization of the 32P-labeled proteins derived from plasma membrane demonstrated that prolactin causes a general increase in the rate of phosphorylation of all 19 phosphoproteins identified by this technique in the plasma membrane preparations. Previous studies have demonstrated that prolactin causes the formation of new ribosomes and polysomal aggregates, and the kinetics of the prolactin-mediated formation of new polysomes (27) is similar to that of the prolactiniuduced phosphorylation of ribosomal proteins.
The selective stimulation of the phosphorylation of four ribosomal proteins in response to prolactin may reflect factors relating to substrate availability as well as the concentration of induced protein kinase. Insulin stimulated the rate of phosphorylation of specific membrane and ribosomal proteins in the undifferentiated mam-mary stem cells, and the maximal stimulation was observed after 16 hours of culture (Fig. 6). This insulin-mediated phosphorylation of membrane and ribosomal proteins occurred concomitantly with the induction by insulin of the cytosol protein kinase (2) and with the phosphorylation of histones and nuclear acidic proteins (13). These results support the concept that the insulin-stimulated phosphorylation of nuclear, ribosomal, and membrane proteins may be mediated by the induced protein kinase.
The previous studies on the action of prolactin on mammary epithelial cells have indicated that prolactin initially interacts, with specific hormone receptors which are part of the superficial cell surface (28,29). Subsequently, both subunits of the cytosol protein kinase are induced, and this induction represents one of the earliest intracellular actions of prolactin (2). The present studies, together with those previously reported (13), indicate that stimulation by prolactin is associated with the phosphorylation of specific proteins of the nucleus, plasma membrane, and ribosomes of mammary cells. The phosphorylation of a variety of proteins in a number of functionally distinct organelles in various compartments of the cell may thus serve to propagate and amplify the initial stimulus of prolactin at the cell surface. Elucidation of the physiological significance of these phosphorylation reactions in the functional activation of the mammary secretory cell is currently under investigation.