Lee Moffitt Cancer Institute and Research Center, University of South Florida, ... Cancer Research and Development Center, Frederick, Maryland 21 702, the ...
THEJOURNAL OF BIOLOGICAL CHEMISTRY
Vol. 267, No. 33, Issue of November 25, pp. 24076-24081,1992 Printed in U.S. A.
Prolactin Receptor Triggering EVIDENCE FOR RAPID TYROSINE KINASE ACTIVATION* (Received for publication, March 3, 1992, and in revised form, May 14, 1992)
Hallgeir RuiSQT,Julie Y. DjeuS, Gerald A.Evans11, Paul A. Kelly**, and William L. Farrare$$ From the SH. Lee Moffitt Cancer Institute andResearch Center, University of South Florida, Tampa, Florida 33612, the 11Biological Carcinogenesis DevelopmentProgram, Program Resources IncJDynCorp, National Cancer Institute, Frederick Cancer Research and Development Center, Frederick, Maryland 21 702, the **Institut National dela Sante et de la Recherche Mkdicale Unite 344, Endocrinologie Moleculaire,Faculte de Medicine Necker, Enfants Malodes, 156 rue de Vaugirard, 75730 Paris Cedex 15, France, and the SCytokine MolecularMechanisms Section, Laboratory of Molecular Immunoregulation, Bioloaical ResoonseModifiers Proaram. National Cancer Institute, Frederick Cancer Research and Development Center, Fredlrick, M&vland 21 702 1
,
The mechanism of action of prolactin (PRL) has remained obscure despite the unveiling of the primary structure of PRL receptors. The present study demonstrates rapid PRL receptor-mediated tyrosine phosphorylation of at least three cellular proteins, designated p120, p97, and p40, in a rat T-lymphoma (Nb2-11C) as revealed by antiphosphotyrosine immunoblotting. One of the phosphotyrosyl proteins, p120, co-purified with activated PRL receptor complexes obtained using either anti-ligand or antireceptor antibodies. Furthermore, in vitro incubation of affinity-purified PRL receptor complexes from PRL-stimulated cells with ATP in the presence of a tyrosine phosphatase inhibitor, resulted in a 10-15fold increase in the phosphotyrosine content o f p120, as revealed by antiphosphotyrosine immunoblotting. Parallel experiments utilizing [yS2P]ATPconfirmed a rapid and time-dependent incorporation of phosphate into p120 in the same affinity-purified PRL receptor complexes. These data provide strong evidencefor the involvement of a tyrosine kinase in PRL signal transduction and suggest the presence of a tyrosine kinase within the activated PRL receptor complex.
growth hormone, several interleukins (IL2 through IL7), granulocyte-macrophage colony-stimulating factor, granulocyte colony-stimulating factor, erythropoietin, and ciliary neurotrophic factor (9-19). To date, two normally occurring forms of PRL receptors have been cloned from mammalian cells. They result from alternate splicing of mRNA and differ only in their cytoplasmic domains (20-24). The long form of the rat PRLreceptor, which comprises 591 amino acids, is capable of transducing signal when transfected into Chinese hamster ovary cells along with a reporter gene construct, whereas a functional role for the short form (291 amino acids) has yet to be demonstrated (25). Despite extensive efforts over the last decades, it hasproven difficult to elucidate the signal transduction mechanisms involved in PRLaction (26). Recently, our laboratory and others have presented evidence that tyrosine kinase activation may be an important early event associated with ligand binding to several hematopoietin receptors, including IL2, IL3, IL4, erythropoietin,and growth hormone (27-32). The present study was initiated to investigate tyrosine kinase involvement in PRL signaling, using the Nb2-11C cell line. This is a preT rat lymphoma cell line, which is dependent on PRL for growth (33). The predominant form of PRL receptor in this cell line has been recently cloned and shown to be a mutant Prolactin (PRL)’ is a multifunctional pituitary hormone of the long form of the rat PRLreceptor (26). It lacks a 198involved in the control of a wide variety of physiological amino acid segment of the cytoplasmic domain, but appears processes in vertebrates, including osmoregulation, reproduc- fully capable of transducing the PRLsignal. The Nb2 cellhas tion, and immune responses, as well as cellular proliferation proven especially valuable in studies of PRL as a growth and differentiation (1-3). The receptors for PRL belong ge- factor, demonstrating effects on growth-related gene trannetically to the recently recognized family of hematopoietin scription within 15-30 min of hormone exposure (34, 35). To or cytokine receptors, whose extracellular domains share sim- date, the only published work on protein phosphorylation in ilar structural elements (4-6). These include 4 paired cysteine response to PRL in target cells are two reports focusing on residues and a Trp-Ser-X-Trp-Ser motif, which have been the phosphorylation state of proteins of M , less than 30,000 postulated to be involved in basic aspects of ligand binding in Nb2 cells several hours downstream of PRL receptor acti(7, 8). An overall homology in the tertiary structure of the vation (36, 37). There are no reports in the literature that extracellular domain has also been predicted (7). In addition clearly describe protein phosphorylation as being important to PRL receptors, this gene family includes receptors for for immediate PRL receptor-associated signaling. In this paper, we present the first direct evidence for the involvement * This project has been funded at least in part with federal funds of tyrosine kinase activation at the level of initial PRL recepfrom the Department of Health and Human Services under Contract tor triggering. No. N01-CO-74102 with Program Resources, Inc./DynCorp. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ll Recipient of Fogarty Fellowship 5F05 TW04300-02. $3 To whom correspondence should be addressed. The abbreviations used are: PRL, prolactin; IL, interleukin; PAGE, polyacrylamide gel electrophoresis; hPRL, human PRL.
EXPERIMENTALPROCEDURES
Materiak-Pituitary hormones, human PRL (NIDDK-hPRL-B-1; AFP-8982C), ovine PRL (NIDDK-oPRL-19; AFP-9221A), human growth hormone (NIDDK-hGH-01; AFP-9755A), ovine growth hormone (NIDDK-oGH-15; AFP7649C), andpolyclonal anti-hPRL antiserum (anti-hPRL-IC-4;AFP-9823102188) were gifts from the National Hormone and Pituitary Program/National Institute of Diabe-
24076
Prolactin Receptor-associated Kinase Tyrosine tes and Digestive and Kidney Diseases (Baltimore). TheNb2 cell line was originally developedby Dr. P. Gout (Universityof British Columbia, Vancouver, Canada), and theNb2-11C clone used in the present studies was obtained from Dr. C. Clevenger (University of Pennsylvania, Philadelphia). Cell Culture and Treatment-Nb2-11C cells were grown in RPMI 1640 medium containing 10% fetal calf serum(Sigma,Cat. No. F2442), 10% gelded horse serum (Sigma, Cat. No. H1895). 2 mM Lglutamine, 5 mM HEPES buffer, pH 7.3, penicillin-streptomycin (50 IU/ml and 50 pg/ml,respectively), and 50 p~ 2-mercaptoethanol. Cells a t a density of 1-1.5 X 10'/ml were made quiescentby incubation for 20-24 h in the above medium without fetal calf serum and 2mercaptoethanol (starvation medium). Viability of 95-100% was ensured beforeproceeding. Forstimulationstudies, cell suspensions were brought to a density of loRcells/ml in the starvation medium and preincubatedfor 20 min at the temperature be to used. Hormones weregenerally added as a 10 X concentration to 1 ml of cells in Eppendorf tubes. This enabledquick pelleting and snap-freezingin a dry ice/methanol bath. Solubilization of Membrane Proteins and IrnmunoprecipitationFrozen pellets from 10' cells were thawed on ice and solubilized in 1 ml of lysis buffer containing 10 mM Tris-HCI, pH 7.6, 5 mM EDTA, 50 mM NaCI, 30 mM sodium pyrophosphate, 50 mM sodium fluoride, 100 p~ sodium orthovanadate, 1% Triton X-100,l mM phenylmethylsulfonyl fluoride, 5 pg/ml aprotinin, 1 pg/ml pepstatin A, and 2 pg/ ml leupeptin. Cell lysates were rotated end over end a t 4 "C for 60 min, and insoluble material was pelleted a t 12,000 X g for 30 min. Duetothe relatively low abundance of PRL-responsivetyrosine phosphoproteins, studies of hormone specificity, time, and dose responses were performed using anti-phosphotyrosine immunoprecipitation to enhance the detection of these species. Depending on the experiment, supernatants were incubated rotating end over end for 2-6 h a t 4 "C with either mouse monoclonal anti-phosphotyrosine antibodies (PY20, ICN Biochemicals, Cat. No. 691371 02RC; 5 pg/ ml), polyclonal rabbit anti-hPRL antiserum (NIDDK anti-hPRL-IC4; 5 pl/ml), normal rabbit serum (5 pl/ml), anti-rat-PRL receptor antibody (U6; IgG,-subtype; 5 pglrnl), or nonspecific mouse monoclonal antibody of subtype IgG, (5 pglml). PY20 mouse monoclonal anti-phosphotyrosineantibodiesandrabbit polyclonal antiserum were precipitated directly with protein A-Sepharose beads, whereas antibodies of IgG,-subtype were precipitated using protein A-Sepharose coated with polyclonal goat anti-mouse IgG, antibodies (Sigma, Cat. No. M8144). Precipitated material was eluted off the beads by boiling in SDS sample buffer for 4 min, subjected to7.5% SDS-PAGE under reducing conditions, and transferred to polyvinylidene difluoride membrane (Immobilion, Millipore, Cat. No. 1PVH 00010). Irnmunoblot Analysis-Blocking of filters was done by incubating in TBS (20 mM Tris-HCI,pH 7.6, 137 mM NaCI)containing 1% bovine serum albumin for a t least 1 h on a rocking platform. All steps were carried out a t room temperature. For detection of phosphotyrosine proteins,blots were incubated for90minwitha different mouse monoclonal anti-phosphotyrosine antibody (UBI, Cat. No. 05321) a t a concentration of 0.1 pg/ml in blocking buffer. The rationale for using a different antiphosphotyrosine antibody for the immunoblotting was that this would reduce the risk of nonspecific detection. However, results were comparable when the monoclonal UBI antibody was used both for precipitation and blotting (data not shown). The blots were washed twice for 5 min in washing buffer (50 mM Tris-HCI,pH 7.6, 200 mM NaCI, 0.05% Tween20), followed by incubation for 45 min with biotinylated, affinity-purified, goat antimouse I& (KierkegaardandPerryLaboratories, Inc., Cat. No. 161802) diluted to 250 ng/ml in washing buffer containing 1% bovine serum albumin.After another two 5-min washes, blots were incubated with horseradishperoxidase-conjugatedstreptavidin(Kierkegaard and Perry Laboratories, Inc., Cat. No. 143000) diluted to 50 ng/ml in washing buffer containing 1% bovine serumalbumin for 30 min, followed by four 5-min washes. The blots were incubated with ECL chemiluminescence substrate mixture according to the manufacturer's instructions (Amersham Corp., Cat. No. RPN2106) for1min before being sealed in plastic bags and exposed to x-ray film for 1-60 min. Affinity Cross-linking-Covalent cross-linking of '"I-hPRL (Du Pont-New England Nuclear, Cat. No. NEX-127) was carried out by incubating 50 X loRlactogen-deprived Nb2-11C cells in 1 ml of RPMI medium a t 4 "C for 2 h with a 1.5 nM concentration of labeled hPRL in the presence or absence of a 200-fold excess of unlabeled hPRL. Cells were washed three times in cold phosphate-buffered saline a t pH 8.3 containing 5 mM MgCI,. The cells were resuspended in 10 ml
24077
of this buffer containing the bifunctional cross-linker his(succinimidy1)suberate (Pierce Chemical Co.. Cat. No. 21579) at a concentration of 2 mM and incubated rotating end over end for 1 h a t 4 "C. The reaction was stopped by adding 400 pl of 1 M ammonium acetate for 10 min, followed hy three washes in phosphate-huffered saline. Cells were then lysed in lysis huffer with proteolysis inhibitors as described above. PRL Receptor-associated Tyrosine K i m w Assay-Quiescent Nh211C cells were exposed to medium or 10 nM hPRI, for 30 min at room temperature. Cell pellets were washed twice in cold medium to eliminate unbound hormone. Cell pellets were lysed in 25 mM HEPES, pH 7.3, 100 mM NaCI, 1% Triton X-100, and proteolysis inhihitors as described above a t a concentration of10" cells/ml. Lysates from stimulated and unstimulated cells were incubated withpolyclonal or monoclonalmouse anti-rat-PRL rabbitanti-hPRLantiserum receptorantihodies or respective controlantibodies atthesame concentrations as described above for 2 h a t 4 "C rotating end over end. Immunoprecipitation was done with protein A-Sepharose beads or anti-mouse-IgC,-coated protein A-Sepharose heads for 30 min at 4 "C. The protein A-Sepharose was pelleted and washed six times in the same buffer but with 0.1% Triton X-100. At the last washing step, pellets were split into aliquots such that each tube contained immunoprecipitated material from approximately 2 X l(r cells and were incubated a t room temperature for 30 min with 3 mM MnCIz and 2 mM sodium orthovanadate. a tyrosine phosphatase inhihitor, in the presence or absence of 15 p~ ATP. After the incubation, the proteinA-Sepharosebeads were washed onceand hound proteins were eluted off hy boiling in SDS sample huffer. Detection of phosphotyrosine-containing proteins was done as described ahove, using antiphosphotyrosine immunoblotting. RESULTS ANDDISCUSSION
PRL Modulation of Levels of Cellular Phosphotyrosyi Protein.+"he tyrosine phosphorylation state of cellular proteins was examined by immunoprecipitation with monoclonal antiphosphotyrosine antibodies of whole cell detergent lysates, followed by SDS-PAGE with subsequent anti-phosphotyrosine immunoblotting, usinga different monoclonal anti-phosphotyrosineantibody to ensurespecificity.Specificitywas also confirmed bythe ahility of phosphotyrosine (10 mM), h u t not phosphoserine or phosphothreonine, to compete with the proteins of interest for binding to the primary antihody (data not shown). Stimulationof cells with PRL for 10 min resulted in a pronounced induction of tyrosinephosphorylation of proteins of M , 120,000,97,000, and 40,000 (Fig 1). This induction was mediated by the PRL receptor since tyrosine phosphorylation was obtained using hoth ovine and h u m a n PRL, as well as the lactogenic human growth hormone, hut not with the nonlactogenic ovine growth hormone, which is M, 100-
-
-p1m
-per
9)-
66-
; 43
-
'OQHC
W
-
-. .
.-
Clrl OPAL hPRL oOH hOH
FIG. 1. PRL receptor-mediated induction of protein tyrosine phosphorylation. Anti-phosphotyrosine immunoblot of antiphosphotyrosine-immunoprecipitated proteins from lysates of Nh2 cells that had been exposed to medium (Ctrl) or 100 nM ovine PRI, (o(;H). or (oPRL),human PRI, (hPRL).ovine growthhormone human growth hormone (h(;H) for 10 min at 37 "C. M, ( X 10")of each protein markers is indicated. Arrouw identifyphosphotyrosyl proteins p120, p97, and p40, and the hrackef marks irnmunoglohulin heavy chains (I&Hc.).
Receptor-ass
24078
:ociated Tyrosine Kinase
Prolactin
unabletointeractwiththePRLreceptor (38,39). These cross-linked to PRL receptors were tested (Fig. 8).Of previPRL-inducedproteintyrosinephosphorylations were ob- ously characterized monoclonal mouse antibodies against rat served at physiological concentrations of the hormone, with PRL receptor (Us, U6, and T1) (40), the antibody LJfi proved EDso values of approximately 200 PM (Fig. 2 A ) . Furthermore, veryefficient in precipitating receptor complexes. Furtherwas ableto the phosphorylation kinetics of proteins p120 and p97 dif- more,polyclonal rabbitanti-hPRLantiserum fered, as can be seen in Fig. 2B. The amount of immunoreac- precipitate identical material with similar efficiency. Autorative phosphotyrosyl residues in p120 peaked as early as 60- diography of immunoprecipitates separated by SDS-PAGE p97 responded more slowly, under reducing conditions revealed a predominant bandof M , 90 s after PRL exposure, whereas with peak phosphorylation a t 5-20 min. The phosphotyrosyl 88,000, which corresponds to the sum of ""I-hPRI, (22,000) and themajor species of Nb2 cell PRI, receptor of M , 62,000bandmigrating a t 40 kDaappearedtorepresentseveral proteins of similar M , and showed a variable degree of phos- 66,000 (26, 41, 42). Specificity of binding of "'I-hPRI, was phorylation between experiments. In general, its time kineticsverified by the ability of excess unlabeled hPRL to compete a somewhat more rapid as seen in whole cell lysate. Additional bands of apparent M , followed an intermediate pattern with induction than that of p97, but peaking a t similar time (10- 121,000, 156,000, and 208,000 were also detected,but no 20 min) (data not shown).Although proteins p120 and p97 in information is presently available on their nature. The abquiescent cells do not contain detectable amounts of phos- sence of any cross-linked species of M , lower than 88,000 photyrosine, p40 appeared tobe phosphorylated a t a low level verified that theM , 42,000 short form of PRL receptor,whose mRNAhas been detected in low levels in Nb2 cells using in unstimulated cells. Ability of Antibodies to Precipitate Chemically Cross-linked polymerase chain reaction technique (26), does not bind PRL PRL Receptor Complexes-In order toexplore whether any of to a significant degree, indicating that PRL-dependent physiological responsesin these cells are mediated through the the observed phosphotyrosyl proteins were associated with the PRL receptor complex, the ability of several antibodies dominant M , 62,000-66,000 variant of the long PRL receptor. Evidence for a PRL Receptor-associated Phosphotyrosyl Protoimmunoprecipitate complexes of '''I-hPRL covalently tein-The polyclonal anti-hPRL antiserumor the monoclonal anti-rat-PRL receptor antibody (U6) were subsequently used toimmunoprecipitate solubilized PRLreceptor complexes from whole cell lysates of stimulated and unstimulated cells. Both strategies revealed that the phosphotyrosyl protein p120 co-purified with the PRL receptor, whereas p97 and p40 did not (Fig. 4). Separate experiments with PRL-stimulat.ed cells ensured that the receptor-associated, tyrosine-phosphorylated p120 co-migrated with the p120, which could be immunoprecipitated with anti-phosphotyrosineantibodies from whole cell lysates(datanotshown).Thecapture of comparable amounts of PRL-induced,tyrosine-phosphorylated, p120 using the two methods supports the view that the two p120 species are identical. However, firm demonstration of such identity awaits peptide mapping. The results presented were 0 H 0 to"2 to-" 10'lO to" 10-8 to" 10-8 obtained using immunoprecipitation of activated receptor
I
Prolactin (MI B
-
Y 100
a
.r(
200-
X
m
E
-208 -156 -121
80
0
97-
-
66-
E 40 1
/
/'
43-
//
-
22-
Cold hPRL
0 0 0.3 0.5
1
2
5
10
20
50
Time (Min)
FIG.2. Induction of tyrosine phosphorylation of p120 and p97 by PRL. A, densitometricanalysis of anti-phosphotyrosine immunoblots of p120 and p97 immunoprecipitated with anti-phosphotyrosineantibodies from lysates of Nb2cells that had been exposed to various concentrations of hPRL for 10 min a t 37 "C. R, timecourse in minutes of the PRL-induced increases in tyrosine phosphorylation of p120 and p97 in Nb2 cells treated with 100 nM hPRL. Note the log scale of the abscissa. Values for optical density (OD; absorbance)obtained by densitometricanalysis were transformed to relative units of maximum stimulation, and error bars represent S.D. of triplicate determinations on representative experiments repeated three times.
+
-
0
"
-
0
"
-
08
-
Whole cell
US U6 T I lgG, ,.hPRL nrs
lysate
clrl
FIG.3. Immunoprecipitation of covalentlycross-linked P R L r e c e p t o r c o m p l e x e s c o n t a i n i n g "'I-hPRL. Whole cell I v sates of Nh2-llC cells to which "''I-hPRI, had been Cross-linked in the presence (+) and absence (-) of 100-fold excess unlaheled ovine PRL were separated on SDS-PAGE under reducing conditions. and complexes containing specifically hound PRI, were detected hy autoradiography. PRL receptor complexes from the same lysates representing 30 times more material were immunoprecipitated using various monoclonal antihodies to rat PRI, receptor ((1.5, (16, TI. and IgC, ctrl) or polyclonal rahhit antiserum to hPRL (nhPHI,) or normal rabbitserumcontrol ( n r s ) . M, ( X lo-') of eachproteinmarker is indicated. Arrows indicate distinct ""I-hPRI.-cross-linked specie9 of varying M,from 88,000 to 208,000.
Receptor-associated Kinase Prolactin Tyrosine
24079 nhPRL as
M,
nrs ctrl
"
-
200 -
97-
-
200 c "p120
97-
-
66-
p66
a "pl20
-
66-
-
p66
43-
" ~
hPRL
-
+
"
-
+
Treatment
"
nhPRL mrPRL-R
FIG.4. Anti-phosphotyrosine immunoblot of solubilized, affinity-purified PRI, receptor complexes separated on SDSPACE under reducing conditions. Hot h polyclonal anti-hPHL antiserumand monoclonal anti-rat-I'RI,receptorantibodies (UC,) were capable of immunoprecipitating a p120 strongly phosphorylated ontyrosine residues in lysates from cells stimulated with 100 nM hPRL ( h P R L + ) for 3 min a t 37 "C, hut not from unstimulated cells (hPRL-). M, ( X lo-') of proteinmarkersareindicated. Arrows identify phosphotyrosyl proteins p120 and p66, and thehrackd marks immunoglobulin heavy chains (IgG,,.). nhPRL, anti-humanPRL antibodies; w P R L - R , anti-rat PRI, regulator antibodies.
-
+
-
+
-
- - + + "
+
+
+
hPRL
+
ATP
M, W a )
200-
-p120
-
"
-
complexes from detergent-solubilized cell lysates clarified by centrifugation at 12,000 x g for 30 min. However, identical ,,rPRL-R Igc, c l r l results were observed when activated receptor complexes were FIG.5. Anti-phosphotyrosine immunoblot from PRI, recepimmunoprecipitated from lysates centrifuged at 105,000 X g for 1 h, ensuring that thePRL receptor complexes were fully tor immune complex tyrosine kinase assay. P R L receptor comsolubilized. As can he further seen in Fig. 4 and other exper- plexes from cells stimulated with (h/'ff/,+) o r without ( h f ' R L - ) 10 nM hPIiL for 30 min at 22 "(1 were immohilized onprotein A iments (e.g. Fig. 5, upper panel),a second proteinof M, 66,000 Sepharose beads using either polyclonal anti-hl'HI. antiserum ( ( 1 . in the receptor complex was weakly phosphorylated on tyroh P R L as)or normal rahhit serum (nrs; upprr p n r l ) . o r monnclonal sine after PRL stimulation. This protein has the described anti-rat-PRLreceptorantibodies (Ufi) or isoty-pe-specific control molecular mass of the Nb2 cell PRL receptor (26, 4 1 , 4 2 ) and antibodies (lower p o n d ) . followed by incuhation in oitro with 15 p~ ATP in the presence of 3 mM Mn" for 30 min at 2'2 'C. The rniddlr is repeatedlyphosphorylatedontyrosinetosomeextent, especially after prolonged PRL stimulation (e.g. Fig. 5 , upper panel indicates whether cells corresponding to the individual lanes had been stimulated with PHIA or not and whether the same immupanel). Further studies are, however, needed to enable us to noprecipitates had been incuhated in the presence or absence of ATP. make firm conclusions regarding the phosphorylation state of M, ( X lo-:') ofeach protein marker is indicated. Arron*sindicate pl20 the activated PRL receptor. and p66. whereas the hrockpts mark immunoglobulin heavv chains P R L Receptor-associated Tyrosine Kinase Actiuity-To ad- (/gC,,c). nrPRL-R. anti-rat PRI, rewlator antihodies. dress the question of whether a tyrosine kinase activity resides rapid andtimewithin the cell-free PRL receptor complex, studies were per- exposedto[y-"P]ATP,demonstratinga formedusinganimmunecomplextyrosinekinaseassay. dependent incorporation of phosphate into ~ 1 2 0Similar . inLysates from cells treated with or without hPRL were incu- corporation of radiolabeledphosphateinto p120 wasalso bated with anti-hPRL antiserum or normal rabbit serum (Fig. observed in receptor complexes immunoprecipitated using the 5 , upper panel), or anti-rat-PRL receptor antibody (U6) or monoclonalanti-ratPRI,receptorantibody (Uti; data not isotype-specific control antibody (Fig. 5 , lower panel), and shown). Altogether, these results clearly imply that a tyrosine immunoprecipitated.Incubation of purified PRLreceptor kinase activity is associated with activated PRL receptors in complexes immobilized on protein A-Sepharose heads with Nh2 cells. A T P in the presence of a tyrosine phosphatase inhibitor for In light of the recent demonstration that the predominant 30 min at room temperature resulted in a significant increase type of PRL receptor in Nh2-11C cells is a deletion mutant inphosphotyrosinecontent of p120,hutonlyinreceptor of thenormallyoccurringlongreceptorform,lacking198 complexes from PRL-stimulated cells. The magnitudeof this amino acids in the cytoplasmic domain (26). experiments are increase above the levels seen in immunoprecipitates that had as currently in progress to demonstrate similar involvement of been incubated in the absence of ATP was 10-15-fold, PRI, receptor-mediatedsignal estimated by densitometry of anti-phosphotyrosine immuno- tyrosinephosphorylationin blots. Immunoprecipitates from unstimulated cells showed no transduction in other target cells. Since it is improbable that the deletion has provided the Nh2 cell PRI, receptor with a tyrosine phosphorylation of p120, and normal rabbit serum or IgG, controls demonstrated the specificity of the observa- unique fashion of signal transduction, the parental long form of inducing tions. Furthermore, a similar degree of in vitro phosphoryla- of the PRL receptor probably also has the capacity protein tyrosine phosphorylation upon activation. In support tion of p120 was also observed in parallel experiments with of a representative of such a notion is our recent observation that the Nb2 cell [y:'2P]ATP.Fig. 6 showstheresults experiment in which activated PRL receptor complexes were PRL receptor is fully functional and has, like the naturally affinity-purified with polyclonal anti-hPRL antibodies and occurringlongreceptorform,theabilitytoinitiate milk "
"
Receptor-associated Kinase Prolactin Tyrosine
24080 .."
~I
-
..
.
.
200"p120
9766 -
"p66
43-
hPRL
+
-
"
32P-ATP
2'
~
+ "
20'
+
~
"
2'
-
+
"
20'
Frc. 6. Autoradiography showing time-dependent phosphorylation of "P from labeled ATP into proteins in affinitypurified PRL receptor complexes,using tyrosine kinase assay conditions. Activated PRL receptor complexes from cells stimulated with (hPHL+)or without (hPHL-) 10 nM hPRL for 30 min a t 22 "C were immobilized on proteinA-Sepharose beads using polyclonal anti-hPRL antiserum (first four lanes) or normal rabbit serum (lost four lanes), followed by in vitro incubation with 1 PM [-y-32P]ATP in the presence of 3 mM Mn2+and 2 mM sodium orthovanadate at 37 "C for 2 and 20 min. After washing, proteins were resolved on SDSPAGE and subjected to autoradiography. M,( X IO-') of each protein marker is indicated.Arrows indicate p120 and p66. Rapid phosphorylation of p120 is seen in activated PRL receptor complexes within 2 min, reaching maximum at 20 min. The slight degree of PRLindependent phosphorylationof a p125 species seen after prolonged incubation in control immunoprecipitates probably reflects nonspecific binding of a highly abundant serine/threonine kinase, since this species is not observed using antiphosphotyrosine immunoblotting.
membrane fractions,3 and thesize of p120 is consistent with that of transmembrane, accessory receptor subunits ('ID-subunits") recently reported for several of thehematopoietin receptors, which fall within the size range of 120,000-140,000. These mayplay important roles in modulation of receptor affinity and mediation of signal transduction and include 8subunits (AIC2A, AIC2R, and KH70) of receptors for granulocyte-macrophagecolony-stimulatingfactor, IL3, and IL5 (47), as well as the gp130 of the IL6 receptor complex (48). Furthermore, a recent report described rapid ligand-induced tyrosine phosphorylationof a proteinof M, 121,000 that could be immunoprecipitated with antibodies to growth hormone p120 is involved receptor (49) and may indicate that the same in both PRL andgrowth hormone receptor signaling. Further work to characterize and identify p120 and to define therole for tyrosine kinase activation in PRL signal transduction is currently in progress. Acknowledgments-We acknowledge the support and critical review of the manuscript hy Dr. J . Oppenheim and Dr. D. Longo. REFERENCES I. Nicoll. C. S. (1980) FPd. Proc. 39,2563-25M 2. Russell, D. H. (19R9)Trends Phnrmocol. S r l . 10.40-44 3. Kelly. P. A, Djiane. J . , Postel-Vlnay. M.4'. & hlerv. M, ( 1 9 9 1 ) fihdrxrine Hru. 12. 23.5-251 4. Razan, d. F. (19R9) Riochem. Riophys. Res. Commun. 164. 7RR795 5. Cosman, I)., Lyman, S. D., Idzerdn. R. I,.. Heckman. M. I>.. Park. 1,. S.. Goodwin. R. G. & March, C . .I. (1990) Trends l h c h p m Sci. 16.265-270 6. Thoreau. E.. Petridou. H.. Kellv. 1'. A,.. Ihane. , . .I, & Mornon. .I. I' ( 1 9 9 1 ) I.'ms'i,Gt. 282,zfi-:ii .. 7. Raznn. .I. F. ( I W O ) I'roc. Notl. Acad. Sci. 87,6934-6998 8. Miyazaki. T., Mrmyama. M.. Yarnada. G . . Hntakeyarna. M. & Taniguchi. 'r. (1991) m l j o J . lo. : m - : % 1 9 ; 9. Leune. I). W.. Swncer. S. A,. Cachlaneaea. G.. Harnrnonds. C.. Collins.. C... He'nzel. W.'J.: Harnard. H.'.Waters. 51. ;J. & &'nod, W . I. (lb87) NofurP 330.537-54 l IO. Hatekeyama. M.. Mori, H., Doi. T. & Tanipuchi. T. (1989) Cell 59, Mi-
protein synthesis in a gene transcription assay.' This implies that the missing segment of 198 amino acids in the cyto84 5 N.,Yonehara. S.. Schreurn. J., Gorman. D. M.. Mantyama. K.. Iahii. plasmic domain is not essentialfor signal transduction,which 11. Itoh. A,. Yahara. I.. Arai. ti. & Miyajima. A. (19%)) Science 247,324-327 is in accordance with studies demonstrating that similar car- 12. Mosley, H..Reckmann. M. l'., Mnrch. C. d . . Idwrda. I
