PI1 SOOZ4.320S(98)00030-7
E.LSEVlER
Life Sciences. Vol. 62, No. 12, pp. 1069-1079, 1998 Copyright0 1998 Eke&x Science Inc. Printed in the USA. All rights reserved 0024-3205/98 $19.00 + .W
A MEMBRANE PROTEIN ASSOCIATED WITH THE PROLACTIN RECEPTOR. STUDIES WITH A PHOTOACTIVATABLE HUMAN GROWTH HORMONE DERIVATIVE N&or
T. H. Masckauchti,
Josh: M. Delfino* and Horatio N. Femindezf
IQUIFIB, Institute de Quimica y Fisicoquimica Biolbgicas, Universidad de Buenos AiresCONICET, Buenos Aires, Argentina.
(Received in final form December 31, 1997) Summary
Prolactin receptor from rat liver (PRL-R, 42 kDa) was cross-linked to a radiolabeled azidophenacyl derivative of human growth hormone ([‘251]AP-hGH) to yield a 63 kDa adduct. In addition, a protein of Mr SO-52 K was detected as a 73 kDa complex. Microsomes incubated with either (a) increasing amounts of [‘251]AP-hGH, or (b) a fixed amount of photoprobe and increasing concentrations of unlabeled hGH, showed that the 73/63 kDa band intensity ratio remains constant (0.71-0.77). Once transferred onto nitrocellulose membranes, only the 42 kDa protein is able to bind [‘251]AP-hGH or [‘?]hGH. Two anti-PRL-R monoclonal antibodies fail to cross-react with proteins of Mr 50-52 K. In membranes solubiliied with 3-[(3-cholamidopropyl)-dimethylammonio]-lpropanesulfonate (CHAPS), a significantly lower amount of the 73 kDa complex is detected. Thus, the SO-52 kDa protein appears to be structurally unrelated to, but is presumably associated with the PRL-R. The 73 kDa complex is also detected under low membrane fluidity conditions (l”C), indicating that PRL-R associates to this SO-52 kDa protein prior to hormone binding. PerfUsion of rat liver with [‘251]AP-hGH shows that this associated protein accompanies the receptor along its intracellular pathway. Key Words: prolactin receptor, associated protein, photoactivatable, cross-linking, reagent
Rat liver PRL-R, a member of a superfamily that also includes the growth hormone and interleukin receptors, consists of a single polypeptide chain of Mr 40-43 K containing oligosaccharide chains. The 291-amino acid sequence, derived from the cDNA (I), predicts a protein with a 24-residue putative transmembrane region and a 57-residue intracelfular portion. Of the three known forms of PRL-R, namely, a long receptor form (59 1 aminoacids), one found * Author to whom correspondence should be addressed: IQUIFIB (UBA-CONICET), Facultad de Farmacia y Bioquimica, Universidad de Buenos Aires, Junin 956, 1113 Buenos Aires, Argentina. Fax: (54-l) 962-5457. E-mail:
[email protected]. 3 Deceased on August 2, 1995.
1070
Prolactin Receptor Associated Protein
on Nb2 cell line (393 aminoacids) most abundant in rat liver tissue. as a PRL-R associated protein in lying beyond the hormone binding
and that Although lymphoid step has
Vol. 62, No. 12, 1998
mentioned above (short form), the latter is by far the the intracellular tyrosine kinase JAK2 was identified cells (2), at present, the signal transduction pathway still to be elucidated.
In our laboratory, a radiolabeled photoactivatable derivative of hGH, namely [L251]APhGH (Mr 21 K), was chemically characterized and its binding and cross-linking to somatogenic and lactogenic sites was assayed (3). This derivative has two photoreactive azidophenacyl groups (AP) attached to residues Cys 182 and Cys 189 of hGH, which were originally linked through a disulfide bridge lying very close to the C-terminus of this protein. Probes located in this area are potentially capable of reacting with neighboring molecules, as can be predicted from the structure of the hGH complexed with the hGH-R extracellular domains (4). In the present work we explore the use of [*2SI]AP-hGH to investigate the association of the hormone-PRL-R complex with neighboring proteins with the aim of elucidating its supramolecular arrangement. Materials and Methods Animals and materials. Normal adult Sprague-Dawley rats (2-3 months of age) of 150-200 g were used. Prolactin is dif%cult to handle because it easily aggregates, hampering chemical derivatization efforts. Therefore hGH, a hormone which interacts specifically with lactogenic receptors was chosen, since it does not present this problem and conditions for its derivatisation are better known. hGH was obtained from Kabi Vitrum Laboratories, Sweden. The derivative AP-hGH was prepared as previously described (3) and radiolabeled using limiting amounts of chloramine-T (5). All chemicals were reagent grade and were obtained from Sigma Chemical Co., St. Louis, MO, USA. Na[“‘I] (15-l 7 Ci/mg) was purchased from New England Nuclear, Boston, MA, USA. Monoclonal antibodies (MAbs) T6 and US raised against PRL-R were obtained from Affinity Bioreagents Inc., Golden, CO, USA. Membrane pur$cation procedures. Plasma membrane fraction was prepared according to Ray (6) and resuspended in an adequate volume of 1 mM NaHC03, 0.5 mM CaC12, pH 7.5 buffer. Purity of this fraction was assessed by its S’nucleotidase specific activity as described elsewhere (7). Golgi and endosomal fractions were obtained from rat liver according to Evans (8) with minor modifications and microsomes were prepared as described by Bonifacino et al. (9). All samples were kept at -20°C until use and their protein content was evaluated (10) after boiling them for 20 min in the presence of 1 M NaOH. PIasma membrane solubilization. Purified plasma membrane fraction was solubilized in 12 mM CHAPS, 25 mM Tris-HC1 and IO n&l MgC12, pH 7.4, at a final protein concentration of 5-7 mgld by constant stirring for 30 min at room temperature. The preparation was then centritiged at 100000 g for 1 h. The cleared supematant containing solubilized proteins was separated and its protein content was estimated. Binding with [1251]AP-hGH was performed immediately a&wards. Binding assays. Membrane samples (0.6-I-2 mg protein) were incubated in flat bottommed plastic tubes in the presence of 1.5-2.0 ng of [‘Z51]AP-hGH (6-lOxlO cpm) in a total volume of 0.5 ml with or without a large excess of unlabeled hGH, as indicated in figure legends. Samples were run in triplicate. Protein content and radioactive derivative concentration was identical in all control and test samples. The incubation medium was 25 mM Tris-HCl, 10 mM MgCI2, 10 niM CaC12, 0.25 mg/ml BSA (bovine serum albumin fraction V), pH 7.4. Samples were incubated in
Vol. 62, No. 12,1998
Prolactin Receptor Associated Protein
1071
the dark for 15-16 h at 25°C. Binding was stopped by addition of 0.5 ml cold buffer, the suspension was then irradiated at 254 nm for 6 min with gentle shaking and finally centritiged for 5 min at 2500 g. The resulting pellet was resuspended in 0.5-I ml of buffer and kept at -20°C until the SDS-PAGE separation. tie1 electrophoresis, autoradiography and densitometry. Samples (0.10-0.15 mg of protein) were boiled at 100°C for 3 min in the presence of 1.5% (w/v) SDS, 5 % glycerol (v/v), 25 mM EDTA, 5 % (v/v) 2-mercaptoethanol and 35 mM Tris-HCl buffer, pH 7.3, and separated by SDS-PAGE on 6.5-12% T (1.5% C, 0.2% (w/v) SDS) gradient gels (11). Gels were then stained with Amido Black, destained and dried. Radioactive bands were visualized by exposure of gels on Kodak X-Omat XAR-5 films at -70°C for 5-20 days with the aid of intensifiring screens. Radioactivity associated with the bands was quantified by densitometric analysis of the autoradiograms. Exposure times of each lane in the autoradiograms were appropriately adjusted so as to fall within the linear dynamic range of the film. Gel lanes were scanned downwards and band intensities were calculated by integration of peak areas using a Shimadzu CS-930 densitometer. Efectroblot analysis. After SDS-PAGE (10% T, 1.5% C, 0.1% (w/v) SDS), rat liver microsomal proteins were blotted onto nitrocellulose membranes for 2 h at 100 V in 25 mM Tris, 193 mM glycine buffer added with 20% (v/v) methanol. Lanes with transferred proteins were cut and immediately blocked with PBS, 0.1% (v/v) Tween-20 and 4% (w/v) dry fat-free milk, pH 7.3 for 2-3 h. Nitrocellulose membranes were then rinsed 4 times for 10 min with 0.10/o (v/v) Tween-20 in PBS and incubated with each of the following: T6 or U5 mouse IgG anti-PRL-R MAbs (4 &ml) (12), [‘*‘I]AP-hGH or [1251]hGH (0.5 &ml) in PBS, 1% (w/v) BSA, 0.1% (v/v) Tween20, pH 7.3 overnight at room temperature. After this step, membranes were rinsed 4 times for 15 min with PBS and 0.1% (v/v) Tween-20, pH 7.3. Membranes incubated with radioiodinated ligands were subjected to autoradiography, while those treated with MAbs were immediately incubated with a donkey anti-immunoglobulin G antibody (l/5000 dilution in PBS, 2% (w/v) BSA and 0.1% (v/v) Tween-20, pH 7.3) for 1 h. Finally, these membranes were rinsed in PBS and revealed by an enzymic chemiluminescent assay (ECL, Amersham Intl., UK). Perfusion experiments. Per&ion of rat liver was carried out immediately after removal of the whole organ. Buffer (Hanks’ balanced salt solution enriched with 1 mM CaC&, 0.2 mM MgQ and 0.1 mg/ml BSA, pH 7.4) was bubbled with carbogen and maintained at 20°C. [“‘I]AP-hGH (final concentration of 5~10~~ M) added to this buffer was perfused through the portal vein at a constant flow rate of 27 ml/min using a peristaltic pump. Then, organs were weighed, rinsed with isotonic saline solution and homogeneized in one volume of storage medium (0.21 M mannitol, 0.07 M sucrose and 20% (w/v) dimethyl sulfoxide). Each homogenate was then irradiated as described above and 1 mM phenyl-methyl-sulfonyl fluoride (PMSF) was added immediately afterwards. In order to estimate the total uptake of the derivative, the radioactivity of the homogenate and the perfusion buffer was measured. This homogenate was the starting material for the purification of plasma membrane, endosome and Golgi fractions. Non-specific uptake of [‘251]AP-hGH was assessed in a parallel experiment (18 min) where a 150-fold molar excess of unlabeled hGH was added to the perfUsion buffer. Results and Discussion Photoderivative [“‘I]AP-hGH (Mr 21 K) has been used to study the supramolecular structure of PRL-R. In rat liver membranes, proteins other than the known PRL-R (Mr 42 K) were detected. In addition to the 63 kDa adduct (hormone-receptor complex), specific covalent complexes of
1072
Prolactin Receptor Associated Protein
Vol. 62, No. 12, 1998
11.01
I
(I. I
Concentration of [‘251]AP-hGH(nM)
/ ,
0
2
Concentration
1
6
’
! 8
of unlabeled hGH (nM)
Fig.
Covalent cross-linking of [‘2SI]AP-hGH to rat liver microsomal proteins as analyzed by SDS-PAGE and autoradiography. (A) Microsomal membranes (0.9 mg of protein) were incubated in the presence of I .7x10-“, 3.8xlO~“, 7.6x1@“, 1.1x 1O-‘“, 1.7x 1O-” and 6.7x 10“’ M [ rz51]AP-hGH as described in Materials and Methods (lanes a-f, respectively) Bo values ([rZ51]AP-hGH bound) ranged between 15.8% and 30.3%. (B) 73/63 kDa band intensity ratio derived from (A) plotted against [‘2SI]AP-hGH concentration. (C) Microsomal membranes (0.9 mg of protein) were incubated in the presence of 1.7x1@” M [‘251]AlJ-hGH, as described before. and 0, 1.7~10-~, 3.4~10~~~6.8~10~~~ 1.36x10-* and 8.5x10-* M unlabeled hGH (lanes a-f, respectively). Bo values ranged between 12.3% and 29.9%. (D) 73/63 kDa band intensity ratio derived from (C) plotted against unlabeled hGH concentration. Experimental points were fitted to a straight line by regression analysis. Bars on the letI indicate the positions of the 73 and 63 kDa bands
Prolactin Receptor Associated Protein
Vol. 62, No. 12,199S
1073
130,73,45 and 35 kDa were visualized a&x analysis of the cross-linked samples by SDS-PAGE and autoradiography (3). Whereas the 130 kDa band probably represents the complex formed by the derivative with the somatogenic receptor, the other bands disappeared in the presence of an excess of prolactin, therefore showing lactogenic nature (3). The 73 and 63 kDa bands were the major products of cross-linking. The detection of the less predominant adducts (130,45 and 35 kDa bands) requires longer exposure times. In this work, we address the issue of the relationship between a protein of 50-52 kDa, integrating the 73 kDa complex, and the PRL-R present in rat liver membranes. When rat liver membrane samples were incubated in the presence of [‘2SI]AP-hGH (Fig. 1 A), the 73163 kDa band intensity ratio remained constant (0.77+0.03, Fig. 1 B) over a 40-fold increase in the concentration of the photoprobe. In addition, when membranes were incubated with a fixed amount of photoprobe and different concentrations of hGH, both 73 and 63 kDa bands were attenuated to the same extent (Fig. 1 C), i.e., the band intensity ratios also remained constant (0.71M.02, Fig. 1 D). Both bands disappeared when an go-fold excess of unlabeled hormone was added (lane e). Thus, both 63 and 73 kDa complexes exhibited identical behavior with regard to their interaction with the ligand. Two anti-Pm-R MAbs exhibiting different reactivity profiles were used to screen for structurally related proteins. h4Ab T6 is known to react with the ligand binding site of rat PRLRs and is capable of inhibiting the interaction of lactogenic hormones with PRL-R in vitro. However, this MAb is fairly restricted in its reactivity, since it does not cross-react with PRL-R from other species or with GH receptor. On the other hand, MAb U5 identifies several mammalian PRL-Rs (12). No microsomal protein in the range 45-68 kDa reacted with either of these two anti-Pm-R MAbs (Fig. 2 A). In agreement with these results, complete absence of other rat liver PRL-R subunits in this Mr range was also reported by others, using several specific ligands, namely, anti-PRL-R MAbs (12,13,14) or [“‘I]hGH (13). After electroblot analysis with [‘251]AP-hGH, we also observed that the only species capable of binding the ligand specitically was the known PRL-R (Fig. 2 B, 1 and 2). The same results were obtained when [‘251]hGH was used instead (Fig. 2 B, 3 and 4). Solubiiition of plasma membrane samples with 12 mM CHAPS prior to incubation with [‘251]AP-hGH resulted in a significant decrease in the 73/63 kDa band intensity ratio (0.11, Fig. 3, lane a), as compared with non-solubilized membranes (0.30, lane c). These bands disappeared when the corresponding samples were incubated in the presence of a large excess of hGH, implying that both complexes are specific (Fig. 3, lanes b and d). This change in the ratio support the idea that a 50-52 kDa protein, which integrates the 73 kDa complex, might not be a hormone receptor, but instead a component loosely associated with the PRL-R in intact membranes. Our results agree well with a previous observation on the disappearance of this band, when crosslinking experiments were performed on microsomes previously solubilized with T&on X-100 (3). The ratio measured on plasma membrane fraction is characteristically lower than the value observed in microsomes. One can speculate that this difference could reflect a change in the relative abundance of the 50-52 kDa protein with respect to the PRL-R. Alternatively, a change in stoichiometty and/or relative geometry between the partner proteins occuring upon internalization could account for this difference. It is worth
noting
that
other
cross-linking
reagents
such as I-ethyl-3-(3-dimethylamino-
1074
Prolactin
Receptor
Associated
Protein
a
55
Ida
Vol. 62, No. 12, 199X
b
32
C
55
k.Da
32
55
kDa
32
Fig. 2 Electroblot analysis of rat liver microsomal proteins incubated with MAbs, [‘251]AP-hGH or [‘*?]hGH. (A) Nitrocellulose membranes containing proteins transferred from an SDS-PAGE of microsomes were incubated with anti-PRL-R MAbs (4 pg/ml) as described in Materials and Methods. An autoradiogram after ECL is shown: MAb US (lane a), MAb T6 (lane b) and MAb T6 in the presence of a IOO-fold molar excess of hGH (lane c). The bar indicates the 42 kDa band. (B) Densitograms corresponding to autoradiograms of similar samples incubated with 0.5 ng/ml ]‘251]AP-hGH (1 and 2) or [i*‘I]hGH (3 and 4) in the absence (I and 3) or in the presence (2 and 4) of a lOOOO-fold molar excess of unlabeled hGH. The arrowhead indicates the position of the 42 kDa peak.
Vol. 62, No. 12, 1998
Prolactin Receptor Associated Protein
a
b
c
1075
d
Fig. 3
Cross-linking products of [“‘I]AP-hGH to proteins solubiliied from rat liver plasma membranes. CHAPS-solubii membranes (0.75 mg of protein) obtained as described in Materials and Methods was incubated with 2x10-i” M [‘251]APhGH in the absence (lane a) or in the presence (lane b) of 600-fold molar excess of unlabeled hGH. Control samples (non-solubiliied membranes) were incubated under the same conditions in the absence (lane c) or in the presence (lane d) of unlabeled hGH. Samples were then irradiated and analyzed by SDS-PAGE and autoradiography. Values of Bo and Bi ([“‘I]AP-hGH bound in the absence and in the presence of unlabeled hGH, respectively) were 30.2% and 20.9% for solubiliied membranes, and 17.5% and 7.4% for control samples. Bars on the lefi indicate the positions of the 73 and 63 kDa bands. propyl)-carbodiimide hydrochloride (EDC) and dimethyl suberimidate hydrochloride @MS) (15) disuccinimidyl suberate (DSS) (16) or N-hydroxysuccinimidyl azidobenzoate (NHSAB) (17) revealed only the 63 kDa complex. A common feature of all these reagents is their e&ctive short span (which does not exceed 12 A). In contrast to this, the AP group in [‘25X]AP-hGH is located near the C-terminus of hGH, a region which is surlke-exposed on the hormone-receptor complex (4) and which can presumably undergo large excursions after the 182-189 disulfide bridge is cleaved and modiied with p-azidophenacyl bromide. As estimated by molecular modeling, the photoreactive head can reach out between 21 and 45 A, making possible the contact with neighboring proteins surrounding the hormone-receptor complex (not shown). In conclusion, our experimental and theoretical evidence support the hypothesis that the 50-52 kDa protein, participating in the 73 kDa complex, could be a PRL-Rap (PRL-R-associated protein). To investigate whether the association of the 50-52 kDa protein with PRL-R exists prior to ligand binding and to diminish membrane fluidity, a low temperature condition was chosen for incubation. Cross-linking experiments on membranes incubated with [‘251J4P-hGH at l°C, reveal& both the 63 and 73 kDa bands (Fig. 4, lane a). The fact that the 73 kDa complex was detected indicates a preassociation of PRL-R with the 50-52 kDa protein. However, the 73163 kDa band intensity ratio at 1°C (0.23*0.04) was signilicantly lower (vO.05, n=4) than the value at room temperature (0.3 lsO.08) by an average of 24&S%. This observation could imply that the
1076
Prolactin Receptor Associated Protein
Vol. 62, No. 12, 1998
:
.“-
a Fig. 4 Cross-linking products of [“?]AP-hGH to proteins of rat liver plasma membranes. under conditions of diminished membrane fluidity. Four identical experiments were carried out, each using a different membrane preparation. An autoradiogram after SDS-PAGE of. one typical experiment is shown. Membrane samples (0.75 mg of protein) were incubated as described in Materials and Methods in the presence of 1.9x 1O+” M [‘251]AP-hGH at 1°C (lanes a and b) or 20°C (lanes c and d). A 500-fold molar excess of unlabeled hGH was present in samples shown in lanes b and d. The ranges for Bo were 40.5-50.8% (1°C) and 47.1-62.7% (2O“C) and for Bi were 5.1-22.0% (1°C) and 15.6-23.4% (20°C). Bars on the left indicate the positions of the 73 and 63 kDa bands. preassociated available.
population
of PRL-Rs amounts
to about three quarters
of the total PRL-Rs
PerfUsion of rat liver with [“$4P-hGH yielded values of 22.5% and 3.4% for total and nonspecific uptakes, respectively. The kinetics of the photoderivative clearance from the petisate (Fig. 5 A) coincides with published data for native hGH (1 S), indicating a similar in vivo uptake by the liver. When [‘*‘IlAP-hGH was perfused for 2, 10 and 18 min, the radioactivity distribution in the subcellular fractions isolated from perfused livers (Fig. 5 B) agreed well with reported data for the internalization of both hGH (19) and PRL (8). The above results validate the use of [‘*‘I]AP-hGH in vivo. The presence of both 63 and 73 kDa in plasma membrane, endosome and Golgi fractions (Fig. 5 C, lanes a-c) indicates a similar fate for both proteins along their intracellular transit. These complexes were also detected when plasma membrane and Golgi fractions isolated from non-perfused rat livers were incubated in vitro with [‘Z51]AP-hGH (Fig. 5 D, lanes a-d), indicating that the 50-52 kDa protein is a component already present in subcellular fractions purified from untreated livers. Independently, this protein was also detected in other tissues, such as Leydig cells and microsomes from mammary gland tumors obtained from rats (not shown). A number of proteins appear to participate in the molecular mechanisms of signal transduction by the PRL-R. Among those, several different PI&Raps have recently been reported: a Mr 48 K membrane protein from rabbit mammary gland (20); and a Mr 52 K protein (21), tyrosine kinase p59@“’(22), JAK2 (23) and serine/threonine kinase c-RAF-l (24) from the pre-T mediated
Vol. 62, No. 12, 1998
0
Prolactin Receptor Associated Protein
+.---
0
7-i_--_
3
-
-~--
6
9
12
15
PMEG
18
PMEG
PMEG
2 10 18 Time of perfusion (min)
Time of perfusion (min)
(0
1077
(D)
a
b
C
a
b
c
d
Fig. 5 Perfusion of rat liver with [‘2’I]AP-hGH. (A) Kinetics of photoderivative uptake. [rz51]AP-hGH (510e9 M) was pertused through the portal vein and fractions of the perfusate (40 pl) were collected every 45 s for 18 min. Radioactivity detected in the @sate after initial equilibration of the system upon injection of the photoderivative into liver (time 0 s) was considered 100%. An exponential curve was fitted to data by non-linear regression. (B) Distribution of radioactivity in subcellular fractions purified from rat liver after perfusion of the photoderivative through the portal vein for 2, 10 and 18 min. Organs were homogenized, Wirradiated, subcellular fractions were isolated and radioactivity was measured. (C) Autoradiograms of SDS-PAGE of cross-linked products of [‘251]AP-hGH with proteins present in plasma membrane (lane a), endosomes (lane b) and Golgi (lane c) fractions purified from liver perfused for 18 min with the photoderivative. Results obtained after 2 and 10 mm were similar. (D) Autoradiograms of SDSPAGE of cross-linked products of [‘251]AP-hGH incubated with plasma membrane (lane a) and Golgi fractions (lane c). The photoderivative (1.9x 1O-l0 M) was incubated with rat liver tractions (0.75 mg of protein) in the absence (lanes a and c) or in the presence of a 250-fold molar excess of unlabeled hGH (lanes b and d). Values of Bo and Bi for plasma membrane fraction were 34.2% and 15.3%, respectively. The corresponding values for Golgi fraction were 87.8% and 25.6%. Bars on the left indicate the positions of the 73 and 63 kDa bands.
1078
Prolactin
Receptor
Associated
Protein
Vol. 62, No. 12, 1998
lymphoma cell line Nb2. Interestingly, tyrosine kinase ~59’~ seems to form a complex with PRLR prior to hormone binding (22). A preformed complex with an associated protein has also been reported in relation to the interleukin-2 receptor (25) another member of the same receptor superfamily as that comprising the PRL-Rs. JAK2 was also identified as associated to GH-R (26). Recently, a c-src-tyrosine kinase of Mr 60 K, associated to PRL-R, was identified in rat hepatocytes and proved to be activated after hormone binding (27). However, [iZJI]AP-hGH is not expected to react with proteins which do not contain at least a significant extracellular domain, therefore the chance that any of the above mentioned proteins could correspond to the PRL-Rap described herein becomes unlikely. The results presented in this paper support the existence of a 50-52 kDa membrane protein which interacts non-covalently with the rat liver PRL-R, but seems to be structurally unrelated to it. This association precedes the hormone binding event, and persists through the intracellular transit of the hormone-receptor complex. To pursue further studies on PRL-R structure, we are currently preparing a new hGH derivative endowed with a photoreactive and radioiodinatable headgroup, capable of transferring the radioactive tag to the target protein after chemical cleavage Acknowledgments This work was supported by grants from CONICET, Fundacion Antorchas, the European Union and University of Buenos Aires, Argentina. We are specially indebted to Drs. J.M. Dellacha, S. Longhi, K. Gomez and L. Retegui for their experimental contributions, and to Dr J.P.F.C. Rossi for his critical reading of the manuscript. References 1. J.M. BOUTIN, C. JOLICOEUR, H. OKAMURA, J. GAGNON, M. EDERY, M. SIBROTA, D. BANVILLE, I. DUSANTER-FOURT, J. DJIANE, P.A. KELLY, Cell 53 69-77 (1988). 2. I. DUSANTER-FOURT, 0. MULLER A. ZIEMIECKI, P. MAYEUX, B. DRUCKER, J. DJIANE, A. W&KS, A.G. HARPUR, S. FISCHER S. GISSELBRECHT, EMBO J. 13 2583-2591 (1994). 3. E.J. ROBETTO, C.A. C AAMANO, H.N. FERNANDEZ, J.M. DELLACHA, Biochim. Biophys. Acta 1013 223-230 (1989). 4. A.M. DE VOS, M. ULTSCH, A.A. KOSSIAKOFF, Protein Data Bank, Brookhaven Nat). Lab., NY, USA, Pub. 71 Code 3HHR (1995). 5. R. MATTERA, D. TURYN, H.N. FERNANDEZ, J.M. DELLACHA, Int. J. Peptide Prot. Res. 19 172-180 (1981). 6. T.K. RAY, Biochim. Biophys. Acta 196 l-9 (1970). 7. N.N. ARONSON, 0. TOUSTER, Methods Enzymol. 31 90-102 (1974). 8. W.H. EVANS, Methods Enzymol. 109 246-257 (1985). 9. J.S. BONIFACINO, S.H. SANCHEZ, A.C. PALADINI, Biochem. J. 194 385-394 (1981). 10. O.H. LOWRY, N.J. ROSEBROUGH, A.L. FARR, R.J. RANDALL, J. Biol. Chem. 193 265-275 (195 1). 1 I. U.K. LAEMMLI, Nature 227 680-685 ( 1970). 12. H. OKAMURA, J. ZACHWIEJA, S. RAGUET. P.A. KELLY, Endocrinology 124 24992508 (1989). 13. M. EMTNER, J. BRANDT, U. JOHANSSON. B. JOUPER, L. FRYKLUND, P. ROOS, J. Endocrinol. 120 40 I-407 ( 1989).
Vol. 62,No. 12,1998
14. M. KATOH, S. RAGUET,
Prolactin Receptor Associated Protein
1079
J. ZACHWIEJA, J. DJIANE, P.A. KELLY, Endocrinology 120 739-749 (1987). 15. CA. C-0, H.N. FERN&IEZ, A.C. PALADINl, Biochem. Biophys. Res. Comm. 115 29-37 (1983). 16. L.A. HALDOSEN, J.A. GUSTAFSSON, Mol. Cell. Endocrinol. 7141-48 (1990). 17. D.W. BORST, M. SAYARE, B.I. POSNER, Mol. Cell. Endocrinol. 39 125-130 (1985). 18. D. TURYN, J.M. DELLACHA, Endocrinology 103 1190-l 195 (1978). 19. M.C. POSTEL-VINAY, C. KAYSER, B. DESBUQUOIS, Endocrinology 111244-251 (1982). 20. M.J. WATERS, N. DANIEL, C. BIGNON, J. DJIANE, J. Biol. Chem. 270 5 136-5 143 (1995). 21. R.A. ERWIN, R.A. KIRKEN, M.G. MALABARBA, W.L. FARRAR, H. RUl, Endocrinology 136 3 5 12-35 18 ( 1995). 22. C.V. CLEVENGER, M.V. MEDAGLIA, Mol. Endocrinol. 8 674-681 (1994). 23. L. DASILVA, O.M. ZACK HOWARD, H. RUI, R.A. KIRKEN, W.L. FARRAR, J. Biol. Chem. 269 18267-18270 (1994). 24. C.V. CLEVENGER, T. TORIGOE, J.C. REED, J. Biol. Chem. 269 5559-5565 (1994). 25. T. TORIGOE, H. SARAGOVI, J. REED, Proc. Natl. Acad. Sci. USA 89 2674-2678 (1992). 26. L.S. ARGETSINGER, G.S. CAMPBELL, X. YANG, B.A. WITTHUHN, 0. SILVENNOINEN, J.N. IHLE, C. CARTER-SU, Cell 74 237-244 (1993). 27. J.J. BERLANGA, J.A.F. VARq J. MARTIN-PEREZ, J.P. GARCIA-RUIZ, Mol. Endocrinol. 9 1461-1467 (1995).