Growth hormone-induced signal tranduction depends on an intact ubiquitin system.

The growth hormone receptor (GHR) is a ubiquitinated cell surface protein. Ligand binding and receptor dimerization activate the cytosolic kinase Jak2. This event initiates signal transduction via STAT proteins. Expression of GHR in a Chinese hamster ovary (CHO) cell line, which exhibits a temperature-sensitive defect in ubiquitin conjugation (CHO-ts20), as well as in wild type cells (CHO-E36) has shown that endocytosis of the receptor requires an intact ubiquitin conjugation system (Strous G. J., van Kerkhof, P., Govers, R., Ciechanover A., and Schwartz, A. L. (1996) EMBO J. 15, 3806-3812). We have now examined the requirement for ubiquitin conjugation in growth factor-mediated signal transduction. In CHO-E36 and in CHO-ts20 cells at the permissive temperature, STAT proteins were activated in a growth factor-dependent fashion. However, no activation of STAT proteins was observed at the nonpermissive temperature in CHO-ts20 cells. Neither tyrosine phosphorylation of GHR nor of Jak2 was inhibited at the nonpermissive temperature. When tyrosine phosphorylation was inhibited following treatment with staurosporin, ubiquitination of the receptor proceeded normally. Furthermore, mutation of GHR phenylalanine-327, which prevents GHR endocytosis, inhibited receptor ubiquitination but allowed normal Jak/STAT-mediated signal transduction. Thus, these data provide evidence that the ubiquitin conjugation system is involved in the Jak/STAT signaling pathway, be it not at the initial stage(s) of Jak2 activity.

tides, which in turn initiates a cascade of events leading to signal transduction in the cell nucleus and down-regulation of the receptor (1,2). Lacking intrinsic tyrosine kinase activity, the GHR recruits and activates a member of the Janus family of cytosolic kinases (Jak2) upon dimerization (3). In addition to the GHR polypeptides and itself, Jak2 phosphorylates special signal transducers and activators of transcription proteins (STATs), which translocate to the nucleus and convey the appropriate signal to specific regulatory DNA-responsive elements (3)(4)(5)(6)(7)(8).
The ubiquitin system has been established as the major regulatory protein degradation system within eukaryotic cells (9 -12). Proteins destined for degradation are covalently modified by conjugation of multiple ubiquitin polypeptides following recognition and modification by members of the E2 (ubiquitin conjugating) and E3 (ligating) protein families. Recently, the ubiquitin conjugation system has been shown to play a role in other regulatory functions such as control of protein kinase activity (13) and receptor endocytosis (14,15). An intact ubiquitin conjugation system is required for prompt ligand-induced GHR endocytosis and degradation (15). Many other cell surface receptors are ubiquitinated upon ligand binding (16 -20), although their dependence upon ubiquitin conjugation for endocytosis has not been addressed. As signal transduction and down-regulation of most cell surface signaling receptors are tightly coupled, we have examined the role of the ubiquitin conjugation system in GHR signal transduction.
We have utilized GHR-transfected CHO-ts20 cells that express a thermolabile ubiquitin activating enzyme, E1 (21). At the nonpermissive temperature ligand-induced endocytosis of the GHR is blocked in these cells (15). Herein, we report that although GH-induced tyrosine phosphorylation of GHR and Jak2 at the cell surface is not affected at the nonpermissive temperature, the Jak2/STAT signaling pathway is completely inhibited.

MATERIALS AND METHODS
Cells and Materials-Cell lines derived from CHO-ts20 and CHO-E36 cells, stably transfected with rabbit GHR-cDNA, were used (15). The GHR-F327A mutant was prepared by inserting a polymerase chain reaction product using a 5Ј-oligonucleotide, containing a Pfl1 site (GATCCCACCCATTGGCCTCAACTGGACTTT) and a 3Ј-oligonucleotide containing a ClaI site and a TTC 3 GCC mutation (F 3 A) (GATCCATCGATGTCTAGCTCGATGGCTTCA). The DNA sequence was confirmed for the entire polymerase chain reaction fragment. Cells were grown in Eagle's minimal essential medium supplemented with 4.5 g/liter glucose, 10% fetal bovine serum, penicillin, and streptomycin, and 0.4 mg/ml geneticin. For experiments the cells were grown on 35-mm plates and used at 75% confluence. The CHO-ts20 clone used in this study expressed approximately 10-fold more receptor than the CHO-E36 clone. To compensate for this, 10 mM sodium butyrate was added to CHO-E36 cells 18 h before use, increasing GHR expression approximately 5-fold (15). As before, the sodium butyrate treatment did not alter the behavior of GHR in any of the parameters examined in this study. Anti-GHR antiserum was raised in rabbits against a fusion protein of glutathione S-transferase and a GHR tail peptide consisting of amino acids 327-493 (15). The fusion protein was expressed in Escherichia coli after transformation with the appropriate pGEX-2T DNA construct (Pharmacia Biotech Inc.). Anti-PY and rabbit antimouse STAT3 were from Upstate Biotechnology, Inc. (Lake Placid, NY), and anti-ISGF3 (monoclonal antibody against the N-terminal 194 amino acid sequence of STAT1) from Transduction Labs (Lexington, KY). Antiserum against Jak2 was raised in rabbits against a synthetic peptide corresponding to the hinge region between domains 1 and 2 of murine Jak2. Antiserum specific for ubiquitin-protein conjugates was a * This work was supported by grants from the Fulbright Program, Grants NWO-902-23-188 and NWO-902-68-244 from the Netherlands Organization for Scientific Research, and grants from the National Institutes of Health. 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. Cell Lysis, Immunoprecipitation, and Immunoblotting-The cells were lysed on ice in 0.3 ml of lysis mix containing 1% Triton X-100, 2 mM Na 3 VO 4 , 1 mM EDTA, 10 g/ml aprotinin, 10 g/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride in PBS. In ubiquitin blotting experiments the cells were lysed in 0.3 ml of boiling lysis buffer containing 1% SDS in PBS in order to minimize isopeptidase activity. The lysate was heated at 100°C for 5 min, sheared to break DNA, and centrifuged for 5 min at 10,000 ϫ g. Immunoprecipitations were carried out in 1% Triton X-100, 0.5% SDS, 0.25% sodium deoxycholate, 0.5% bovine serum albumin, 1 mM EDTA, 2 mM Na 3 VO 4 , 10 g/ml leupeptin, 10 g/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride in PBS. The reactions were incubated for 2 h at 0°C with 5 l of specific rabbit anti-GHR antiserum; protein A-agarose (Repligen Co., Cambridge, MA) was used to isolate the immune complexes. The immunoprecipitates were washed twice with the same buffer and once with 10-fold diluted PBS; immune complexes were analyzed by 7.5% polyacrylamide gel electrophoresis in the presence of SDS.
After transfer to polyvinylidene difluoride paper, the blots were immunostained using anti-ubiquitin conjugate antibody, anti-GHR (after stripping of the blots), and anti-PY (15). The antigens were visualized using the ECL system (Amersham Corp.).
Transcription Factor Complexation Assays-Either total cell lysates or nuclear extracts from 2 ϫ 10 8 cells were incubated with a 32 P-labeled oligonucleotide known as the c-sis-inducible element (SIE) of the c-fos gene promoter as described (22,23). To prepare radiolabeled SIE two oligonucleotides were synthesized (5Ј-GTCGACATTTCCCGTA-AATCGTCGA-3Ј and 5Ј-TCGACGATTTACGG-3Ј) and hybridized, and the gap was filled using the Klenow reaction in the presence of [␣-32 P]ATP and unlabeled dCTP, dGTP, and dTTP. Transcription factor complex formation and electrophoresis were performed as described (22).

RESULTS
In order to determine whether the ubiquitin conjugation system is involved in signal transduction of the GHR, CHO-ts20, and CHO-E36 cells stably transfected with GHR were incubated at the permissive (30°C) and nonpermissive (42°C) temperatures, and the cell and nuclear extracts were examined for GH-inducible band shifts using 32 P-labeled SIE. As seen in Fig. 1A, GH induced a clear signal in CHO-E36 cells at both the permissive and nonpermissive temperatures and in CHO-ts20 cells incubated at the permissive temperature. Closer examination of the band pattern (Fig. 1B) reveals that the specific signal is composed of three bands (a, b, and c). If excess unlabeled SIE was added, all three bands disappeared (Fig. 1B,  lanes 2 and 3); if antibodies against STAT1 were present the two lower bands disappeared and a new more slowly migrating complex became visible; anti-STAT3 had no effect, probably because our antibody did not interfere with STAT complex formation in CHO cells. These observations together with data from literature indicate that the three-band pattern represents homo-and heterodimers of STAT1 and STAT3 (5). In nuclear extracts from CHO-E36 cells the lower band (Fig. 1B, c) was lacking (Fig. 1A).
If CHO-ts20 cells were incubated at the nonpermissive temperature in the presence of GH, no STAT complexes were detectable. This observation was consistently made over a temperature range of 40 -42°C, and the STAT complex formation coincided fully with the capacity of the CHO-ts20 cells to activate ubiquitin. The weak signal at lower electrophoretic mobility in this lane did not disappear if the incubation was carried out in the presence of excess unlabeled probe. If CHO-ts20 cells were used transfected with a mutant DNA (GHR-F327A), the same results were obtained as with wild type GHR. No STAT complex was observed in these cells following incubation at the nonpermissive temperature in the presence of GH (Fig. 1C). Together, the data show that an intact ubiquitin conjugation system is necessary for GH-mediated Jak/STAT signaling.
The F327A mutation in the GHR was previously shown to affect ligand-induced endocytosis, not signal transduction (24).
As seen in Fig. 1C, CHO-ts20 cells transfected with GHR-F327A DNA show normal signaling via STAT pathway upon GH addition at the permissive temperature. The mutated GHR in these cells is not ubiquitinated upon GH addition at the permissive temperature. 2 Thus, although endocytosis and ubiquitination of the GHR were inhibited in the GHR-F327A mutant, Jak/STAT signal transduction occurred normally. This result also shows that ubiquitination of the receptor tail is not required for signal transduction.
Following dimerization of the GHR polypeptides, tyrosine phosphorylation of both GHR and Jak2 is the initial event in the signaling cascade. Therefore, we examined whether an intact ubiquitin conjugation system is essential for this event. In all cases a substantial phosphotyrosine signal was observed 2 R. Govers and G. J. Strous, unpublished observation.

Signal Transduction and Ubiquitin System 41
on both the mature GHR and Jak2 upon the addition of GH. In the case of Jak2 (Fig. 2, right panel) an extra diffuse and slower migrating PY-labeled band is visible, most probably originating from co-immunoprecipitation of GHR. A somewhat diminished PY signal was observed in both GHR and Jak2 in CHO-ts20 and in CHO-E36 cells following incubation at temperatures above 40°C. Occasionally the decrease in signal was more prominent in the CHO-ts20 than in CHO-E36 cells. This variability is likely due to temperature-dependent activity of Jak2. Thus, it is unlikely that the ubiquitin conjugating system is required for Jak2 recruitment nor for its action. We also examined the extent to which the GHR-F327A mutant was phosphorylated in response to ligand. Both receptor and Jak2 phosphorylation were similar to that observed with the wild type receptor (not shown). Because both GHR tyrosine phosphorylation and ubiquitination appear to occur at the cell surface, it is important to determine whether the two events are independent. Thus, to eliminate indirect effects of phosphorylation on receptor ubiquitination, we utilized the phosphokinase inhibitor staurosporin. As seen in Fig. 3, staurosporin inhibited the ligandinduced GHR phosphorylation almost completely. However, the GH stimulation of GHR ubiquitination was unaffected. Of note is that no PY signal is visible at the position of ubiquitinated GHR (Fig. 3, middle panel), indicating that ubiquitinated GHR is not phosphorylated. Taken together with the results described above, we conclude that ubiquitin conjugation and tyrosine phosphorylation of the GHR (and Jak2) are independent events. DISCUSSION The ubiquitin conjugating system plays a pivotal role in a diverse array of regulatory events including cell cycle progression, DNA repair, and transcriptional control. Many of these processes are dependent upon both ubiquitin conjugation to target proteins as well as the subsequent degradation of the protein-ubiquitin moieties via the proteasome. In the present study we demonstrate that the ubiquitin conjugating system is involved in signal transduction via the Jak/STAT pathway. This pathway mediates transduction of a wide variety of extracellular signals via receptors for interferon, many cytokines, and GH (3-5, 7, 8, 25). Using cells that display an intact GH-GHR pathway yet contain a temperature-sensitive muta-tion in the initial enzyme of the ubiquitin conjugation pathway, we have demonstrated that inactivation of the ubiquitin pathway inhibits GH-induced stimulation the Jak/STAT pathway. Our data do not define at which stage(s) ubiquitin conjugation is involved in the signaling cascade, although the data render it unlikely that the initial steps that likely occur at the cell surface (receptor dimerization, Jak2 recruitment, and Jak2 activation) are affected by the ubiquitin system. Previously, we have shown that the ubiquitin system is involved in receptor endocytosis (15). These results together with the current data suggest that the cell surface is the likely site of GHR signaling. Dependent upon the GH concentration, temperature, and cell type, ligand-induced endocytosis of the GHR occurs with a half-time of 30 -60 min (26 -28). Tyrosine phosphorylation, however, occurs maximally within 15-20 min (Ref. 29 and Fig.  2). Because no phosphotyrosine-labeled GHR was found to be ubiquitinated and because the two events appear to be independent, it is tempting to speculate that phosphorylation and subsequently Jak/STAT transduction occur first, following which the receptor is subject to endocytosis mediated by the ubiquitin conjugation system.
As proposed for the role of ubiquitination in ligand-induced endocytosis of the GHR, ubiquitination in addition to protein phosphorylation may act as a control point in regulating some step(s) in the Jak/STAT signaling cascade. However, because many basic cellular processes are regulated subsequent to ubiquitin conjugation, use of the E1 mutant cell line at the nonpermissive temperature does not exclude the possibility that other E1-dependent events are critical in the control of this signaling pathway.
Recently, Allevato and colleagues (24) have reported that phenylalanine 346 within the rat GHR cytoplasmic tail is critical for ligand-induced internalization but not essential for transcriptional control. We have made the analogous mutation in the rabbit receptor, which results in defective endocytosis as well as ubiquitination. 2 These results further support the notion that the ubiquitin conjugating system acts on the signaling pathway downstream from the tyrosine phosphorylating action of Jak2. Recently, Chen et al. (13) have shown that the protein kinase involved in the activation of the transcription factor NFB depends on the presence and function of several factors of the ubiquitin conjugation system. Thus, the ubiquitin system appears to be involved in a variety of growth control and signal transduction pathways. Future studies will focus on the mechanism(s) involved in modulation of the Jak/STAT pathway by the ubiquitin system. Acknowledgments-We thank Dennis Lebbing for excellent technical FIG. 2. Tyrosine phosphorylation of GHR and Jak2 at permissive and nonpermissive temperature. Effect of temperature. CHO-ts20 or CHO-E36 cells expressing GHR were cultured in serum-free medium overnight. Cells were then incubated for 60 min at the indicated temperature, followed by 20 min in the absence or the presence of 8 nM GH. Half of the cell extracts were used for immunoprecipitation using anti-GHR (left panel), and the other half were used for immunoprecipitation using anti-Jak2 (right panel); both blots were probed with anti-PY antiserum as described under "Materials and Methods." Reprobing of the left panel with anti-GHR and of the right panel with anti-Jak2 showed equal intensities of both antigens in each lane (not shown).
FIG. 3. Tyrosine phosphorylation of GHR/Jak2 is not required for GH-induced receptor ubiquitination. After overnight serumfree incubation, CHO-ts20 cells expressing GHR were kept at 30°C, preincubated with or without 1 M staurosporin for 15 min, incubated in the absence or the presence of 8 nM GH for 20 min, and then lysed in hot lysis buffer as described under "Materials and Methods." After immunoprecipitation half of the samples were used for the anti-ubiquitin blot, and the other half were used for the anti-PY blot. Reprobing of the PY-blot with anti-GHR showed equal intensity bands for GHR (left panel).
Signal Transduction and Ubiquitin System 42