The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hunter, M. G.
Right arrow Articles by Avalos, B. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hunter, M. G.
Right arrow Articles by Avalos, B. R.
Right arrowPubmed/NCBI databases
*Gene*GEO Profiles
*HomoloGene*Nucleotide
*Protein*UniGene
*Substance via MeSH
The Journal of Immunology, 2004, 173: 5036-5045.
Copyright © 2004 by The American Association of Immunologists

Loss of SHIP and CIS Recruitment to the Granulocyte Colony-Stimulating Factor Receptor Contribute to Hyperproliferative Responses in Severe Congenital Neutropenia/Acute Myelogenous Leukemia1

Melissa G. Hunter*, Anand Jacob{ddagger}, Lynn C. O’Donnell*, Amanda Agler{dagger}, Lawrence J. Druhan*, K. Mark Coggeshall§ and Belinda R. Avalos2,*

* Bone Marrow Transplantation Program, The Arthur G. James Cancer Hospital and Richard Solove Research Institute, {dagger} Department of Molecular Genetics, Ohio State University, Columbus, OH 43210; {ddagger} University of Pennsylvania School of Medicine, Philadelphia, PA 19104; and § Department of Immunobiology and Cancer, Oklahoma Medical Research Foundation, Oklahoma City, OK 73014


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Mutations in the G-CSF receptor (G-CSFR) in patients with severe congenital neutropenia (SCN) are postulated to contribute to transformation to acute myelogenous leukemia (AML). These mutations result in defective receptor internalization and sustained cellular activation, suggesting a loss of negative signaling by the G-CSFR. In this paper we investigated the roles of SHIP and cytokine-inducible Src homology 2 protein (CIS) in down-modulating G-CSFR signals and demonstrate that loss of their recruitment as a consequence of receptor mutations leads to aberrant signaling. We show that SHIP binds to phosphopeptides corresponding to Tyr744 and Tyr764 in the G-CSFR and that Tyr764 is required for in vivo phosphorylation of SHIP and the formation of SHIP/Shc complexes. Cells expressing a G-CSFR form lacking Tyr764 exhibited hypersensitivity to G-CSF and enhanced proliferation, but to a lesser degree than observed with the most common mutant G-CSFR form in patients with SCN/AML, prompting us to investigate whether suppressor of cytokine signaling proteins also down-modulate G-CSFR signals. G-CSF was found to induce the expression of CIS and of CIS bound to phosphopeptides corresponding to Tyr729 and Tyr744 of the G-CSFR. The expression of CIS was prolonged in cells with the SCN/AML mutant G-CSFR lacking Tyr729 and Tyr744, which also correlated with increased G-CSFR expression. These findings suggest that SHIP and CIS interact with distal phosphotyrosine residues in the G-CSFR to negatively regulate G-CSFR signaling by limiting proliferation and modulating surface expression of the G-CSFR, respectively. Novel therapeutic approaches targeting inhibitory pathways that limit G-CSFR signaling may have promise in the treatment of patients with SCN/AML.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Granulocyte CSF and its receptor (G-CSFR)3 critically regulate neutrophil numbers to maintain myeloid homeostasis (1). Ligand binding to the G-CSFR induces receptor dimerization and phosphorylation of the G-CSFR itself as well as multiple other signaling proteins that ultimately culminate in extension of cell survival and cellular proliferation (1, 2, 3, 4, 5, 6, 7, 8, 9). Activation of inhibitory proteins that limit G-CSFR signals must also occur to prevent sustained cellular proliferation and leukocytosis. In patients with severe congenital neutropenia (SCN) transforming to acute myelogenous leukemia (AML), acquired mutations in the G-CSFR have been detected (10). These mutations induce prolonged receptor surface expression, sustained cellular activation, and enhanced growth responses and are postulated to play a role in the development of AML (7, 11, 12, 13, 14).

Currently, little is known about the mechanisms that down-regulate mitogenic signaling by the G-CSFR. SHP-1 and the suppressors of cytokine signaling (SOCS) protein SOCS3 have recently been implicated in negative regulation of G-CSFR signaling (15, 16). SOCS proteins can inhibit cytokine signaling by directly binding to either Jak kinases or phosphotyrosine residues in the cytoplasmic domains of cytokine receptors to compete for binding with STAT molecules (17, 18, 19, 20, 21, 22, 23, 24, 25). Our laboratory previously reported that the distal tail of the G-CSFR mediates tyrosine phosphorylation of the Src homology 2-containing inositol 5'-phosphatase SHIP and the formation of SHIP/Shc protein complexes, the appearance of which correlated with a decrease in proliferative signaling (6). For several other immune and cytokine receptors, SHIP has been shown to negatively regulate proliferative signaling (26, 27, 28, 29). The 5'-phosphatase activity of SHIP specifically hydrolyzes inositol (1,3,4,5)-tetraphosphate and the PI3K metabolite phosphoinositol (3,4,5)-trisphosphate (PI(3,4,5)P3) at the 5' phosphate group. Hydrolysis of the latter results in decreased activation of the serine/threonine kinase Akt and induction of apoptosis (30, 31).

We have previously shown that G-CSF activates the PI3K/Akt pathway to promote cell survival (6, 7) and that activation of this pathway is prolonged in SCN/AML, conferring increased resistance to apoptosis. We also demonstrated prolonged surface expression of the G-CSFR in SCN/AML due to defective internalization of truncated G-CSFR forms (11). Because cells from patients with SCN/AML exhibit sustained activation of the PI3K/Akt pathway, we were interested in determining whether this resulted from a loss of recruitment of SHIP to the truncated G-CSFR.

In this study we have examined the role of SHIP as well as the SOCS family member, cytokine-inducible Src homology 2 protein (CIS), in limiting G-CSFR proliferative signaling. We show that both proteins bind to phosphotyrosine residues in the distal tail of the G-CSFR, which is deleted in patients with SCN/AML, to down-regulate proliferative signaling. Our data also suggest that CIS may function as a chaperone to route the G-CSFR intracellularly for targeted degradation.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Reagents

Reagents for the maintenance of cell lines were obtained from Invitrogen Life Technologies (Grand Island, NY). [Methyl-3H]thymidine was purchased from Amersham Biosciences (Piscataway, NJ). Abs to Shc and CIS were obtained from Upstate Biotechnology (Saranac, NY) and Santa Cruz Biotechnology (Santa Cruz, CA), respectively. The rabbit polyclonal Ab recognizing SHIP was generated from a GST fusion protein corresponding to residues 886–955 of p130 SHIP, as previously reported (32). The 4G10 anti-phosphotyrosine Ab was a gift from Dr. B. Druker (Oregon Health and Science University, Portland, OR). Expression vectors containing the cDNAs for mouse SOCS1, human SOCS3, and mouse CIS were gifts from Dr. A. Yoshimura (Kyusnu University, Maidashi, Fukuoka, Japan). Synthetic phosphopeptides corresponding to sequences surrounding each tyrosine residue in the cytoplasmic domain of the G-CSFR as well as a phosphopeptide specific for the phospho-ITIM region of Fc{gamma}RIIB were purchased from Quality Controlled Biochemicals (Hopkinton, MA). Unless indicated, all other reagents were obtained from Sigma-Aldrich (St. Louis, MO).

DNA constructs

Details for the construction of pCDM8-wild-type (WT)-G-CSFR and pCDM8-{Delta}716 have been previously described (6, 11). Tyrosine to phenylalanine (Y->F) mutations at amino acid positions 744 and 764 of the WT G-CSFR were introduced, and the corresponding G-CSFR forms were designated Y744F and Y764F, respectively. Y744F and Y764F were constructed by the introduction of an A to T point mutation at nt 2465 or 2525, respectively, in the WT-G-CSFR cDNA by overlap extension PCR (33). The oligonucleotides used to generate these mutants were forward primer Y744F F1 (5'-GGCCAGGGCACTTTCTCCGCTGTGAC-TCCACTC-3'), forward primer Y764F F2 (5'-GGCCAGGGCACTTTCTCCGCTGTGA-CTCCACTC-3'), reverse primer Y744F R1 (5'-GATGGGAGTCACAGCGGAGAAAGTGCCC-3'), and reverse primer Y764F R2 (5'-GGAACCAGAAATTCTCAAAGCTTTTGGGGCTGG-3'); the underlined nucleotides indicate the positions of the point mutations. To generate Y744F, primer F3 containing a 5' restriction site for BamHI and corresponding to nt 2252–2268 of the WT-G-CSFR cDNA was used in conjunction with the Y744F R1 to amplify WT-G-CSFR cDNA. In a separate PCR, WT-G-CSFR cDNA was amplified with Y744F F1 primer and primer R3 that was designed to contain an XhoI restriction site and corresponded to nt 2581–2596 of the WT-G-CSFR cDNA. The products from the two initial PCRs were combined and amplified by PCR using the external primers F3 and R3. The PCR product was digested with XhoI and BamHI and then subcloned into the XhoI and BamHI sites of pBluescript SK+. An internal 280-bp fragment in the cloned product was excised from pBluescript SK+ by Cfr10I and BstEII digestion, gel-purified, and subcloned into the pCDM8-WT plasmid, replacing WT sequences. This plasmid was designated pCDM8-Y744F.

To generate the Y764F mutant, primer pairs Y764F F2/R3 and F3/Y764F R2 were used in initial PCRs, and the resulting PCR products were further amplified with oligonucleotides F3 and R3. Using the same cloning strategy used to generate pCDM8-Y744F, pCDM8-Y746F was constructed.

Cells

Parental Ba/F3 cells and Ba/F3 cells stably transfected with eukaryotic expression vectors containing the WT, Y744F, Y764F, or {Delta}716 G-CSFR forms were maintained in RPMI 1640 medium containing 10% FBS, 2 mM glutamine, and 10% WEHI-3-conditioned medium as a source of IL-3. Details for the generation of stably transfected Ba/F3 cells have previously been described (6, 11). Stable transfectants were selected by antibiotic resistance, then screened for equivalent receptor numbers and dissociation constants by Scatchard analysis. Receptor expression was also analyzed using flow cytometry. Single clones as well as pools of three clones exhibiting similar receptor numbers were analyzed in the functional assays described below. Data from pools of three clones for each transfectant are reported. For studies with COS cells, COS-7 cells were grown in DMEM supplemented with 10% FBS. The cells were transiently transfected by calcium phosphate precipitation of plasmid DNA (11).

Akt in vitro kinase assay

G-CSFR transfectants were serum- and cytokine-deprived for 4 h, then stimulated with 100 ng/ml G-CSF for the indicated times and lysed. Protein concentrations were determined and 0.75 mg protein was immunoprecipitated using immobilized anti-Akt1 Ab (Cell Signaling Technology, Beverly, MA). Quantitation of Akt activity was determined using the phosphorylated glycogen synthase kinase 3 (pGSK3) in vitro kinase assay kit (Cell Signaling Technology). Briefly, immune complexes were washed in kinase buffer, then incubated in kinase buffer supplemented with 1 µg of GSK3 fusion protein and 200 µM ATP and incubated at 30°C for 30 min. The reactions were stopped by the addition of SDS-sample buffer, resolved on 10% polyacrylamide gels, then transferred to nitrocellulose. Membranes were immunoblotted with a rabbit Ab recognizing the pGSK3 protein. Membranes were subsequently stripped and reblotted with anti-Akt Ab to control for the amount of immunoprecipitated Akt loaded. The intensities of the pGSK3 and total Akt bands were quantified using densitometry, and the fold increase in Akt activity for each time point was calculated by comparison of pGSK3 levels with total Akt levels.

Proliferation studies

Proliferation of stably transfected Ba/F3 cells was analyzed as previously described (6, 7, 11). Briefly, the cells were serum and cytokine deprived for 2–4 h in RPMI 1640, 1% BSA, and 2 mM glutamine. After starvation, the cells were washed and resuspended in RPMI 1640 containing 10% FBS and 2 mM glutamine. A total of 5 x 103 cells/well were seeded in 96-well microtiter plates with varying concentrations of G-CSF (0.002–2000 pM). Duplicate plates were seeded with cells, and the cells were grown in the presence of IL-3 without G-CSF at 37°C in 5% CO2 for a total of 72 h. The plates were pulsed with 0.5 µCi/well [methyl-3H]thymidine during the last 8 h of incubation. Samples were harvested onto glass-fiber filters and counted in scintillation fluid.

Induction of membrane-targeted CD8-SHIP expression

The cDNA for CD8-SHIP was cloned into the mifepristone inducible vector pGene (Invitrogen Life Technologies, Carlsbad, CA), stably transfected along with the regulatory plasmid pSwitch into {Delta}716 transfectants, and selected for hygromycin and zeocin resistance to generate {Delta}716/CD8-SHIP transfectants. Cells transfected with empty pGene and pSwitch vectors ({Delta}716/Vec) were generated as a negative control. Cells were incubated with 5 x 10–9 M mifepristone for 3 h, then washed and incubated in growth medium. To confirm the expression of the membrane-targeted CD8-SHIP, cells treated with mifepristone were washed in HBSS containing 1% BSA and 0.1% NaN3, then incubated on ice for 1 h with 5 µg of anti-CD8-FITC (BD Pharmingen, San Diego, CA). The cells were then washed and analyzed using the FACSCalibur cytometer and CellQuest software (BD Pharmingen). At the indicated times after mifepristone induction, cell viability was also measured by trypan blue exclusion.

Immunoprecipitations and Western blot analyses

Cells were serum- and cytokine-deprived for 4 h in RPMI 1640 medium containing 1% BSA and 2 mM glutamine. The cells were then stimulated with 100 ng/ml G-CSF for 5 min and lysed in buffer containing 1% Nonidet P-40 (Roche, Indianapolis, IN), 1 mM EDTA (pH 8.0), 20 mM Tris (pH 8.0), 150 mM NaCl, 0.15 U/ml aprotinin, 10 µg/ml leupeptin, 10 µg/ml pepstatin A, and 1 mM sodium vanadate, then cleared of insoluble material. Protein concentrations in whole cell lysates were determined using the bicinchoninic acid assay (Pierce, Rockford, IL). For immunoprecipitations, 1 mg of protein from each sample was precleared with protein A agarose (Invitrogen Life Technologies) before immunoprecipitation with the indicated Abs. The immune complexes were washed three times, resuspended in SDS-PAGE sample buffer, and resolved on 10% acrylamide gels. Western blot analyses were performed as previously described (7).

Phosphopeptides and Far Western blotting

Phosphopeptides were synthesized by F-moc chemistry, purified by reverse phase HPLC, and analyzed for purity by ion spray mass spectrometry by the vendor. For each sample, 1 mg of cell lysate was incubated with 1 µM biotinylated phosphopeptide in TN1 buffer (50 mM Tris (pH 8.0), 10 mM EDTA, 10 mM sodium pyrophosphate, 10 mM NaF, 1% Nonidet P-40, 125 mM NaCl, 10 mM vanadate, 10 µg/ml aprotinin, and 10 µg/ml leupeptin) for 2 h at 4°C. The protein-peptide complexes were captured on streptavidin agarose beads (Invitrogen Life Technologies) in TN1 buffer containing 0.1% BSA for 20 min at 4°C, washed five times in TN1 buffer, and boiled. Samples were resolved on 10% SDS-polyacrylamide gels, transferred to nitrocellulose, and immunoblotted. For Far Western analyses, cell lysates from unstimulated transfectants were immunoprecipitated with rabbit polyclonal anti-SHIP Ab. The immune complexes were separated by SDS-PAGE and transferred to nitrocellulose, then blotted with biotinylated phosphopeptides. To detect the protein:phosphopeptide interactions the membranes were blotted with streptavidin conjugated to HRP and visualized by ECL (Amersham Biosciences).

Northern blot analyses

Total RNA was extracted from 1 x 107 cells using TRIzol (Invitrogen Life Technologies) according to the manufacturer’s instructions. RNA samples (10 µg) were denatured in a 25% deionized glyoxal solution for 1 h at 50°C, separated on 1% agarose gels, and transferred to nylon membranes (Schleicher & Schuell, Keene, NH). The membranes were prehybridized with Express Hyb (BD Clontech, Palo Alto, CA) for 10 min, then incubated with 32P-labeled cDNA probes corresponding to CIS, SOCS1, or SOCS3. Membranes were washed initially at room temperature in 2x SSC and 0.05% SDS, then in 0.1x SSC and 0.1% SDS at 65°C. The dried membranes were cross-linked with a UV Stratalinker (Stratagene, La Jolla, CA), then exposed to a PhosphorImager screen (Molecular Dynamics, Sunnyvale, CA, and the signals were quantitated with ImageQuant software (Molecular Dynamics). Membranes were stripped in 0.1x SSC with 0.5% SDS at 100°C before reprobing.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Analysis of in vitro binding of SHIP to G-CSFR-derived phosphopeptides

We previously reported that phosphorylation of SHIP and the formation of SHIP/Shc protein complexes were mediated by C-terminal sequences present in the WT G-CSFR (6). This region contains three of the four cytoplasmic tyrosine residues in the G-CSFR (Fig. 1A). Because SHIP was previously shown to directly interact with Shc and with the phospho-ITIM motif (ITIM) of Fc{gamma}RIIB to modulate negative signaling in B cells (32, 34, 35), we were interested in determining whether phosphorylated tyrosine residues in the cytoplasmic domain of the G-CSFR also mediated the recruitment of SHIP.



View larger version (39K):
[in this window]
[in a new window]
  		FIGURE 1. Schematic diagram of WT and mutant G-CSFR forms and sequences of phosphopeptides. Shown are the WT and mutant G-CSFR forms containing either single tyrosine (Y) to phenylalanine (F) substitutions at residue 744 or 764, designated Y744F and Y764F, or a premature stop codon resulting in the {Delta}716 truncation mutant isolated from patients with SCN/AML. The {Delta}716 G-CSFR lacks the C-terminal 98 aa. The transmembrane (TM) domain and conserved box 1, 2, and 3 cytoplasmic sequences are indicated. B, Sequences of the phosphopeptides are shown with the centrally located phosphotyrosine (pY) underlined. The N terminus of each peptide contained a six-carbon spacer (hexamer) that was biotinylated. The C-terminal end of each peptide was amidylated to prevent proteolytic cleavage.

 
To address this, synthetic phosphopeptides of 13 aa in length, corresponding to sequences surrounding each of the four cytoplasmic tyrosine residues in the G-CSFR, which were biotinylated at the N-terminal end and amidylated at the C terminus, were generated (Fig. 1B). The phosphopeptides were incubated with whole-cell lysates from unstimulated or G-CSF-stimulated Ba/F3 cells stably transfected with the WT G-CSFR. Peptide-protein complexes were captured on streptavidin-agarose beads and analyzed by Western blotting using Abs recognizing SHIP and Shc. As shown in Fig. 2, SHIP bound only to phosphopeptides corresponding to the two most distal tyrosine residues in the G-CSFR (pY744 and pY764). No complexes containing SHIP were detected when lysates were incubated with phosphopeptides corresponding to the more proximal tyrosine residues in the G-CSFR. SHIP also bound to a phosphopeptide specific for the phospho-ITIM region of Fc{gamma}RIIB (positive control), but not to streptavidin-agarose beads alone (negative control). SHIP appeared to preferentially bind to pY744 and weakly associate with pY764. Detection of SHIP in lysates from both unstimulated and G-CSF-stimulated WT cells indicates that an inducible accessory molecule is not required for interaction of SHIP with the phosphopeptides and suggests that SHIP may directly bind to the G-CSFR (Fig. 2A).



View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 2. Identification of SHIP and Shc in peptide-protein complexes generated with G-CSFR-derived phosphopeptides. WT cells were unstimulated (0) or stimulated (+) with 100 ng/ml G-CSF for 5 min, lysed, then incubated with biotinylated phosphopeptide pY704, pY729, pY744, pY764, or phospho-ITIM (pITIM). Peptide-protein complexes were collected on streptavidin-agarose beads. A, The samples were immunoblotted with anti-SHIP Ab. B, Immunoblot with anti-Shc Ab. Lysates incubated with streptavidin agarose beads alone (Neg) were used as a negative control. The presence of SHIP and Shc in initial whole-cell lysates (WCL) was confirmed (last lane). A representative blot from three independent experiments is shown. C, Far Western analysis of cell lysates immunoprecipitated with rabbit polyclonal SHIP Ab, then blotted with either phosphopeptide pY744 (left) or pY764 (right). Cell lysates immunoprecipitated with normal rabbit serum (NRS) were blotted with the identical phosphopeptides and are shown as negative controls.

 
Shc was also found to bind to pY744 and pY764 in the G-CSFR. Similar to SHIP, this interaction was observed to occur independently of protein modification or the presence of accessory proteins, as indicated by the presence of Shc in peptide-protein complexes from unstimulated cells (Fig. 2B). Notably, two distinct isoforms of Shc were consistently detected with the pY764 phosphopeptide, but not with the pY744phosphopeptide.

Far Western analyses were performed to confirm that SHIP could bind directly to the G-CSFR. As shown in Fig. 2C, both pY744 and pY764 bound to SHIP immune complexes. The interaction between pY764 and SHIP (Fig. 2C, right panel) appeared to be weaker than that observed between pY744 and SHIP (Fig. 2C, left panel). As expected, the phosphopeptides did not interact with normal rabbit serum, which served as a negative control.

In vivo requirements for G-CSFR-mediated SHIP phosphorylation and formation of SHIP/Shc complexes

To determine whether our observations in vitro correlated with in vivo events, we examined tyrosine phosphorylation of SHIP in Ba/F3 cells stably transfected with G-CSFR mutants containing tyrosine to phenylalanine (Y->F) mutations at Tyr744 or Tyr764 (Y744F and Y764F, respectively) and the truncated {Delta}716 G-CSFR, which lacks the C-terminal 98 aa, including the tyrosine residues at aa positions 729, 744, and 764, and was derived from a patient with SCN/AML. PCR and flow cytometry were used to confirm G-CSFR expression and expression of the appropriate receptor form in the stable transfectants. G-CSFR numbers and ligand binding affinities for the stable transfectants were also determined using Scatchard analyses of equilibrium binding data. Single clones and pools of three clones for each transfectant that expressed equivalent G-CSFR numbers and binding affinities were analyzed. Cells stably transfected with the Y->F G-CSFR mutants expressed similar receptor numbers as cells transfected with the WT G-CSFR (data not shown). We have previously extensively characterized the expression and binding properties of the WT G-CSFR in Ba/F3 cells (6, 11).

As shown in Fig. 3A, SHIP was phosphorylated in response to G-CSF in WT and Y744F cells, but not in Y764F or {Delta}716 cells. Additionally, a 52-kDa tyrosine-phosphorylated protein was observed to consistently coassociate with SHIP in response to G-CSF stimulation in WT and Y744F transfectants, but not in Y764F or {Delta}716 transfectants. Because we previously demonstrated that Shc associated with SHIP in WT cells in response to G-CSF stimulation (6), studies were performed to determine whether the 52-kDa phosphoprotein that coimmunoprecipitated with SHIP in G-CSF-stimulated Y744F cells was also Shc. As shown in Fig. 4A, a 52-kDa phosphoprotein corresponding to tyrosine-phosphorylated Shc along with a 145-kDa tyrosine-phosphorylated band consistent with the Mr of SHIP could consistently be detected in WT and Y744F cells after G-CSF stimulation, but not in G-CSF-treated Y764F cells or {Delta}716 cells. In some experiments these proteins could be faintly detected in Y764F cells, but always at markedly reduced levels compared with WT and Y744F transfectants. To confirm that the 145-kDa protein that coimmunoprecipitated with Shc in WT and Y744F cells was SHIP, the same blot was stripped and reblotted with an Ab recognizing SHIP (Fig. 4B).



View larger version (53K):
[in this window]
[in a new window]
  		FIGURE 3. G-CSF-induced tyrosine phosphorylation of SHIP in WT and mutant G-CSFR-expressing cells. A, SHIP was immunoprecipitated from unstimulated (0) and G-CSF stimulated (+) cells, and the samples were immunoblotted with the 4G10 anti-phosphotyrosine Ab. Arrows indicate phosphorylated SHIP (upper) and a coimmunoprecipitating 52-kDa protein (lower). B, To confirm equal protein loading of SHIP, the blot in A was stripped and reblotted with anti-SHIP Ab. A representative blot from three independent experiments is shown.

 


View larger version (45K):
[in this window]
[in a new window]
  		FIGURE 4. Association of SHIP with Shc in response to G-CSF in cells expressing WT and mutant G-CSFR forms. A, Shc was immunoprecipitated from unstimulated (0) or G-CSF stimulated (+) cells and immunoblotted with the 4G10 anti-phosphotyrosine Ab. Arrows indicate the migration of tyrosine-phosphorylated Shc (lower) and a 145-kDa protein (upper). B, The blot in A was stripped and reblotted with anti-SHIP Ab. C, Equal loading of Shc was confirmed by stripping the blot in A and reblotting with anti-Shc Ab. A representative blot from three independent experiments is shown.

 
Regulation of Akt activation by G-CSF in Y->F G-CSFR transfectants

We previously reported that activation of Akt was prolonged in cells expressing the truncated {Delta}716 G-CSFR form from patients with SCN/AML. We were therefore interested in investigating whether loss of SHIP association with the Y->F G-CSFR mutants also induced prolonged activation of Akt when the Y->F transfectants were stimulated with G-CSF. For these experiments, a nonradioactive method was used to measure Akt activity by analysis of Akt-induced phosphorylation of the substrate GSK3.

In our earlier studies using radioactive in vitro kinase assays, we showed that Akt was rapidly and transiently activated within 5 min of G-CSF stimulation in cells expressing the WT G-CSFR, and that activated Akt was undetectable by 30 min (7). In {Delta}716 cells, G-CSF stimulation resulted in both delayed and prolonged activation of Akt that was still apparent 2 h later. Similar to our findings, Dong et al. (36) also reported delayed as well as prolonged activation of Akt in {Delta}716 cells. These investigators showed that activation of Akt in WT G-CSFR-expressing cells occurred within 5 min of G-CSF stimulation and was rapidly extinguished within 30 min, whereas the onset of activation of Akt in {Delta}716 cells was delayed beyond 5 min, and persistent activation was still detected at 2 h (36). Using a nonradioactive method and the substrate GSK3 to measure Akt activity, G-CSF stimulation of Y764F cells was found to induce both delayed and prolonged activation of Akt, similar to our earlier observations with {Delta}716 cells (Fig. 5). Although we could not detect activation of Akt after treatment of Y764F cells with G-CSF at 10 min (data not shown), activated Akt could still be detected in these cells 2 h after G-CSF stimulation at a time when Akt activity had returned to near basal levels in WT cells. Although activation of Akt in WT cells could be detected at 30 min after G-CSF stimulation in some experiments, Akt activity consistently returned to near basal levels in WT cells by 2 h. The time differences in detection of the onset of Akt activation in G-CSF-stimulated WT cells in these experiments compared with our earlier studies may relate to differences in the assays used. Delayed and prolonged activation of Akt in Y764F cells was reproducible in multiple independent experiments (n = 5), and there was no evidence for G-CSF-mediated prolonged activation of Akt in Y744F cells. In some experiments, activation of Akt in Y744F cells appeared less robust than that in WT cells.



View larger version (34K):
[in this window]
[in a new window]
  		FIGURE 5. G-CSF-induced activation of Akt in Y->F G-CSFR mutants. A, Cells were serum and cytokine deprived for 4 h, then stimulated with 100 ng/ml G-CSF for the indicated times and lysed. Akt was immunoprecipitated from equivalent amounts of protein (0.75 mg) from each lysate, and the immune complexes were subjected to in vitro kinase assays using GSK3 as a substrate. Samples were resolved by SDS-PAGE, transferred to nitrocellulose, and immunoblotted with an Ab to phosphorylated GSK3 (upper panel). Membranes were stripped and reblotted with anti-Akt Ab to control for the quantity of immunoprecipitated Akt. A representative blot from five independent experiments is shown. B, The membranes in A were subjected to densitometric analysis using the Gel-Doc system (Bio-Rad, Hercules, CA) with Quantity One software. Bar graphs indicate the fold increase in Akt activity with respect to total Akt levels.

 
Proliferative responses of cells expressing Y->F G-CSFR mutants

We previously reported that the class IV alternatively spliced G-CSFR variant, which has an altered C-terminal tail and is overexpressed in cells from some patients with de novo AML, transduces hyperproliferative responses (6, 37). We also showed that in cells expressing this splice variant neither tyrosine phosphorylation of SHIP nor formation of SHIP/Shc protein complexes occurred in response to G-CSF (6). We postulated that a loss of inhibitory signaling through SHIP contributed to the enhanced growth of class IV transfectants. Because G-CSF also failed to induce tyrosine phosphorylation of SHIP and the formation of SHIP/Shc protein complexes in {Delta}716 cells and Y764F cells similar to our previous observations with class IV transfectants, which also hyperproliferate in response to G-CSF, additional studies were performed to examine the growth response of cells expressing Y->F G-CSFR mutants. As we previously reported (6, 11), a shift to the left in the G-CSF dose-response curve was observed with {Delta}716 transfectants compared with WT cells (Fig. 6). In addition, a left-shifted G-CSF dose-response curve was observed with cells expressing the Y764F G-CSFR. Notably, a 1-log shift in the dose-response curve was observed with Y764F cells compared with a 2-log shift to the left consistently observed with {Delta}716 cells, suggesting that other intermediates whose recruitment is independent of Tyr764 also contribute to negative signaling to limit G-CSF-stimulated growth responses. No significant differences in growth responses to G-CSF were observed between Y744F and WT cells, suggesting a nonessential role for Tyr744 in the recruitment of other inhibitory proteins.



View larger version (21K):
[in this window]
[in a new window]
  		FIGURE 6. Proliferative responses of G-CSFR transfectants. Cells were serum- and cytokine-deprived for 4 h before cytokine stimulation. DNA synthesis was measured by [3H]thymidine uptake in the presence of G-CSF. The data shown are the average of two experiments and are expressed as a percentage of the maximum cpm.

 
Effects of membrane-targeted SHIP on cell survival

Because membrane localization of SHIP has been shown to be crucial to activation of its phosphatase activity and for PI (3,4,5)P3 hydrolysis (38), we reasoned that expression of a membrane-targeted form of SHIP, which bypasses the need for receptor-mediated recruitment, should decrease proliferative signaling. We therefore examined the effects of enforced plasma membrane localization of SHIP on proliferation of Ba/F3 cells expressing the {Delta}716 G-CSFR by overexpressing a human CD8-SHIP chimera. The CD8-SHIP chimera is a chimeric protein with the CD8{alpha} extracellular and transmembrane domains fused to SHIP. To avoid toxic cellular effects from constitutive expression of membrane-localized SHIP, we used an inducible expression system in which the expression of CD8-SHIP was under the control of a mifepristone-inducible hybrid promoter containing GAL4 upstream activating sequences and the adenovirus E1b TATA box. Induction of surface expression of CD8-SHIP was monitored by flow cytometric analysis of CD8. After treatment of {Delta}716/CD8-SHIP cells with mifepristone for 3 h and subsequent culture for 18 h in growth medium, CD8 surface expression could be detected (Fig. 7A, right, open curve). Surface expression of CD8 could not be detected on {Delta}716/CD8-SHIP cells in the absence of the inducer (data not shown) or on mifepristone-treated {Delta}716/Vec cells (Fig. 7A, left, shaded curve). As shown in Fig. 7B, mifepristone (inducer) treatment had no effect on the growth of {Delta}716 cells transfected with empty vector alone ({Delta}716/Vec) and stimulated with G-CSF. In contrast, the expression of membrane-targeted CD8-SHIP in {Delta}716 cells significantly inhibited their growth in response to G-CSF with a >50% reduction in viability observed by 42 h. Induction of CD8-SHIP expression in cells expressing the WT G-CSFR also resulted in decreased cell survival (data not shown). The decreased viability of {Delta}716 cells observed when SHIP is targeted to the membrane suggests that the enhanced growth of {Delta}716 cells is in part mediated by loss of membrane recruitment of SHIP due to deletion of the domain in the distal cytoplasmic region of the G-CSFR containing Tyr764 that is required for binding of SHIP to the G-CSFR.



View larger version (55K):
[in this window]
[in a new window]
  		FIGURE 7. Decreased cell viability after induction of membrane-targeted SHIP. SHIP was targeted to the membrane by an in-frame fusion with the extracellular and transmembrane domains for human CD8. The cDNA for CD8-SHIP was cloned into the mifepristone-inducible vector pGene and was stably transfected along with the regulatory plasmid pSwitch into {Delta}716 cells ({Delta}716/CD8-SHIP). Cells transfected with empty pGene and pSwitch vectors ({Delta}716/Vec) were generated as a negative control. Cells (3 x 106) were incubated with 5 x 10–9 M mifepristone for 3 h to induce CD8-SHIP expression, then washed and incubated in growth medium for 18 h. CD8-SHIP expression was monitored by CD8 surface expression using anti-CD8 FITC and a FACSCalibur cytometer. A, Analysis of CD8 surface expression after mifepristone treatment in {Delta}716/Vec (left, shaded curve) and {Delta}716/CD8-SHIP (right, open curve) cells (1 x 106) showing inducible CD8 expression in {Delta}716/CD8-SHIP cells only. B, At the indicated times after treatment with mifepristone, cell viability and number were measured by trypan blue exclusion in {Delta}716/Vec (left) and {Delta}716/CD8-SHIP (right) cells. Absolute viable cell numbers at 42 h for {Delta}716/CD8-SHIP and {Delta}716/Vec cells were 2.36 x 106 and 44.5 x 106, respectively.

 
Analysis of SOCS proteins in G-CSFR signaling

Our observations suggested that other negative regulators of proliferation in addition to SHIP must be recruited to the distal cytoplasmic tail of the G-CSFR. Because activation of STAT5 in {Delta}716 cells was previously reported to be prolonged (14), which we have also confirmed (data not shown), we reasoned that defective recruitment of a SOCS family member to the G-CSFR could mediate the sustained activation of STAT5 observed in {Delta}716 cells and could account for the 1-log difference in the dose-response curves between Y764F and {Delta}716 cells. To determine whether this was the case, we initially examined the expression of SOCS family members in cells stimulated with G-CSF, because expression of SOCS proteins has been shown to be induced by cytokine treatment of cells (16, 20, 25). In response to G-CSF stimulation, mRNA transcripts corresponding to CIS were detected in both WT and {Delta}716 cells (Fig. 8A). Maximal expression of CIS in WT cells was observed 30 min after G-CSF stimulation, and by 2 h CIS transcripts were undetectable. In contrast, prolonged expression of CIS in response to G-CSF was observed in {Delta}716 cells. CIS mRNA transcripts could still be detected in {Delta}716 cells 3 h after G-CSF stimulation. Although the expression of SOCS1 and SOCS3 in {Delta}716 cells was also found to be prolonged, the differences in their expression levels in WT and {Delta}716 cells were consistently less than the differences observed in CIS expression (data not shown).



View larger version (44K):
[in this window]
[in a new window]
  		FIGURE 8. Prolonged expression of CIS in {Delta}716 cells in response to G-CSF stimulation. Ba/F3 cells transfected with WT G-CSFR, {Delta}716 G-CSFR or empty vector (CDM8) were serum- and cytokine-deprived for 4 h, then stimulated with 100 ng/ml G-CSF for the indicated times. A, Total RNA was isolated, and RNA samples (10 µg) were denatured with glyoxal, separated on 1% agarose gels, and transferred onto nylon membranes (upper panel). The membranes were incubated with a 32P-labeled probe corresponding to CIS at 65°C, washed stringently, and exposed to a PhosphorImager screen. CDM8 cells stimulated with 20 ng/ml IL-3 (last lane, right) are shown as a positive control. Equivalent RNA loading was confirmed by stripping the blot in the upper panel and blotting with a 32P-labeled GAPDH probe (lower panel). B, Whole cell lysates were immunoblotted with an anti-CIS Ab (upper panel). The blot was stripped in the upper panel and reblotted with Ab recognizing SHIP to confirm equivalent protein loading.

 
Because the regulation of SOCS1 and SOCS3 expression by G-CSF and, more recently, binding of SOCS3 to the G-CSFR have previously been reported (25, 16, 39), and given our findings with CIS transcript levels, we were interested in further examining the role of CIS in G-CSFR signaling. Similar to our observations with CIS transcript levels, CIS protein expression was found to be prolonged in {Delta}716 cells in response to G-CSF stimulation. As shown in Fig. 8B, a 37-kDa protein corresponding to CIS could be detected 1 h after G-CSF stimulation in both WT and {Delta}716 cells (Fig. 8B). However, CIS protein expression was rapidly down-regulated in WT cells, whereas a significant amount of CIS protein could still be detected at 3 h after G-CSF stimulation in {Delta}716 cells.

CIS associates with the G-CSFR via Tyr729 and Tyr744

Because CIS was previously shown to directly bind to the receptors for erythropoietin (EPO) and IL-3 (25), we were interested in determining whether CIS also directly binds to the G-CSFR. As shown in Fig. 9, using whole-cell lysates from unstimulated and G-CSF-stimulated cells and the identical synthetic phosphopeptides used to examine SHIP and Shc interactions with the G-CSFR, CIS was found to associate with Tyr729 and Tyr744 of the G-CSFR, but not with Tyr704 or Tyr764.



View larger version (16K):
[in this window]
[in a new window]
  		FIGURE 9. CIS binds to phosphopeptides corresponding to tyrosine residues 729 and 744 of the G-CSFR. Whole cell lysates (WCL) from unstimulated (0) or G-CSF-stimulated (+) {Delta}716 cells were incubated with biotinylated phosphopeptides specific for sequences surrounding each of the cytoplasmic tyrosine residues in the G-CSFR. The peptide-protein complexes were captured on streptavidin-conjugated agarose beads and immunoblotted with anti-CIS Ab. WCL from cells stimulated with G-CSF (G) or IL-3 for 90 min are shown as positive controls. WCL incubated with agarose beads alone were included as a negative control (Neg).

 
To circumvent the problems posed by the requirement for prior cytokine stimulation to detect CIS protein expression and the low G-CSFR numbers in transfected Ba/F3 cells (40), COS-7 cells were used to examine the in vivo requirements for interaction of CIS with the G-CSFR. We have previously extensively used COS-7 cells to examine G-CSFR internalization and degradation, and we have shown that they provide an excellent in vivo model system for studying G-CSFR trafficking (11). Because COS-7 cells do not constitutively express CIS, the full-length CIS cDNA was transfected along with the WT or {Delta}716 G-CSFR cDNA, and the interactions between CIS and the G-CSFR were investigated. As shown in Fig. 10A (left panel), CIS protein was undetectable in whole-cell lysates from cells transfected with empty vector alone. In cells doubly transfected with both the WT G-CSFR and CIS cDNAs, CIS coprecipitated with the WT receptor (Fig. 10A, right panel). In contrast, CIS did not associate with the receptor in cells expressing the {Delta}716 G-CSFR form. Notably, the association of CIS with WT G-CSFR was independent of ligand stimulation. In Far Western analyses, we were unable to demonstrate direct binding of CIS to the G-CSFR (data not shown), suggesting that CIS and the G-CSFR may indirectly interact and that this interaction may occur through an accessory molecule.



View larger version (67K):
[in this window]
[in a new window]
  		FIGURE 10. CIS associates with WT, but not the {Delta}716, G-CSFR and modulates G-CSFR expression. COS-7 cells were transiently transfected for 72 h with 20 µg of empty vector alone (Vec) or 20 µg of expression vectors containing the cDNAs for CIS, WT G-CSFR, {Delta}716 G-CSFR (left panel); both WT G-CSFR and CIS (WT/CIS); or {Delta}716 G-CSFR and CIS ({Delta}716/CIS). The cells were serum-starved for 4 h at 37°C, then stimulated with 100 ng/ml G-CSF for 15 min and lysed. Equivalent amounts of WCL from WT/CIS and {Delta}716/CIS were immunoprecipitated with anti-G-CSFR Ab. WCL are shown in the left panels, and G-CSFR immunoprecipitates in the right panel. A, Immunoblot with anti-CIS Ab. B, The blot in A was stripped and reblotted with anti-STAT5 Ab. C, The blot in A was reblotted with anti-G-CSFR Ab. D, G-CSFR expression in WT and {Delta}716 cells with or without constitutive CIS expression in cell lysates immunoprecipitated with anti-G-CSFR Ab and immunoblotted with the same Ab.

 
CIS has previously been shown to inactivate STAT5 signaling in several growth factor receptor systems by binding to tyrosine residues in the growth factor receptors (25). In the case of the EPO-R, STAT5 was shown to directly bind to phosphotyrosine residues in the EPO-R cytoplasmic domain. CIS was shown to directly compete with STAT5 for binding to the EPO-R to down-regulate EPO-R signaling. CIS was also found to route the EPO-R to the proteasome for degradation. We therefore investigated whether STAT5, which is activated in response to G-CSF, also binds to the G-CSFR. We were unable to detect association of STAT5 with either the WT or {Delta}716 G-CSFR form (Fig. 10B, right panel) despite its presence in whole-cell lysates (Fig. 10B, left panel), which is consistent with a previous report by Nicholson et al. (4). In WT cells coexpressing CIS, G-CSF treatment resulted in decreased expression of both the G-CSFR and CIS (Fig. 10, A and C, right panel). In contrast, G-CSFR expression remained unchanged during the same time period in {Delta}716/CIS cells treated with G-CSF, suggesting that CIS interacts with the distal G-CSFR to down-regulate its surface expression. Although coexpression of CIS enhanced the level of receptor expression in both {Delta}716/CIS and WT/CIS transfectants compared with cells transfected with receptor only ({Delta}716 only or WT only), as shown in Fig. 10D, ligand stimulation resulted in a decrease in expression of the WT receptor form only.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Acquired mutations in the G-CSFR that introduce a premature stop codon and result in the expression of a truncated form of the G-CSFR lacking the distal cytoplasmic tail have been identified in nearly all patients with SCN transforming to AML (10, 41). Truncated G-CSFR forms from these patients are not properly internalized, and as a result, surface expression of the mutant receptor is prolonged (11). In addition, the truncated G-CSFR induces ligand hypersensitivity, increased resistance to apoptosis, and enhanced proliferative responses to G-CSF (7, 11). Cells bearing the truncated G-CSFR forms exhibit sustained activation of STAT5 and Akt (7, 14). Collectively, these observations suggest that truncations in the G-CSFR in SCN/AML delete a distal inhibitory domain that down-modulates G-CSFR proliferative signals. In this study we have investigated the signaling intermediates that are recruited by the region deleted in patients with SCN/AML to negatively regulate G-CSFR signaling.

Previous studies by our laboratory demonstrated the presence of a domain in the distal 98 aa of the G-CSFR that is required for tyrosine phosphorylation of SHIP and association of SHIP with Shc (6). In other receptor systems, SHIP has been shown to negatively regulate the PI3K/Akt pathway by virtue of its 5'-phosphatase activity that dephosphorylates the PI3K metabolite, PI(3,4,5)P3, which is required for activation of Akt (28, 29, 30, 31, 32). Membrane localization of SHIP was shown to be necessary for its tyrosine phosphorylation and the subsequent formation of SHIP/Shc protein complexes (32, 34, 35, 38). We postulated a similar role for SHIP in down-regulating G-CSFR signaling.

Using synthetic phosphopeptides, we show that SHIP binds to a region in the G-CSFR that is deleted in patients with SCN/AML. SHIP bound in vitro to phosphopeptides corresponding to the two most distal tyrosine residues (pY744 and pY764) in the G-CSFR, both of which are deleted in SCN/AML. Studies with G-CSFR mutants containing tyrosine to phenylalanine (Y->F) mutations at residues 744 and 764 demonstrated an in vivo requirement for Tyr764 for the association of SHIP and SHIP/Shc complex formation, suggesting that Tyr764 mediates membrane localization of SHIP. Activation of Akt was also found to be delayed and prolonged in Y764F cells, consistent with a role for this tyrosine residue in inhibition of the PI3K-Akt survival pathway via SHIP recruitment. The observation that Akt activation is delayed and prolonged in Y764F cells, but not in Y744F cells, supports a role for Tyr764, but not for Tyr744, in the in vivo recruitment of SHIP to the G-CSFR complex. The reasons for the discrepancy in our in vitro and in vivo observations with pY744 and the Y744F mutant, respectively, could be due to changes in the secondary structure of the full-length G-CSFR which may make Tyr744 unavailable for in vivo phosphorylation of Shc and SHIP and their subsequent association.

Cells expressing the Y764F G-CSFR mutant were found to hyperproliferate in response to G-CSF, consistent with an inhibitory role for SHIP, whereby SHIP is directly recruited to the G-CSFR cytoplasmic tail and phosphorylated, resulting in SHIP/Shc complex formation and decreased mitogenic responses. Through the expression of a membrane-targeted CD8-SHIP chimeric protein in {Delta}716 cells, which inhibited cell growth, we demonstrate that loss of SHIP recruitment to the {Delta}716 G-CSFR contributes to the enhanced growth of these cells. However, inspection of the dose-response curve for Y764F cells revealed a persistent 1-log difference in ligand sensitivity between the Y764F and {Delta}716 G-CSFR forms, indicating that a loss of SHIP recruitment alone via loss of Tyr764 could not fully explain the increased ligand sensitivity and enhanced growth of cells expressing the truncated {Delta}716 G-CSFR. These data indicate that in addition to SHIP, other negative regulators of proliferation must be recruited to distal tyrosine residues in the G-CSFR that are deleted in patients with SCN/AML.

Recently, the SOCS proteins (SOCS1–8 and CIS) were identified as a novel class of cytokine-inducible proteins that negatively regulate signaling by cytokine receptors (17, 18, 19, 20, 21, 22, 23, 24, 25). These proteins have been shown to inhibit cytokine signaling via multiple mechanisms, including direct binding to Jaks, competition with STATs for binding to receptor cytoplasmic tyrosine residues, and receptor routing to the ubiquitin-proteasome pathway.

Because STAT5 activation is prolonged in SCN/AML cells bearing truncated G-CSFR forms (14), we reasoned that loss of recruitment of a SOCS family member by deletion of the region in the G-CSFR required for its recruitment, might also contribute to the enhanced growth of SCN/AML cells and might underlie the prolonged STAT5 activation observed in these cells. Notably, in the case of the EPO receptor (EPO-R) system, CIS was shown to compete with STAT5 for binding to the EPO-R to down-regulate EPO-R signaling (25).

We initially investigated the effect of G-CSF on the induction of mRNA transcripts and protein for SOCS1–3 and CIS in cells expressing the WT G-CSFR and the truncated {Delta}716 G-CSFR form. G-CSF induced the expression of transcripts for SOCS1 and -3 (data not shown) as well as CIS in both WT and {Delta}716 cells, but their expression was sustained in {Delta}716 cells, which was most marked for CIS. Previous investigators have not reported induction of CIS expression by G-CSF. Yoshimura et al. (25) were unable to detect expression of CIS in the NFS-60 murine myeloid leukemia cell line after stimulation with G-CSF. These discrepancies could reflect differences in the cell lines used (25).

To further investigate the role of CIS in G-CSFR signaling, we initially examined whether CIS binds to the G-CSFR and the specific receptor sequence(s) that is required for interaction with CIS. CIS bound to the WT G-CSFR, but not the truncated {Delta}716 G-CSFR form. Using phosphopeptides corresponding to sequences surrounding each of the four cytoplasmic tyrosine residues in the G-CSFR, binding of CIS was found to be predominantly mediated by pY729 and, to a lesser extent, pY744, both of which are deleted in the {Delta}716 G-CSFR form. Hornter et al. (16) recently reported that Tyr729 of the G-CSFR also mediates the binding of SOCS3.

CIS was previously shown to bind to a YXXL sequence in the receptors for EPO and IL-3 (25). Despite the presence of an identical YXXL motif in the G-CSFR at Tyr729 and a similar motif at Tyr744 corresponding to YXXR, we were unable to demonstrate a direct interaction between CIS and phosphopeptides corresponding to these tyrosine residues in the G-CSFR by Far Western analysis. It is therefore possible that CIS interacts with the G-CSFR through an accessory molecule.

Using COS-7 cells that do not endogenously express CIS, we also show that CIS expression increases the expression of both the WT and truncated G-CSFR forms. The expression of the WT, but not the truncated, G-CSFR was found to decrease in response to G-CSF stimulation. CIS was found to coprecipitate with the WT G-CSFR, but not the truncated {Delta}716 form, consistent with our observations with the phosphopeptide pull-down experiments, in which CIS bound to pY729 and pY744, both of which are deleted in the {Delta}716 G-CSFR.

Our data confirm that STAT5 does not directly bind to the G-CSFR, as previously reported by Nicholson et al. (4). Because CIS does not compete with STAT5 for binding to the G-CSFR, another mechanism is probably responsible for the CIS-mediated extinction of G-CSFR-induced STAT5 signaling. CIS-mediated degradation of the G-CSFR signaling complex containing STAT5 may represent one mechanism by which this occurs. Failure to recruit CIS to the {Delta}716 G-CSFR, which lacks CIS binding sites, would result in prolonged expression of both CIS and the {Delta}716 G-CSFR, as we previously observed (11), and also lead to sustained STAT5 activation, which we and others have demonstrated in {Delta}716 cells (data not shown (14)). It is possible that CIS acts as a chaperone to target the G-CSFR to the proteasome where CIS, the G-CSFR, and other associated proteins, such as STAT5, are degraded to down-regulate intracellular signaling (42, 43, 44, 45).

In this article we have demonstrated that the expression of a truncated form of the G-CSFR, such as occurs in patients with SCN/AML, leads to a loss of recruitment of multiple signaling pathways that negatively regulate G-CSFR-mediated mitogenesis. We show that tyrosine residues present in the distal cytoplasmic tail of the G-CSFR that are deleted in the truncated receptor play critical roles in the recruitment of growth inhibitory molecules to the G-CSFR signaling complex. Future studies with additional G-CSFR mutants including G-CSFR forms containing multiple Y->F substitutions should help to further clarify these interactions.

Our data also provide additional evidence for the importance of inhibitory signaling in limiting proliferative signals emanating from growth factor receptors. Both SHIP and the SOCS family of proteins have been shown to negatively regulate hemopoiesis in mice (46, 47, 48). Targeted disruption of SHIP in mice produces a syndrome characterized by splenomegaly, pulmonary infiltration with myeloid cells, progenitor cell growth factor hypersensitivity, and shortened life spans (46). In BCR/abelson kinase-transformed cell lines and primary chronic myelogenous leukemia cells, SHIP expression has been reported to be reduced or absent (49, 50). Targeted disruption of SOCS1 and -3 is also associated with severe abnormalities in hemopoiesis (47, 48). SOCS1-deficient mice die at 3 wk of age with an accumulation of immature monocytes and granulocytes in the lung and spleen (47), whereas deletion of the SOCS3 gene is lethal, due to abnormalities in fetal erythropoiesis (48). Fetal erythroid progenitor cells from SOCS3–/– mouse embryos are hypersensitive to EPO. Together, these observations underscore the importance of negative signaling and suggest a potential role for novel therapeutics targeting inhibitory pathways.


   Footnotes
 
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.

1 This work was supported by Grants R29CA75226 and R01CA82859 from the National Cancer Institute (Bethesda, MD). Back

2 Address correspondence and reprint requests to Dr. Belinda R. Avalos, Bone Marrow Transplantation Program, Ohio State University, A437A Starling-Loving Hall, 320 West 10th Avenue, Columbus, OH 43210. E-mail address: avalos-1{at}medctr.osu.edu Back

3 Abbreviations used in this paper: G-CSFR, G-CSF receptor; AML, acute myelogenous leukemia; EPO, erythropoietin; EPO-R, EPO receptor; p, phosphorylated; PI(3,4,5)P3, phosphoinositol (3,4,5)-trisphosphate; SCN, severe congenital neutropenia; SOCS, suppressor of cytokine signaling; WT, wild type; CIS, cytokine-inducible Src homology 2 protein; pGSK3, phosphorylated glycogen synthase kinase 3. Back

Received for publication January 22, 2004. Accepted for publication August 12, 2004.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Avalos, B. R.. 1996. Molecular analysis of the granulocyte colony-stimulating factor receptor. Blood 88:761.[Free Full Text]
  2. Ziegler, S. F., T. A. Bird, K. K. Morella, B. Mosley, D. P. Gearing, H. Baumann. 1993. Distinct regions of the human granulocyte colony-stimulating factor receptor cytoplasmic domain are required for proliferation and gene induction. Mol. Cell. Biol. 13:2384.[Abstract/Free Full Text]
  3. Nicholson, S. E., A. C. Oates, A. G. Harper, A. Ziemiecki, A. F. Wilks, J. E. Layton. 1994. Tyrosine kinase Jak1 is associated with the granulocyte-colony-stimulating factor receptor and both become tyrosine phosphorylated after receptor activation. Proc. Natl. Acad. Sci. USA 91:2985.[Abstract/Free Full Text]
  4. Nicholson, S. E., R. Starr, U. Novak, D. J. Hilton, J. E. Layton. 1996. Tyrosine residues in the granulocyte colony-stimulating factor (G-CSF) receptor mediate G-CSF-induced differentiation of murine myeloid leukemic (M1) cells. J. Biol. Chem. 271:26947.[Abstract/Free Full Text]
  5. Chakraborty, A., D. J. Tweardy. 1998. STAT3 and G-CSF-induced myeloid differentiation. Leukemia Lymph. 30:433.
  6. Hunter, M. G., B. R. Avalos. 1998. Phosphatidylinositol 3'-kinase and SH2-containing inositol phosphatase (SHIP) are recruited by distinct positive and negative growth-regulatory domains in the granulocyte colony-stimulating factor receptor. J. Immunol. 160:4979.[Abstract/Free Full Text]
  7. Hunter, M. G., B. R. Avalos. 2000. Granulocyte colony-stimulating factor receptor mutations in severe congenital neutropenia transforming to acute myelogenous leukemia confer resistance to apoptosis and enhance cell survival. Blood 95:2132.[Abstract/Free Full Text]
  8. Ward, A. C., J. L. Monkhouse, X. F. Csar, I. P. Touw, P. A. Bello. 1998. The Src-like tyrosine kinase Hck is activated by granulocyte colony-stimulation factor (G-CSF) and docks to the activated G-CSF receptor. Biochem. Biophys. Res. Commun. 251:117.[Medline]
  9. Corey, S. J., A. L. Burkhardt, J. B. Bolen, R. L. Geahlen, L. S. Tkatch, D. J. Tweardy. 1994. Granulocyte colony-stimulating factor receptor signaling involves the formation of a three-component complex with Lyn and Syk protein-tyrosine kinases. Proc. Natl. Acad. Sci. USA 91:4683.[Abstract/Free Full Text]
  10. Dong, F., R. K. Brynes, N. Tidow, K. Welte, B. Löwenberg, I. P. Touw. 1995. Mutations in the gene for the granulocyte colony-stimulating-factor receptor in patients with acute myeloid leukemia preceded by severe congenital neutropenia. N. Engl. J. Med. 333:487.[Abstract/Free Full Text]
  11. Hunter, M. G., B. R. Avalos. 1999. Deletion of a critical internalization domain in the G-CSFR in acute myelogenous leukemia preceded by severe congenital neutropenia. Blood 93:440.[Abstract/Free Full Text]
  12. Ward, A. C., Y. M. van Aesch, A. M. Schelen, I. P. Touw. 1999. Defective internalization and sustained activation of truncated granulocyte colony-stimulated factor receptor found in severe congenital neutropenia/acute myeloid leukemia. Blood 93:447.[Abstract/Free Full Text]
  13. Touw, I. P., F. Dong. 1996. Severe congenital neutropenia terminating in acute myeloid leukemia: disease progression associated with mutations in the granulocyte-colony stimulating factor receptor gene. Leuk. Res. 20:629.[Medline]
  14. Hermans, M. H. A., C. Antonissen, A. C. Ward, A. E. M. Mayhen, R. E. Ploemacher, I. P. Touw. 1999. Sustained receptor activation and hyperproliferation in response to granulocyte colony-stimulating factor (G-CSF) in mice with a severe congenital neutropenia/acute myeloid leukemia-derived mutation in the G-CSF receptor gene. J. Exp. Med. 189:683.[Abstract/Free Full Text]
  15. Ward, A. C., S. P. Oomen, L. Smith, J. Gits, D. van Leeuwen, A. A. Soede-Bobok, C. A. Erpelinck-Verschueren, T. Yi, I. P. Touw. 2000. The SH2 domain-containing protein tyrosine phosphatase SHP-1 is induced by granulocyte colony-stimulating factor (G-CSF) and modulates signaling from the G-CSF receptor. Leukemia 14:1284.[Medline]
  16. Hortner, M., U. Nielsch, L. M. Mayr, J. A. Johnston, P. C. Heinrich, S. Haan. 2002. Suppressor of cytokine signaling-3 is recruited to the activated granulocyte-colony stimulating factor receptor and modulates its signal transduction. J. Immunol. 169:1219.[Abstract/Free Full Text]
  17. Masuhara, M., H. Sakamoto, A. Matsumoto, R. Suzuki, H. Yasukawa, K. Mitsui, T. Wakioka, S. Tanimura, A. Sasaki, H. Misawa, et al 1997. Cloning and characterization of novel CIS family genes. Biochem. Biophys. Res. Commun. 239:439.[Medline]
  18. Minamoto, S., K. Ikegame, K. Ueno, M. Narazaki, T. Naka, H. Yamamoto, T. Matsumoto, H. Saito, S. Hosoe, T. Kishimoto. 1997. Cloning and functional analysis of new members of STAT induced STAT inhibitor (SSI) family: SSI-2 and SSI-3. Biochem. Biophys. Res. Commun. 237:79.[Medline]
  19. Naka, T., M. Narazaki, M. Hirata, T. Matsumoto, S. Minamoto, A. Aono, S. Nishimoto, T. Kajita, T. Taga, K. Yoshizaki, et al 1997. Structure and function of a new STAT-induced STAT inhibitor. Nature 387:924.[Medline]
  20. Nicholson, S. E., T. A. Willson, A. Farley, R. Starr, J.-G. Zhang, M. Baca, W. S. Alexander, D. Metcalf, D. J. Hilton, N. A. Nicola. 1999. Mutational analyses of the SOCS proteins suggest a dual domain requirement but distinct mechanisms for inhibition of LIF and IL-6 signal transduction. EMBO J. 18:375.[Medline]
  21. Nicholson, S. E., D. De Souza, L. J. Fabri, J. Corbin, T. A. Willson, J.-G. Zhang, A. Silva, M. Asimakis, A. Farley, A. D. Nash, et al 2000. Suppressor of cytokine signaling-3 preferentially binds to the SHP-2-binding site on the shared cytokine receptor subunit gp130. Proc. Nat. Acad. Sci. USA 97:6493.[Abstract/Free Full Text]
  22. Nicos, N., C. J. Greenhalgh. 2000. The suppressors of cytokine signaling (SOCS) proteins: Important feedback inhibitors of cytokine action. Exp. Hematol. 28:1105.[Medline]
  23. Saski, A., H. Yasukawa, T. Shouda, T. Kitamura, I. Dikic, A. Yoshimura. 2000. CIS3/SOCS-3 suppresses erythropoietin (EPO) signaling by binding the EPO receptor and JAK2. J. Biol. Chem. 275:29338.[Abstract/Free Full Text]
  24. Schmitz, J., M. Weissenbach, S. Haan, P.C. Heinrich, F. Schaper. 2000. SOCS3 exerts its inhibitory function on interleukin-6 signal transduction through the SHP2 recruitment site of gp130. J. Biol. Chem. 275:12848.[Abstract/Free Full Text]
  25. Yoshimura, A., T. Ohkubo, T. Kiguchi, N. A. Jenkins, D. J. Gilbert, N. G. Copeland, T. Hara, A. Miyajima. 1995. A novel cytokine-inducible gene CIS encodes an SH2-containing protein that binds to tyrosine-phosphorylated interleukin 3 and erythropoietin receptors. EMBO J. 14:2816.[Medline]
  26. Lioubin, M. N., P. A. Algate, T. Schichkwann, K. Carlberg, R. Aebersold, L. R. Rohrschneider. 1996. p150SHIP, a signal transduction molecule with inositol polyphosphate-5-phosphatase activity. Genes Dev. 10:1084.[Abstract/Free Full Text]
  27. Chacko, G. W., S. Tridandapani, J. Damen, L. Liu, G. Krystal, K. M. Coggeshall. 1996. Negative signaling in B-lymphocytes induces tyrosine phosphorylation of the 145-kDa inositol polyphosphate 5-phosphatase, SHIP. J. Immunol. 157:2234.[Abstract]
  28. Ono, M., S. Bolland, R. Tempst, J. V. Ravetch. 1996. Role of the inositol phosphatase SHIP in negative regulation of the immune system by the receptor Fc{gamma}RIIB. Nature 383:263.[Medline]
  29. Damen, J. E., L. Liu, P. Rosten, R. K. Humphries, A. B. Jefferson, P. W. Majerus, G. Krystal. 1996. The 145-kDa protein induced to associate with Shc by multiple cytokines is an inositol tetraphosphate and phosphatidylinositol 3,4,5-trisphosphate 5'-phosphatase. Proc. Natl. Acad. Sci. USA 93:1689.[Abstract/Free Full Text]
  30. Liu, Q., T. Sasaki, I. Kozieradzki, A. Wakeman, A. Itie, D. J. Dumont, J. M. Penninger. 1999. SHIP is a negative regulator of growth factor-mediated PKB/Akt activation and myeloid cell survival. Genes Dev. 13:786.[Abstract/Free Full Text]
  31. Jacob, A., D. Cooney, S. Tridandapani, T. Kelley, K. M. Coggeshall. 1999. Fc{gamma}RIIb modulation of surface immunoglobulin-induced Akt activation in murine B cells. J. Biol. Chem. 274:13704.[Abstract/Free Full Text]
  32. Tridandapani, S., T. Kelley, M. Pradhan, D. Cooney, L. B. Justement, K. M. Coggeshall. 1997. Recruitment and phosphorylation of SH2-containing inositol phosphatase and Shc to the B-cell Fc{gamma} immunoreceptor tyrosine-based inhibition motif peptide motif. Mol. Cell Biol. 17:4305.[Abstract]
  33. Ho, S. N., H. D. Hunt, R. M. Horton, J. K. Pullen, L. R. Pease. 1989. Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77:51.[Medline]
  34. Pradhan, M., K. M. Coggeshall. 1997. Activation-induced bi-dentate interaction of SHIP and Shc in B-lymphocytes. J. Cell. Biochem. 67:32.[Medline]
  35. Tridandapani, S., M. Pradhan, J. R. LaDine, S. Garber, C. L. Anderson, K. M. Coggeshall. 1999. Protein interactions of Src homology 2 (SH2) Domain-containing inositol phosphatase (SHIP): association with Shc displaces SHIP from Fc{gamma}RIIb in B cells. J. Immunol. 162:1408.[Abstract/Free Full Text]
  36. Dong, F., A. C. Larner. 2000. Activation of Akt kinase by granulocyte colony-stimulating factor (G-CSF): evidence for the role of a tyrosine kinase activity distinct from the Janus kinases. Blood 95:1656.[Abstract/Free Full Text]
  37. White, S. M., M. H. Alarcon, D. J. Tweardy. 2000. Inhibition of granulocyte colony-stimulating factor-mediated myeloid maturation by low level expression of the differentiation-defective class IV granulocyte colony-stimulating factor receptor isoform. Blood 95:3335.[Abstract/Free Full Text]
  38. Phee, H., A. Jacob, K. M. Coggeshall. 2000. Enzymatic activity of the SH2 domain-containing inositol phosphatase is regulated by a plasma membr