Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pelletier, S. Articles by Ihle, J. N. Search for Related Content PubMed PubMed Citation Articles by Pelletier, S. Articles by Ihle, J. N.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, November 2006, p. 8527-8538, Vol. 26, No. 22
0270-7306/06/$08.00+0     doi:10.1128/MCB.01035-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Two Domains of the Erythropoietin Receptor Are Sufficient for Jak2 Binding/Activation and Function

Stéphane Pelletier, Sébastien Gingras, Megumi Funakoshi-Tago, Sherié Howell, and James N. Ihle^*

Department of Biochemistry, St. Jude Children's Research Hospital, Memphis, Tennessee 38105

Received 8 June 2006/ Returned for modification 14 July 2006/ Accepted 1 September 2006

Top Abstract Introduction Materials and Methods Results Discussion References

 
Biochemical and genetic studies have shown that Jak2 is an essential component of EpoR signal transduction which is required for normal erythropoiesis. However, whether Jak2 is the sole direct mediator of EpoR signal transduction remains controversial. To address this issue, we have used an extensive and systematic mutational analysis across the EpoR cytoplasmic tail and transmembrane domain with the goal of determining whether mutants that negatively affected EpoR biological activity but retained Jak2 activation could be identified. Analysis of over 40 mutant receptors established that two large domains in the membrane-proximal region, which include the previously defined Box1 and Box2 domains as well as a highly conserved glycine among cytokine receptors, are required for Jak2 binding and activation and to sustain biological activity of the receptor. Importantly, none of the mutants that lost the ability to activate Jak2 retained the ability to bind Jak2, thus questioning the validity of models of receptor reorientation for Jak2 activation. Also, no correlation was made between cell surface expression of the receptor and its ability to bind Jak2, thus questioning the role of Jak2 in trafficking the receptor to the plasma membrane. Collectively, the results suggest that Jak2 is the sole direct signaling molecule downstream of EpoR required for biological activity.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Activation of the erythropoietin receptor (EpoR) by its ligand erythropoietin (Epo) is critical for the production of erythrocytes (reviewed in reference 13). Embryos lacking either EpoR or Epo die in utero at embryonic day 12.5 due to a severe anemia (16). Fetal liver cells from these embryos have normal numbers of committed erythrocyte colony-forming progenitors (CFU-e), and a small fraction of the cells survive to terminally differentiate, indicating that Epo signaling is required primarily for proliferation and/or survival of the progenitors during differentiation.

One of the biochemical consequences of EpoR ligation is the activation of the receptor-associated tyrosine kinase Janus kinase 2 (Jak2) (15). The requirement for Jak2 in Epo signal transduction was demonstrated most definitively by the consequences of Jak2 gene deletion being similar to those of deletion of EpoR or Epo (11, 12). A variety of studies have led to a model in which Jak2 associates with EpoR at a membrane-proximal region of the cytoplasmic domain. The binding of Jak2 to the receptor has been hypothesized to be required for EpoR processing and cell surface expression (5). It has been hypothesized further that binding of Epo induces a conformational change in the cytoplasmic domain of the receptor and allows the juxtaposition of Jak2 molecules in a manner that facilitates their transphosphorylation within the activation loop, resulting in the activation of the kinase (2, 14). Once activated, Jak2 phosphorylates tyrosine residues in the receptor as well as a number of additional sites on Jak2 (reviewed in references 6 and 13). The ability of the receptor site-specific phosphorylations to recruit mediators of signal transduction has been the focus of a wide variety of studies on EpoR. Although this issue has been studied extensively, considerable controversy regarding which sites and mediators play essential, nonredundant roles in Epo signal transduction remains.

In addition to Jak2, it has been hypothesized that essential mediators that function independently of Jak2 may exist. In particular, several studies have suggested that other cytoplasmic tyrosine kinases, particularly of the Src family of kinases, may play critical roles in Epo signal transduction (reviewed in reference 13). To address this issue, we have undertaken an extensive analysis of mutant receptors with the primary goal of identifying mutants that had lost significant function but retained the ability to activate Jak2. Such mutants would be of considerable value for the identification of Jak2-independent, critical mediators of Epo signaling. As detailed here, we were unable to identify such mutants, leading to the conclusion that the sole function of the Epo receptor is to activate Jak2. Second, the properties of a variety of mutants challenge the hypothesis that Jak2 binding is required for EpoR processing and cell surface expression. Last, the studies fail to support the conclusion that receptor mutants exist that while binding Jak2 are unable to mediate its activation and thus challenge the model that receptor cytoplasmic reorientation is required for aligning Jak2 molecules for activation to occur.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Antibodies and reagents. All reagents used in this study, except those mentioned below, were from Fisher Scientific. M2 monoclonal and anti-Flag polyclonal antibodies were from Sigma Chemicals (St. Louis, MO). Anti-phospho-Jak2 (Tyr1007/1008) was from Bio-Source International. Anti-phospho-extracellular signal-regulated kinase 1 and 2 (ERK1/2) (Thr202/Tyr204), phospho-AKT (Ser473 and Thr308), phospho-Stat5 (Tyr694), and ERK1/2 antibodies were from Cell Signaling Technologies.

Puromycin was purchased from InvivoGen (San Diego, CA). Cell culture media, L-glutamine, and antibiotics were purchased from Gibco-BRL. Fetal bovine serum (FBS) was purchased from Life Technologies (Gaithersburg, MD). RPMI 1640 was purchased from Cell Grow (Herndon, VA). Complete protease inhibitor tablets and horseradish peroxidase-conjugated antihemagglutinin (anti-HA) (clone 3F10) were from Roche Diagnostics (Indianapolis, IN). Laemmli buffer (2x) and polyvinylidene difluoride and nitrocellulose membranes were from Bio-Rad Laboratories (Hercules, CA). Chemiluminescence reagents were purchased from Amersham (Piscataway, NJ). QuikChange XL-II mutagenesis kits were from Stratagene (La Jolla, CA). Sequencing services were performed and mutagenesis oligonucleotides were produced by the Hartwell Center of St. Jude Children's Research Hospital (Memphis, TN).

Cell culture. The interleukin-3 (IL-3)-dependent myeloid cell line DA3 was grown in RPMI 1640 supplemented with heat-inactivated FBS, L-glutamine, penicillin, streptomycin, and mouse recombinant IL-3 (1.81 ng/ml). Stably expressing DA3 cell lines were grown in RPMI 1640 supplemented with heat-inactivated FBS, L-glutamine, penicillin, streptomycin, mouse recombinant IL-3 (1.81 ng/ml), and puromycin (1.81 µg/ml). HEK293T and NIH 3T3 cells were cultured in Dulbecco modified essential medium, high glucose, supplemented with FBS, L-glutamine, ampicillin, and streptomycin. Cell lines were maintained in the exponential growth phase unless indicated otherwise.

Plasmids and expression vectors. The cDNA encoding wild-type (WT) EpoR was obtained from Alan D'Andrea. EpoR cDNA was subcloned into pBS-KS(+), where all of the different mutants of EpoR were generated using conventional molecular biology techniques, and then subcloned into pMSCV-puromycin (BD Biosciences, Palo Alto, CA). The WT and the kinase-inactive mutant of Jak2 (the K882R mutant) were subcloned into pMSCV-hygromycin (BD Biosciences).

Retroviral infections and generation of stable cell lines. HEK293T cells were transfected with the various retroviral vector constructs (6.5 µg) together with the helper virus Phi2 (6.5 µg) by use of the CaPO₄ precipitation method. Twenty-four hours after transfection, the medium was changed to RPMI 1640, and supernatants were harvested every 4 to 6 h for 72 h and filtered through a 0.45-µm filter. Viral titers were determined by infection of NIH 3T3 cells with various concentrations of the viral supernatants and Polybrene (6 µg/ml) and then selected using 1.81 µg/ml puromycin. The titer was calculated as the ratio of puromycin-resistant cells to nonselected cells. Multiplicity of infection was determined as the ratio of viral particles versus the number of cells. DA3 cells were infected twice with the different viruses and Polybrene (6 µg/ml) and then selected with puromycin (1.81 µg/ml). Stable cell lines derived from Jak2^–/– mouse embryonic fibroblasts (MEFs) were derived by following the same procedure and selected using puromycin (1.81 µg/ml) and hygromycin B (200 µg/ml). Fetal liver cells were infected as described below under "Colony formation assay."

EpoR cell surface expression. Exponentially growing DA3 cell lines expressing various active and inactive mutants of EpoR were washed twice and resuspended in binding buffer (RPMI 1640, 10% heat-inactivated FBS, and 50 mM HEPES, pH 7.2) at a density of 1 x 10⁷ cells/ml. Aliquots (100 µl) of cell suspension were incubated with 0.1 nM ¹²⁵I-Epo (specific activity of 3,531 Ci/mmol; Amersham Biosciences) in the absence (total binding) or the presence (nonspecific binding) of a 200-fold excess of unlabeled recombinant human Epo. The final volume of the reaction was 150 µl. After an overnight incubation at 4°C, the binding reaction was terminated by sedimenting the cells through 500 µl of dibutyl phthalate oil in an Eppendorf centrifuge at 8,000 rpm. The tubes were frozen on dry ice, and the cell pellets were counted using a gamma counter. Assuming a dissociation constant of 0.5 nM for all receptor mutants, the receptor cell surface density was estimated and reported as estimated receptor number/cell. Cell viability was greater than 95% as assessed by trypan blue exclusion assays. Estimations of cell surface expression of WT EpoR in WT or Jak2-deficient MEFs were performed similarly except that adherent cells were left in a tissue culture plate and washed three times with ice-cold phosphate-buffered saline (PBS) after the overnight incubation in the reaction buffer before the bound ¹²⁵I-Epo was extracted with 500 µl of 0.1 N NaOH.

Growth assays. Exponentially growing DA3 cells expressing the different types of EpoR were washed twice with PBS, deprived of IL-3 for 16 h, and then stimulated with 5 U/ml (unless indicated) of recombinant human Epo for the indicated times. Living cells were counted using a Beckman Coulter VI-Cell (Beckman Coulter, Fullerton, CA). The results (except where indicated) are expressed as increase in cell number (stimulation [n-fold]) compared to the number of cells plated on day 0, from triplicates.

Immunoprecipitation and Western blot analysis. Cells were maintained in the exponential growth phase and then starved for 12 to 16 h prior to their stimulation with erythropoietin (1 U/ml), unless specified in the figure legend. After stimulation, both treated and nontreated cells were washed twice with ice-cold PBS and lysed in NP-40 lysis buffer (50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 50 mM sodium fluoride, 5 mM EDTA, 40 mM ß-glycerophosphate, 1 mM sodium orthovanadate, 1% NP-40 [Igepal-CA-630] supplemented with protease inhibitors [Complete; Roche]) at 4°C for 30 min. Lysates were cleared by centrifugation at 12,000 rpm (Baxter) at 4°C for 5 min. The cleared lysates were mixed with an equal volume of 2x Laemmli buffer and denatured at 95°C for 5 min. Samples were run on either 7.5 or 10% sodium dodecyl sulfate-polyacrylamide gels and transferred to nitrocellulose membranes. For immunoprecipitation assays, cells were washed twice with ice-cold PBS and lysed in NP-40 lysis buffer (50 mM Tris-HCl, 150 mM NaCl, and 1% NP-40 supplemented with protease inhibitors [Complete; Roche]). Lysates were cleared by centrifugation as described above and subjected to immunoprecipitation using EZ-view M2 agarose conjugated from Sigma. Immunoprecipitated protein complexes were eluted using the Flag peptide (Sigma) as described by the manufacturer. Samples were run on either 7.5 or 10% sodium dodecyl sulfate-polyacrylamide gels and transferred to nitrocellulose or polyvinylidene difluoride membranes. For Western blotting, membranes were blocked at room temperature for 1 h with 5% nonfat milk in TBST (50 mM Tris-HCl, pH 7.6, 150 mM NaCl, 0.1% Tween). Membranes were washed three times with TBST and probed with the various phosphospecific antibodies overnight at 4°C or for 2 h at room temperature for other antibodies.

Colony formation assay. Embryonic day 12.5 EpoR^–/– fetal liver cells were isolated as previously described (12) and transduced with the different viruses (same titers) and RetroNectin (5 µg/cm²) in 24-well plates by using the spin infection protocol described by the manufacturer (TaKaRa, Japan). Twenty-four hours after infection, cells were washed twice and resuspended in MethoCult M3334 and plated at a density of 50,000 cells per 3.5-cm petri dish (two dishes per construct). Dishes were then incubated at 37°C, 95% humidity, and 5% CO₂ for 3 to 4 days. Colonies (CFU-e) were stained with benzidine and counted as described previously (12).

Statistical analysis and sequence alignment. Protein sequence alignments were performed using Vector NTI software 9.0.0 with the blosum62mt2 matrix. Statistical analyses were performed using Graph Pad Prizm 3.0 software. Sigmoid curves were fitted to concentration-response data to generate estimates of EC₅₀ and E_max values as follows: response = (E_maxxⁿ)/(EC₅₀ + xⁿ), where x is the agonist concentration, E_max is the maximal response, n is the Hill coefficient, and EC₅₀ is the concentration of agonist producing half-maximal stimulation. Unless mentioned in the figure legend, all experiments were performed at least three times and representative data are presented.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Generation of a series of EpoR mutations and analysis of biological activity. Genetic and biochemical evidence strongly supports the concept that Jak2 is an essential component of EpoR signal transduction. However, controversy exists regarding the role of additional kinases or signaling proteins for EpoR function, independent of Jak2 function. To address this question, we used a genetic approach in which we generated an array of receptor mutations in the transmembrane domain (TM), the juxtamembrane region, and the cytoplasmic tail and examined them for function with the goal of identifying mutants that retained the ability to associate with and activate Jak2 but had lost biological activity. Our analysis focused on the membrane-proximal half of the cytoplasmic domain, which lacks any tyrosines, since previous studies (17), both in cell lines and in vivo, demonstrated that the distal-half region of the cytoplasmic tail and receptor tyrosines are not critical for function.

The first series of mutants were derived from the systematic replacement of five amino acids with five alanines, starting with histidine 249 next to the transmembrane domain and continuing this replacement throughout the minimal cytoplasmic tail. A carboxyl-terminal Flag tag (-c-Flag) was added to facilitate the detection of the mutants. From this, we obtained 25 mutants that were named based on the number of the first substituted amino acid (Fig. 1A). Once the cDNAs encoding the mutant receptors were subcloned into a retroviral vector containing a puromycin resistance gene and transduced into an IL-3-dependent myeloid cell line (DA3), stable cell lines were established. All of the mutant receptors were expressed at comparable levels in the individual cell lines and were at levels comparable to that seen with cells expressing the wild-type receptor or receptors that had the distal half of the cytoplasmic domain deleted (EpoR-H) or the distal half deleted and the remaining tyrosine (Y³⁴³) mutated to a phenylalanine (EpoR-HM) (Fig. 1B). A variety of the cell lines were also examined for levels of cell surface expression of receptors by binding of labeled Epo and were found to have comparable levels of cell surface expression (see Fig. 7).

View larger version (58K): [in this window] [in a new window]

FIG. 1. Outline of the WT and the H, HM, and alanine scanning mutant receptors. (A) Schematic representation of the WT and the H and HM mutant receptors and sequence alignment of the various alanine scanning mutants. D1, extracellular domain 1, which is required for Epo binding; D2, extracellular domain 2, which is required for Epo binding; JM, juxtamembrane domain; Box1, highly conserved amino acid sequence in cytokine receptor family; Box2, highly conserved amino acid sequence in cytokine receptors; Y-F, Y343F. The alanine scanning mutants are named based on the number of the first amino acid changed by an alanine. The number in parentheses is the amino acid position in the alignment for the immature form of the polypeptide. (B) Whole-cell extracts, prepared from exponentially growing DA3 cell lines expressing or not expressing (pMSCVpuro) the various EpoR constructs, were subjected to immunoprecipitation (IP) using the anti-Flag antibody (M2) and analyzed by Western blotting (WB) using the M2 antibody. EpoR-WT, DA3 cells expressing WT EpoR; EpoR-WT-c-Flag, DA3 cell line expressing WT-c-Flag EpoR; 249, DA3 cell line expressing the EpoR-HM mutant 249; etc.

View larger version (14K): [in this window] [in a new window]

FIG. 7. Jak2 association with EpoR does not affect EpoR cell surface expression. (A) Cell surface expression of biologically active (WT, H, HM, +1A, +2A, 289, 294, and 309 mutants) or inactive (249, 254, 264, 269, 274, 279, 284, 299, 304, and W258A mutants) mutant receptors in DA3 cells. (B) Cell surface expression of WT EpoR in WT or Jak2-deficient MEFs. Receptor density at the cell surface was determined as described in Materials and Methods, and results are expressed as means ± standard errors of the means.

The mutant receptor-expressing cell lines were initially tested for the ability to proliferate in response to Epo (Fig. 2A). Introduction of mutations from the transmembrane domain (mutant 249) up to amino acid 288 and from amino acid 299 to 308 resulted in a loss of function of the receptor, while mutations elsewhere had little or no effect on the biological activity of the receptor. Among the cell lines that had the ability to proliferate in response to Epo, concentration-response curves were assessed in order to identify the more sensitive or less sensitive receptors (Fig. 2B). The majority had concentration-response curves identical to that seen with the EpoR-HM mutant, while mutants 289 and 309 required higher concentrations of Epo for comparable growth. Interestingly, these mutants were on the boundaries of mutants that were inactive/active. Together, the results identified two domains (domains 1 and 2) that are critical for receptor function (Fig. 3A).

View larger version (22K): [in this window] [in a new window]

FIG. 2. Two functional domains are required for the biological activity of EpoR. (A) Exponentially growing DA3 cell lines were starved for 12 to 16 h and then treated with Epo (5 U/ml). Cell numbers were counted every 24 h over a period of 3 days. The results are expressed as increase in cell number (stimulation [n-fold]) compared to the number of cells plated on day 0 (for Epo stimulation) and represent the means of triplicate results for each time point and each cell line. (B) Exponentially growing DA3 cell lines that responded to Epo (as presented in panel A) were starved for 12 to 16 h and then stimulated with increasing concentrations of Epo (0.001 to 5 U/ml) for 3 days. Cell numbers were counted after 3 days. The results are expressed in percentages of maximal response and represent the means of triplicate results for each concentration. (C) EpoR-deficient fetal liver cells were infected with empty or EpoR-expressing retroviruses and subjected to in vitro colony assays with Epo (5 U/ml). Benzidine-positive CFU-e were scored at days 2 to 3. Results are expressed as the percentages of colonies observed compared to WT EpoR transduction and represent the means ± standard errors of the means from three independent experiments. Descriptions of designations are given in the legend for Fig. 1.

View larger version (65K): [in this window] [in a new window]

FIG. 3. Domain 1 contains a highly conserved glycine which is essential for biological activity. (A) Schematic representation of domains 1 and 2, required for biological activity of the receptor. Abbreviations are defined in the legend for Fig. 1A. (B) Sequence alignment of various cytokine receptors and species variants, illustrating the high degree of conservation of glycine 277 (black arrow) and proline 262 (open arrowhead). Abbreviations: GHR, growth hormone receptor; PRLR, prolactin receptor; TSLPR, thymic stromal lymphopoietin receptor. Prefixes: h, Homo sapiens; m, Mus musculus; x, Xenopus laevis; ss, Sus scrofa; cerf, Cervus elaphus; dr, Danio rerio; oa, Ovis aries; r, Rattus norvegicus; t, Tetraodon nigroviridis; c, Canis familiaris; b, Bos taurus. (C) Whole-cell extracts prepared from exponentially growing DA3 cell lines expressing or not expressing (pMSCVpuro) the various EpoR constructs were subjected to immunoprecipitation (IP) using the anti-Flag antibody (M2) and analyzed by Western blotting (WB) using the M2 antibody. (D) See the legend for Fig. 2A for details.

The above-described experiments utilized a long-term IL-3-dependent myeloid cell line; consequently, it was important to also assess function in normal hematopoiesis. We therefore evaluated a number of the mutants for the ability to rescue the EpoR-deficient fetal liver cells in CFU-e and erythrocyte burst-forming unit colony formation (Fig. 2C). Among the mutants examined, there was a complete correlation between the ability to support Epo-induced proliferation of DA3 cells and the ability to rescue Epo-induced CFU-e colony formation.

As another approach to obtain a series of mutants to screen for Jak2-independent functions, we focused on a glycine residue (G²⁷⁷) that is highly conserved in cytokine receptors. This glycine is within the domain 1 region defined by the above-described alanine scanning approach and is between the regions that have been defined as Box1 and Box2 in other studies (6) (Fig. 3B). For our analysis, we mutated G²⁷⁷ to various amino acids, cloned the mutants into retroviral vectors, and derived stable DA3 cell lines expressing the receptor mutants. The various mutants were expressed at comparable levels in the cell lines (Fig. 3C), although the mutants had a range of biological activities when assessed for the ability to support Epo-induced cell growth (Fig. 3D). In particular, the G²⁷⁷W mutation has little effect on the ability to support Epo-induced cell growth, while the G²⁷⁷A and G²⁷⁷C mutations significantly compromised receptor function, although they allowed some growth. Last, a number of the mutations fully inactivated receptor function (G²⁷⁷V, G²⁷⁷P, G²⁷⁷R, G²⁷⁷Q, and G²⁷⁷E).

As a third source of EpoR mutations to incorporate into our studies, we constructed mutants based on previous studies that had focused on an 11-amino-acid EpoR cytosolic juxtamembrane domain (2). A single alanine residue scanning across this region identified the W²⁵⁸A mutant, which has been reported to have the interesting properties of not affecting the binding of Jak2 to the receptor and showing a reduced ability to activate Jak2 while completely inactivating receptor function (2) (Fig. 4A). More interestingly, insertion of one or two alanines after R²⁵¹ identified mutants (the +1A and +2A mutants) (Fig. 4A) that were reported to inactivate receptor function but that retained the ability to bind and activate Jak2. On the basis of these mutations, it was proposed that binding of Epo induced a reorientation of Jak2 kinases to allow activation. In our studies, all of the receptors were expressed at comparable levels (Fig. 4B) and the W²⁵⁸A mutation inactivated the receptor function as predicted (Fig. 4C and D). However, neither the +1A mutant nor the +2A mutant was significantly impaired in biological activity (Fig. 4C and D). The possible basis for the difference in these results is discussed below, as is the significance of the results for receptor-mediated reorientation models.

View larger version (24K): [in this window] [in a new window]

FIG. 4. Outline and biological activities of the W258A, +1A, and +2A EpoR mutants and the EpoR-R2(TM) and EpoR-gp130(TM) transmembrane swap mutants. (A) Schematic representation of W258A, +1A, +2A, EpoR-R2(TM), and EpoR-gp130(TM) mutants and sequence alignment of EpoR compared to the transmembrane swap mutants. (B) Whole-cell extracts, prepared from exponentially growing DA3 cell lines expressing or not expressing (pMSCVpuro) the various EpoR constructs, were subjected to immunoprecipitation (IP) using the anti-Flag antibody (M2) and analyzed by Western blotting (WB) using the M2 antibody. (C) Exponentially growing DA3 cells were starved for 12 to 16 h and then stimulated with increasing concentrations of Epo (0.001 to 10 U/ml) for 3 days. Cell numbers were counted after 3 days, and the results are expressed as the increase (n-fold) in cell number. (D) EpoR-deficient fetal liver cells were infected with empty or EpoR-expressing retroviruses and subjected to in vitro colony assays with Epo (5 U/ml). Benzidine-positive CFU-e were scored at days 2 and 3. Results are expressed as the percentages of colonies observed compared to WT EpoR transduction and represent the means ± standard errors of the means from three independent experiments. Designations and abbreviations are defined in the legend for Fig. 1.

Finally, we generated two mutants in which the transmembrane domain of EpoR was substituted with the transmembrane domain of either gp130 or the R2 subunit of the gamma interferon (IFN-) receptor (IFN-R2) (Fig. 4A). These mutations were predicated on previous studies (14) that suggested that the transmembrane domain is critical in orienting the juxtamembrane domain and allowing full receptor activation. The sequences of the three membrane domains are quite divergent, and the IFN-R2 transmembrane domain is predicted to be one amino acid shorter (Fig. 4A). Somewhat surprisingly, EpoR containing the transmembrane domain of either IFN-R2 or gp130 was fully biologically active (Fig. 4C and D), suggesting the absence of a requirement for a specific amino acid sequence in the transmembrane domain for precise orientation and function.

Correlation between the ability to bind and activate Jak2 with biological activity of the receptor. Having defined a series of mutations that affected the biological activity of EpoR, we determined the ability to activate Jak2 as assessed by tyrosine phosphorylation of the activation loop tyrosine (Y¹⁰⁰⁷/Y¹⁰⁰⁸). The results obtained with the alanine scanning mutations are illustrated in Fig. 5. As demonstrated by the results, there was a correlation between the retention of biological activity and the ability to activate Jak2. More specifically, none of the mutants that lost biological activity retained the ability to activate Jak2. As illustrated in Fig. 5, comparable results were obtained with the G²⁷⁷ substitution mutations and again there was a correlation between the extent of retention of biological function and the activation of Jak2. For example, G²⁷⁷C retains some biological activity and weakly activates Jak2, G²⁷⁷A retains more biological activity and more strongly activates Jak2, and G²⁷⁷W retains full biological activity and fully activates Jak2. Consistent with previous studies, Jak2 was not activated by the W²⁵⁸A mutant whereas the +1A and +2A mutants fully activated Jak2 as well as downstream signaling (data not shown). Last, both transmembrane substitutions were fully able to activate Jak2 (data not shown). The results, as discussed below, are consistent with the hypothesis that the sole function of EpoR is to facilitate the activation of Jak2, which in turn activates critical downstream signaling events independently of tyrosine phosphorylation and/or association of other signaling molecules to the receptor.

View larger version (53K): [in this window] [in a new window]

FIG. 5. Correlation between Jak2 activation and biological activity of the receptor. DA3 cell lines expressing the various alanine scanning and G277 mutants were starved for 12 to 16 h and then stimulated with Epo (1 U/ml) for the indicated times. Whole-cell extracts were analyzed by immunoblotting using an anti-phospho-Y1007/1008 Jak2 antibody. The same extracts were also analyzed by immunoblotting using an anti-Jak2 antibody. WB, Western blotting.

A variety of the mutants were also examined for the ability to bind to Jak2 (Fig. 6). As illustrated in Fig. 6, there was a complete correlation between the ability to bind Jak2 and the ability to activate Jak2 and biological activity among all of the alanine scanning mutants examined (Fig. 6A) and the G²⁷⁷ substitution mutants examined (Fig. 6B). With the mutants that had reduced responses to Epo (mutants 289 and 309), the ability to detect mutant EpoR binding to Jak2 was significantly decreased. One mutant (mutant 294) that retained biological activity was reduced in its ability to bind Jak2 under the conditions of the assay. It was proposed previously that EpoR mutants that can bind but cannot activate Jak2 following Epo binding exist, including the W²⁵⁸A mutant (2). However, the W²⁵⁸A mutant either does not bind Jak2 or binds Jak2 with considerably less affinity (Fig. 6C) than was reported previously for this mutant (2).

View larger version (55K): [in this window] [in a new window]

FIG. 6. Correlation between Jak2 binding and biological activity of the receptor. IL-3-deprived DA3 cell lines expressing or not expressing (pMSCVpuro or MSCV) the EpoR-HM mutant (HM) and various biologically inactive and some active alanine scanning mutants (A), G277 mutants (B), and the W258 mutant (C) were left untreated or treated with Epo (1 U/ml) for 15, 30, or 60 min before solubilization. Cell lysates were subjected to immunoprecipitation (IP) using the anti-Flag antibody (M2). EpoR and associated Jak2 were analyzed by Western blotting (WB) using the M2 and anti-Jak2 antibodies, respectively.

It has been speculated that interaction of Jak2 with the cytoplasmic tail of EpoR in the Golgi body is required for the proper processing and cell surface expression of the receptor. Since all biologically inactive mutant receptors were unable to associate with Jak2, the possibility that some critical direct mediators of EpoR signal transduction may not be activated due to the improper cell surface targeting of the mutants existed. To test that possibility, we estimated the cell surface expression levels of various biologically inactive and active EpoR mutants to determine whether a correlation between cell surface expression and association with Jak2 could be established. Consistent with previous reports (9) but in contrast to more-recent speculations (5), no correlation between cell surface expression of the receptors and ability to bind Jak2 has been established (Fig. 7A). Moreover, no major differences exist between EpoR cell surface expression levels from WT and Jak2-deficient MEFs (Fig. 7B).

Downstream signaling from mutant Epo receptors. The above-described studies demonstrated a striking correlation of biological activity with the binding and activation of Jak2. However, it is possible that activation of some signal transduction pathways may be receptor mediated by Jak2-independent mechanisms but that their activation is not sufficient for maintaining biological activity. We therefore examined a series of mutants for the ability to activate specific signaling pathways. Initially, we examined the abilities of signaling proteins that have been reported to be recruited to the receptor complex by receptor tyrosines through Src homology 2 domains and become tyrosine phosphorylated. Among these predicted proteins, p85, the regulatory subunit of phosphatidylinositol 3 kinase (PI3K) activity, was not detectably phosphorylated in DA3 cells (data not shown). As predicted from previous studies (10), the induction of tyrosine phosphorylation of p52-Shc was dependent upon the distal region of the receptor (Fig. 8A). Stat5 phosphorylation was rapidly and strongly induced in cells expressing the wild-type receptor or the carboxyl-truncated EpoR-H mutant receptor, consistent with previous results (Fig. 8B). However, in DA3 cells expressing the truncated, tyrosine-less HM mutant receptor, there was a delayed and weak induction of tyrosine phosphorylation of Stat5, suggesting that Stat5 can be recruited, albeit less efficiently, to the receptor complex through mechanisms not involving Src homology 2 domain recognition of EpoR Y³⁴³. This activation was nevertheless dependent upon Jak2, as demonstrated by the complete correlation of Stat5 phosphorylation with Jak2 activation among the alanine scanning mutations (Fig. 8C).

View larger version (39K): [in this window] [in a new window]

FIG. 8. Proximal downstream signaling from EpoR mutants. (A) IL-3-deprived DA3 cell lines expressing WT EpoR and the EpoR-H and EpoR-HM mutants were left untreated or treated with Epo (2 U/ml) for the indicated times before solubilization. Cell lysates were subjected to immunoprecipitation (IP) with an anti-Shc antibody, and tyrosine phosphorylation was analyzed by Western blotting (WB) using antiphosphotyrosine antibody 4G10. Immunoprecipitated Shc proteins were also analyzed by Western blotting using an anti-Shc antibody. (B and C) IL-3-deprived DA3 cell lines expressing WT EpoR and the EpoR-H and EpoR-HM mutants (B) and alanine scanning mutants (C) were left untreated or treated with Epo (1 U/ml) for the indicated times. The activation of Stat5 was monitored by immunoblotting with phosphospecific antibodies against the activating tyrosine phosphorylation site (Tyr694). Descriptions of designations are given in the legend for Fig. 1.

Epo stimulation is known to activate a number of signaling pathways by less defined mechanisms, including the activation of Erk1/2. As illustrated in Fig. 9A, phosphorylation of Erk1/2 is rapidly and strongly induced in response to Epo in cells expressing the wild-type receptor or the truncated H mutant receptor. Induction of phosphorylation is also observed with the truncated, tyrosine-less mutant (the HM mutant), although it is slightly delayed and marginally reduced. To explore the possibility that this activation may be mediated by Jak2-independent mechanisms, we assessed induced phosphorylation in cells expressing the spectrum of alanine scanning mutations (Fig. 9B). As shown here, there was a complete correlation between Erk1/2 phosphorylation and Jak2 activation or, conversely, no Erk1/2 phosphorylation was observed with cells expressing mutant receptors that were not able to bind and activate Jak2 (Fig. 9B). Similar results were obtained in the analysis of Epo-induced serine phosphorylation of Akt (Fig. 9B).

View larger version (48K): [in this window] [in a new window]

FIG. 9. Distal downstream signaling from EpoR. IL-3-deprived DA3 cell lines expressing WT EpoR and the EpoR-H and EpoR-HM mutants (A) and alanine scanning mutants (B) were left untreated or treated with Epo (1 U/ml) for the indicated times. The activating phosphorylation of Erk1/Erk2 and Akt was monitored by immunoblotting of lysate proteins with phosphospecific antibodies as indicated. (C) Exponentially growing Jak2–/– MEFs stably expressing or not expressing Jak2-WT-c-HA or the kinase-inactive Jak2-KD-c-HA mutant (KD, kinase dead) and with or without EpoR-WT-c-Flag and HM mutants were starved for 12 to 16 h and then stimulated with Epo for the indicated times. The activating phosphorylation of Erk1/Erk2 and Akt was monitored by immunoblotting of lysate proteins with phosphospecific antibodies as indicated. WB, Western blotting. Descriptions of designations are given in the legend for Fig. 1.

To further establish the essential role for Jak2 in the activation of Erk1/2 and Akt, we utilized Jak2-deficient MEFs. As illustrated in Fig. 8C, introduction of the Epo receptor (either the wild type or the HM mutant) into Jak2-deficient MEFs and stimulation with Epo did not result in Erk1/2 phosphorylation or in the serine phosphorylation of Akt. However, introduction of wild-type Jak2 into the Jak2-deficient MEFs with either wild-type or HM mutant Epo receptors restored the Epo-induced activation of Erk1/2 and Akt. Activation of both required the kinase activity of Jak2, since introduction of a kinase-dead Jak2 mutant did not restore the ability to activate either Erk1/2 or Akt.

Top Abstract Introduction Materials and Methods Results Discussion References

 
Genetic and biochemical evidence has demonstrated the essential role that Jak2 plays in EpoR function, but a number of studies have also suggested that there may exist additional, non-Jak2-dependent functions of EpoR (13). Therefore, the primary goals of these experiments were to provide evidence for non-Jak2-dependent functions and to obtain receptor mutants that would allow approaches to characterize the proteins involved. To these ends, we examined a large spectrum of receptor mutants in an effort to identify mutations that affected receptor function without affecting the ability of the receptor to associate with and activate Jak2. The striking conclusion from the results is that such a mutation did not exist among a large collection of mutations that encompassed the entire functional region of the cytoplasmic domain. This is further supported by the observation that there were no detectable downstream signaling events in Epo-stimulated Jak2-deficient fibroblasts expressing EpoR. Therefore, we would hypothesize that the only essential function of EpoR is to bind and activate Jak2, which in turn activates critical downstream signaling events independently of other receptor domains. However, we cannot rule out the possibility that another cofactor binds within the same domains that are required for Jak2 binding, although, given the number of mutants examined, we feel that this possibility is unlikely.

The results also address several concepts related to the role of EpoR-Jak2 interaction. In particular, it was proposed that the N-terminal domain of Jak2 was required for Golgi body processing and cell surface expression of EpoR (5), and this was viewed as a novel type of quality control of receptor kinase signaling (1). Our results, obtained with a variety of mutant EpoR constructs that fail to interact with Jak2, fail to demonstrate a role for Jak2 in the cell surface expression of EpoR, since all of the mutant receptors examined were expressed at the cell surface comparably to those that bound Jak2. More directly, we also demonstrated that EpoR is expressed at the cell surface in Jak2-deficient fibroblasts at levels comparable to those seen with wild-type fibroblasts.

The hypothesis that Jak2 was required for EpoR cell surface expression was also not supported by experiments performed several years ago (9). These studies demonstrated that EpoR mutants that lack the transmembrane domain and the cytoplasmic domain were processed properly and secreted. Consistent with our results, Jak2 has also been shown to not be required for the cell surface expression of the growth hormone receptor (3). Similarly, it has been demonstrated that Jak3 is not required for membrane expression of its partner cytokine receptor, the _c receptor (4). The basis for the difference between conclusions of Huang et al. (5) and those of other studies is not known, but it should be noted that the studies of Huang et al. utilized only the Jak2-deficient human long-term fibrosarcoma cell line 2A and/or transient expression.

Our studies also address a previously proposed model dealing with the role of the juxtamembrane domain in orienting receptor-associated Jak2 molecules as required for activation (2). Specifically, it was proposed that a hydrophobic motif containing L²⁵³, I²⁵⁷, and W²⁵⁸ is critical for Jak2 activation and that the -helical structure given by this region is essential for the ability of the associated and activated Jak2 to phosphorylate the receptor. The hypothesis was based on the observation that alanine insertions after R²⁵¹ would be predicted to change the orientation of these residues and thus the orientation of the associated Jak2. In the experiments presented, insertion of one (the +1A mutant) or two (the +2A mutant) alanines was shown to still allow Jak2 association and activation but this activation was not associated with receptor phosphorylation or biological activity. In contrast to previously reported studies, we have found that both the +1A and +2A mutants are fully biologically active; thus, the central observation supporting the reorientation theory was not substantiated in DA3 cells, suggesting the possibility of cell line differences. However, we have also examined these mutations in Ba/F3 cells and obtained dose-response curves comparable to those shown for DA3 cells. In this regard, it should be noted that in the studies described by Constantinescu et al. (2) the complete dose-response curves were not presented, and the slight shift in responses we also observe without the full dose-response curve suggested a much more profound effect on biological activity than is present.

The model proposed by Constantinescu et al. (2) also relied on the observation that mutation of W²⁵⁸ to alanine did not abolish binding of Jak2 but impaired the ability to activate Jak2. Consistent with their results, we found that the W²⁵⁸A mutation inactivates receptor function. However, in contrast to the data that were indicated in the article as "data not shown" stating that the mutant still bound Jak2, we find that this mutant fails to bind Jak2. Again, the basis for the differences is not known. Importantly, in a wide spectrum of mutants which we have identified that affect receptor function, all those that have lost activity have lost the ability to associate with Jak2. Moreover, the identification of two relatively large domains that, when mutated, can eliminate Jak2 binding suggests that the structure that binds Jak2 is more complex than a simple hydrophobic patch of a proposed -helical domain in the juxtamembrane region.

The model proposed by Constantinescu et al. (2) has the central concept that EpoR is a dimer and that binding of Epo does not affect the extent of receptor aggregation but rather transduces a conformational change through the transmembrane to the juxtamembrane region. This conformation changes the orientation of the associated Jak2 molecules such that tyrosine phosphorylation of the activation loop could occur. The assumption that EpoR remains a dimer following ligand binding may be incorrect. In particular, studies that examined the status of the tyrosine-phosphorylated forms of EpoR following Epo binding demonstrated that much of the receptor is found in disulfide-linked oligomeric forms (9). Recent studies (8) utilized a cysteine-scanning mutagenesis of the extracellular, juxtamembrane, and transmembrane domains to demonstrate that introduction of additional opportunities for disulfide linking of receptors could result in receptor activation. Therefore, an alternative model is one in which Jak2 activation is facilitated by the formation of oligomeric complexes of receptor dimers and in which it is the increased concentration of Jak2 molecules brought about by aggregation that results in Jak2 transphosphorylation and activation.

Subsequent studies by Seubert et al. (14) also proposed a role for the transmembrane domain in correctly orienting the juxtamembrane region to allow Jak2 activation. These studies relied on the use of fusions of the leucine zipper, coiled-coil domain of the Put3 transcription factor with the transmembrane domain and the cytoplasmic domain of EpoR. Deletions of the N-terminal region of the EpoR transmembrane domain in these constructs were shown to differentially affect activation of Jak2 or downstream signaling events. From the conclusion of these studies, it would not have been predicted that one could substitute highly unrelated transmembrane domains from other cytokine receptors and retain function, as was seen in our studies. In particular, it should be noted that the gp130 receptor associates with Jak1 and not Jak2. Since the most active of the Put3-EpoR fusion proteins activated Jak2 or Stat5 only weakly relative to controls, it is possible that the results are relevant only to the fusion proteins that were used and are not indicative of the role of the transmembrane domain in EpoR.

One of the central paradigms in cytokine signaling has been that receptor-mediated activation of associated tyrosine kinases results in receptor tyrosine phosphorylation and that this is essential for recruitment and the subsequent activation of critical signaling pathways. In the case of EpoR, this concept was initially challenged by the derivation of a strain of mice containing a genetically modified receptor that lacked the distal half of the cytoplasmic domain and had a mutation of the only residual tyrosine (Y³⁴³) to a phenylalanine. This mutation marginally affected erythropoiesis, and the effects seen could be ascribed to the loss of the ability to efficiently recruit and activate Stat5a/b (17). Consistent with these and other studies (13), the ability to recruit and activate Stat5a/b was only weak, with delayed kinetics. The basis for the low level of activation in these studies is not known but may be related to levels of expression of the transduced mutants that were higher than the levels of expression of the endogenous wild-type receptor or genetically modified endogenous loci. Irrespectively, as demonstrated by our results, the ability of EpoR to weakly activate Stat5a/b in the absence of receptor tyrosines was completely correlated with the ability to recruit and activate Jak2. Thus, activation may occur through the recruitment to the complex by sites of tyrosine phosphorylation of Jak2 or, alternatively, activated Jak2 may activate other receptor systems that have the ability to recruit and activate Stat5a/b.

Activation of the mitogen-activated kinases Erk1 and Erk2 has been seen consistently in studies of EpoR signal transduction, as reviewed previously (13). However, the pathways involved in activation have been more controversial and have included studies that suggested that Jak2 may not be involved. Our studies demonstrate that the distal region of the receptor and receptor tyrosines are not required for activation, although their absence did consistently result in a slightly delayed activation. Erk1/2 activation, however, was absolutely correlated with the ability of the receptor to bind and activate Jak2. In particular, our studies demonstrated an essential role for a receptor tyrosine (Y³⁴³) in the recruitment of Shc but ruled out the requirement for this recruitment in a pathway that results in Erk1/2 activation. Our results are not consistent with studies that implicated receptor Y⁴⁷⁹ in Erk1/2 activation. One study (7) focused on the role of Y⁴⁷⁹ in recruiting the regulatory subunit of PI3K, p85, and the requirement for PI3K activity for Erk1/2 activation. Our studies do not support a role for receptor tyrosines, and we have not found that tyrosine phosphorylation of p85 is inducible in our cell system (data not shown). More recently, it has been proposed that EpoR Y⁴⁷⁹ couples to Erk1/2 activation through the recruitment of phospholipase C and that, in the cell systems used, inhibition of PI3K activity did not affect Erk1/2 activation (7).

The possibility also exists that, as a consequence of activating Jak2, tyrosine phosphorylation sites on Jak2 or a receptor-associated substrate are required for the subsequent recruitment and activation of signaling pathways. With regard to receptor-associated substrates, it has been proposed previously that insulin receptor substrate 1 (IRS1) or IRS2 may play an essential role in Epo-regulated erythropoiesis (13). However, neither IRS1 nor IRS2 protein is expressed in DA3 cells. Consequently, neither is required for Epo-induced cell proliferation. Additionally, mice lacking both IRS1 and IRS2 have been derived and have not been reported to have defects in erythropoiesis. If an essential receptor-associated substrate existed, we would have anticipated identifying an EpoR mutant that had lost the ability to associate with the substrate but retained the ability to bind and activate Jak2. Clearly, such a mutant was not observed. Thus, it is unlikely that these insulin receptor-associated substrates play an essential role in Epo signaling.

 
This work was supported by Cancer Center CORE grant CA21765, by grants RO1 DK42932 and PO1 HL53740 to J.N.I., and by the American Lebanese Syrian Associated Charities (ALSAC).

 
* Corresponding author. Mailing address: Department of Biochemistry, St. Jude Children's Research Hospital, 332 North Lauderdale, Memphis, TN 38105. Phone: (901) 495-2474. Fax: (901) 525-8025. E-mail: jihle.mcb{at}stjude.org.

Published ahead of print on 18 September 2006.

Top Abstract Introduction Materials and Methods Results Discussion References

 

1
1. Bonifacino, J. S. 2002. Quality control of receptor-kinase signaling complexes. Dev. Cell 2:1-2.[CrossRef][Medline]
2
2. Constantinescu, S. N., L. J. Huang, H. Nam, and H. F. Lodish. 2001. The erythropoietin receptor cytosolic juxtamembrane domain contains an essential, precisely oriented, hydrophobic motif. Mol. Cell 7:377-385.[CrossRef][Medline]
3
3. He, K., X. Wang, J. Jiang, R. Guan, K. E. Bernstein, P. P. Sayeski, and S. J. Frank. 2003. Janus kinase 2 determinants for growth hormone receptor association, surface assembly, and signaling. Mol. Endocrinol. 17:2211-2227.[Abstract/Free Full Text]
4
4. Hofmann, S. R., A. Q. Lam, S. Frank, Y.-J. Zhou, H. L. Ramos, Y. Kanno, D. Agnello, R. J. Youle, and J. J. O'Shea. 2004. Jak3-independent trafficking of the common chain receptor subunit: chaperone function of Jaks revisited. Mol. Cell. Biol. 24:5039-5049.[Abstract/Free Full Text]
5
5. Huang, L. J., S. N. Constantinescu, and H. F. Lodish. 2001. The N-terminal domain of Janus kinase 2 is required for Golgi processing and cell surface expression of erythropoietin receptor. Mol. Cell 8:1327-1338.[CrossRef][Medline]
6
6. Ihle, J. N., W. Thierfelder, S. Teglund, D. Stravapodis, D. Wang, J. Feng, and E. Parganas. 1998. Signaling by the cytokine receptor superfamily. Ann. N. Y. Acad. Sci. 865:1-9.[CrossRef][Medline]
7
7. Klingmuller, U., H. Wu, J. G. Hsiao, A. Toker, B. C. Duckworth, L. C. Cantley, and H. F. Lodish.1997 . Identification of a novel pathway important for proliferation and differentiation of primary erythroid progenitors.Proc. Natl. Acad. Sci. USA 94:3016-3021.[Abstract/Free Full Text]
8
8. Lu, X., A. W. Gross, and H. F. Lodish.2006 . Active conformation of the erythropoietin receptor: random and cysteine-scanning mutagenesis of the extracellular juxtamembrane and transmembrane domains. J. Biol. Chem. 281:7002-7011.[Abstract/Free Full Text]
9
9. Miura, O., and J. N. Ihle. 1993. Dimer- and oligomerization of the erythropoietin receptor by disulfide bond formation and significance of the region near the WSXWS motif in intracellular transport. Arch. Biochem. Biophys. 306:200-208.[CrossRef][Medline]
10
10. Miura, Y., O. Miura, J. N. Ihle, and N. Aoki. 1994. Activation of the mitogen-activated protein kinase pathway by the erythropoietin receptor. J. Biol. Chem. 269:29962-29969.[Abstract/Free Full Text]
11
11. Neubauer, H., A. Cumano, M. Muller, H. Wu, U. Huffstadt, and K. Pfeffer.1998 . Jak2 deficiency defines an essential developmental checkpoint in definitive hematopoiesis. Cell 93:397-409.[CrossRef][Medline]
12
12. Parganas, E., D. Wang, D. Stravopodis, D. J. Topham, J. C. Marine, S. Teglund, E. F. Vanin, S. Bodner, O. R. Colamonici, J. M. van Deursen, G. Grosveld, and J. N. Ihle. 1998. Jak2 is essential for signaling through a variety of cytokine receptors. Cell 93:385-395.[CrossRef][Medline]
13
13. Richmond, T. D., M. Chohan, and D. L. Barber.2005 . Turning cells red: signal transduction mediated by erythropoietin. Trends Cell Biol. 15:146-155.[CrossRef][Medline]
14
14. Seubert, N., Y. Royer, J. Staerk, K. F. Kubatzky, V. Moucadel, S. Krishnakumar, S. O. Smith, and S. N. Constantinescu. 2003. Active and inactive orientations of the transmembrane and cytosolic domains of the erythropoietin receptor dimer. Mol. Cell 12:1239-1250.[CrossRef][Medline]
15
15. Witthuhn, B. A., F. W. Quelle, O. Silvennoinen, T. Yi, B. Tang, O. Miura, and J. N. Ihle. 1993. JAK2 associates with the erythropoietin receptor and is tyrosine phosphorylated and activated following stimulation with erythropoietin.Cell 74:227-236.[CrossRef][Medline]
16
16. Wu, H., X. Liu, R. Jaenisch, and H. F. Lodish.1995 . Generation of committed erythroid BFU-E and CFU-E progenitors does not require erythropoietin or the erythropoietin receptor. Cell 83:59-67.[CrossRef][Medline]
17
17. Zang, H., K. Sato, H. Nakajima, C. McKay, P. A. Ney, and J. N. Ihle. 2001. The distal region and receptor tyrosines of the Epo receptor are nonessential for in vivo erythropoiesis. EMBO J. 20:3156-3166.[CrossRef][Medline]

---

Molecular and Cellular Biology, November 2006, p. 8527-8538, Vol. 26, No. 22
0270-7306/06/$08.00+0     doi:10.1128/MCB.01035-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Jia, Y., Warin, R., Yu, X., Epstein, R., Noguchi, C. T. (2009). Erythropoietin signaling promotes transplanted progenitor cell survival. FASEB J. 23: 3089-3099 [Abstract] [Full Text]  
• Sathyanarayana, P., Dev, A., Fang, J., Houde, E., Bogacheva, O., Bogachev, O., Menon, M., Browne, S., Pradeep, A., Emerson, C., Wojchowski, D. M. (2008). EPO receptor circuits for primary erythroblast survival. Blood 111: 5390-5399 [Abstract] [Full Text]  
• Grebien, F., Kerenyi, M. A., Kovacic, B., Kolbe, T., Becker, V., Dolznig, H., Pfeffer, K., Klingmuller, U., Muller, M., Beug, H., Mullner, E. W., Moriggl, R. (2008). Stat5 activation enables erythropoiesis in the absence of EpoR and Jak2. Blood 111: 4511-4522 [Abstract] [Full Text]  
• Wernig, G., Gonneville, J. R., Crowley, B. J., Rodrigues, M. S., Reddy, M. M., Hudon, H. E., Walz, C., Reiter, A., Podar, K., Royer, Y., Constantinescu, S. N., Tomasson, M. H., Griffin, J. D., Gilliland, D. G., Sattler, M. (2008). The Jak2V617F oncogene associated with myeloproliferative diseases requires a functional FERM domain for transformation and for expression of the Myc and Pim proto-oncogenes. Blood 111: 3751-3759 [Abstract] [Full Text]  
• Funakoshi-Tago, M., Pelletier, S., Moritake, H., Parganas, E., Ihle, J. N. (2008). Jak2 FERM Domain Interaction with the Erythropoietin Receptor Regulates Jak2 Kinase Activity. Mol. Cell. Biol. 28: 1792-1801 [Abstract] [Full Text]  
• Murray, P. J. (2007). The JAK-STAT Signaling Pathway: Input and Output Integration. J. Immunol. 178: 2623-2629 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pelletier, S. Articles by Ihle, J. N. Search for Related Content PubMed PubMed Citation Articles by Pelletier, S. Articles by Ihle, J. N.