Hanson Institute, Division of Human Immunology, Institute of Medical and Veterinary Science, Adelaide, and Department of Medicine, Adelaide University, South Australia
Correspondence: Thomas J. Gonda, Bionomics Ltd., 31 Dalgleish Street, Thebarton, South Australia 5031. E-mail: tgonda{at}bionomics.com.au
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-subunit (GMR), which is not expressed in CFU-E. Here, we show that coexpression of hßc EC mutants and EpoR in BaF-B03 cells, which lack GMR, results in factor-independent proliferation and JAK2 activation. Mutant receptors that cannot activate JAK2 fail to produce a functional interaction. As there is no detectable phosphorylation of hßc on intracellular tyrosine residues, EpoR displays constitutive tyrosine phosphorylation. These observations suggest that JAK2 activation mediates cross-talk between EC mutants of hßc and EpoR. The implications of these data are discussed as are our findings that activated hßc mutants can functionally interact with certain other cytokine receptors.
Key Words: erythropoietin receptor • factor independence • JAK2 • tyrosine phosphorylation
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
It is within this context that we have considered the possible mechanisms underlying recent observations from our in vivo studies of constitutively activated mutants of hßc (see refs. [13
, 14
] for review). In particular, transgenic [15
] and bone marrow reconstitution [16
] models have revealed that hßc, rendered constitutively active by mutations in the extracellular domain, can induce a profound erythrocytosis in mice. At a cellular level, the erythrocytosis was a result of massively increased numbers and Epo independence of late erythroid progenitors [erythroid colony-forming units (CFU-E)]. However, our previous studies had indicated that this class of hßc mutant [extracellular (EC)] interacted with and required the GM-CSF receptor subunit (GMR) for activity [17
, 18
]. As it is thought that CFU-E do not express GMR [19
], we wondered whether other CRs could also interact with the EC mutants and functionally substitute for GMR.
In this report, we have used BAF-B03 cells (which do not express GMR) to examine the ability of several other CRs to functionally "complement" EC mutants. Our major focus was on the EpoR, as this is the predominant CR in CFU-E and as interactions between EpoR and wild-type (WT) GM-CSF/IL-3Rß subunits have previously been reported [11
, 12
]. We also examined cross-talk between the hßc EC mutants and Mpl, the receptor for Tpo, as in several cases, the mice expressing hßc mutants also exhibited thrombocytosis [15
, 16
]. In view of the results obtained with these two CRs, we tested three further CRs in initial attempts to examine the specificity of functional interactions between the hßc EC mutants and heterologous CRs.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2 Producer cells transfected with retrovirus containing WT human GMR (hGMR), hßc, or hßc-activated mutants, I374N and FI
, were maintained as described previously [20
]. The BAF-B03 subline [21
] of the mouse IL-3-dependent pro-B cell line, Ba/F3, was maintained as described in Jenkins et al. [22
].
Cytokines
hGM-CSF was a gift from Professor Angel Lopez (Hanson Institute, Adelaide, S.A., Australia). Recombinant murine (rm)IL-3 was produced from a baculovirus vector supplied by Dr. Andrew Hapel (John Curtin School of Medical Research, Canberra, A.C.T., Australia). Recombinant human (rh)Epo was purchased from Janssen Cilag (Lane Cove, N.S.W., Australia). rhTpo was purchased from PeproTech Inc. (Rocky Hill, NJ). Human GH (hGH) was a gift from Professor Michael Waters (University of Queensland, St. Lucia, Australia). hG-CSF was obtained from Dr. Andrew Zannettino (Division of Haematology, Institute of Medical and Veterinary Science, Adelaide, Australia). Human IL-6 and soluble human IL-6R were gifts from Dr. Richard Simpson (Ludwig Institute for Cancer Research, Melbourne, Victoria, Australia).
CR expression constructs
The hßc and EpoR mutants used in this study are illustrated in Figure 1
. The hßc mutants I374N and FI have been described previously [22
, 23
]. WT and mutant hßc were cloned into the retroviral expression vector pRufNeo [24
]. hGMR was cloned into the retroviral expression vector pRufPuro [22
]. WT EpoR cDNA and cDNAs encoding mutant EpoR, EpoR321 (see ref. [25
]), and EpoRW282R (see ref. [26
]), cloned in the retroviral expression vector MSCV 2.2, were a gift from Professor Peter Klinken (University of Western Australia, Nedlands). N-terminal Flag-tagged human TpoR (Mpl) was a gift from Nathanial Albanese (Hanson Institute). hGH receptor (hGHR) cDNA, cloned in the retroviral expression vector MSCV IRES Neo, was a gift from Professor Michael Waters (University of Queensland). Human G-CSF receptor (hGCSFR) cDNA cloned in pEF-BOS was a gift from Dr. Judy Layton (Ludwig Institute for Cancer Research). N-terminal Flag-tagged human gp-130 (hgp-130) was a gift from Dr. Doug Hilton (Walter and Eliza Hall Institute of Medical Research, Melbourne, Victoria, Australia). The PGKHygro plasmid has been described by Mortensen et al. [27
].
 View larger version (27K): [in a new window] |
Figure 1. hßc and EpoR constructs. Schematic illustrating the WT and mutant CR constructs used in this work. Each vertical rectangle represents a receptor subunit with the extracellular domains at top and intracellular domains below the cell membrane (gray bar). Y, The positions of intracellular tyrosine residues; solid bars, conserved Boxes 1 and 2 regions. Amino acid (aa) substitutions are indicated in single letter code by residue number. Deletions are indicated by dashed lines and the duplication in FI, by light shading. Note that the double mutants used in this study, as described in the text, combined the individual mutations depicted here.
|
, were generated by polymerase chain reaction (PCR)-based mutagenesis. Briefly, to remove the core nine aa Box 1 sequence, KWEEKIPNP, from hßc, two fragments were initially synthesized. A 5' Box 1 fragment was generated by PCR (25 cycles of 94°C 30'', 62°C 30'', 72°C 3') with primers RCR (CTTCGAAAACCACACTGCTCGGAC) and 5' Box 1 (GAACAGGTGGCTCTTGCTTCTGCGCAGCCTGTA). A 3' Box 1 fragment was generated similarly with primers RCF1 (TTGGGGACTCTGCTGACC) and 3' Box 1 (GGGTACAGGCTGCGCAGAAGCAAGAGCCACCTGTTC). The 5' Box 1 fragment and the 3' Box 1 fragment were mixed, and a second round of PCR (20 cycles of 94°C 30'', 62°C 30'', and 72°C 4') with RCR and RCF1 generated full-length hßc, lacking the 9 aa Box 1 sequence hßcbox1. PCR products were purified, cut with BamHI/HindIII, and cloned back into pRufNeo.
Antibodies/reagents
Monoclonal hßc antibodies were a gift from Professor Angel Lopez (Hanson Institute). Rabbit polyclonal antibody directed against the N-terminal region of mouse EpoR (mEpoR) was a gift from Professor Peter Klinken and Dr. Peta Tilbrook (University of Western Australia). hGHR N-terminal antibody was a gift from Michael Waters (University of Queensland). hGCSFR N-terminal antibody was a gift from Dr. Judy Layton (Ludwig Institute for Cancer Research). Antiphosphotyrosine (4G10) and anti-JAK2 antibodies were purchased from Upstate Biotechnology (Lake Placid, NY). Antiactive JAK2 was purchased from Affinity BioReagents Inc. (Golden, CO). Anti-Flag antibody, M2, was purchased from Sigma Chemical Co. (Castle Hill, N.S.W., Australia). Goat anti-mouse immunoglobulin G (IgG) and anti-rabbit IgG antibodies conjugated to horseradish peroxidase were purchased from Pierce (Rockford, IL). AG-490 JAK2 kinase inhibitor was purchased from Biomol Reasearch Laboratories Inc. (Plymouth Meeting, PA).
Electroporation and infection
Approximately 20 µg plasmid DNA for mEpoR, hMpl, and hGHR was electroporated at 270V/975 µF with a BioRad Gene Pulser II (Hercules, CA) into 107 BAF-B03 cells. Constructs for hGCSFR and hgp-130 were coelectroporated with pGKHygro in a ratio of 10 µg:1 µg as described above. Cells were allowed to recover overnight before drug selection. Where appropriate, cells were further selected for their ability to grow in the cytokine for the specific transfected receptor. Surface expression of the CRs was assessed by flow cytometry as described previously [28
]. Following this selection, BAF-B03 populations expressing the various CRs were infected with retroviruses encoding hGMR, WT hßc, I374N, or FI as described previously [20
]. Surface expression of hßc was assessed and if necessary, cells sorted for hßc expression as described previously [22
].
Cell proliferation assays
BAF-B03 cells were washed twice in PBS and cultured with or without appropriate cytokine. To maintain cell viability, cultures were split as required (usually when density reached between 1 and 1.5x106 cells/ml), and the dilution factor was accounted for in the determination of total cell number. Viable cell numbers were assessed by Trypan blue dye exclusion in a haemocytometer.
Immunoprecipitation and Western blot analysis
Cell lysates were prepared and immunoprecipitated as described previously [28
]. To prepare total cell lysates, cells were treated similarly, except after the initial PBS wash, cells were resuspended in 1x reducing sodium dodecyl sulfate (SDS) sample buffer, sonicated with a Microson ultrasonic cell disruptor (Farmingdale, NY) with a microtip probe for 20 s, boiled for 2 min, and centrifuged at 13,000 g for 5 min. The lysate was fractionated by SDS-polyacrylamide gel electrophoresis (PAGE), and Western blot analysis was carried out as described previously [28
].
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, could allow the hßc EC mutants to function and therefore be a potential cofactor for activated forms of hßc in CFU-E of mice expressing these mutants. To test this, we expressed WT hßc and the activated mutants I374N [22
] and FI [23
] (see also Fig. 1
) in BAF-B03 cells in the presence of mEpoR and tested for receptor activity in the absence of ligand, i.e., for factor independence. To ensure functional EpoR expression, BAF-B03 cells were selected for growth in Epo after infection with a mEpoR retrovirus; subsequently, they were infected with retrovirus encoding the WT or mutant forms of hßc and enriched by fluorescence-activated cell sorting for hßc expression. Expression of both receptor types was confirmed by flow cytometry after appropriate antibody staining (Fig. 2 A
). Note that as we have reported previously [22
], neither I374N nor FI alone could confer factor independence on this cell type (Fig. 2B)
. These cells could all, as expected, proliferate in the presence of Epo (Fig. 2C)
. Importantly, Figure 2D
shows that only the cells expressing mutant hßc could proliferate in the absence of any added cytokine, indicating that EpoR could functionally replace the GMR subunit in this context.
 View larger version (20K): [in a new window] |
Figure 2. Cooperation of EC mutants and EpoR in BaF-B03 cells. (A) Surface expression of mEpoR and hßc as assessed by flow cytometry of BAF-B03 cell populations used in C and D below. Similar levels of surface expression of WT hßc, I374N, and FI were seen in BAF-B03 cells used in the experiment shown in B (data not shown). (B) Proliferation of BAF-B03 cells expressing I374N (triangles) or FI (circles), grown with murine IL-3 (mIL-3; open shapes) or without cytokines (closed shapes). (C and D) Proliferation of BAF-B03 cell populations coexpressing mEpoR and WT hßc (squares), I374N (triangles), or FI (circles) grown with or without rhEpo as indicated.
|
 View larger version (43K): [in a new window] |
Figure 3. Coimmunoprecipitation of hßc and EpoR from BAF-B03 cell extracts with anti-hßc antibodies. (A) Immunoprecipitation of hßc from cells coexpressing hßc and mEpoR. (B) Immunoprecipitation of hßc from cells expressing hßc only. (C) Control immunoprecipitation from cells expressing only mEpoR. In each case, cells were starved of cytokine (-), starved and then stimulated with rhEpo (+), or grown continuously in rhEpo (c). In each case, immunoprecipitates were fractionated by SDS-PAGE and immunoblotted (IB) with antibodies to hßc or mEpoR as indicated.
|
have shown that these mutant receptors do not demonstrate detectable, constitutive tyrosine phosphorylation [28
]. Nevertheless, as EpoR associates with and activates JAK2 following ligand stimulation [29
, 30
], we also examined the possibility that interaction of the EC mutants with EpoR might induce hßc phosphorylation; however, there was no significant increase in tyrosine phosphorylation of hßc over and above the level observed for the WT receptor in the absence of ligand (Fig. 4 A
).
 View larger version (36K): [in a new window] |
Figure 4. Tyrosine phosphorylation status of hßc and EpoR in BaF-B03 cells. (A) Anti-hßc immunoprecipitation of extracts derived from BAF-B03 cells coexpressing hßc (WT or mutant as indicated) and mEpoR. Cells were starved of cytokine (-) or starved and then stimulated with rhEpo or hGM-CSF (+) as indicated. The right-most lane (*) shows phosphorylation of hßc in BAF-B03-expressing hGMR in addition to hßc in response to a pulse of hGM-CSF. (B) Anti-mEpoR immunoprecipitation of extracts derived from BAF-B03 cells coexpressing hßc (WT or mutant) and mEpoR. In each case, cells were starved of cytokine (-) or starved and then stimulated with rhEpo (+). Immunoprecipitates were fractionated by SDS-PAGE and immunoblotted (IB) with antiphosphotyrosine (4G10), anti-hßc, or anti-mEpoR antibodies as indicated.
|
but not WT hßc. In fact, the level of EpoR tyrosine phosphorylation observed is similar to that seen in cells grown continuously in the presence of Epo, as distinct from cells deprived of Epo and then restimulated (T. J. Blake and T. J. Gonda, unpublished observations).
Given these observations, we next asked whether JAK2 was indeed activated by the EpoR-I374N/FI interaction and further, whether JAK2 activation (by either receptor) or EpoR tyrosine phosphorylation was necessary for factor-independent growth of BAF-B03 cells. Figure 5 A
shows that constitutive activation of JAK2 could be detected in BAF-B03 cells coexpressing EpoR and mutant but not WT hßc. (JAK2 activation was not detected in BAF-B03 cells expressing either receptor alone; data not shown.) The requirement of JAK2 activity for factor-independent growth was further supported by the observation that the growth of cells expressing mutant hßc plus EpoR was blocked by the JAK inhibitor AG490 (Fig. 5B)
.
 View larger version (33K): [in a new window] |
Figure 5. Activation of JAK2 in cells coexpressing EC mutants and EpoR. (A) JAK2 phosphorylation in BaF-B03 cells coexpressing mEpoR and hßc (WT or mutant). Cells were starved of cytokine (-) or starved and then stimulated with rhEpo (+). Extracts were fractionated by SDS-PAGE and immunoblotted with antibodies to phosphorylated (active) JAK2 or total JAK2 as indicated. (B) Proliferation assay of BAF-B03 cells coexpressing mEpoR and WT hßc (squares), I374N (triangles), or FI (circles) grown with or without rhEpo and in the presence or absence of 25 mM AG490 JAK2 inhibitor. Surface expression of receptors was confirmed by flow cytometry (data not shown).
|
constitutive mutant of hßc. Similarly, no factor-independent growth of cells coexpressing EpoR and I374Nbox1 or FIbox1 was observed (Fig. 6B)
, which is consistent with the observation that these mutants do not confer factor independence in the presence of GMR (data not shown). Moreover, no JAK2 activation was detected in the absence of growth factor in cells expressing any of these combinations of receptors (Fig. 6C
and 6D)
. As expected, robust JAK2 activation was detected in response to IL-3 in cells expressing the W282R mutant (Fig. 6C)
and in response to Epo in cells expressing WT EpoR (Fig. 6D)
. That the hßc Box 1 mutation abolished JAK2 activation is confirmed by the lack of phosphorylated JAK2 in response to hGM-CSF in cells expressing hGMR and hßcbox1 (left-most lane of Fig. 6D
). Taken together, these data strongly suggest that activation of JAK2 molecules associated with each receptor is essential for functional signaling by the EpoR/I374N or EpoR/FI complexes. Most likely, JAK2 activation results from a reciprocal, trans interaction between JAK2 molecules and/or the Box 1 regions of each receptor (see Discussion).
 View larger version (41K): [in a new window] |
Figure 6. Cross-talk requires JAK2 activation. (A) Proliferation assay of BAF-B03 cells coexpressing mEpoRW282R and hßc WT (squares), I374N (triangles), or FI (circles), grown with or without mIL-3. (B) Proliferation assay of BAF-B03 cells coexpressing mEpoR and hßcbox1 (squares), I374Nbox1 (triangles), or FIbox1 (circles), grown with or without rhEpo. (C) JAK2 phosphorylation in BaF-B03 cells coexpressing mEpoRW282R and hßc (WT or mutant). Cells were starved of cytokine (-) or starved and then stimulated with mIL-3 (+). (D) JAK2 phosphorylation in BaF-B03 cells coexpressing mEpoR and hßc (WT or mutant) with a deletion removing the Box 1 sequence. Cells were starved of cytokine (-) or starved and then stimulated with rhEpo or hGM-CSF (+) as indicated. Extracts were fractionated by SDS-PAGE and immunoblotted with antibodies to phosphorylated (active) JAK2 or JAK2 as indicated. The left-most lane (*) shows the absence of JAK2 phosphorylation in BaF-B03 cells expressing hGMR and hßcbox1 in response to a pulse of hGM-CSF.
|
321; Fig. 1
) that retains the Box 1 and Box 2 motifs but removes all of the tyrosine residues. This mutant has previously been reported to exhibit severely reduced proliferation [25
]. In combination with I374N or FI, EpoR321 conferred factor-independent growth but at a significantly slower rate than did WT mEpoR (Fig. 7A
). However, the reduced growth rate was similar to that observed when growth was mediated via Epo stimulation of the truncated EpoR (Fig. 7B) consistent with phosphorylation of tyrosines on EpoR contributing to the signaling through the hßc/EpoR complex.
 View larger version (23K): [in a new window] |
Figure 7. Cooperation of EC mutants with truncated EpoR. Proliferation assay of BAF-B03 cells coexpressing mEpoR (solid shapes) or mEpoR mutant 321 (open shapes) and hßc WT (squares), I374N (triangles), or FI (circles), grown without added cytokine (A) or with added rhEpo (B). Surface expression of receptors was confirmed by flow cytometry (data not shown).
|
frequently showed thrombocytosis in addition to erythrocytosis [15
, 16
]. Given the key role of Tpo in platelet production [32
], we also asked whether the TpoR (Mpl) could, like EpoR, interact with the activated EC hßc mutants. Figure 8A
shows that this was indeed the case, i.e., that BAF-B03 cells coexpressing Mpl and I374N or FI could proliferate in the absence of added cytokine. This result in turn led us to examine the ability of other CRs to functionally complement I374N and FI to determine whether there was any specificity to such interactions. Therefore, we generated BAF-B03 cells expressing GHR, GCSFR, and the gp130 signaling subunit that is common to a group of CRs including that for IL-6. In each case, the cells were selected for growth in the appropriate cytokine before infection with retroviruses carrying WT or mutant forms of hßc. In the case of gp130, this was achieved by adding the sIL-6R subunit along with IL-6. GHR allowed survival of I374N- and FI-expressing BAF-B03 cells (Fig. 8B)
and slow growth over extended periods of time (data not shown). In contrast, coexpression of neither GCSFR (Fig. 8C)
nor gp130 (Fig. 8D)
was sufficient for factor-independent growth of cells expressing mutant hßc.
 View larger version (20K): [in a new window] |
Figure 8. Cooperation of EC mutants with other CRs. (A) Proliferation assay of BAF-B03 cells coexpressing Mpl and WT hßc (squares), I374N (triangles), or FI (circles), grown with or without rhTpo. (B) Proliferation assay of BAF-B03 cells coexpressing GHR and WT hßc (squares), I374N (triangles), or FI (circles), grown with or without GH. (C) Proliferation assay of BAF-B03 cells coexpressing GCSFR and hßc WT (squares), I374N (triangles), or FI (circles), grown with or without G-CSF. (D) Proliferation assay of BAF-B03 cells coexpressing gp-130 receptor and WT hßc (squares), I374N (triangles), or FI (circles), grown with or without sIL-6R subunit and IL-6. Surface expression of receptors was confirmed by flow cytometry (data not shown).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-subunit of the GM-CSFR, allowing factor-independent JAK2 activation, survival, and proliferation of cells expressing mutant hßc.
We have previously shown that the EC mutants required association with the GMR for their activity [17
, 18
]. In addition, we showed that in contrast to WT hßc, the mutant forms of hßc could constitutively associate with murine GMR (mGMR) [17
]. This suggested that the conformation of the mutant hßc provided an accessible binding site for mGMR and that the induced association may have been sufficient for activation. However, the association of EpoR with hßc is not sufficient for activation, as there is constitutive association with EpoR for all forms (WT and mutant) of hßc. This implies that the activation is a result of the altered hßc conformation, which results in activation of normally inactive intracellular kinases associated with a preformed hßc/EpoR complex.
We have addressed the requirements for a functional interaction between EC mutants and the complementing CR subunit using mutants of the EpoR containing targeted mutations in the membrane-proximal intracellular region. We have found that JAK2 activation per se and the regions associated with JAK2 activation in the EC hßc mutant and EpoR are necessary for functional complex formation as measured by factor independence. This and the fact that deletion of Box 1 from I374N or FI or that the EpoR W282R mutation completely abolishes JAK2 activation are suggestive of bidirectional trans activation of receptor-associated JAK2 molecules. However, we have not directly shown that JAK2 molecules associated with each receptor are activated. As the Box 1 region of hßc is also essential for the function of EC mutants in the presence of GMR (T. J. Blake and T. J. Gonda, unpublished observations), this would suggest by analogy that GMR might have associated kinase activity and possibly an association with JAK2. Indeed Ogata et al. [33
] reported an association of JAK2 with the related IL-5R subunit; however, Quelle et al. [34
] failed to detect any association between GMR and JAK2. Alternatively, GMR may provide a distinct JAK-activating function—be it another kinase or otherwise. We cannot rule out the possibility that a similar JAK-activating function is provided by EpoR and by hßc and is affected by the W282R substitution and Box 1 deletions, respectively.
Another issue addressed directly by our findings is the role and requirement of receptor tyrosine phosphorylation for signaling. Mutant hßc is not detectably phosphorylated when functioning in complexes with mGMR [28
], and GMR does not undergo tyrosine phosphorylation. That these mutants could signal in the absence of tyrosine phosphorylation is consistent with the observation that signaling still occurs when ligand-dependent tyrosine phosphorylation of hßc is abolished by phenylalanine substitution [35
] or by mutation of certain extracellular cysteine residues [36
]. We have previously suggested that the lack of hßc phosphorylation reflects the nature of the complex formed by the EC mutants [13
]. Similarly, in the present case, the mutant hßc subunits did not become tyrosine-phosphorylated in the presence of EpoR. This may indicate a lack of activation of EpoR-associated JAK2 (i.e., that only hßc-associated JAK2 is activated) or incorrect rotational apposition of the EpoR-associated JAK2 molecule and the phosphorylation sites on associated hßc.
The fact that JAK2 is activated by these mutants (see Fig. 5A
and ref. [28
]) suggests that a suitable substrate in proximity (i.e., EpoR in a hßc/EpoR complex) could become phosphorylated, and importantly, we did observe constitutive phosphorylation of EpoR when coexpressed with I374N or FI. This is in contrast to the findings of Hanazono et al. [8
], who reported that GM-CSF or IL-3 could not induce tyrosine phosphorylation of EpoR in UT-7 cells. Despite EpoR displaying constitutive phosphorylation on tyrosine in the presence of EC hßc mutants, loss of the EpoR tyrosine residues did not totally abolish activity of the EC mutant/EpoR complexes. This is consistent with EpoR providing a tyrosine-independent function similar to GMR. However, phosphorylation of EpoR appears to be important for full activation of the hßc/EpoR complex, as the 321 mutant displays a reduced ability to signal in association with the EC hßc mutants. Several previous studies also indicated an important role for tyrosine phosphorylation in EpoR signaling [37
38
39
40
], and in fact, the 321 EpoR truncation used here displays a reduced level of growth in the presence of ligand (see Fig. 7
and ref. [25
]). However, we cannot rule out the possibility that there are other functions provided by the EpoR cytoplasmic domain beyond residue 321 that are needed for full function.
The ability of Mpl to complement the EC hßc mutants in this system also demonstrates a functional interaction and is consistent with the report of Ooi et al. [9
] that Tpo induces tyrosine phosphorylation of hßc (although we have not determined whether hßc mutants are tyrosine-phosphorylated in the presence of Mpl). Taken together with our previous observations that mice expressing constitutive EC hßc mutants display expansion of the megakaryocyte lineage (in addition to the erythroid and myeloid lineages), this prompted us to investigate the interaction of hßc constitutive mutants with Mpl and other CRs. The ability to cooperate with the hßc mutants was restricted to EpoR, Mpl, and GHR. The selective abilities of the various CR molecules to functionally complement the hßc EC mutants in BaF-B03 cells raise the question of what determines the specificity of these interactions. One possibility is that the specificity reflects the structures of the extracellular domains of the CRs, which may in turn determine their ability to physically interact with hßc. For example, EpoR, Mpl, GHR, and GMR have extracellular domains that are comprised entirely of conserved CR family modules(s), although GMR has an N-terminal extension. In contrast, gp130 and GCSFR have fibronectin III-like domains between the CR module (CRM) and the transmembrane domain (reviewed in ref. [41
]). This may restrict association between heterologous CRMs and thus limit the spectrum of interactions. Alternatively, the specificity may be a function of the intracellular domains. A precise rotational orientation of the associated JAK kinases and/or functional motifs in the membrane-proximal regions may be required for receptor complex activation as suggested by Constantinescu et al. [29
]. Another common feature of the intracellular domains of hßc, EpoR, GHR, and Mpl is that they all predominantly associate with and activate JAK2 [42
], and GCSFR and gp130 predominantly use JAK1 [42
, 43
]. Thus, the specificity may be related to the ability of particular receptor-associated JAKs to transactivate each other. However, other CRs use heterologous JAK family members; for example, the IL-2Rß subunit associates with JAK1, and the 
c subunit associates with JAK3 [44
], so there is no clear precedent for such a requirement.
Finally, we should consider the possible biological significance of these interactions. The present studies were initiated to examine the notion that cross-talk may be responsible for the effects of EC hßc mutants on the erythroid and megakaryocytic lineages in vivo. Although the data presented here are certainly consistent with this hypothesis, studies using primary cells are necessary for confirmation. To this end, we are examining the ability of EC mutants to affect growth of erythroid progenitors from EpoR-/- mice.
The other facet of this question is whether functional cross-talk occurs between CRs during a ligand response and whether this has a physiological role. Scott et al. [45 ] have argued that such interactions are not physiologically significant, as progenitors from ßc/ßIL-3-deficient mice exhibit responses to G-CSF, Epo, and SCF in an in vitro colony-forming assay that is comparable with progenitors from WT animals. Moreover, no defects in myelopoiesis have been reported in the EpoR- and Mpl-deficient mice under steady-state conditions [46 47 48 ]. However, we have shown that constitutive ß-subunit signaling activates EpoR (rather than the converse). Such cross-talk between the WT hßc and EpoR would be consistent with the observation that CFU-E frequency is reduced in mice carrying null mutations in the GM-CSF or IL-3 genes [49 ]. Thus, a complete characterization of the physiological significance of this cross-talk requires a more detailed analysis of the frequencies and growth-factor responses of a range of haemopoietic progenitors from ß-subunit and EpoR-deficient mice.
It is also possible that the effects of CR interactions may only become apparent in vivo under conditions of physiological challenge. To date, there have been limited studies with receptor-deficient mouse models under conditions requiring rapid emergency haemopoiesis. It is possible that in this context, there is an amplification of the initial signal via cross-talk. Until such situations are addressed, the significance of CR cross-talk in vivo is still an open question.
Current address of Richard J. D’Andrea: Child Health Research Institute, Adelaide Women’s and Children’s Hospital, South Australia.
Current address of Thomas J. Gonda: Bionomics Ltd., Thebarton, South Australia.
Received July 5, 2002; revised August 23, 2002; accepted August 26, 2002.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
subunit for activation J. Biol. Chem. 274,8669-8677 and ßc subunit, respectively, and are activated upon IL-5 stimulation Blood 91,2264-2271This article has been cited by other articles:
 |
|
 |
|
  L. Sautina, Y. Sautin, E. Beem, Z. Zhou, A. Schuler, J. Brennan, S. I. Zharikov, Y. Diao, J. Bungert, and M. S. Segal Induction of nitric oxide by erythropoietin is mediated by the {beta} common receptor and requires interaction with VEGF receptor 2 Blood, January 28, 2010; 115(4): 896 - 905. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Rundqvist, E. Rullman, C. J. Sundberg, H. Fischer, K. Eisleitner, M. Stahlberg, P. Sundblad, E. Jansson, and T. Gustafsson Activation of the erythropoietin receptor in human skeletal muscle Eur. J. Endocrinol., September 1, 2009; 161(3): 427 - 434. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Brines, G. Grasso, F. Fiordaliso, A. Sfacteria, P. Ghezzi, M. Fratelli, R. Latini, Q.-w. Xie, J. Smart, C.-j. Su-Rick, et al. Erythropoietin mediates tissue protection through an erythropoietin and common {beta}-subunit heteroreceptor PNAS, October 12, 2004; 101(41): 14907 - 14912. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
