J Biol Chem, Vol. 275, Issue 7, 5124-5130, February 18, 2000

The Role of Disulfide-linked Dimerization in Interleukin-3 Receptor Signaling and Biological Activity^*

Fei Le, Frank Stomski, Joanna M. Woodcock^§, Angel F. Lopez, and Thomas J. Gonda^¶

From the Hanson Centre for Cancer Research, The Institute of Medical and Veterinary Science, Adelaide, South Australia 5000, Australia

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cysteine residues 86 and 91 of the  subunit of the human interleukin (hIL)-3 receptor (hc) participate in disulfide-linked receptor subunit heterodimerization. This linkage is essential for receptor tyrosine phosphorylation, since the Cys-86  Ala (Mc4) and Cys-91  Ala (Mc5) mutations abolished both events. Here, we used these mutants to examine whether disulfide-linked receptor dimerization affects the biological and biochemical activities of the IL-3 receptor. Murine T cells expressing hIL-3R and Mc4 or Mc5 did not proliferate in hIL-3, whereas cells expressing wild-type hc exhibited rapid proliferation. However, a small subpopulation of cells expressing each mutant could be selected for growth in IL-3, and these proliferated similarly to cells expressing wild-type hc, despite failing to undergo IL-3-stimulated hc tyrosine phosphorylation. The Mc4 and Mc5 mutations substantially reduced, but did not abrogate, IL-3-mediated anti-apoptotic activity in the unselected populations. Moreover, the mutations abolished IL-3-induced JAK2, STAT, and AKT activation in the unselected cells, whereas activation of these molecules in IL-3-selected cells was normal. In contrast, Mc4 and Mc5 showed a limited effect on activation of Erk1 and -2 in unselected cells. These data suggest that whereas disulfide-mediated cross-linking and hc tyrosine phosphorylation are normally important for receptor activation, alternative mechanisms can bypass these requirements.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The receptors for human granulocyte-macrophage colony-stimulating factor (GM-CSF),¹ IL-3 and IL-5, are members of the hematopoietin receptor superfamily, often termed the cytokine receptor family. The high affinity GM-CSF·IL-3·IL-5 receptor complexes are composed of specific  chains that bind GM-CSF·IL-3·IL-5 with low affinity (1-3) and a common  chain (c) that converts the binding to high affinity by forming a heterodimer with the  subunit (4). The oligomerization of  and  subunits mediated by ligand binding is thought to initiate signal transduction.

It is generally accepted that whereas the cytoplasmic domains of both  and  subunits are required for receptor activation, in some cases dimerization of the cytoplasmic domain of the  subunit alone is sufficient. This was demonstrated by a study in which a chimeric receptor consisting of the extracellular domain of the erythropoietin receptor and the intracellular domain of c or _IL-3 was constructed. Erythropoietin induced proliferation signals in Ba/F3 cells through the chimeric receptors, indicating that the homodimerization of the  subunit is sufficient for receptor activation (5). Likewise, a mutant GM-CSF receptor  chain (GM-CSFR), in which the cytoplasmic domain was replaced with that of the c, formed a high affinity receptor with the normal c and transduced proliferative signals, again indicating the importance of the dimerization of the  cytoplasmic domain (6). In contrast, the c mutant in which the cytoplasmic domain was substituted with that of GM-CSFR forms a high affinity GM-CSFR with the GM-CSFR but was unable to induce a proliferation signal, indicating that the dimerization of the GM-CSFR cytoplasmic domain was not sufficient for signaling (6).

Despite the lack of an intrinsic tyrosine kinase in the receptors, GM-CSF, IL-3, and IL-5 induce rapid tyrosine phosphorylation of various cellular proteins, including the  subunits themselves. It is now known that cytoplasmic JAK family tyrosine kinases associate with the  subunit via conserved membrane proximal regions known as "box-1" and "box-2" (7, 8). The activated JAK kinases phosphorylate tyrosine residues of many signaling proteins, among which are the latent cytoplasmic transcription factors known as signal transducer and activator of transcription (STAT). The activation of JAKs and/or STATs is important for many if not all activities of cytokine receptors (7, 9, 10).

The Ras-Raf-MAP kinase pathway is another major signaling pathway activated in response to GM-CSF and IL-3 (11-13). It has been demonstrated that a membrane-distal region of c is required for activation of the Ras-Raf-MAP kinase pathway (11). Circumstantial evidence implies that JAK2 is also required for activation of this pathway by IL-3 and GM-CSF. Deletion of the box-1 region in c (which is believed to be the site of JAK2 association with c) or expression of a dominant negative form of JAK2 prevented phosphorylation of molecules involved in Ras activation (Shc and SHP2) (7). Moreover, these same manipulations also blocked IL-3/GM-CSF-induced activation of the c-fos promoter (7, 14), which is also blocked by dominant negative Ras (14). This pathway may also be important for cell survival (15-17) (see also "Discussion").

Signaling pathways that promote cell survival also include the phosphatidylinositide-3'-OH kinase (PI3K)-AKT pathway. The AKT kinase is a general mediator of cytokine-induced survival and has been shown to suppress the apoptotic death of a number of cell types induced by a variety of stimuli, including growth factor withdrawal, cell cycle discordance, loss of cell adhesion, and DNA damage (18-21). Thus, a signaling pathway has been defined in which cytokine receptor activation leads to the sequential activation of PI3K and AKT, which then promotes cell survival and blocks apoptosis.

Our previous study (22) showed that Cys-86 and Cys-91 but not other Cys residues of human c (hc) participated in disulfide-linked receptor subunit heterodimerization. We also showed that this linkage is essential for receptor phosphorylation because alanine substitutions of residues 86 (termed Mc4) and 91 (Mc5), but not 100 (Mc7), abolished not only IL-3-induced disulfide-linked IL-3 receptor subunit dimerization but also tyrosine phosphorylation of hc, without affecting IL-3 binding. To investigate whether or not disulfide-mediated cross-linking of IL-3 receptor subunits and phosphorylation of hc are essential for receptor signaling, we have now examined the effects of these cysteine mutations on biological and biochemical activities of the IL-3 receptor when expressed in the murine T cell line CTL-EN. We have found that hc mutants Mc4 and Mc5 barely induce cell proliferation compared with wild-type (WT) or Mc7. Interestingly, a subpopulation of cells expressing each of the former mutants could be selected for growth in IL-3, which, despite the absence of detectable hc tyrosine phosphorylation, proliferated at similar rates to cells expressing WT hc. Our data also show that Mc4 and Mc5 (but not Mc7) impaired IL-3-facilitated protection against apoptosis and failed to induce IL-3-stimulated JAK2, STAT, and AKT activation in unselected cells. However, Mc4 and Mc5 had a less severe effect on IL-3-induced Erk1/2 MAP kinase activation in the unselected cells. In contrast, in the IL-3-selected cells, the cysteine mutations had no effects on IL-3-induced activation of any of the signaling molecules that we examined.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Cell Culture-- The ecotropic 2 retrovirus packaging cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat-inactivated fetal calf serum (FCS), 2 mM L-glutamine, and antibiotics. The IL-2-dependent mouse T cell line, CTL-EN (a derivative of CTLL-2, described in Ref. 23), was maintained in CTLL medium (DMEM supplemented with 10% FCS, 2 mM L-glutamine, antibiotics, 50 µM -mercaptoethanol, 5% conditioned medium from MLA cells, and 100 units/ml bacterially synthesized mouse IL-2.

Construction of Expression Plasmids-- Cysteine mutants Mc4, Mc5, Mc7, or WT hc cDNAs (22) were inserted between the BamHI and HpaI restriction sites of the pRufHygro retroviral vector. The pRufHygro retroviral vector was constructed by replacing the MC1Neo cassette of pRufNeo (24) with a phosphoglycerate kinase/hygromycin resistance cassette from pPGKHygro (25).

Transfection and Infection Procedures-- The ecotropic packaging cell line 2 was transfected by a standard calcium phosphate transfection procedure as described (26) with 10 µg of retroviral plasmid containing WT or cysteine mutant hc per 10-cm dish. After overnight incubation the cells were shocked with 2.5 ml of 15% glycerol in DMEM for 4 min followed by a further 24 h incubation and then selected in hygromycin (Roche Molecular Biochemicals) at 200 µg/ml. The selected cells were sorted for expression of hc, and collected in DMEM. These sorted 2 cells expressing WT or the cysteine mutant hc were then used to infect CTL-EN cells previously infected with a pRUFNeo retrovirus vector (24) encoding the  chain of human IL-3 receptor (hIL-3R), using procedures described previously (26). The infected CTL-EN cells were selected in hygromycin at 600 µg/ml, and expression of WT or cysteine mutant hc on the surface of the hygromycin-resistant cells was examined by flow cytometry.

Cell Sorting and Analysis of Receptor Subunit Expression by Flow Cytometry-- Cells expressing WT or cysteine mutant hc were collected by cell sorting on a FACStar^PLUS flow cytometer (Coulter, Hialeah, FL). Briefly, cells were washed and resuspended in cold DMEM supplemented with 5% FCS. Cells were incubated with the anti-hc monoclonal antibody 8B8 (27) for 20 min on ice, washed, and subsequently incubated with fluorescein isothiocyanate-conjugated anti-mouse IgG (Silenus, Hawthorn, Victoria, Australia) for 20 min on ice. After washing and resuspension in medium, the cells were sorted, and the positive cells collected in CTLL medium.

Radiolabeling of Human IL-3 and Binding Assays-- The procedures used for radiolabeling of human IL-3 with ¹²⁵I and performing saturation-binding assays have been described previously (28, 29).

Cell Proliferation Assays-- Infected CTL-EN cells expressing hIL-3R and WT or cysteine mutant hc were washed three times with PBS, and triplicate samples of 5 × 10³ cells were cultured in a 96-well microtiter plate with or without IL-3 for 72 h. Cell proliferation was measured by the CellTiter 96 Non-Radioactive Cell Proliferation Assay (Promega, Madison, WI). The data were normalized with respect to proliferation of the same cells cultured in 100 units/ml of IL-2.

Apoptosis Assays-- Apoptotic cells were detected using the Annexin-V-Fluos Staining kit (Roche Molecular Biochemicals) as per manufacturer's instructions. Briefly, cells (5 × 10⁵) were washed with PBS and centrifuged at 2,000 × g for 5 min. The cell pellets were resuspended in 100 µl of labeling solution (1:50 diluted Annexin-V-Fluos labeling reagent in 10 mM HEPES buffer (pH 7.4) and 1 µg/ml of propidium iodide) and then added to 0.4 ml of incubation buffer (10 mg of HEPES (pH 7.4), 140 mM NaCl, 5 mM CaCl₂). The cells were analyzed on a flow cytometer using a 488-nm excitation and a 515-nm band pass filter.

Immunoprecipitation-- CTL-EN cells coexpressing hIL-3R and WT or cysteine mutant hc (4 × 10⁷) were cultured overnight in the absence of cytokines. Cells were washed with cold PBS followed by stimulation at 37 °C for 5-15 min as indicated with varying concentrations of IL-3 (0-10 ng/ml) and then lysed on ice in lysis buffer (50 mM HEPES (pH 7.5), 150 mM NaCl, 10% glycerol, 1% Nonidet P-40, 0.1% SDS, 0.1% sodium deoxycholate 2 mM sodium orthovanadate, 1 mM phenylmethylsulfonyl fluoride, 1 mM EDTA, 1 mM EGTA, 2 mg/ml iodoacetamide, 50 mM sodium fluoride, 10 mM sodium pyrophosphate, 0.2 mg/ml trypsin inhibitor (Roche Molecular Biochemicals) and Complete^TM protease inhibitor (Roche Molecular Biochemicals)) for 15 min. Insoluble materials were removed by centrifugation, and cell lysates were incubated with rabbit anti-JAK2 antibody² or anti-hc monoclonal antibody 8E4 (30) for 2 h at 4 °C. Immune complexes were precipitated with 75 µl of protein A-Sepharose (Amersham Pharmacia Biotech) for 90 min at 4 °C, washed twice with lysis buffer, and boiled in 1× reducing SDS loading buffer.

Western Blotting-- Immunoprecipitates, or cell lysates prepared as described above, were separated by SDS-polyacrylamide gel electrophoresis and electrophoretically transferred to Protran^R nitrocellulose transfer membranes (Schleicher & Schuell). Membranes were incubated with a blocking solution (3% bovine serum albumin in TBS-T (50 mM Tris-HCl (pH 7.4), 135 mM NaCl and 0.1% Tween 20)) at room temperature for 1 h and then incubated overnight with antibody in the same solution at 4 °C. The antibodies used were anti-JAK2 (as above); anti-phosphotyrosine (4G10; Upstate Biotechnology, Inc., Lake Placid, NY); anti-hc (1C1; Ref. 28); anti-Erk1/2 (Zymed Laboratories Inc., San Francisco, CA); anti-phospho-Erk1/2 (Promega, Madison, WI); and anti-phospho-AKT and anti-AKT (Santa Cruz Biotechnology, Santa Cruz, CA). The membranes were then washed three times in TBS-T solution and incubated with anti-mouse or anti-rabbit secondary antibodies (as appropriate) coupled with horseradish peroxidase (Pierce). Membranes were washed in TBS-T three times and subjected to enhanced chemiluminescence detection as per the manufacturer's instructions (Pierce). Before reprobing, membranes were stripped in 50 mM Tris (pH 7.4), 2% SDS, 100 mM -mercaptoethanol at 55 °C for 10 mim, washed three times in TBS-T, and blocked in TBS-T containing 3% bovine serum albumin.

Electrophoretic Mobility Shift Analysis (EMSA) of STATs-- Nuclear extracts from CTL-EN cells expressing hIL-3R and WT or cysteine mutant hc were prepared as described by Jenkins et al. (31). These were mixed with radiolabeled double-stranded oligonucleotides corresponding to the prolactin-responsive element in the bovine -casein promoter, and EMSA was performed essentially as described by Barry et al. (32).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Expression of WT or Cysteine Mutant hc in Murine CTL-EN Cells-- To introduce WT or cysteine mutant hc into CTL-EN cells expressing hIL-3R, the cells were infected using 2 cells producing the corresponding RUFHygro retroviruses. The infected CTL-EN cells were selected in hygromycin in IL-2-containing medium, and cell surface expression of hc was assessed by FACS analysis. Fig. 1 shows that WT and cysteine mutant c were expressed at comparable levels. Previous studies demonstrated that the cysteine mutations had no effect on the ligand binding in HEK293T cells (22). To confirm this finding in CTL-EN cells, we performed binding assays with ¹²⁵I-IL-3. The results of Fig. 2 show that all receptors, either WT or the cysteine mutants, had similar high affinity binding and were present in similar numbers, in agreement with previous results in HEK293T cells (22) and the FACS analyses of Fig. 1.

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Surface expression of WT or mutant (as indicated) hc in sorted CTL-EN/IL-3R cells as measured by flow cytometry and indirect immunofluorescence. Dashed lines represent cells stained with an irrelevant control antibody, and solid lines indicate staining with an anti-hc antibody 8B8.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Scatchard transformation of saturation binding studies performed on CTL-EN cells coexpressing hIL-3R and WT or cysteine mutant hc as indicated. Binding assays were performed with 125I-labeled IL-3 over a concentration range of 10 pM to 10 nM. The data were analyzed using the LIGAND program, and the lines indicate the high affinity binding component for WT or Cys mutant hc as indicated.

Effects of the hc Cysteine Mutations on IL-3-induced Proliferation-- The growth of CTL-EN cells expressing WT hc or the cysteine mutants cultured in hIL-3 (10 ng/ml) was monitored over a 3-day period. The results of Fig. 3A show that cells expressing Mc4 and Mc5 mutants did not exhibit detectable growth, whereas cells expressing the Mc7 mutant grew similarly to those expressing WT hc. Interestingly, it was found that although most cells expressing Mc4 and Mc5 mutants could not grow or survive in IL-3, a small subset (we estimate 1%) of these cells grew out over a period of several weeks from IL-3-containing cultures. To distinguish these subsets, which were maintained in medium containing hIL-3, we will term these cells "IL-3-selected cells," whereas the bulk population of (mutant) hc-expressing cells, maintained in IL-2, will hereafter be termed "unselected cells." It was found that the IL-3-selected cells proliferated at similar rates to the cells expressing WT hc (Fig. 3B) and did not grow or survive in medium lacking cytokines (data not shown). Moreover, polymerase chain reaction amplification and sequencing of hc from genomic DNA of the IL-3-selected cells confirmed the presence of the cysteine mutations, i.e. these cells were not genetic revertants.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Proliferation of unselected CTL-EN cells (A) or IL-3-selected CTL-EN cells (B) coexpressing hIL-3R and WT or cysteine mutant hc as indicated. 5 × 103 cells were plated in triplicate, and cell proliferation was measured at each time point as described under "Experimental Procedures." The absorbance was normalized to that of the same cells cultured in IL-2 for 3 days. The mean and standard error of each triplicate is shown.

Cysteine Mutations Mc4 and Mc5 Markedly Impair hIL-3-facilitated Protection against Apoptosis-- To examine whether the Mc4 and Mc5 mutations have an effect on cell survival, unselected CTL-EN cells expressing WT or cysteine mutant hc were cultured in different concentrations of IL-3 for 48 h, following which the proportions of apoptotic cells were determined by annexin-V staining and FACS analysis. The results in Fig. 4A show that IL-3 at either low or high concentrations effectively protected the cells expressing WT or Mc7 from apoptosis in a dose-dependent manner. The cysteine mutations Mc4 and Mc5 significantly impaired, but importantly, did not completely abrogate the ability of IL-3 to protect the cells against apoptosis. This result was confirmed in the time course experiment shown in Fig. 4B. The impairment by Mc4 and Mc5 of hIL-3-facilitated protection against apoptosis was evident at 48 h and was even more marked by 72 h.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Apoptosis of unselected CTL-EN cells coexpressing hIL-3R and WT or cysteine mutant hc. A, the cells were incubated with the indicated concentrations of hIL-3 for 48 h and then stained with annexin-V and analyzed by flow cytometry to detect the percentage of annexin-V-stained cells. B, the same cell populations were incubated with 1 ng/ml hIL-3 or without hIL-3 (control). At each time point as indicated the cells were stained with annexin-V and analyzed by flow cytometry.

The Cysteine Mutations Mc4 and Mc5 Prevent Tyrosine Phosphorylation of hc in CTL-EN Cells-- We previously showed that cysteine mutations Mc4 and Mc5 abrogated disulfide-linked IL-3 receptor dimerization and abolished tyrosine phosphorylation of hc in response to IL-3 when expressed in HEK293T cells. To test whether this was also the case in the IL-3-selected CTL-EN cells, we performed Western blot analysis to detect tyrosine phosphorylation of hc. The results of Fig. 5 show that the cysteine mutations Mc4 and Mc5 abolished IL-3-induced receptor tyrosine phosphorylation, which was, however, readily detected in cells expressing WT or Mc7.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   The cysteine mutations Mc4 and Mc5 prevent IL-3-induced tyrosine phosphorylation of hc in IL-3-selected CTL-EN cells coexpressing hIL-3R and WT or cysteine mutant hc. Upper panel, the cells were stimulated with 10 ng/ml hIL-3 for 5 min or left unstimulated, as indicated. After cell lysis, proteins were immunoprecipitated with anti-hc monoclonal antibody 8E4, and the immunoprecipitates were separated by SDS-polyacrylamide gel electrophoresis and analyzed by Western blotting (WB) with anti-phosphotyrosine antibody 4G10. Lower panel, the filters were stripped and reprobed with anti-c monoclonal antibody, 1C1.

Effects of Cysteine Mutations Mc4 and Mc5 on JAK2 and STAT Activation-- The activity of JAK2 is believed to be necessary for all the biological functions of IL-3 and GM-CSF (7). We therefore examined the effects of the Mc4 and Mc5 mutations on JAK2 and STAT activation. In the case of unselected cells, it was found that phosphorylation of JAK2 was observed in cells expressing WT or Mc7 after stimulating the cells with 10 ng/ml hIL-3 but not in cells expressing the Mc4 or Mc5 mutants (Fig. 6A). Although no JAK2 activation was detected when the cells expressing WT or Mc7 were stimulated with a low concentration of hIL-3 (1 ng/ml), we found that IL-3 at a concentration of 1 ng/ml was able to support proliferation of cells expressing WT or Mc7 (data not shown; see also below).

View larger version (47K): [in this window] [in a new window]   Fig. 6.   Effects of cysteine mutations on IL-3-induced activation of JAK2 and STATs in unselected (A and B) or IL-3-selected CTL-EN cells (C and D). A, the unselected CTL-EN cells expressing WT or cysteine mutant hc plus hIL-3R were incubated with the indicated concentration of hIL-3 for 5 min. Lysates from the cells were subject to immunoprecipitation with anti-JAK2 antibodies followed by Western blotting (WB) with anti-phosphotyrosine antibodies 4G10 (upper panel) or anti-JAK2 antibodies (lower panel). B, the unselected cells expressing WT or cysteine mutant hc plus IL-3R were incubated with different concentrations of IL-3 for 15 min as indicated. Nuclear extracts prepared from the cells were subjected to EMSA using a -casein promoter oligonucleotide probe. C and D, IL-3-selected cells were analyzed as for A and B, respectively.

One class of effectors of JAK2 are the STAT transcription factors, which are phosphorylated and activated by JAK2. We therefore examined nuclear extracts from unselected cells expressing WT or mutant hc for the presence of STAT DNA binding activity by performing EMSAs. After stimulation of the cells with 10 ng/ml IL-3, extracts from cells expressing WT or Mc7 contained a protein complex that specifically bound to a -casein oligonucleotide probe containing a DNA-binding site for STATs 1, 3, and 5 (Fig. 6B). However, no STAT activation was detected in cells expressing the Mc4 or Mc5 mutants (Fig. 6B) after stimulation by IL-3. Weak STAT activation was also detected in cells expressing WT or Mc7 stimulated with 1 ng/ml IL-3. The most likely explanation for this, considering that no JAK2 activation was detected with this concentration of IL-3, is simply that EMSA detection of STAT activity is more sensitive than detection of JAK2 tyrosine phosphorylation.

Because, as mentioned above, a subpopulation of cells expressing the cysteine mutants Mc4 or Mc5 could be selected for growth in IL-3, we investigated whether or not JAK2 and STATs were activated in IL-3-selected cells expressing these mutants. Fig. 6, C and D, shows that JAK2 and STAT activation was detected following stimulation with IL-3 at 10 ng/ml not only in cells expressing WT or Mc7 but, in contrast to the unselected cells, also in cells expressing Mc4 or Mc5. As for Mc7 and WT, weak activation of STAT but not JAK2 was also detected when a low concentration of IL-3 was used.

Effects of Cysteine Mutations Mc4 and Mc5 on Ras-Raf-MAP Kinase Pathway-- One of the major signaling pathways activated in response to cytokines is the Ras-Raf-MAP kinase pathway (11, 12). To examine whether the cysteine mutations have an effect on hc-mediated activation of Erk1/2 MAP kinases, Western blot analyses of cell lysates from both unselected and IL-3-selected cells with an antibody specific for activated, i.e. phosphorylated, Erk1/2 MAP kinases were conducted. Fig. 7A shows that phosphorylation of Erk1/2 MAP kinases was detected in unselected cells expressing WT or Mc7 mutant hc when the cells were stimulated with 1 or 10 ng/ml hIL-3. Moreover, Erk1/2 phosphorylation was also clearly detected when cells expressing Mc4 or Mc5 were stimulated with 10 ng/ml hIL-3, although little or no phosphorylation above background was detected with 1 ng/ml hIL-3. In the case of IL-3-selected cells, activation of Erk1/2 MAP kinases was readily detectable in cells expressing WT and all of the cysteine mutant forms of hc, regardless of whether the cells were stimulated with a low or high concentration of hIL-3 (Fig. 7B).

View larger version (24K): [in this window] [in a new window]   Fig. 7.   Effects of the cysteine mutations on IL-3-induced Erk1/2 MAP kinase phosphorylation. A, unselected CTL-EN cells expressing WT or cysteine mutant hc plus hIL-3R were stimulated with the indicated concentrations of hIL-3 for 15 min. Cell lysates were immunoblotted with anti-phospho-Erk1/2 (upper panel) or anti-Erk1/2 (lower panel). B, IL-3-selected cells were analyzed exactly as in A. WB, Western blotting.

Effects of Cysteine Mutations Mc4 and Mc5 on AKT Kinase-- The activation of the PI3K and its downstream effector AKT has been shown to promote cell survival and suppress apoptosis (18-20). We therefore examined the levels of activated AKT kinase in both unselected and IL-3-selected CTL-EN cells expressing WT or mutant hc. Western blot analysis of cell lysates from unselected cells using an antibody specific for phosphorylated AKT showed that AKT was activated in cells expressing WT or Mc7 but not in cells expressing Mc4 or Mc5 following stimulation with IL-3 at either 1 or 10 ng/ml (Fig. 8A). In contrast, when the IL-3-selected cells were similarly treated and analyzed, activated AKT kinase was detected in cells expressing either WT or any of the mutant forms of hc (Fig. 8B).

View larger version (24K): [in this window] [in a new window]   Fig. 8.   Effects of the cysteine mutations on IL-3-induced AKT kinase activation. A, unselected CTL-EN cells expressing WT or cysteine mutant hc plus hIL-3R as indicated were stimulated for 15 min with the IL-3 at the indicated concentrations. The cell lysates were immunoblotted with anti-phospho-AKT (upper panel) or anti-AKT antibodies (lower panel). B, IL-3-selected cells expressing WT or mutant hc were analyzed as in A. WB, Western blotting.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

We previously showed (28) the presence of IL-3-induced disulfide-linked and non-disulfide-linked heterodimers, suggesting two levels of IL-3R and c interaction, a noncovalent one and one that is mediated by Cys-Cys bridging of the receptors. Furthermore, substitution of alanine for cysteines 86 (Mc4) or 91 (Mc5) abolished disulfide-linked IL-3 receptor dimerization and tyrosine phosphorylation of c subunit but had no effect on high affinity binding between ligand and receptor, nor on IL-3-dependent coimmunoprecipitation of  and  subunits (22). This suggests that disulfide-linked dimerization is essential for receptor phosphorylation. The present studies were designed to determine whether blocking disulfide-linked dimerization has effects on biological and biochemical functions of the IL-3 receptor. To this end, we utilized populations of (normally) IL-2-dependent CTL-EN cells expressing WT or cysteine mutant forms of hc along with hIL-3R.

We have shown here that the cysteine mutations Mc4 and Mc5 almost completely inhibited IL-3-induced proliferation of CTL-EN cells. Another cysteine mutant, Mc7, which showed normal disulfide-linked dimerization and tyrosine phosphorylation in transfected HEK 293T cells (22), behaved like WT hc and had no effect on cell proliferation. Interestingly, despite the fact that most of the CTL-EN cells expressing Mc4 or Mc5 could not proliferate in IL-3, a small fraction survived and continued to proliferate in IL-3 at similar rates to cells expressing WT hc and Mc7. Nevertheless, and unlike WT- and Mc7-expressing cells, IL-3-selected cells expressing Mc4 or Mc5 did not undergo IL-3-induced hc phosphorylation.

Analyses of several key intracellular signaling molecules showed that most of these, JAK2, STATs, and Akt, were not activated in response to IL-3 in unselected cells expressing Mc4 or Mc5 but were activated in the corresponding IL-3-selected populations. Thus triggering of these pathways correlated well with the ability of the cells to proliferate in IL-3; this finding was not totally unexpected since considerable evidence indicates that JAK2 activation is an essential and primary effector of receptor function (7, 33). More unexpectedly, the Erk1/2 MAP kinases were activated by IL-3 in both populations, although it appeared that higher concentrations may be necessary in the case of the unselected Mc4- or Mc5-expressing cells.

The fact that IL-3-selected cells expressing the Mc4 and Mc5 mutants could proliferate normally in response to IL-3 without detectable hc phosphorylation provides independent confirmation of other studies that dissociate these two properties. Okuda et al. (34), Itoh et al. (35), and Guthridge et al. (36) all reported that a mutant of hc in which all eight cytoplasmic tyrosines were replaced by phenylalanine could still transduce proliferative signals, albeit less efficiently than WT hc. In addition, our own studies with constitutively activated point mutants of hc (31) showed that one class of such mutants failed to show detectable tyrosine phosphorylation but still elicited rapid proliferation and activation of JAK2, STATs, and Erk1/2. These results also highlight the possibility that serine/threonine phosphorylation of hc may contribute to some of the activities of the IL-3 receptor.

We previously proposed a model for the activated, disulfide-linked GM-CSF, IL-3, and IL-5 receptor complexes in which the activated receptor consists of two cytokine molecules, two  chains and two c chains, i.e. of two disulfide-linked - dimers (22, 37). Formation of this hexameric receptor complex would result in juxtaposition of two c molecules with their associated JAK kinases, facilitating JAK2 activation and inducing receptor phosphorylation. This model would explain why the Mc4 and Mc5 mutants, which cannot form disulfide-linked - dimers, are unable to induce JAK2 activation and proliferation in the unselected cells. However, the ability of the IL-3-selected cells expressing Mc4 or Mc5 to proliferate normally and activate JAK2 strongly suggests that there can exist other forms of the activated receptor, in addition to the disulfide-linked complex. As we have suggested for certain constitutive hc mutants (23), it may be that - dimers can be active in some circumstances. Alternatively, the Mc4 and Mc5 mutants may be able to form a spatially "incorrect" non-covalently linked higher order complex in which JAK2 activation does not result in hc phosphorylation.

Analyses of intracellular signaling by the cysteine mutant receptors in unselected and IL-3-selected CTL-EN cells strongly supports the essential role of JAK2 activation in the induction of proliferation and an important role in maintaining cell survival. Our data are also consistent with an important role for Akt in IL-3-mediated maintenance of cell viability (19, 21), since the absence of IL-3-induced activation of Akt in unselected Mc4- or Mc5-expressing cells correlated with markedly increased apoptotic death. More surprising was our finding that IL-3 induced activation of Erk1/2 even in the unselected cells, i.e. in the absence of detectable JAK2 activation. (We believe that assaying JAK2 activation by its tyrosine phosphorylation is relatively insensitive; however, the EMSA detection of STAT activation appears to be a much more sensitive, albeit surrogate, assay. This, too, was negative in the unselected Mc4- or Mc5-expressing cells.) While it is generally thought that JAK2 activation is essential for activation of the Ras/Raf/Erk pathway, the evidence for this, in the case of the GM-CSF/IL-3 receptors, is mostly circumstantial (see "Introduction"), although more direct evidence has been published for other cytokine receptors (38, 39). However, a mutant of the thrombopoietin receptor, Mpl, has been described which activates Erk1/2 in the absence of detectable JAK or STAT activation, and dominant negative JAK2 only partially inhibited Erk activation by gp130 (40), supporting the notion that cytokine receptors may be able to activate Erk kinases by a JAK-independent mechanism (41). The apparently lower sensitivity of Erk1/2 activation to IL-3 in the unselected cells needs further study but is consistent with the utilization of an alternate, less efficient pathway than in IL-3-selected or WT-expressing cells.

The activation of Erk1/2 in unselected Mc4- and Mc5-expressing CTL-EN cells is also interesting in the light of the partial resistance of these cells, in the presence of IL-3, to apoptosis. There are apparently contradictory reports on the importance of Erk1/2 activation in the anti-apoptotic activity of the GMR/IL3R/IL5R system and in hemopoietic cells in general (discussed in Ref. 42), but several studies have shown that activation of these molecules can at least contribute to anti-apoptotic function (15, 17). Thus, the ability of Mc4 and Mc5 to promote some survival of the unselected cells may be due, at least in part, to activation of Erk1/2; future studies with specific inhibitors could clarify this issue, but our results nevertheless suggest that the anti-apoptotic activity of IL-3 probably involves multiple pathways (42, 43).

An intriguing question raised by our present studies relates to the emergence of the IL-3-selected subpopulations of CTL-EN cells expressing Mc4 and Mc5. That is, in what way(s) do the selected cells differ from the bulk of the population? The trivial explanations of genetic reversion or spontaneous factor independence have been ruled out. It is formally possible that culture in IL-3 has selected for a second-site mutation in hc that somehow compensates for the Mc4 and Mc5 mutations (without restoring hc tyrosine phosphorylation). However the fact that with both Mc4 and Mc5, IL-3-responsive cells grew out in each of two independent experiments suggests it is far more likely that there is an intrinsic difference in the selected cells themselves. One possible explanation is that (i) the selected cells represent those that express the highest levels of a particular accessory signaling molecule and (ii) high levels of this molecule can enhance the efficiency of signaling by an - dimer or non-covalent higher order complex (see above). Although there are no obvious candidates for such a molecule, we note that SH2-B has recently been shown to interact with and enhance activation of JAK2 (44); high levels of a molecule such as this could readily account for the properties of the IL-3-selected cells.

	ACKNOWLEDGEMENTS

We thank Tim Blake for assistance with Western blotting and immunoprecipitation techniques and Dr. Tim Hercus for assistance in the preparation of IL-2.

	FOOTNOTES

^* This work was supported by grants from the National Health and Medical Research Council of Australia (to T. J. G. and to A. F. L.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Present address: Dept of Biochemistry and Molecular Biology, Australian National University, Canberra, Australian Capital Territory 0200, Australia.

^§ Research Fellow of the Anti-cancer Foundation of the Universities of South Australia.

^¶ Principal Research Fellow of the National Health and Medical Research Council of Australia. To whom correspondence should be addressed: Hanson Centre for Cancer Research, The Institute of Medical and Veterinary Science, Frome Road, Adelaide, South Australia 5000, Australia; Tel.: 61-8-8222-3305; Fax: 61-8-8232-4092; E-mail: tom.gonda@imvs.sa.gov.au.

² T. J. Blake and T. J. Gonda, unpublished data.

	ABBREVIATIONS

The abbreviations used are: GM-CSF, granulocyte-macrophage colony-stimulating factor; IL-3 and -5, interleukin 3 and interleukin 5; GM-CSFR, GM-CSF receptor  chain; c, human common  subunit of the GM-CSF, IL-3 and IL-5 receptors; hc, human c; WT, wild type; PI3K, phosphatidylinositide-3'-OH; DMEM, Dulbecco's modified Eagle's medium; FCS, fetal calf serum; PBS, phosphate-buffered saline; FACS, fluorescence-activated cell sorter; MAP, mitogen-activated protein; EMSA, electrophoretic mobility shift analysis; STAT, signal transducer and activator of transcription.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES