Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Suppression of interleukin‐3‐induced gene expression by a C‐terminal truncated Stat5: role of Stat5 in proliferation.

Suppression of interleukin‐3‐induced gene expression by a C‐terminal truncated Stat5: role of... The EMBO Journal vol.15 no.10 pp.2425-2433, 1996 Suppression of interleukin-3-induced gene by a C-terminal truncated Stat5: expression role of Stat5 in proliferation Wakao2'3, transduction. Structure-function analyses of the ,3-subunit Alice L.-F.Mui1, Hiroshi receptors have defined shared by the IL3, IL5 and GM-CSF Toshio Kitamura2 and Taisei Kinoshita2'3, domains necessary for activation of specific intracellular Atsushi Miyajima2'3 et al., 1992; Sato et al., signalling pathways (Sakamaki Departments of 'Molecular Biology and 2Cell Signalling, region is required for activation 1993). A membrane-distal DNAX Research Institute for Molecular and Cellular Biology, kinase cascade and induction of of the Ras/Raf-l/MAP 901 California Ave, Palo Alto, CA 94304, USA and 3Institute of A second, membrane-proximal, domain is needed c-fos. Molecular and Cellular Bioscience, The University of Tokyo, The Ras pathway for c-myc, pim-J and cis induction. 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113, Japan of apoptosis, appears to be important for prevention 'Corresponding author pathways directed by the receptor's mem- whereas the for DNA brane-proximal domain appear necessary Interleukin-3 (IL3) was shown recently to utilize the of synthesis (Kinoshita et al., 1995b). Maintenance long- transcription factor Stat5, but the genes regulated by both anti- term proliferation requires the contribution of this pathway and the biological consequence of Stat5 apoptotic and DNA synthesis signals. activation remained to be determined. In order to on the role of the Janus Recently, attention has focused study the role of Stat5 in IL3 signalling, we constructed action. Jak kinase (Jak) family of kinases in cytokine a dominant-negative Stat5 protein by C-terminal trunc- kinases were first described as functioning downstream of in IL3-dependent ation, and inducibly expressed it an have the interferon receptors, but many cytokine receptors Stat5 induc- cell line. The effect of dominant-negative now been shown to associate physically with and activate of IL3 early response genes was tion on expression specific Jak kinases (Ihle et al., 1994). Jak2, the Jak family and of several including examined, expression genes, kinase important for IL3 and GM-CSF action, associates cis, osm and pim-1 was inhibited profoundly. The with the 1-chain via the membrane-proximal domain. Jak expression of c-fos was also reduced, but to a lesser kinases, in turn, regulate latent, cytoplasmic transcription extent. While activated Ras alone (though not StatS factors termed Stats (Signal transducers and activators of could induce c-fos, maximal expression required alone) Stats dimerize and transcription). Upon phosphorylation, the action of both Ras and Stat5. Interestingly, although translocate to the nucleus where they bind to DNA the membrane-proximal region of the IL3 receptor sequences, most of which are related to the gamma 1-chain is responsible for both Jak2-StatS activation interferon activated site (GAS), a regulatory ele- (IFN-y) affected by and c-myc induction, c-myc levels were not ment in the promoter of genes (Darnell IFN-y-inducible the dominant-negative Stat5. Thus, the signals directed the Stat et Presently, seven members comprise al., 1994). which is essential by this membrane-proximal domain, of family and each functions in the signalling pathways for a DNA signal, can be separ- transducing synthesis For specific cytokines (Darnell et al., 1994). example, further into Stat5 or c-myc pathways. The net ated and et while IFN-a activates Statl Stat2 (Darnell al., effect of dominant-negative Stat5 expression was par- et Mui et 1994), IL3 utilizes StatS (Azam al., 1995; al., tial inhibition of IL3-dependent growth. This provides 1995) and Stat6 (Quelle et al., 1995). the first direct evidence that Stat5 is involved in identified as Since Statl and Stat2 originally were regulation of cell proliferation. of certain IFN- factors which regulate transcription Keywords: cell proliferation/c-myc/interleukin-3/Stat5 in to responsive genes, the targets of these Stats response the IFN stimulation are known. In contrast, target genes the Stat are of subsequent members of family largely Introduction is since unknown. For StatS, one defined target 1-casein, StatS was first purified as a prolactin-responsive regulator the differentiation and func- Cytokines govern growth, Other of this gene (Wakao et al., 1994). cytokines, tional of cells in the haemopoietic and immune activity Mui et GM- including IL3 (Azam et al., 1995; al., 1995) Of interleukin-3 (IL3) has a broad range systems. these, et Mui et IL5 on cells at various of CSF al., 1995; al., 1995), of function, with actions stages (Barahmand-pour et IL2 et Wakao et The (Mui al., 1995), (Lin al., 1995; al., haemopoietic development (Arai etal., 1990). receptor et and of a a-subunit and a 1995), erythropoeitin (Ep) (Wakao al., 1995) growth for IL3 consists ligand-specific later found also to hormone et were 13-subunit which is also found in the (Waxman al., 1995), high affinity receptor other do utilize StatS. most of these cytokines for two related IL5 and However, complexes functionally cytokines, and the factor not induce genes regulated colony-stimulating (GM- ,B-casein expression, granulocyte-macrophage these factors have not been Both subunits StatS in to et 1993). belong by response CSF) (Miyajima al., 1992, to characterize the role of StatS in both identified. In order to the cytokine receptor superfamily and, although a StatS reconstitute the IL3 we constructed dominant-negative are required to high affinity binding, signalling, dominant StatS has to the role in The use of this negative 1-chain is believed perform greater signal protein. Press ( Oxford University A.L.-F.Mui et al. in AStatS- expression of the truncated AStat5 product lanes 4 and but not transfected (Figure 2A, 5) parental The effect of AStat5 induction on the level cells (lane 2). 6000- was then of endogenous StatS-DNA binding activity C.) shift measured by electrophoretic mobility analysis U) 4000- DNA (EMSA) using the Stat5 target sequence, ,-casein 01) 0) et As shown in GAS, as a probe (Mui al., 1995). Figure ._ n 2B, induction of AStat5 greatly diminished the 3-casein 2000- in cells lane GAS binding activity Ba/F3 (Figure 2B, 4). This inhibition could be reversed the constitutive by OI of full-length, wild-type StatS (Figure 2B, overexpression lane 5). vector AStat5 on a Fig. 1. Effect of transient AStat5 expression Stat5-responsive Effect of 4Stat5 expression on various transfected as described reporter plasmid. Ba/F3 cells were transiently 1L3-activated signalling pathways in with the luciferase reporter Materials and methods Stat5-responsive The effect of AStatS expression on various IL3-activated the test DNAs: vector plasmid, ,B-casein-luc, and (pME18S), pME- stimulated or not with 10 IL3 AStat5 or pME-AStatl. Cells were ng/ml signalling pathways was examined in order to test the h for luciferase determination. Luciferase for 6 and extracted activity specificity of AStatS action. As shown in Figure 3A, is in units. Results are the of activity plotted arbitrary average AStatS induction profoundly inhibited the tyrosine in triplicate measurements three independent experiments. StatS. which phosphorylation state of endogenous AStatS, act is not phosphorylated itself, may thus by competing not only allowed identification of several StatS-regulated with wild-type StatS at the level of the 1L3-receptor to the or for access genes, but has also revealed a previously unappreciated complex, either for binding receptor for the role of StatS in IL3-dependent proliferation. to the tyrosine kinase responsible phosphorylation. However, AStatS expression had no measurabe effect on the of Vav or Shc. AStatS tyrosine phosphorylation Jak2, Results induction also did not alter the observed co-precipitation Expression of a dominant-negative Stat5 protein of a 90 kDa phosphoprotein with the 85 kDa tyrosine is for Stat dimer subunit of nor did it inhibit the Since tyrosine phosphorylation necessary phosphoinositol-3'-kinase, we reasoned that a truncation upward mobility shift, an indication of ligand-induced formation, carboxy-terminal which removes most of the C-terminal tyrosines, including phosphorylation, of Raf-1 kinase and p70 ribosomal S6- reduced the residue shown to be in kinase (S6K). However, AStat5 expression slightly tyrosyl phosphorylated response to stimulation et al., 1994), would MAP kinase (MAPK) activity by 30% (Figure 3B). Since prolactin (Gouilleux a not be reversed of produce a StatS protein that might function as dominant this inhibition could by overexpression of this the cDNA not of MAPK negative. As an initial test hypothesis, wild-type StatS (data shown), regulation was after Materials is not a normal function of StatS. More for StatSB truncated Tyr683 (see activity probably and and transfected into the IL3- probably, the extremely high levels of truncated AStatS methods) transiently cell with a protein achieved resulted in interactions not normally dependent line, Ba/F3, together Stat5-respon- luciferase Cells that were transiently encountered by wild-type StatS. In order to reduce the sive, reporter plasmid. transfected with the mutant StatS construct (AStatS) level of this interference, the amount of AStatS protein the expressed significantly less luciferase activity than control expressed was titrated down (by increasing concentra- 1 full MAPK cells that were transfected with the vector alone (Figure 1). tion of tetracycline). At ng/ml tetracycline, all For further analysis, AStatS was cloned downstream activity was restored (Figure 3C). For subsequent a and unless otherwise of tetracycline (tet) operator-controlled promoter experiments, indicated, AStat5 protein into cells to was induced in the presence of 1 ng/ml tetra- transfected Ba/F3 previously engineered partially the transactivator (tTA; or repressed by the addition of 1 ,ug/ml express tetracycline-regulated cycline fully Gossen and Bujard, 1992). Since tet operators are bacterial tetracycline. In addition, as a control for non-specific in origin, endogenous, mammalian promoters should not interference, the ability of wild-type StatS to reverse be affected directly by activation of the tTA transcription the phenotype of any observed AStatS-inhibited event factor. Several clones were chosen based on was checked. independent their of AStatS Before the effect of AStatS on tetracycline-dependent expression protein. investigating expression One of these clones was infected further with a retrovirus IL3-induced genes, we examined its effect on the induction harbouring wild-type StatS in order to overexpress the of a gene regulated by another Stat. Since IRF-I is in 2A wild-type protein these cells. Figure shows the regulated by IFN-y through Statl- (Darnell et al., 1994) AStatS and StatS as and not full-length expression, assessed by StatS-dependent pathways, AStatS should not Western analysis with an anti-StatS monoclonal antibody, inhibit the induction of IRF-I in response to IFN-y. As in the variously manipulated Ba/F3 cells. The level of expected, AStat5 indeed did not alter IRF-I IFN-y-induced endogenous, wild-type StatS was detected as a faint band expression (Figure 3D). in both parental Ba/F3-tTA and Ba/F3-AStatS cells (Figure lanes as 2A, 1-4) and, expected, the intensity of this band Effect of AStat5 expression on induction of was enhanced in cells infected by the StatS greatly 1L3-responsive genes retrovirus (Figure 2A, lane 5). Culturing these cells in The effect of AStatS expression on IL3 induction of early inducing conditions (no tetracycline) resulted in high level was then assessed response genes by Northern blot analysis 2426 Role of Stat5 in cell proliferation Un cn U_) C/) 4- 4- C), C.) C) 'a aa C 4 ;; + co + Q - cn L3 *-+*'-++ Induction Induction -+ -m .+ full length Stat5 Stat5-DNA AStat5 complex 1 2 34 5 1 2 34 5 Western EMSA Fig. 2. Effect of AStatS expression on endogenous StatS DNA binding activity. Cell lysates from Ba/F3 cells expressing tetracycline-inducible AStatS were subjected to Westem analysis with an anti-Stat5 monoclonal antibody (A) and gel mobility shift analysis (B) with a labelled J-casein GAS oligonucleotide probe (Mui et al., 1995), respectively. Lanes 1 and 2 are from parental Ba/F3-tTA cells, lanes 3 and 4 are from AStat5 transfectant clone 23. Lane S represents lysate from AStatS clone 23 which had been infected with a retrovirus harbouring full-length, wild-type StatSB in order to overexpress wild-type Stat5 in these cells. Samples in lanes I and 3 are from cells grown in non-inducing (-) conditions (1 tetracycline), gg/ml while those in lanes 2, 4 and 5 are from cells grown for 24 h in inducing (+) conditions (no tetracycline). All cells were harvested from cultures growing exponentially in 1L3. A B anti-protein blot anti-p-tyr - + induction I- + I induction - + 1+ + IL3 77]I F- +1 +1 77741 1L3 activity [I7 rx;-+ _MAPK EZI E: .. Stat5 =_I tet 0 10 100 1000 ( ng/mL) 0.1 1.0 Jak2 4S'" * UV4II anti-stat5 -:, vavLX activity IMAPK p85 D induction - + Shc 0 0.5 1.0 0.5 1.0l IFN/(hr) _*-eIRF-1 _0 _m wo Raf-1 GAPDH S6K induction on IL3-induced activation of signalling molecules. AStatS expression in Ba/F3- Fig. 3. Specificity of AStatS action. (A) Effect of AStatS not 12 h culture or with, 1 tetracycline, respectively, in growth medium (10 ng/ml IL3, 10% AStatS cells was induced (+), or (-), by without, gg/ml in RPMI were then washed free of IL3 and cultured in 10% FCS in RPMI 1640 containing tetracycline as appropriate. After 8 h, FCS 1640). Cells et 1995). Antibodies to cells (2X 107 were stimulated with 1 ng/ml 1L3 for 10 min and detergent lysates prepared as described (Mui al., cells/point) and Shc were used to immunoprecipitate Jak2 Vav the 85 kDa subunit of phosphoinositol-3'-kinase (p85, UBI) (Transduction) StatS, (UBI), (UBI), were then resolved SDS-PAGE and by immunoblotting with anti- the molecule from the cell lysates. The immunoprecipitates by analysed designated ribosomal S6-kinase (S6K), 1 of cell lysate was resolved directed by (clone 4G10, UBI). For Raf-1 kinase (Raf-l) and p70 gg phosphotyrosine with antibodies either Raf-l or S6K. Effect of AStatS expression on MAPK activity. SDS-PAGE and subjected to immunoblotting recognizing (B) was and et in lysates prepared as in (A). (C) Effect of tetracycline concentration on MAP kinase activity determined quantitated (Samuels al., 1993) and MAPK AStatS was induced in varying concentrations of tetracycline, and AStatS protein detected as in Figure AStat5 expression activity. protein of IRF-1. AStatS expression in Ba/F3- I or MAPK determined as above. (D) Effect of AStatS induction on IFN-y-stimulated expression activity 1 in medium. Cells (2X 10 cells/ AStatS cells was induced or not (-), by 24 h culture without, or with, ,ug/ml tetracycline, respectively, growth (+), 1 time and RNA extracted the RNeasy extraction kit. The were then stimulated with zg/ml IFN-y for the indicated length of using Qiagen point) IRF-J and GAPDH RNA was analysed by Northern analysis with probes. 2427 A.L.-F.Mui et al. + wt parental \Stat5 \Stat5 Induction - ++ F.----l +---- 1-- r 00.5 2 3 0 05 2 .5 1 2 3 0 0.5 1 2 3 IL3 (hr) 0 0.5 1 0 u. \Stat5 expression ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. .. .. .. ... .... .. .. .... .... .. .. _ ... ,~0 ° [tet] (.ig/mL) -o6 cisIX cis pim-1 ~~4,'41 pim-1 Id-1 _ __ _ Id-1 osm osm c-Fos 1 c-Fos c-myc i]1 ........_ _... __ _ _ _ _. _ ........ c-myc |_ A cyclin 02 cyclin 02 cvclin D3 cyclin GAPDH * GAPD)HE | 4. Effect of induction on of IL3-inducible or Ba/F3-AStat5 + wild- Fig. AStat5 expression genes. (A) Ba/F3-tTA (parental), Ba/F3-AStat5 (AStat5) + wt cells were cultured for 24 h under AStat5 or conditions as in 2. Cells were then type Stat5 (AStat5 Stat5) inducing (+) non-inducing (-) Figure washed free of IL3 and starved for 8 h in 10% FCS in RPMI or I as These cells (2X 107 cells/ lacking containing ,ug/ml tetracycline appropriate. were then stimulated for the indicated of time with 10 IL3 before RNA was extracted as in 2. RNA was point) length ng/ml Figure (10 jig/lane) on a and to Northern with various random DNA Ba/F3-AStat5 cells were separated formaldehyde gel subjected analysis prime-labelled probes. (B) induced for 24 h as in with concentrations of before IL3 for 8 h. The cells were then stimulated for 1 h with (A) varying tetracycline deprivation 1 ng/ml IL3 and RNA for Northern as above. prepared analysis two is (Figure 4, left panel). Parental or AStatS Ba/F3 cells were separated further into at least pathways, one which in or conditions as and another which is for cultured AStatS-inducing non-inducing StatS dependent responsible and then stimulated with As induction. The inhibition of before, IL3. previously c-myc expression by c-fos resulted in the AStatS, on the other hand, was unexpected, since analysis reported, IL3 stimulation rapid appearance of a number of of the truncation mutants had the genes, including cis, pim-l, osm, c-fos, receptor assigned region and and those D2 and D3 required for induction to a membrane-distal Id-l c-myc encoding cyclins c-fos second, et Kinoshita et Yoshimura domain that is also for Raf and MAP (Sato al., 1993; al., 1995b; necessary Ras, et 1995 and et kinase activation (Sato et The data al., M.Takagi al., unpublished). Although al., 1993). present c-myc and cyclin D2 and D3 was not altered suggest that signals from both and expresssion membrane-proximal in cells induced to express AStatS, the expression levels membrane-distal domains may be required for full induc- of cis, pim-J, osm, Id-i and c-fos were tion of c-fos. In fact, both human (Treisman, 1986) and significantly in reduced, suggesting that these five genes are regulated by murine (Renz et al., 1985) c-fos promoters contain, of addition to the Ras/Raf/MAP StatS-dependent pathways. Importantly, overexpression kinase-responsive element, Stat5 restored the induction StatS et Gouilleux et wild-type gene pattern (Figure potential (Schmitt al., 1991; al., to that of control cells. In the 1995) binding sites (human: 5'-TGCactGAA, mouse: 4, right panel) addition, degree of inhibition of gene expression correlated with 5'-ATCtccGAA), and co-transfection of wild-type StatS the amount of AStat5 protein expressed. Decreasing the greatly augmented the luciferase activity from a c-fos- levels of AStatS protein by increasing the concentration promoter reporter plasmid (Figure SA) in Ba/F3 cells. of The StatS alone is tetracycline (see Figure 3C) diminished the degree of signal insufficient for c-fos, since cis, pim-l, and inhibition truncation which lack the osm, Id-l c-fos (Figure 4B). receptor mutants, domain needed The sensitivity of cis, pim-i, osm and Id-i to AStatS for activation of the Ras/Raf/MAP cascade but retain the expression is consistent with studies which mapped the region required for the Jak2-Stat5 pathway, cannot induce of the IL3/GM-CSF region receptor 5-chain responsible (Sato et al., 1993; Quelle et al., 1994; Mui et al., c-fos for their induction et (Sato al., 1993; Yoshimura et al., 1995). However, the Ras-mediated signal is also not 1995; to the same membrane- sufficient for full of A.Miyajima, unpublished) expression c-fos. Previously, we proximal domain required for Jak2 (Quelle et al., 1994) established Ba/F3 cell lines the expressing constitutively and StatS (Mui et activation. actvated form of al., 1995) Interestingly, Ras (Rasvall2) under the control of a although c-myc induction also required the membrane- dexamethasone (dex)-responsive, mouse mammary domain et proximal (Sato al., 1993), c-myc expression did tumour virus (MMTV) promoter. In Ba/F3 cells expressing not to be appear regulated by StatS. Thus the signals GM-CSF receptor truncation mutants membrane- lacking from the can emanating membrane-proximal region be distal domains, induction of RasvalI2 protein is able to 2428 Role of Stat5 in cell proliferation sion of full-length, wild-type Stat5 (Figure 6B). As one test of the specificity of AStat5 inhibition, we compared its effect on IL3- or IL4-supported viability of Ba/F3 cells. IL4 utilizes Stat6 and not Stat5 (Hou et al., 1994; data not shown) and thus, as expected, AStat5 expression inhibits the IL3 but not the IL4 response (Figure 6C). As another test of the specificity of growth inhibition, the X,, 10000- colony-stimulating factor-I receptor (CSF-1R) was intro- duced into either parental tTA or AStat5 Ba/F3 cells. Since _ IL3 CSF- 1 stimulation did not activate Stat5 detectably in - IL3 these cells (data not shown), AStat5 should not inhibit CSF- 1-driven [3H]thymidine incorporation. As shown in 5000- CZM Figure 6D, although AStat5 induction inhibits [3H]- -j thymidine incorporation in response to it does not IL3, alter the response to CSF-1. Significantly, even at the highest levels of AStat5 vector Stat5B vector Stat5B test DNA protein (when cells are induced by complete removal of tetracycline), growth is not fully inhibited. To examine c-fos luc .)-casein luc reporter DNA this in more detail, cell cycle analyses were carried out which revealed that although induction of AStat5 retards the rate of entry into and the percentage of cells entering S-phase, it does not prevent GI/S progression completely (Figure 6E-G). rasva[12 Discussion parental (IL3) tDex) Specificity of AStat5 action 0 0.5 1.0 0 12 24 time(hr) To the of the investigate contribution Jak2-Stat5 pathway .4- c-fos to IL3 signalling, we expressed a C-terminal truncated t N N - GAPDH o4w version of Stat5, AStat5, in Ba/F3 cells. Expression of AStat5 in Ba/F3 cells interfered with the function of 1 2 3 4 5 6 endogenous Stat5; it decreased the levels both of luciferase Fig. 5. Regulation of expression. (A) Ba/F3 cells were c-fos transiently activity from a Stat5-responsive reporter plasmid and of transfected with either the f-casein-luciferase or c-fos-luciferase Stat5-DNA binding activity, detected by EMSA of cell reporter plasmids and either vector (pME18S) or pME-Stat5B. After the recovery period, cells were divided and cultured for 6 h in IL3 or lysates. AStat5 apparently acts as a dominant-negative by no factor. Luciferase activity was determined from cell extracts. Data interfering at the level of either the receptor or the are the means of tfiplicate measurements. Similar results were kinase responsible for the phosphorylation. However, an obtained in two separate experiments. (B) Parental Ba/F3 cells (lanes in of important consideration interpreting results experi- 1-3) were stimulated with IL3 or pMAM-RasValI2_expressing Ba/F3 is the ments utilizing dominant-interfering molecules, cells (lanes 4-6) were treated with dexamethasone as described et (Kinoshita al., 1995b). Total RNA was isolated and subjected to of action. For AStat5 associate specificity example, may Northern with c-fos as in 4. analysis probe Figure with the IL3R and activation of other stably prevent which the same domain of the In signals require receptor. compensate for the lack of a GM-CSF-generated, Ras an to address this we tested but found attempt question, activation signal and collaborate with GM-CSF to no effect of AStat5 on various other support signalling pathways. long-term proliferation. Two Ras-regulated genes which At extremely high expression levels of AStat5, we did may mediate its anti-apoptotic action, bcl-2 and bcl-x, are observe a slight inhibition of MAPK so we carefully dex to the same as the level of to eliminate this induced by extent observed with IL3 adjusted down AStat5 potential stimulation et we cannot rule out the that (Kinoshita al., 1995a). However, although problem. However, possibility dex treatment resulted in other either known ones we did not check or c-fos expression (Figure 5B, pathways, lanes the amount of mRNA did not rise to the unknown ones which have not been are 4-6), c-fos yet characterized, same level as that observed in IL3-stimulated cells affected AStat5. (Figure non-specifically by of action lanes unlike bcl-2 and the We also assessed the specificity AStat5 by 5B, 1-3). Thus, bcl-x, expression the of more than mere Ras activation. the of Stat5 to reverse observed c-fos requires testing ability wild-type The of to phenotypes. ability wild-type protein compensate Effect of AStat5 on for the action of molecules is a com- expression IL3-dependent dominant-negative control. For our proliferation monly accepted experiments, however, if the fact that AStat5 does not inhibit or this control must be the caveat that AStat5 Despite c-myc qualified by D-cyclin it DNA non-specifically interferes with unrelated pathways by expression, compromised IL3-dependent of the then with cultures continuous occupation receptor, wild-type synthesis (Figure 6A), AStat5-expressing than controls after 24 h Stat5 still be able to reverse the inhibition containing 40% fewer cells may by (data with This inhibition is on the levels of AStat5 from a stable not shown). dependent preventing forming complex AStat5 protein expressed (i.e. dependent on the concentra- the Definitive evidence for the of the receptor. specificity thus tion of and could be reversed inhibitory effects we report here tetracycline) by overexpres- may require targeted 2429 A.L.-EMui et al. 0 10000 o l { . 8000D CL i .*- Ai, tTA- ---- not iniduced Ec 0 p.-s- s000 .. -,r- - tTA induced .\Stat5 0 -0W- s/t * oStatS + vwt P.0 * t%Stat5 - not nduced .E 4000 IUD' 1 0- 00 Stat5 - induced 1- >, 2000 u01 0.1 1 10 . 1 i 10 100 nor IL.3 concentration (ng/mL) tetracycline concentration (ng/mL) C D E.o L not induced IC.E E _not induced r- induced 0, 0 0 6... IL3 1L4 [jjjFj[jjI CTA-CSF1R Statb-CSF1R F 3 I[~1 cn 7; t'rA not induced 4) u) t.TA- iriduced 'b a Stat5 - not induced 40- o1 ..0 _-- Stat5 indLJced 0'... _ .- ., -10 0 1 0 20 30 -1 0 0 1 0 2 0 30 + time (hrs) + time (hrs'. IL3 IL3 -.--- .- 20Or G2/M a. IT U) I S1 ID a A/ o9 -10 0 1 0 20 30 (hrs) time 1L3 6. Effect of AStat5 on Parental Ba/F3-tTA and Ba/F3-AStat5 cells were cultured under Fig. expression proliferation. (A) (OL 0) (U @) inducing in 2. The and (O @) or non-inducing (C1 U) conditions as described Figure cells were then deprived of IL3 as before their IL3 growth response was assessed in a [3H]thymidine assay. (B) Comparison of the [3H]thymidine incorporation of Ba/F3-AStat5 and Ba/F3-AStat5 StatS (Cl) wild-type (U) Effect of on and maintenance of cell cells at various concentrations of tetracycline. (C) AStat5 expression IL3 IL4 Ba/F3 viability. Ba/F3-AStat5 as and their to 1 or was measured in an Alamar Blue cells were induced (U) or not (LI),to express AStat5 above, response ng/ml IL3 IL4 (Alamar colorimetric as described of the of Biosciences) assay (Ho et al., 1995). (D) Comparison [3H]thymidine incorporation CSF-lR-expressing Ba/F3-tTA and Ba/F3-AStat5 cells in to IL3 or under or conditions. F and Cell status S response CSF-l, inducing (U) non-inducing (El) (E, G) cycle (GO/GI, and G2/M of Ba/F3-tTA and Ba/F3-AStatS cells cultured under or phase respectively) parental (El 0) (U @) inducing (O @) non-inducing (OL U) = h conditions as in After 12 h of the cells were washed free of IL3 time -10 and restimulated with I IL3 at (A). induction, (at ) ng/ml = 0 time h. of and osm was since disruption both StatS genes (Mui et al., 1995) expressed regulating cis, pim-J, Id-l expected, in murine cells. However, corroborative evidence as these to the same domain genes map membrane-proximal detailed below that the we of the n-chain as those for Jak2-Stat5 activation. suggests phenomena report receptor here from result specific inhibition of the StatS pathway. In contrast, although this membrane-proximal domain is also responsible for c-myc induction, c-myc levels were Identification of a not affected Stat5-regulated genes using by AStat5 expression. These data dissociate dominant Stat5 induction from activation. protein c-myc StatS However, as c-myc Induction of five genes, cis, pim-J, Id-], osm and c-fos, induction is sensitive to tyrosine kinase inhibitors was inhibited by AStat5 expression. A role for StatS in (Kinoshita et is al., 1995b), c-myc induced by either a 2430 Role of Stat5 in cell proliferation tyrosine kinase distinct from Jak2, or alternatively, through important for proliferation, since certain IL2R mutants a Jak2-dependent pathway which is distinct from that which lack the ability to activate StatS are still able to responsible for Stat5 activation. stimulate mitogenesis. Close examination of their data, Inhibition of gene expresssion could be a direct result however, shows that the relevant mutant, H-4, is reduced, of AStat5 on transcription from their promoters, or an by a statistically significant 15%, in its ability to support indirect effect, i.e. transcription requires the action of a IL2-driven growth. Since growth can still occur at a StatS-regulated gene product rather than Stat5 itself. To reduced level, Fujii et al. conclude that StatS is not investigate these possibilities, the available promoter important for proliferation. However, we and others sequences of murine pim-J, cis and osm were also (Damen et al., 1995; Friedmann et al., 1995) propose that examined for potential Stat5 binding sites. Elements con- StatS activation is required for a maximal proliferative forming to the Stat5 GAS consensus were found for response. the c-fos, cis and osm promoters, and DNA fragments Several reasons potentially may explain the incomplete containing this region were able to confer Stat5-dependent nature of the growth inhibition. The simplest explanation a of AStatS results in transcription on reporter gene (Figure SA and Yoshimura is that insufficiently high expression et the other hand, although a canonical This al., 1996). On incomplete blockade of StatS action. hypothesis of inhibition would be GAS motif occurs in the human pim-] promoter, none would predict that the degree could be found in the published murine pim-J 5'-flanking lower of IL3 but, as Figure 6A shows, greater at doses A occur. the other the data are sequence (Yip-Schneider et al., 1995). StatS binding this does not On hand, be of The first arises site may either outside the published 5'-flanking consistent with two other possibilities. or from the GAS consensus. the that StatS regulates many target region it may deviate from observation the promoter activity of are for Alternatively, StatS may affect genes. Some of these important proliferation (i.e. pim-1 indirectly. while like cis et others, (Yoshimura al., 1995), may c-fos) Interestingly, cis, pim-J, Id-], osm and c-fos display have a role. During the course negative growth regulatory varying degrees of inhibition by AStatS. This probably of these genes are not expressed normal IL3 signalling, reflects the process of signal integration that provides the simultaneously since other factors in addition to StatS are combinatorial versatility in gene regulation. Athough GAS involved in their regulation. The global inhibition of all sites may be present in a gene promoter, the extent to which StatS-regulated genes by AStatS thus results in antagonistic they are utilized depends on many other considerations. For signals, and the net effect of these interactions is partial example, slight variations in the GAS sequence may alter growth inhibition. However, a second possibility is that a its relative affinity for available Stat dimers (Zhang et al., full proliferative response to cytokines, including IL3, elements and factors that 1995). Additional promoter results from the contribution of several nominally inde- mRNA will regulate stability also contribute to the net pendent pathways. Previous studies using IL3/GM-CSFR expression level of a Stat-responsive gene. Futhermore, ,-chain deletion mutants defined a membrane-proximal the homo- DNA potential for or heterodimer formation, serine/ domain which is important for synthesis (Kinoshita threonine and interaction other The effect of on phosphorylation with non- et al., 1995b). partial inhibitory AStat5 Stat will the a DNA synthesis, therefore, further dissects the signals from proteins modify potential ability of Stat complex to transactivate even if it binds to a given the membrane-proximal domain; inhibition of one of promoter (Bluyssen et al., 1995; Eilers et al., 1995). The these the StatS diminishes but does not signals, pathway, regulation of c-fos expression provides a good illustration eliminate IL3-driven proliferation. of some of these complexities. First, both StatS and Ras The mechanism by which AStat5 inhibits DNA synthesis pathways are required for full induction of Second, is not clear. StatS may regulate a gene(s) directly required c-fos. although Ras alone can induce a low level of c-fos, StatS for progression into S-phase or, alternatively, control a activation, by itself, cannot. gene(s) whose product contributes to establishment of the S-phase competence. None of cyclin genes, however, Dominant-negative Stat5 inhibition of appear to be inhibited by AStatS. Among those genes which 1L3-dependent proliferation which are inhibited by AStatS, the mechanism by The is of decreased cis and Id-i contributes to net result of AStatS expression inhibition IL3- expression growth driven DNA have been is not obvious. Cis is to synthesis. Stat proteins recently inhibition immediately thought in the control of studies which associated with of a implicated proliferation by be down-regulation proliferative that cells transformed by HTLV-I (Migone et al., et while Id-I and related reported signal (Yoshimura al., 1995), 1995), bcr-abl (Danial et al., 1995) or src (Yu et al., genes have been variously reported to be involved either in 1995) exhibit constitutively activated Stat-DNA binding differentiation (Voronova and Lee, 1994) or progression GI of the serine/ activity. Similar conclusions have been derived from (Hara et al., 1994). The induction pim-1 are to activate on the other correlates with mito- analysis of receptor mutants which unable threonine kinase, hand, who have examined either in several et the StatS pathway. Investigators genesis systems (Lilly al., 1992). Moreover, et or IL2R et with in EpR (Damen al., 1995) (Friedmann al., pim-i synergizes myc leukaemogenesis (van mutants have correlated the to Lohuizen et and mast cells isolated from 1995) receptors' ability al., 1989), pim- to activate StatS with their to transduce a I-deficient mice exhibit an impaired response IL3 ability mitogenic in a manner similar to that of Ba! the StatS mutants (Domen et al., 1993) signal. Significantly, pathway-defective in their cells induced to AStatS. constitutive are F3 express However, again only partially impaired proliferative on cell is similar to of did not rescue the capacity: this limited effect growth expression pim-] proliferative action. of AStatS-inhibited cells not that observed with AStatS Other investigators (Fujii response (data shown). Thus, in have that StatS is not the for the inhibition et reported key target responsible growth may al., 1995), contrast, 2431 A.L.-F.Mui et aL either a minimal or c-fos promoter and test DNAs, in pME be an as unidentified 18S, yet StatS-responsive gene, or growth P-casein at room in temperature RPMI supplemented with 10 mg/ml DEAE- inhibition be the sum of several may interacting events. dextran. After a 12 h in recovery period 1L3-containing medium, cells A more detailed of which are survey genes differentially were deprived of IL3 for 12 h before a 6 h stimulation period with in cells induced to should help expressed express AStatS 10 no ng/ml IL3 or factor. Cytoplasmic extracts were prepared by three of freeze-thaw in 0.25 M Tris to resolve this cycles normalized question. pH 7.5, by protein concentration and for luciferase assayed activity the of this (Promega). During preparation manuscript, two reports that correlated the of certain receptor appeared inability Thymidine incorporation assay mutants to activate Stat5 with their ability to impaired For the 2x 104 assay, cells were plated per well in a 96-well plate in a et 1995; Friedmann serial dilutions of in RPMI + I support mitogenic signal (Damen al., IL3 5% FCS and jg/ml tetracycline as After 24 I of appropriate. h, ,tCi [3H]thymidine was added for 2 h et and others have observed constitutive activa- al., 1995), and the cells were harvested onto glass fibre filters and incorporated tion of the Jak-Stat in transformed and factor- pathway [3H]thymidine quantitated by liquid scintillation counting. cells et 1995; Migone et al., independent (Danial al., Yu et we provide the first 1995; al., 1995). However, Acknowledgements direct evidence that StatS is for proliferation. important The use of a Stat protein has also dominant-negative We thank Drs G.Krystal, K.Moore and M.McMahon for their critical allowed us to several and reading of this manuscript, and M.Gossen and H.Bujard for the tetra- identify Stat5-regulated genes, cycline repressor plasmids. A.Miyajima was the partially supported by this can be extended to of other Stats technique analysis Ministry of Education, Science and Culture of Japan. DNAX Research and other Even though the identity of cytokine systems. Institute is supported by Schering-Plough Corporation. the Stats which are activated in each is becoming cytokine well the fundamental question of what role these defined, References Stats in function remains to be answered. perform cytokine Arai,K., Lee,F., Miyajima,A., Miyatake,S., Arai,N. and Yokota,T. (1990) coordinators of Cytokines: immune and inflammatory responses. Annu. Materials and methods Rev. Biochem., 59, 783-836. Azam,M., Erdument-Bormage,H., Krieder,B.L., Xia,M., Quelle,F., Materials Basu,R., Saris,C., Tempst,P., Ihle,J.N. and Schindler,C. (1995) The kind repressor-transactivator system plasmids were the tetracycline Interleukin-3 signals through multiple isoforms of Stat5. EMBO J., of Drs M.Gossen and The retroviral is a H.Bujard. vector, pMX, gifts 14, 1402-1411. derivative of BOSC 23 and unpublished). (ATCC) pBabe (T.Kitamura, Barahmand-pour,F., Meinke,A., Eilers,A., Gouilleux,F., Groner,B. and cells were maintained as described. All other materials were Ba/F3 Decker,T. (1995) Colony-stimulating factors and interferon-gamma obtained as described. previously activate a protein related to MGF-Stat 5 to cause formation of the differentiation-induced factor in cells. FEBS myeloid Lett., 360, 29-33. Construction of AStat5 Bluyssen,A.A.R., Muzaffar,R., Vlieststra,R.J., Van der Made,A.C.J., Stat5B was truncated after the XhoI-NotI Tyr683 by replacing fragment Leung,S., Stark,G.R., Kerr,I.M., Trapman,J. and Levy,D.E. (1995) in et with an XhoI-NotI-digested pME18S-StatSB (Mui al., 1995) Combinatorial association and abundance of components of interferon- PCR with 5'-TGACCAAGGAGAACCTCGTGT-3' fragment amplified stimulated gene factor 3 dictate the selectivity of interferon responses. and 5'-TAGTTATGCGGCCGCTAGTAGTACTTAGAATATA- (sense) Proc. Natl Acad. Sci. USA, 92, 5645-5649. CTT-3' The (antisense) primers. resulting construct, pME18S-AStatS, Damen,J.E., Wakao,H., Miyajima,A., Krosl,J., was Humphries,R.K., confirmed sequencing. The AStat5 cDNA was excised with by Cutler,R.L. and Krystal,G. (1995) Tyrosine 343 in the erythropoietin EcoRI-NotI and blunt-end into the XbaI site of ligated pUHDIO-3-hygro receptor positively regulates erythropoietin-induced cell proliferation and with a PGK promoter-driven [pUHDIO-3 (Gossen Bujard, 1992) and StatS activation. EMBO J., 14, 5557-5568. resistance inserted at the gene PvuII site]. hygromycin Danial,N., Pernis,A. and Rothman,P. (1995) Jak-STAT signaling induced by the v-abl oncogene. Science, 269, 1875-1877. Production of stable Ba/F3 cell clones AStat5-expressing Darnell,J.J., and Stark,G.R. (1994) Jak-STAT pathways Ba/F3 cells the tTA were transfection Kerr,I.M. and expressing protein generated by transcriptional activation in response to IFNs and other with and which had been modified extracellular pUHD15-1 (Gossen Bujard, 1992) signaling proteins. Science, 264, 1415-1421. at the XhoI site to include a resistance under the control puromycin gene of an Domen,J., van der Lugt,N.M., Laird,P.W., Saris,C.J., SV40 isolated dilution in Clarke,A.R., early promoter. Clones, by limiting 1.5 in Hooper,M.L. and Berns,A. (1993) Impaired interleukin-3 response in IL3 media [10 ng/ml IL3, 10% fetal ,ug/ml puromycin growth calf serum 50 in RPMI Pim-l-deficient bone marrow-derived mast cells. Blood, 82, 1445- mercaptoethanol 1640], were tested (FCS), ,uM for functional tTA transient 1452. protein by transfection with the tetracycline luciferase Eilers,A., Georgellis,D., Klose,G., Schindler,C., Ziemiecki,A., Harpur, reporter plasmid, pUHC 13-3. One clone operator-controlled A.G., Wilks,A.F. and Decker,T. (1995) Differentiation-regulated which well in this was then transfected with pUHDIO- serine responded assay phosphorylation of STAT1 promotes GAF activation in and selected in 1.4 I puro- macrophages. 3-hygro-AStatS mg/ml hygromycin, gg/ml Mol. Cell. Biol., 15, and 1 in IL3 medium. Tetracycline 3579-3586. mycin ,ug/ml tetracycline growth Friedmann,M., Migone,T., Russell,S. and Leonard,W. (1995) was 2 with a half medium change. To identify Different replaced every days clones which IL2R beta tyrosines couple to at least two signalling pathways and clones were washed express AStatS, hygromycin-resistant free of and cultured for 24 h in synergistically mediate IL2 induced proliferation. Proc. Natl Acad. tetracycline the absence or presence of Cell were as Sci. USA, 93, 2077-2082. tetracycline. lysates prepared described (Mui et al., 1995) and to Western with an anti-StatS Fujii,H., Nakagawa,Y., Schindler,U., Kawahara,A., Mori,H., subjected analysis monoclonal antibody Gouilleux,F., Groner,B., Ihle,J.N., Minami,Y., Miyazaki,T. Several clones were analysed, and data and (Transduction Lab). independent from one Taniguchi,T. (1995) Activation of Stat5 by interleukin 2 requires clone are shown in Figure 1. To overexpress a representative StatS in carboxyl-terminal region of the interleukin 2 receptor beta chain but AStat5 transfectants, one of these clones full-length wild-type was co-cultured with a viral line is not essential for the proliferative signal transduction, Proc. Natl packaging (BOSC 23) which had been transfected with StatS cloned into a retroviral vector, pMX. Acad. Sci. USA, 92, 5482-5486. wild-type After 24 h co-culture in 8 the Ba/F3 cells were cloned Gossen,M. and Bujard,H. (1992) Tight control of gene expression in jig/ml polybrene, mammalian cells by tetracycline-responsive promoters. Proc. dilution. Several clones were isolated which overexpressed Natl by limiting Acad. Sci. USA, 89, 5547-555 1. StatS as assessed Western of cell lysates; the data full-length by analysis from one Gouilleux,F., Wakao,H., Mundt,M. and Groner,B. (1994) of these clones are shown in 1. Prolactin Figure induces phosphorylation of Tyr694 of StatS (MGF), a prerequisite for Transient transfection assay DNA binding and induction of transcription. EMBO J., 13,43614369. Ba/F3 cells were at mF 960 and 270 V with 15 of Gouilleux,F., Pallard,C., Dusanter,F.I., Wakao,H., Haldosen,L.A., pg electroporated of a luciferase under the control of and Groner,B. (1995) Prolactin, growth hormone, Norstedt,G., Levy,D. reporter plasmids consisting gene 2432 Role of Stat5 in cell proliferation erythropoietin and granulocyte-macrophage colony stimulating factor genesis in pimn-1 transgenic mice: cooperation with c-mvc and N-mvc induce MGF-Stat5 DNA binding activity. EMBO J., 14, 2005-2013. in murine leukemia virus-induced tumors. Cell, 56, 673-682. Hara,E., Nojima,H., Ide,T., Campisi,J., Okayama,H. and Voronova,A.F. and Lee,F. (1994) The E2A and tal-1 helix-loop-helix Yamaguchi,T., Oda,K. (1994) Id-related genes encoding helix-loop-helix proteins proteins associate in vivo and are modulated by Id proteins during are required for GI progression and are repressed in senescent human interleukin 6-induced myeloid differentiation. Proc. Natl Acad. Sci. fibroblasts. J. Biol. C/ewn., 269, 2139-2145. 5952-5956. USA. 91, Ho,A.S.-Y., Wei.S.H.-Y., Miyajima,A. and Moore,K.W. Gouilleux,F. and Groner,B. (1994) Mammary gland factor Mui,A.L.-F.. Wakao,H., (1995) Functional regions of the mouse interleukin-10 receptor (MGF) is a novel member of the cytokine regulated transcription cytoplasmic domain. Mol. Cell. Biol., 15, 5043-5053. factor gene family and confers the prolactin response. EMBO J., 13, Hou,J., Schindler,U., Henzel,W.J.H., HoT., Brasseur.M. and McKnight,S. 2182-2191. (1994) An interleukin-4-induced transcription factor: IL-4 Stat. Wakao,H., HaradaN., Kitamura,T., Mui,A.L. and Miyajima,A. (1995) Science. 1701-1706. Interleukin-2 and erythropoietin activate STatS/MGF via distinct Ihle,J.N., Witthuhn,B.A., Quelle,F.W., Yamamoto,K., Thierfelder,W.E., pathways. EMBO J., 14, 2527-2535. Kreider,B. and (1994) Signaling by the cytokine Waxman,D.J., Ram,P.A., Park,S.H. and Choi,H.K. (1995) Intermittent Silvennoinen,O. receptor superfamily: JAKs and STATs. Trends Biochem. Sci., 19, plasma growth hormone triggers tyrosine phosphorylation and nuclear 222-227. translocation of a liver expressed, StatS-related DNA binding protein. Kinoshita,T., Yokota,T., Arai,K. and Miyajima,A. (1995a) Regulation of J. Biol. Chemn., 270, 13262-13270. Bcl-2 expression by oncogenic Ras protein in hematopoietic cells. Yip-Schneider,M.T., and Broxmeyer,H.E. (1995) Tran- Horie,M. Oncogene, 10, 2207-2212. scriptional induction of pim- 1 protein kinase gene expression by Kinoshita,T.. Yokota,T., Arai,K. and Miyajima,A. (1995b) Suppression interferon gamma and posttranscriptional effects on costimulation of apoptotic death in hematopoietic cells by signalling through the with steel factor. Blood, 85, 3494-3502. IL-3/GM-CSF receptors. EMBO J., 14, 266-275. Yoshimura,A., Ohkubo,T., Kiguchi,T., Jenkins,N.A., Gilbert,D.J., A novel Lilly,M., Le,T., Holland,P. and Hendrickson,S.L. (1992) Sustained Copeland,N.G., Hara,T. and Miyajima,A. (1995) cytokine- that binds to expression of the pim- 1 kinase is specifically induced in myeloid cells inducible gene CIS encodes an SH2-containing protein by cytokines whose receptors are structurally related. Oncogene, 7, tyrosine phosphorylated interleukin 3 and erythropoietin receptors. 727-732. EMBO J., 14, 2816-2826. of shared receptor motifs and common Lin,J.X. et al. (1995) The role Yoshimura,A., Ichichara,M., Kinjyo,l., Moriyama,M.. Copeland,N.G., and Mouse Stat proteins in the generation of cytokine pleiotropy and redundancy Gilbert,D.J., Jenkins,N.A., Hara,T. Miyajima,A. (1996) Ininunitv, 2, 331-339. M: immediate induced multiple cytokines by IL-2, IL-4, IL-7, IL-13, and IL-I5. oncostatin an early gene by Mulloy,J., O'Shea,J., Franchini,G. and a common EMBO J., 15, 1055-1063. Migone,T., Lin,J., Cereseto,A., through JAK-STAT pathway. activated Jak-STAT pathways in T Larner,A., Carter-Su,C., Schartz,J. and Leonard,W. (1995) Constitutively Yu,C., Meyer,D., Campbell,G., 79-81. Enhanced of a Stat3-related cells transformed wiht HTLV-1. Science, 269, Jove,R. (1995) DNA-binding activity Kitamura,T., Yokota,T. and Arai,K. (1992) protein in cells transformed by the src oncoprotein. Science, 269, Miyajima,A., Harada,N., and transduction. Annu. Rei: Immunol., 10, 81-83. Cytokine receptors signal 295-331. Li,H., Schindler,C. and Chen-Kiang,S. (1995) Zhang,X., Blenis,J., Miyajima,A., Mui,A.L.-F., and (1993) Requirement of serine phosphorylation for formation of STAT- Ogorochi,T. Sakamaki,K. for colony-stimulating factor, Science, 267, 1990-1994. Receptors granulocyte-macrophage promoter complexes. interleukin 3 and interleukin 5. Blood, 1960-1974. 82, Mui,A.L., O'Farrell,A., Harada,N. and Miyajima,A. (1995) Received on December 4, 1995; rev,ised on Februarv 9, Wakao,H., colony stimulating factor and Interleukin-3, granulocyte-macrophage interleukin-5 signals through two STAT5 homologs. EMBO J., 14, 1166-1175. Quelle,F.W., Sato,N., Witthuhn,B.A., Inhorn,R.C., Eder,M., Miyajima,A., and Ihle,J.N. (1994) JAK2 associates with the beta c chain Griffin,J.D. of the receptor for granulocyte-macrophage colony-stimulating factor, Mol. Cell. and its activation requires the membrane-proximal region. Biol., 14, 4335-4341. Quelle,F.W., Shimoda,K., Thierfelder,W., Fischer,C., Kim,A., Ruben, S.M., Cleveland,J.L., Pierce,J.H., Keegan,A.D., Nelms,K., Paul,W.E. of murine Stat6 and human Stat and Ihle,J.N. (1995) Cloning Stat6, in to IL4 and IL3 proteins that are tyrosine phosphorylated responses but are not for Mol. Cell. 3336-3343. required mitogenesis. Biol., 15, and Muller,R. (1985) Renz,M., Neuberg,M., Kurz,C., Bravo,R. of in mouse fibroblasts: identification Regulation c-fos transcription of DNase sites and upstream sequences. I-hypersensitive regulatory EMBOJ., 4,3711-3716. Kitamura,T. and Miyajima,A. (1992) Critical Sakamaki,K., Miyajima,I., of the human GM- domains of the common beta subunit cytoplasmic transduction and IL-3 and IL-5 for CSF, receptors growth signal EMBO 3541-3549. phosphorylation. J., 11, tyrosine and Conditional Samuels,M., Weber,J., Bishop,J. McMahon,M. (1993) activation of the activated transformation of cells and rapid mitogen an human raf-1 protein kinase cascade by estradiol-dependent protein 6241-6252. kinase. Mol. Cell. Biol., 13. and Sato,N., Sakamaki,K., Terada,N., Arai,K. Miyajima,A. (1993) Signal transduction the GM-CSF two distinct by high affinity receptor: of the common f subunit for cytoplasmic regions responsible differentiation. EMBO 4181-4189. J., 12, Beta-casein and Groner,B. Schmitt,N.M., Doppler,W., Ball,R.K. (1991) the hormone-mediated relief is regulated by gene promoter activity and a nuclear of transcriptional repression mammary-gland-specific 3745-3755. factor. Mol. Cell. Biol., 11, Identification of a site that mediates Treisman,R. (1986) protein-binding of the to serum factors. Cell, transcriptional response c-fos gene 46, 567-574. van Verbeek,S., Krimpenfort,P., Domen,J., Saris,C., Lohuizen,M., to and Radaszkiewicz,T. Berns,A. (1989) Predisposition lymphoma- http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The EMBO Journal Springer Journals

Suppression of interleukin‐3‐induced gene expression by a C‐terminal truncated Stat5: role of Stat5 in proliferation.

Download PDF
9 pages

Loading next page...
 
/lp/springer-journals/suppression-of-interleukin-3-induced-gene-expression-by-a-c-terminal-oILwRUdFGU

References (43)

Publisher
Springer Journals
Copyright
Copyright © European Molecular Biology Organization 1996
ISSN
0261-4189
eISSN
1460-2075
DOI
10.1002/j.1460-2075.1996.tb00600.x
Publisher site
See Article on Publisher Site

Abstract

The EMBO Journal vol.15 no.10 pp.2425-2433, 1996 Suppression of interleukin-3-induced gene by a C-terminal truncated Stat5: expression role of Stat5 in proliferation Wakao2'3, transduction. Structure-function analyses of the ,3-subunit Alice L.-F.Mui1, Hiroshi receptors have defined shared by the IL3, IL5 and GM-CSF Toshio Kitamura2 and Taisei Kinoshita2'3, domains necessary for activation of specific intracellular Atsushi Miyajima2'3 et al., 1992; Sato et al., signalling pathways (Sakamaki Departments of 'Molecular Biology and 2Cell Signalling, region is required for activation 1993). A membrane-distal DNAX Research Institute for Molecular and Cellular Biology, kinase cascade and induction of of the Ras/Raf-l/MAP 901 California Ave, Palo Alto, CA 94304, USA and 3Institute of A second, membrane-proximal, domain is needed c-fos. Molecular and Cellular Bioscience, The University of Tokyo, The Ras pathway for c-myc, pim-J and cis induction. 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113, Japan of apoptosis, appears to be important for prevention 'Corresponding author pathways directed by the receptor's mem- whereas the for DNA brane-proximal domain appear necessary Interleukin-3 (IL3) was shown recently to utilize the of synthesis (Kinoshita et al., 1995b). Maintenance long- transcription factor Stat5, but the genes regulated by both anti- term proliferation requires the contribution of this pathway and the biological consequence of Stat5 apoptotic and DNA synthesis signals. activation remained to be determined. In order to on the role of the Janus Recently, attention has focused study the role of Stat5 in IL3 signalling, we constructed action. Jak kinase (Jak) family of kinases in cytokine a dominant-negative Stat5 protein by C-terminal trunc- kinases were first described as functioning downstream of in IL3-dependent ation, and inducibly expressed it an have the interferon receptors, but many cytokine receptors Stat5 induc- cell line. The effect of dominant-negative now been shown to associate physically with and activate of IL3 early response genes was tion on expression specific Jak kinases (Ihle et al., 1994). Jak2, the Jak family and of several including examined, expression genes, kinase important for IL3 and GM-CSF action, associates cis, osm and pim-1 was inhibited profoundly. The with the 1-chain via the membrane-proximal domain. Jak expression of c-fos was also reduced, but to a lesser kinases, in turn, regulate latent, cytoplasmic transcription extent. While activated Ras alone (though not StatS factors termed Stats (Signal transducers and activators of could induce c-fos, maximal expression required alone) Stats dimerize and transcription). Upon phosphorylation, the action of both Ras and Stat5. Interestingly, although translocate to the nucleus where they bind to DNA the membrane-proximal region of the IL3 receptor sequences, most of which are related to the gamma 1-chain is responsible for both Jak2-StatS activation interferon activated site (GAS), a regulatory ele- (IFN-y) affected by and c-myc induction, c-myc levels were not ment in the promoter of genes (Darnell IFN-y-inducible the dominant-negative Stat5. Thus, the signals directed the Stat et Presently, seven members comprise al., 1994). which is essential by this membrane-proximal domain, of family and each functions in the signalling pathways for a DNA signal, can be separ- transducing synthesis For specific cytokines (Darnell et al., 1994). example, further into Stat5 or c-myc pathways. The net ated and et while IFN-a activates Statl Stat2 (Darnell al., effect of dominant-negative Stat5 expression was par- et Mui et 1994), IL3 utilizes StatS (Azam al., 1995; al., tial inhibition of IL3-dependent growth. This provides 1995) and Stat6 (Quelle et al., 1995). the first direct evidence that Stat5 is involved in identified as Since Statl and Stat2 originally were regulation of cell proliferation. of certain IFN- factors which regulate transcription Keywords: cell proliferation/c-myc/interleukin-3/Stat5 in to responsive genes, the targets of these Stats response the IFN stimulation are known. In contrast, target genes the Stat are of subsequent members of family largely Introduction is since unknown. For StatS, one defined target 1-casein, StatS was first purified as a prolactin-responsive regulator the differentiation and func- Cytokines govern growth, Other of this gene (Wakao et al., 1994). cytokines, tional of cells in the haemopoietic and immune activity Mui et GM- including IL3 (Azam et al., 1995; al., 1995) Of interleukin-3 (IL3) has a broad range systems. these, et Mui et IL5 on cells at various of CSF al., 1995; al., 1995), of function, with actions stages (Barahmand-pour et IL2 et Wakao et The (Mui al., 1995), (Lin al., 1995; al., haemopoietic development (Arai etal., 1990). receptor et and of a a-subunit and a 1995), erythropoeitin (Ep) (Wakao al., 1995) growth for IL3 consists ligand-specific later found also to hormone et were 13-subunit which is also found in the (Waxman al., 1995), high affinity receptor other do utilize StatS. most of these cytokines for two related IL5 and However, complexes functionally cytokines, and the factor not induce genes regulated colony-stimulating (GM- ,B-casein expression, granulocyte-macrophage these factors have not been Both subunits StatS in to et 1993). belong by response CSF) (Miyajima al., 1992, to characterize the role of StatS in both identified. In order to the cytokine receptor superfamily and, although a StatS reconstitute the IL3 we constructed dominant-negative are required to high affinity binding, signalling, dominant StatS has to the role in The use of this negative 1-chain is believed perform greater signal protein. Press ( Oxford University A.L.-F.Mui et al. in AStatS- expression of the truncated AStat5 product lanes 4 and but not transfected (Figure 2A, 5) parental The effect of AStat5 induction on the level cells (lane 2). 6000- was then of endogenous StatS-DNA binding activity C.) shift measured by electrophoretic mobility analysis U) 4000- DNA (EMSA) using the Stat5 target sequence, ,-casein 01) 0) et As shown in GAS, as a probe (Mui al., 1995). Figure ._ n 2B, induction of AStat5 greatly diminished the 3-casein 2000- in cells lane GAS binding activity Ba/F3 (Figure 2B, 4). This inhibition could be reversed the constitutive by OI of full-length, wild-type StatS (Figure 2B, overexpression lane 5). vector AStat5 on a Fig. 1. Effect of transient AStat5 expression Stat5-responsive Effect of 4Stat5 expression on various transfected as described reporter plasmid. Ba/F3 cells were transiently 1L3-activated signalling pathways in with the luciferase reporter Materials and methods Stat5-responsive The effect of AStatS expression on various IL3-activated the test DNAs: vector plasmid, ,B-casein-luc, and (pME18S), pME- stimulated or not with 10 IL3 AStat5 or pME-AStatl. Cells were ng/ml signalling pathways was examined in order to test the h for luciferase determination. Luciferase for 6 and extracted activity specificity of AStatS action. As shown in Figure 3A, is in units. Results are the of activity plotted arbitrary average AStatS induction profoundly inhibited the tyrosine in triplicate measurements three independent experiments. StatS. which phosphorylation state of endogenous AStatS, act is not phosphorylated itself, may thus by competing not only allowed identification of several StatS-regulated with wild-type StatS at the level of the 1L3-receptor to the or for access genes, but has also revealed a previously unappreciated complex, either for binding receptor for the role of StatS in IL3-dependent proliferation. to the tyrosine kinase responsible phosphorylation. However, AStatS expression had no measurabe effect on the of Vav or Shc. AStatS tyrosine phosphorylation Jak2, Results induction also did not alter the observed co-precipitation Expression of a dominant-negative Stat5 protein of a 90 kDa phosphoprotein with the 85 kDa tyrosine is for Stat dimer subunit of nor did it inhibit the Since tyrosine phosphorylation necessary phosphoinositol-3'-kinase, we reasoned that a truncation upward mobility shift, an indication of ligand-induced formation, carboxy-terminal which removes most of the C-terminal tyrosines, including phosphorylation, of Raf-1 kinase and p70 ribosomal S6- reduced the residue shown to be in kinase (S6K). However, AStat5 expression slightly tyrosyl phosphorylated response to stimulation et al., 1994), would MAP kinase (MAPK) activity by 30% (Figure 3B). Since prolactin (Gouilleux a not be reversed of produce a StatS protein that might function as dominant this inhibition could by overexpression of this the cDNA not of MAPK negative. As an initial test hypothesis, wild-type StatS (data shown), regulation was after Materials is not a normal function of StatS. More for StatSB truncated Tyr683 (see activity probably and and transfected into the IL3- probably, the extremely high levels of truncated AStatS methods) transiently cell with a protein achieved resulted in interactions not normally dependent line, Ba/F3, together Stat5-respon- luciferase Cells that were transiently encountered by wild-type StatS. In order to reduce the sive, reporter plasmid. transfected with the mutant StatS construct (AStatS) level of this interference, the amount of AStatS protein the expressed significantly less luciferase activity than control expressed was titrated down (by increasing concentra- 1 full MAPK cells that were transfected with the vector alone (Figure 1). tion of tetracycline). At ng/ml tetracycline, all For further analysis, AStatS was cloned downstream activity was restored (Figure 3C). For subsequent a and unless otherwise of tetracycline (tet) operator-controlled promoter experiments, indicated, AStat5 protein into cells to was induced in the presence of 1 ng/ml tetra- transfected Ba/F3 previously engineered partially the transactivator (tTA; or repressed by the addition of 1 ,ug/ml express tetracycline-regulated cycline fully Gossen and Bujard, 1992). Since tet operators are bacterial tetracycline. In addition, as a control for non-specific in origin, endogenous, mammalian promoters should not interference, the ability of wild-type StatS to reverse be affected directly by activation of the tTA transcription the phenotype of any observed AStatS-inhibited event factor. Several clones were chosen based on was checked. independent their of AStatS Before the effect of AStatS on tetracycline-dependent expression protein. investigating expression One of these clones was infected further with a retrovirus IL3-induced genes, we examined its effect on the induction harbouring wild-type StatS in order to overexpress the of a gene regulated by another Stat. Since IRF-I is in 2A wild-type protein these cells. Figure shows the regulated by IFN-y through Statl- (Darnell et al., 1994) AStatS and StatS as and not full-length expression, assessed by StatS-dependent pathways, AStatS should not Western analysis with an anti-StatS monoclonal antibody, inhibit the induction of IRF-I in response to IFN-y. As in the variously manipulated Ba/F3 cells. The level of expected, AStat5 indeed did not alter IRF-I IFN-y-induced endogenous, wild-type StatS was detected as a faint band expression (Figure 3D). in both parental Ba/F3-tTA and Ba/F3-AStatS cells (Figure lanes as 2A, 1-4) and, expected, the intensity of this band Effect of AStat5 expression on induction of was enhanced in cells infected by the StatS greatly 1L3-responsive genes retrovirus (Figure 2A, lane 5). Culturing these cells in The effect of AStatS expression on IL3 induction of early inducing conditions (no tetracycline) resulted in high level was then assessed response genes by Northern blot analysis 2426 Role of Stat5 in cell proliferation Un cn U_) C/) 4- 4- C), C.) C) 'a aa C 4 ;; + co + Q - cn L3 *-+*'-++ Induction Induction -+ -m .+ full length Stat5 Stat5-DNA AStat5 complex 1 2 34 5 1 2 34 5 Western EMSA Fig. 2. Effect of AStatS expression on endogenous StatS DNA binding activity. Cell lysates from Ba/F3 cells expressing tetracycline-inducible AStatS were subjected to Westem analysis with an anti-Stat5 monoclonal antibody (A) and gel mobility shift analysis (B) with a labelled J-casein GAS oligonucleotide probe (Mui et al., 1995), respectively. Lanes 1 and 2 are from parental Ba/F3-tTA cells, lanes 3 and 4 are from AStat5 transfectant clone 23. Lane S represents lysate from AStatS clone 23 which had been infected with a retrovirus harbouring full-length, wild-type StatSB in order to overexpress wild-type Stat5 in these cells. Samples in lanes I and 3 are from cells grown in non-inducing (-) conditions (1 tetracycline), gg/ml while those in lanes 2, 4 and 5 are from cells grown for 24 h in inducing (+) conditions (no tetracycline). All cells were harvested from cultures growing exponentially in 1L3. A B anti-protein blot anti-p-tyr - + induction I- + I induction - + 1+ + IL3 77]I F- +1 +1 77741 1L3 activity [I7 rx;-+ _MAPK EZI E: .. Stat5 =_I tet 0 10 100 1000 ( ng/mL) 0.1 1.0 Jak2 4S'" * UV4II anti-stat5 -:, vavLX activity IMAPK p85 D induction - + Shc 0 0.5 1.0 0.5 1.0l IFN/(hr) _*-eIRF-1 _0 _m wo Raf-1 GAPDH S6K induction on IL3-induced activation of signalling molecules. AStatS expression in Ba/F3- Fig. 3. Specificity of AStatS action. (A) Effect of AStatS not 12 h culture or with, 1 tetracycline, respectively, in growth medium (10 ng/ml IL3, 10% AStatS cells was induced (+), or (-), by without, gg/ml in RPMI were then washed free of IL3 and cultured in 10% FCS in RPMI 1640 containing tetracycline as appropriate. After 8 h, FCS 1640). Cells et 1995). Antibodies to cells (2X 107 were stimulated with 1 ng/ml 1L3 for 10 min and detergent lysates prepared as described (Mui al., cells/point) and Shc were used to immunoprecipitate Jak2 Vav the 85 kDa subunit of phosphoinositol-3'-kinase (p85, UBI) (Transduction) StatS, (UBI), (UBI), were then resolved SDS-PAGE and by immunoblotting with anti- the molecule from the cell lysates. The immunoprecipitates by analysed designated ribosomal S6-kinase (S6K), 1 of cell lysate was resolved directed by (clone 4G10, UBI). For Raf-1 kinase (Raf-l) and p70 gg phosphotyrosine with antibodies either Raf-l or S6K. Effect of AStatS expression on MAPK activity. SDS-PAGE and subjected to immunoblotting recognizing (B) was and et in lysates prepared as in (A). (C) Effect of tetracycline concentration on MAP kinase activity determined quantitated (Samuels al., 1993) and MAPK AStatS was induced in varying concentrations of tetracycline, and AStatS protein detected as in Figure AStat5 expression activity. protein of IRF-1. AStatS expression in Ba/F3- I or MAPK determined as above. (D) Effect of AStatS induction on IFN-y-stimulated expression activity 1 in medium. Cells (2X 10 cells/ AStatS cells was induced or not (-), by 24 h culture without, or with, ,ug/ml tetracycline, respectively, growth (+), 1 time and RNA extracted the RNeasy extraction kit. The were then stimulated with zg/ml IFN-y for the indicated length of using Qiagen point) IRF-J and GAPDH RNA was analysed by Northern analysis with probes. 2427 A.L.-F.Mui et al. + wt parental \Stat5 \Stat5 Induction - ++ F.----l +---- 1-- r 00.5 2 3 0 05 2 .5 1 2 3 0 0.5 1 2 3 IL3 (hr) 0 0.5 1 0 u. \Stat5 expression ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. .. .. .. ... .... .. .. .... .... .. .. _ ... ,~0 ° [tet] (.ig/mL) -o6 cisIX cis pim-1 ~~4,'41 pim-1 Id-1 _ __ _ Id-1 osm osm c-Fos 1 c-Fos c-myc i]1 ........_ _... __ _ _ _ _. _ ........ c-myc |_ A cyclin 02 cyclin 02 cvclin D3 cyclin GAPDH * GAPD)HE | 4. Effect of induction on of IL3-inducible or Ba/F3-AStat5 + wild- Fig. AStat5 expression genes. (A) Ba/F3-tTA (parental), Ba/F3-AStat5 (AStat5) + wt cells were cultured for 24 h under AStat5 or conditions as in 2. Cells were then type Stat5 (AStat5 Stat5) inducing (+) non-inducing (-) Figure washed free of IL3 and starved for 8 h in 10% FCS in RPMI or I as These cells (2X 107 cells/ lacking containing ,ug/ml tetracycline appropriate. were then stimulated for the indicated of time with 10 IL3 before RNA was extracted as in 2. RNA was point) length ng/ml Figure (10 jig/lane) on a and to Northern with various random DNA Ba/F3-AStat5 cells were separated formaldehyde gel subjected analysis prime-labelled probes. (B) induced for 24 h as in with concentrations of before IL3 for 8 h. The cells were then stimulated for 1 h with (A) varying tetracycline deprivation 1 ng/ml IL3 and RNA for Northern as above. prepared analysis two is (Figure 4, left panel). Parental or AStatS Ba/F3 cells were separated further into at least pathways, one which in or conditions as and another which is for cultured AStatS-inducing non-inducing StatS dependent responsible and then stimulated with As induction. The inhibition of before, IL3. previously c-myc expression by c-fos resulted in the AStatS, on the other hand, was unexpected, since analysis reported, IL3 stimulation rapid appearance of a number of of the truncation mutants had the genes, including cis, pim-l, osm, c-fos, receptor assigned region and and those D2 and D3 required for induction to a membrane-distal Id-l c-myc encoding cyclins c-fos second, et Kinoshita et Yoshimura domain that is also for Raf and MAP (Sato al., 1993; al., 1995b; necessary Ras, et 1995 and et kinase activation (Sato et The data al., M.Takagi al., unpublished). Although al., 1993). present c-myc and cyclin D2 and D3 was not altered suggest that signals from both and expresssion membrane-proximal in cells induced to express AStatS, the expression levels membrane-distal domains may be required for full induc- of cis, pim-J, osm, Id-i and c-fos were tion of c-fos. In fact, both human (Treisman, 1986) and significantly in reduced, suggesting that these five genes are regulated by murine (Renz et al., 1985) c-fos promoters contain, of addition to the Ras/Raf/MAP StatS-dependent pathways. Importantly, overexpression kinase-responsive element, Stat5 restored the induction StatS et Gouilleux et wild-type gene pattern (Figure potential (Schmitt al., 1991; al., to that of control cells. In the 1995) binding sites (human: 5'-TGCactGAA, mouse: 4, right panel) addition, degree of inhibition of gene expression correlated with 5'-ATCtccGAA), and co-transfection of wild-type StatS the amount of AStat5 protein expressed. Decreasing the greatly augmented the luciferase activity from a c-fos- levels of AStatS protein by increasing the concentration promoter reporter plasmid (Figure SA) in Ba/F3 cells. of The StatS alone is tetracycline (see Figure 3C) diminished the degree of signal insufficient for c-fos, since cis, pim-l, and inhibition truncation which lack the osm, Id-l c-fos (Figure 4B). receptor mutants, domain needed The sensitivity of cis, pim-i, osm and Id-i to AStatS for activation of the Ras/Raf/MAP cascade but retain the expression is consistent with studies which mapped the region required for the Jak2-Stat5 pathway, cannot induce of the IL3/GM-CSF region receptor 5-chain responsible (Sato et al., 1993; Quelle et al., 1994; Mui et al., c-fos for their induction et (Sato al., 1993; Yoshimura et al., 1995). However, the Ras-mediated signal is also not 1995; to the same membrane- sufficient for full of A.Miyajima, unpublished) expression c-fos. Previously, we proximal domain required for Jak2 (Quelle et al., 1994) established Ba/F3 cell lines the expressing constitutively and StatS (Mui et activation. actvated form of al., 1995) Interestingly, Ras (Rasvall2) under the control of a although c-myc induction also required the membrane- dexamethasone (dex)-responsive, mouse mammary domain et proximal (Sato al., 1993), c-myc expression did tumour virus (MMTV) promoter. In Ba/F3 cells expressing not to be appear regulated by StatS. Thus the signals GM-CSF receptor truncation mutants membrane- lacking from the can emanating membrane-proximal region be distal domains, induction of RasvalI2 protein is able to 2428 Role of Stat5 in cell proliferation sion of full-length, wild-type Stat5 (Figure 6B). As one test of the specificity of AStat5 inhibition, we compared its effect on IL3- or IL4-supported viability of Ba/F3 cells. IL4 utilizes Stat6 and not Stat5 (Hou et al., 1994; data not shown) and thus, as expected, AStat5 expression inhibits the IL3 but not the IL4 response (Figure 6C). As another test of the specificity of growth inhibition, the X,, 10000- colony-stimulating factor-I receptor (CSF-1R) was intro- duced into either parental tTA or AStat5 Ba/F3 cells. Since _ IL3 CSF- 1 stimulation did not activate Stat5 detectably in - IL3 these cells (data not shown), AStat5 should not inhibit CSF- 1-driven [3H]thymidine incorporation. As shown in 5000- CZM Figure 6D, although AStat5 induction inhibits [3H]- -j thymidine incorporation in response to it does not IL3, alter the response to CSF-1. Significantly, even at the highest levels of AStat5 vector Stat5B vector Stat5B test DNA protein (when cells are induced by complete removal of tetracycline), growth is not fully inhibited. To examine c-fos luc .)-casein luc reporter DNA this in more detail, cell cycle analyses were carried out which revealed that although induction of AStat5 retards the rate of entry into and the percentage of cells entering S-phase, it does not prevent GI/S progression completely (Figure 6E-G). rasva[12 Discussion parental (IL3) tDex) Specificity of AStat5 action 0 0.5 1.0 0 12 24 time(hr) To the of the investigate contribution Jak2-Stat5 pathway .4- c-fos to IL3 signalling, we expressed a C-terminal truncated t N N - GAPDH o4w version of Stat5, AStat5, in Ba/F3 cells. Expression of AStat5 in Ba/F3 cells interfered with the function of 1 2 3 4 5 6 endogenous Stat5; it decreased the levels both of luciferase Fig. 5. Regulation of expression. (A) Ba/F3 cells were c-fos transiently activity from a Stat5-responsive reporter plasmid and of transfected with either the f-casein-luciferase or c-fos-luciferase Stat5-DNA binding activity, detected by EMSA of cell reporter plasmids and either vector (pME18S) or pME-Stat5B. After the recovery period, cells were divided and cultured for 6 h in IL3 or lysates. AStat5 apparently acts as a dominant-negative by no factor. Luciferase activity was determined from cell extracts. Data interfering at the level of either the receptor or the are the means of tfiplicate measurements. Similar results were kinase responsible for the phosphorylation. However, an obtained in two separate experiments. (B) Parental Ba/F3 cells (lanes in of important consideration interpreting results experi- 1-3) were stimulated with IL3 or pMAM-RasValI2_expressing Ba/F3 is the ments utilizing dominant-interfering molecules, cells (lanes 4-6) were treated with dexamethasone as described et (Kinoshita al., 1995b). Total RNA was isolated and subjected to of action. For AStat5 associate specificity example, may Northern with c-fos as in 4. analysis probe Figure with the IL3R and activation of other stably prevent which the same domain of the In signals require receptor. compensate for the lack of a GM-CSF-generated, Ras an to address this we tested but found attempt question, activation signal and collaborate with GM-CSF to no effect of AStat5 on various other support signalling pathways. long-term proliferation. Two Ras-regulated genes which At extremely high expression levels of AStat5, we did may mediate its anti-apoptotic action, bcl-2 and bcl-x, are observe a slight inhibition of MAPK so we carefully dex to the same as the level of to eliminate this induced by extent observed with IL3 adjusted down AStat5 potential stimulation et we cannot rule out the that (Kinoshita al., 1995a). However, although problem. However, possibility dex treatment resulted in other either known ones we did not check or c-fos expression (Figure 5B, pathways, lanes the amount of mRNA did not rise to the unknown ones which have not been are 4-6), c-fos yet characterized, same level as that observed in IL3-stimulated cells affected AStat5. (Figure non-specifically by of action lanes unlike bcl-2 and the We also assessed the specificity AStat5 by 5B, 1-3). Thus, bcl-x, expression the of more than mere Ras activation. the of Stat5 to reverse observed c-fos requires testing ability wild-type The of to phenotypes. ability wild-type protein compensate Effect of AStat5 on for the action of molecules is a com- expression IL3-dependent dominant-negative control. For our proliferation monly accepted experiments, however, if the fact that AStat5 does not inhibit or this control must be the caveat that AStat5 Despite c-myc qualified by D-cyclin it DNA non-specifically interferes with unrelated pathways by expression, compromised IL3-dependent of the then with cultures continuous occupation receptor, wild-type synthesis (Figure 6A), AStat5-expressing than controls after 24 h Stat5 still be able to reverse the inhibition containing 40% fewer cells may by (data with This inhibition is on the levels of AStat5 from a stable not shown). dependent preventing forming complex AStat5 protein expressed (i.e. dependent on the concentra- the Definitive evidence for the of the receptor. specificity thus tion of and could be reversed inhibitory effects we report here tetracycline) by overexpres- may require targeted 2429 A.L.-EMui et al. 0 10000 o l { . 8000D CL i .*- Ai, tTA- ---- not iniduced Ec 0 p.-s- s000 .. -,r- - tTA induced .\Stat5 0 -0W- s/t * oStatS + vwt P.0 * t%Stat5 - not nduced .E 4000 IUD' 1 0- 00 Stat5 - induced 1- >, 2000 u01 0.1 1 10 . 1 i 10 100 nor IL.3 concentration (ng/mL) tetracycline concentration (ng/mL) C D E.o L not induced IC.E E _not induced r- induced 0, 0 0 6... IL3 1L4 [jjjFj[jjI CTA-CSF1R Statb-CSF1R F 3 I[~1 cn 7; t'rA not induced 4) u) t.TA- iriduced 'b a Stat5 - not induced 40- o1 ..0 _-- Stat5 indLJced 0'... _ .- ., -10 0 1 0 20 30 -1 0 0 1 0 2 0 30 + time (hrs) + time (hrs'. IL3 IL3 -.--- .- 20Or G2/M a. IT U) I S1 ID a A/ o9 -10 0 1 0 20 30 (hrs) time 1L3 6. Effect of AStat5 on Parental Ba/F3-tTA and Ba/F3-AStat5 cells were cultured under Fig. expression proliferation. (A) (OL 0) (U @) inducing in 2. The and (O @) or non-inducing (C1 U) conditions as described Figure cells were then deprived of IL3 as before their IL3 growth response was assessed in a [3H]thymidine assay. (B) Comparison of the [3H]thymidine incorporation of Ba/F3-AStat5 and Ba/F3-AStat5 StatS (Cl) wild-type (U) Effect of on and maintenance of cell cells at various concentrations of tetracycline. (C) AStat5 expression IL3 IL4 Ba/F3 viability. Ba/F3-AStat5 as and their to 1 or was measured in an Alamar Blue cells were induced (U) or not (LI),to express AStat5 above, response ng/ml IL3 IL4 (Alamar colorimetric as described of the of Biosciences) assay (Ho et al., 1995). (D) Comparison [3H]thymidine incorporation CSF-lR-expressing Ba/F3-tTA and Ba/F3-AStat5 cells in to IL3 or under or conditions. F and Cell status S response CSF-l, inducing (U) non-inducing (El) (E, G) cycle (GO/GI, and G2/M of Ba/F3-tTA and Ba/F3-AStatS cells cultured under or phase respectively) parental (El 0) (U @) inducing (O @) non-inducing (OL U) = h conditions as in After 12 h of the cells were washed free of IL3 time -10 and restimulated with I IL3 at (A). induction, (at ) ng/ml = 0 time h. of and osm was since disruption both StatS genes (Mui et al., 1995) expressed regulating cis, pim-J, Id-l expected, in murine cells. However, corroborative evidence as these to the same domain genes map membrane-proximal detailed below that the we of the n-chain as those for Jak2-Stat5 activation. suggests phenomena report receptor here from result specific inhibition of the StatS pathway. In contrast, although this membrane-proximal domain is also responsible for c-myc induction, c-myc levels were Identification of a not affected Stat5-regulated genes using by AStat5 expression. These data dissociate dominant Stat5 induction from activation. protein c-myc StatS However, as c-myc Induction of five genes, cis, pim-J, Id-], osm and c-fos, induction is sensitive to tyrosine kinase inhibitors was inhibited by AStat5 expression. A role for StatS in (Kinoshita et is al., 1995b), c-myc induced by either a 2430 Role of Stat5 in cell proliferation tyrosine kinase distinct from Jak2, or alternatively, through important for proliferation, since certain IL2R mutants a Jak2-dependent pathway which is distinct from that which lack the ability to activate StatS are still able to responsible for Stat5 activation. stimulate mitogenesis. Close examination of their data, Inhibition of gene expresssion could be a direct result however, shows that the relevant mutant, H-4, is reduced, of AStat5 on transcription from their promoters, or an by a statistically significant 15%, in its ability to support indirect effect, i.e. transcription requires the action of a IL2-driven growth. Since growth can still occur at a StatS-regulated gene product rather than Stat5 itself. To reduced level, Fujii et al. conclude that StatS is not investigate these possibilities, the available promoter important for proliferation. However, we and others sequences of murine pim-J, cis and osm were also (Damen et al., 1995; Friedmann et al., 1995) propose that examined for potential Stat5 binding sites. Elements con- StatS activation is required for a maximal proliferative forming to the Stat5 GAS consensus were found for response. the c-fos, cis and osm promoters, and DNA fragments Several reasons potentially may explain the incomplete containing this region were able to confer Stat5-dependent nature of the growth inhibition. The simplest explanation a of AStatS results in transcription on reporter gene (Figure SA and Yoshimura is that insufficiently high expression et the other hand, although a canonical This al., 1996). On incomplete blockade of StatS action. hypothesis of inhibition would be GAS motif occurs in the human pim-] promoter, none would predict that the degree could be found in the published murine pim-J 5'-flanking lower of IL3 but, as Figure 6A shows, greater at doses A occur. the other the data are sequence (Yip-Schneider et al., 1995). StatS binding this does not On hand, be of The first arises site may either outside the published 5'-flanking consistent with two other possibilities. or from the GAS consensus. the that StatS regulates many target region it may deviate from observation the promoter activity of are for Alternatively, StatS may affect genes. Some of these important proliferation (i.e. pim-1 indirectly. while like cis et others, (Yoshimura al., 1995), may c-fos) Interestingly, cis, pim-J, Id-], osm and c-fos display have a role. During the course negative growth regulatory varying degrees of inhibition by AStatS. This probably of these genes are not expressed normal IL3 signalling, reflects the process of signal integration that provides the simultaneously since other factors in addition to StatS are combinatorial versatility in gene regulation. Athough GAS involved in their regulation. The global inhibition of all sites may be present in a gene promoter, the extent to which StatS-regulated genes by AStatS thus results in antagonistic they are utilized depends on many other considerations. For signals, and the net effect of these interactions is partial example, slight variations in the GAS sequence may alter growth inhibition. However, a second possibility is that a its relative affinity for available Stat dimers (Zhang et al., full proliferative response to cytokines, including IL3, elements and factors that 1995). Additional promoter results from the contribution of several nominally inde- mRNA will regulate stability also contribute to the net pendent pathways. Previous studies using IL3/GM-CSFR expression level of a Stat-responsive gene. Futhermore, ,-chain deletion mutants defined a membrane-proximal the homo- DNA potential for or heterodimer formation, serine/ domain which is important for synthesis (Kinoshita threonine and interaction other The effect of on phosphorylation with non- et al., 1995b). partial inhibitory AStat5 Stat will the a DNA synthesis, therefore, further dissects the signals from proteins modify potential ability of Stat complex to transactivate even if it binds to a given the membrane-proximal domain; inhibition of one of promoter (Bluyssen et al., 1995; Eilers et al., 1995). The these the StatS diminishes but does not signals, pathway, regulation of c-fos expression provides a good illustration eliminate IL3-driven proliferation. of some of these complexities. First, both StatS and Ras The mechanism by which AStat5 inhibits DNA synthesis pathways are required for full induction of Second, is not clear. StatS may regulate a gene(s) directly required c-fos. although Ras alone can induce a low level of c-fos, StatS for progression into S-phase or, alternatively, control a activation, by itself, cannot. gene(s) whose product contributes to establishment of the S-phase competence. None of cyclin genes, however, Dominant-negative Stat5 inhibition of appear to be inhibited by AStatS. Among those genes which 1L3-dependent proliferation which are inhibited by AStatS, the mechanism by The is of decreased cis and Id-i contributes to net result of AStatS expression inhibition IL3- expression growth driven DNA have been is not obvious. Cis is to synthesis. Stat proteins recently inhibition immediately thought in the control of studies which associated with of a implicated proliferation by be down-regulation proliferative that cells transformed by HTLV-I (Migone et al., et while Id-I and related reported signal (Yoshimura al., 1995), 1995), bcr-abl (Danial et al., 1995) or src (Yu et al., genes have been variously reported to be involved either in 1995) exhibit constitutively activated Stat-DNA binding differentiation (Voronova and Lee, 1994) or progression GI of the serine/ activity. Similar conclusions have been derived from (Hara et al., 1994). The induction pim-1 are to activate on the other correlates with mito- analysis of receptor mutants which unable threonine kinase, hand, who have examined either in several et the StatS pathway. Investigators genesis systems (Lilly al., 1992). Moreover, et or IL2R et with in EpR (Damen al., 1995) (Friedmann al., pim-i synergizes myc leukaemogenesis (van mutants have correlated the to Lohuizen et and mast cells isolated from 1995) receptors' ability al., 1989), pim- to activate StatS with their to transduce a I-deficient mice exhibit an impaired response IL3 ability mitogenic in a manner similar to that of Ba! the StatS mutants (Domen et al., 1993) signal. Significantly, pathway-defective in their cells induced to AStatS. constitutive are F3 express However, again only partially impaired proliferative on cell is similar to of did not rescue the capacity: this limited effect growth expression pim-] proliferative action. of AStatS-inhibited cells not that observed with AStatS Other investigators (Fujii response (data shown). Thus, in have that StatS is not the for the inhibition et reported key target responsible growth may al., 1995), contrast, 2431 A.L.-F.Mui et aL either a minimal or c-fos promoter and test DNAs, in pME be an as unidentified 18S, yet StatS-responsive gene, or growth P-casein at room in temperature RPMI supplemented with 10 mg/ml DEAE- inhibition be the sum of several may interacting events. dextran. After a 12 h in recovery period 1L3-containing medium, cells A more detailed of which are survey genes differentially were deprived of IL3 for 12 h before a 6 h stimulation period with in cells induced to should help expressed express AStatS 10 no ng/ml IL3 or factor. Cytoplasmic extracts were prepared by three of freeze-thaw in 0.25 M Tris to resolve this cycles normalized question. pH 7.5, by protein concentration and for luciferase assayed activity the of this (Promega). During preparation manuscript, two reports that correlated the of certain receptor appeared inability Thymidine incorporation assay mutants to activate Stat5 with their ability to impaired For the 2x 104 assay, cells were plated per well in a 96-well plate in a et 1995; Friedmann serial dilutions of in RPMI + I support mitogenic signal (Damen al., IL3 5% FCS and jg/ml tetracycline as After 24 I of appropriate. h, ,tCi [3H]thymidine was added for 2 h et and others have observed constitutive activa- al., 1995), and the cells were harvested onto glass fibre filters and incorporated tion of the Jak-Stat in transformed and factor- pathway [3H]thymidine quantitated by liquid scintillation counting. cells et 1995; Migone et al., independent (Danial al., Yu et we provide the first 1995; al., 1995). However, Acknowledgements direct evidence that StatS is for proliferation. important The use of a Stat protein has also dominant-negative We thank Drs G.Krystal, K.Moore and M.McMahon for their critical allowed us to several and reading of this manuscript, and M.Gossen and H.Bujard for the tetra- identify Stat5-regulated genes, cycline repressor plasmids. A.Miyajima was the partially supported by this can be extended to of other Stats technique analysis Ministry of Education, Science and Culture of Japan. DNAX Research and other Even though the identity of cytokine systems. Institute is supported by Schering-Plough Corporation. the Stats which are activated in each is becoming cytokine well the fundamental question of what role these defined, References Stats in function remains to be answered. perform cytokine Arai,K., Lee,F., Miyajima,A., Miyatake,S., Arai,N. and Yokota,T. (1990) coordinators of Cytokines: immune and inflammatory responses. Annu. Materials and methods Rev. Biochem., 59, 783-836. Azam,M., Erdument-Bormage,H., Krieder,B.L., Xia,M., Quelle,F., Materials Basu,R., Saris,C., Tempst,P., Ihle,J.N. and Schindler,C. (1995) The kind repressor-transactivator system plasmids were the tetracycline Interleukin-3 signals through multiple isoforms of Stat5. EMBO J., of Drs M.Gossen and The retroviral is a H.Bujard. vector, pMX, gifts 14, 1402-1411. derivative of BOSC 23 and unpublished). (ATCC) pBabe (T.Kitamura, Barahmand-pour,F., Meinke,A., Eilers,A., Gouilleux,F., Groner,B. and cells were maintained as described. All other materials were Ba/F3 Decker,T. (1995) Colony-stimulating factors and interferon-gamma obtained as described. previously activate a protein related to MGF-Stat 5 to cause formation of the differentiation-induced factor in cells. FEBS myeloid Lett., 360, 29-33. Construction of AStat5 Bluyssen,A.A.R., Muzaffar,R., Vlieststra,R.J., Van der Made,A.C.J., Stat5B was truncated after the XhoI-NotI Tyr683 by replacing fragment Leung,S., Stark,G.R., Kerr,I.M., Trapman,J. and Levy,D.E. (1995) in et with an XhoI-NotI-digested pME18S-StatSB (Mui al., 1995) Combinatorial association and abundance of components of interferon- PCR with 5'-TGACCAAGGAGAACCTCGTGT-3' fragment amplified stimulated gene factor 3 dictate the selectivity of interferon responses. and 5'-TAGTTATGCGGCCGCTAGTAGTACTTAGAATATA- (sense) Proc. Natl Acad. Sci. USA, 92, 5645-5649. CTT-3' The (antisense) primers. resulting construct, pME18S-AStatS, Damen,J.E., Wakao,H., Miyajima,A., Krosl,J., was Humphries,R.K., confirmed sequencing. The AStat5 cDNA was excised with by Cutler,R.L. and Krystal,G. (1995) Tyrosine 343 in the erythropoietin EcoRI-NotI and blunt-end into the XbaI site of ligated pUHDIO-3-hygro receptor positively regulates erythropoietin-induced cell proliferation and with a PGK promoter-driven [pUHDIO-3 (Gossen Bujard, 1992) and StatS activation. EMBO J., 14, 5557-5568. resistance inserted at the gene PvuII site]. hygromycin Danial,N., Pernis,A. and Rothman,P. (1995) Jak-STAT signaling induced by the v-abl oncogene. Science, 269, 1875-1877. Production of stable Ba/F3 cell clones AStat5-expressing Darnell,J.J., and Stark,G.R. (1994) Jak-STAT pathways Ba/F3 cells the tTA were transfection Kerr,I.M. and expressing protein generated by transcriptional activation in response to IFNs and other with and which had been modified extracellular pUHD15-1 (Gossen Bujard, 1992) signaling proteins. Science, 264, 1415-1421. at the XhoI site to include a resistance under the control puromycin gene of an Domen,J., van der Lugt,N.M., Laird,P.W., Saris,C.J., SV40 isolated dilution in Clarke,A.R., early promoter. Clones, by limiting 1.5 in Hooper,M.L. and Berns,A. (1993) Impaired interleukin-3 response in IL3 media [10 ng/ml IL3, 10% fetal ,ug/ml puromycin growth calf serum 50 in RPMI Pim-l-deficient bone marrow-derived mast cells. Blood, 82, 1445- mercaptoethanol 1640], were tested (FCS), ,uM for functional tTA transient 1452. protein by transfection with the tetracycline luciferase Eilers,A., Georgellis,D., Klose,G., Schindler,C., Ziemiecki,A., Harpur, reporter plasmid, pUHC 13-3. One clone operator-controlled A.G., Wilks,A.F. and Decker,T. (1995) Differentiation-regulated which well in this was then transfected with pUHDIO- serine responded assay phosphorylation of STAT1 promotes GAF activation in and selected in 1.4 I puro- macrophages. 3-hygro-AStatS mg/ml hygromycin, gg/ml Mol. Cell. Biol., 15, and 1 in IL3 medium. Tetracycline 3579-3586. mycin ,ug/ml tetracycline growth Friedmann,M., Migone,T., Russell,S. and Leonard,W. (1995) was 2 with a half medium change. To identify Different replaced every days clones which IL2R beta tyrosines couple to at least two signalling pathways and clones were washed express AStatS, hygromycin-resistant free of and cultured for 24 h in synergistically mediate IL2 induced proliferation. Proc. Natl Acad. tetracycline the absence or presence of Cell were as Sci. USA, 93, 2077-2082. tetracycline. lysates prepared described (Mui et al., 1995) and to Western with an anti-StatS Fujii,H., Nakagawa,Y., Schindler,U., Kawahara,A., Mori,H., subjected analysis monoclonal antibody Gouilleux,F., Groner,B., Ihle,J.N., Minami,Y., Miyazaki,T. Several clones were analysed, and data and (Transduction Lab). independent from one Taniguchi,T. (1995) Activation of Stat5 by interleukin 2 requires clone are shown in Figure 1. To overexpress a representative StatS in carboxyl-terminal region of the interleukin 2 receptor beta chain but AStat5 transfectants, one of these clones full-length wild-type was co-cultured with a viral line is not essential for the proliferative signal transduction, Proc. Natl packaging (BOSC 23) which had been transfected with StatS cloned into a retroviral vector, pMX. Acad. Sci. USA, 92, 5482-5486. wild-type After 24 h co-culture in 8 the Ba/F3 cells were cloned Gossen,M. and Bujard,H. (1992) Tight control of gene expression in jig/ml polybrene, mammalian cells by tetracycline-responsive promoters. Proc. dilution. Several clones were isolated which overexpressed Natl by limiting Acad. Sci. USA, 89, 5547-555 1. StatS as assessed Western of cell lysates; the data full-length by analysis from one Gouilleux,F., Wakao,H., Mundt,M. and Groner,B. (1994) of these clones are shown in 1. Prolactin Figure induces phosphorylation of Tyr694 of StatS (MGF), a prerequisite for Transient transfection assay DNA binding and induction of transcription. EMBO J., 13,43614369. Ba/F3 cells were at mF 960 and 270 V with 15 of Gouilleux,F., Pallard,C., Dusanter,F.I., Wakao,H., Haldosen,L.A., pg electroporated of a luciferase under the control of and Groner,B. (1995) Prolactin, growth hormone, Norstedt,G., Levy,D. reporter plasmids consisting gene 2432 Role of Stat5 in cell proliferation erythropoietin and granulocyte-macrophage colony stimulating factor genesis in pimn-1 transgenic mice: cooperation with c-mvc and N-mvc induce MGF-Stat5 DNA binding activity. EMBO J., 14, 2005-2013. in murine leukemia virus-induced tumors. Cell, 56, 673-682. Hara,E., Nojima,H., Ide,T., Campisi,J., Okayama,H. and Voronova,A.F. and Lee,F. (1994) The E2A and tal-1 helix-loop-helix Yamaguchi,T., Oda,K. (1994) Id-related genes encoding helix-loop-helix proteins proteins associate in vivo and are modulated by Id proteins during are required for GI progression and are repressed in senescent human interleukin 6-induced myeloid differentiation. Proc. Natl Acad. Sci. fibroblasts. J. Biol. C/ewn., 269, 2139-2145. 5952-5956. USA. 91, Ho,A.S.-Y., Wei.S.H.-Y., Miyajima,A. and Moore,K.W. Gouilleux,F. and Groner,B. (1994) Mammary gland factor Mui,A.L.-F.. Wakao,H., (1995) Functional regions of the mouse interleukin-10 receptor (MGF) is a novel member of the cytokine regulated transcription cytoplasmic domain. Mol. Cell. Biol., 15, 5043-5053. factor gene family and confers the prolactin response. EMBO J., 13, Hou,J., Schindler,U., Henzel,W.J.H., HoT., Brasseur.M. and McKnight,S. 2182-2191. (1994) An interleukin-4-induced transcription factor: IL-4 Stat. Wakao,H., HaradaN., Kitamura,T., Mui,A.L. and Miyajima,A. (1995) Science. 1701-1706. Interleukin-2 and erythropoietin activate STatS/MGF via distinct Ihle,J.N., Witthuhn,B.A., Quelle,F.W., Yamamoto,K., Thierfelder,W.E., pathways. EMBO J., 14, 2527-2535. Kreider,B. and (1994) Signaling by the cytokine Waxman,D.J., Ram,P.A., Park,S.H. and Choi,H.K. (1995) Intermittent Silvennoinen,O. receptor superfamily: JAKs and STATs. Trends Biochem. Sci., 19, plasma growth hormone triggers tyrosine phosphorylation and nuclear 222-227. translocation of a liver expressed, StatS-related DNA binding protein. Kinoshita,T., Yokota,T., Arai,K. and Miyajima,A. (1995a) Regulation of J. Biol. Chemn., 270, 13262-13270. Bcl-2 expression by oncogenic Ras protein in hematopoietic cells. Yip-Schneider,M.T., and Broxmeyer,H.E. (1995) Tran- Horie,M. Oncogene, 10, 2207-2212. scriptional induction of pim- 1 protein kinase gene expression by Kinoshita,T.. Yokota,T., Arai,K. and Miyajima,A. (1995b) Suppression interferon gamma and posttranscriptional effects on costimulation of apoptotic death in hematopoietic cells by signalling through the with steel factor. Blood, 85, 3494-3502. IL-3/GM-CSF receptors. EMBO J., 14, 266-275. Yoshimura,A., Ohkubo,T., Kiguchi,T., Jenkins,N.A., Gilbert,D.J., A novel Lilly,M., Le,T., Holland,P. and Hendrickson,S.L. (1992) Sustained Copeland,N.G., Hara,T. and Miyajima,A. (1995) cytokine- that binds to expression of the pim- 1 kinase is specifically induced in myeloid cells inducible gene CIS encodes an SH2-containing protein by cytokines whose receptors are structurally related. Oncogene, 7, tyrosine phosphorylated interleukin 3 and erythropoietin receptors. 727-732. EMBO J., 14, 2816-2826. of shared receptor motifs and common Lin,J.X. et al. (1995) The role Yoshimura,A., Ichichara,M., Kinjyo,l., Moriyama,M.. Copeland,N.G., and Mouse Stat proteins in the generation of cytokine pleiotropy and redundancy Gilbert,D.J., Jenkins,N.A., Hara,T. Miyajima,A. (1996) Ininunitv, 2, 331-339. M: immediate induced multiple cytokines by IL-2, IL-4, IL-7, IL-13, and IL-I5. oncostatin an early gene by Mulloy,J., O'Shea,J., Franchini,G. and a common EMBO J., 15, 1055-1063. Migone,T., Lin,J., Cereseto,A., through JAK-STAT pathway. activated Jak-STAT pathways in T Larner,A., Carter-Su,C., Schartz,J. and Leonard,W. (1995) Constitutively Yu,C., Meyer,D., Campbell,G., 79-81. Enhanced of a Stat3-related cells transformed wiht HTLV-1. Science, 269, Jove,R. (1995) DNA-binding activity Kitamura,T., Yokota,T. and Arai,K. (1992) protein in cells transformed by the src oncoprotein. Science, 269, Miyajima,A., Harada,N., and transduction. Annu. Rei: Immunol., 10, 81-83. Cytokine receptors signal 295-331. Li,H., Schindler,C. and Chen-Kiang,S. (1995) Zhang,X., Blenis,J., Miyajima,A., Mui,A.L.-F., and (1993) Requirement of serine phosphorylation for formation of STAT- Ogorochi,T. Sakamaki,K. for colony-stimulating factor, Science, 267, 1990-1994. Receptors granulocyte-macrophage promoter complexes. interleukin 3 and interleukin 5. Blood, 1960-1974. 82, Mui,A.L., O'Farrell,A., Harada,N. and Miyajima,A. (1995) Received on December 4, 1995; rev,ised on Februarv 9, Wakao,H., colony stimulating factor and Interleukin-3, granulocyte-macrophage interleukin-5 signals through two STAT5 homologs. EMBO J., 14, 1166-1175. Quelle,F.W., Sato,N., Witthuhn,B.A., Inhorn,R.C., Eder,M., Miyajima,A., and Ihle,J.N. (1994) JAK2 associates with the beta c chain Griffin,J.D. of the receptor for granulocyte-macrophage colony-stimulating factor, Mol. Cell. and its activation requires the membrane-proximal region. Biol., 14, 4335-4341. Quelle,F.W., Shimoda,K., Thierfelder,W., Fischer,C., Kim,A., Ruben, S.M., Cleveland,J.L., Pierce,J.H., Keegan,A.D., Nelms,K., Paul,W.E. of murine Stat6 and human Stat and Ihle,J.N. (1995) Cloning Stat6, in to IL4 and IL3 proteins that are tyrosine phosphorylated responses but are not for Mol. Cell. 3336-3343. required mitogenesis. Biol., 15, and Muller,R. (1985) Renz,M., Neuberg,M., Kurz,C., Bravo,R. of in mouse fibroblasts: identification Regulation c-fos transcription of DNase sites and upstream sequences. I-hypersensitive regulatory EMBOJ., 4,3711-3716. Kitamura,T. and Miyajima,A. (1992) Critical Sakamaki,K., Miyajima,I., of the human GM- domains of the common beta subunit cytoplasmic transduction and IL-3 and IL-5 for CSF, receptors growth signal EMBO 3541-3549. phosphorylation. J., 11, tyrosine and Conditional Samuels,M., Weber,J., Bishop,J. McMahon,M. (1993) activation of the activated transformation of cells and rapid mitogen an human raf-1 protein kinase cascade by estradiol-dependent protein 6241-6252. kinase. Mol. Cell. Biol., 13. and Sato,N., Sakamaki,K., Terada,N., Arai,K. Miyajima,A. (1993) Signal transduction the GM-CSF two distinct by high affinity receptor: of the common f subunit for cytoplasmic regions responsible differentiation. EMBO 4181-4189. J., 12, Beta-casein and Groner,B. Schmitt,N.M., Doppler,W., Ball,R.K. (1991) the hormone-mediated relief is regulated by gene promoter activity and a nuclear of transcriptional repression mammary-gland-specific 3745-3755. factor. Mol. Cell. Biol., 11, Identification of a site that mediates Treisman,R. (1986) protein-binding of the to serum factors. Cell, transcriptional response c-fos gene 46, 567-574. van Verbeek,S., Krimpenfort,P., Domen,J., Saris,C., Lohuizen,M., to and Radaszkiewicz,T. Berns,A. (1989) Predisposition lymphoma-

Journal

The EMBO Journal – Springer Journals

Published: May 1, 1996

There are no references for this article.