2664 V Calvo, M. Wood, C. Gjertson, T. Vik and B. E. Bierer Eur. J. Immunol. 1994. 24: 2664-2671

Victor Calvoon, Michael Wood.., Carl Gjertson., Terry VikoBvo and Barbara E. BiereroAw

Division of Pediatric Oncology., Dana-Farber Cancer Institute, Boston, The Children's Hospital., Boston, Harvard University Boston., the Hematology- Oncology DivisionA, Department of Medicine, Brigham and Women's Hospital, Boston and Departments of Pediatrics. and Medicine., Harvard Medical School, Boston

Activation of 70-kDa S6 kinase, induced by the cytokines interleukin-3 and erythropoietin and inhibited by rapamycin, is not an absolute requirement for cell proliferation*

The cytokines interleukin (1L)-3 and erythropoietin (EPO) are critical regulators of the proliferation and differentiation of cells of the hematopoietic system, but their intracellular mechanisms of action are not fully understood. Binding of IL-3 to the IL-3 receptor (IL-3R) and binding of EPO to the EPOR both induce changes in intracellular tyrosine and serinekhreonine phosphorylation; the phosphorylation of a number of polypeptides appears to be a shared response upon cytokine stimulation. We have previously shown that binding of IL-2 to the IL-2R activates the 70-kDa (p70) S6 kinase, a serinekhreonine kinase whose activity is regulated by serinekhreonine phosphorylation; the immunosuppres- sant rapamycin inhibits IL-2-dependent proliferation and IL-2-triggered activa- tion of p70 S6 kinase.We, therefore, sought to examine whether induction of p70 S6 kinase activity is a conserved response upon cytokine triggering, and whether this activity is essential for cell proliferation. Proliferation of the IL-3-dependent pro-B cell line BaF3 transfected with the EPOR (Ba/F3-EPOR) can be supported by either IL-3 or EPO. In this cell line, both IL-3 and EPO induced p70 S6 kinase activity; rapamycin inhibited both the IL-3 and EPO-induced activation of the 70-kDa S6 kinase as well as cellular proliferation. Thus. p70 S6 kinase activation appears to be a common intermediate triggered by the stimulation of IL-3, EPO, and IL-2 receptors. The Friend spleen focus-forming virus gpSS renders the EPOR constitutively active, and confers growth factor independence on cells expressing EPOR. Ba/F3-EPOR cotransfected with gp5.5 (BaF3- EpoRgpSS) and the erythroleukemia cell line MEL, which also expresses both t h e EPOR and gpSS, were analyzed. Rapamycin inhibited the activation of p70 S6 kinase in both cell lines. However, rapamycin inhibited proliferation of BdF3- EpoRgp5S but not of MEL cells despite inhibition of p70 S6 kinase activity in both cells. Thus, p70 S6 kinase activation is not an absolute requirement for cell proliferation. These results are discussed in relation to the role of the activation of the 70-kDa S6 kinase activation pathway in the regulation of cell cycle progression.

[I 123701

Present address: Instituto de Investigaciones Biomedicas, Madrid, Spain.

c' Prcsent address: Riley Hospital for Children, Indiana U Med Ccnter, Indianapolis, IN.

*' M. W. is a Howard Hughes Medical Institute Postdoctoral Fcllow. B. E. B. is the recipient of an Established Investigator Award from the American Heart Association. This work was supported in part by grant A132514 from the National Institutes of Health.

Correspondence: Barbara E. Bierer, Division of Pediatric Onco- logy, Room 171OA. Dana-Farber Canccr Institute, 44 Binney Strect. Boston. MA 0211.5, USA (Fax: 6 17 632-5144)

Abbreviations: CsA: Cyclosporin A EPO: Erythropoietin FKBP: FKS06 Binding protein rsk: YO-kDa ribosomal S6 kinase

Key words: Protein kinases / Cytokines / Rapamycin / Prolifera- tion

1 Introduction

The cytokines interleukin (1L)-3 and erythropoietin (EPO) regulate the differentiation and proliferation of hemato- poietic cells. The cell surface receptors for these cytokines and for 1L-2 belong to the hematopoietic superfamily of receptors that do not possess a tyrosine kinase domain in their cytoplasmic regions [ 1, 21. The signal transduction mechanisms that mediate the cellular responses triggered by IL-3, EPO, and IL-2 include the activation of both tyrosine and serinehhreonine protein kinases [ 1, 21. Bind- ing of TL-2 to its receptor has been shown to stimulate the activity of the 70-kDa (p70) S6 kinase [3-51. This ser- inehhreonine kinase is responsible for phosphorylation of the ribosomal S6 protein [6-101, which appears to partici- pate in the regulation of the G1 cell cycle progression of somatic cells [3, 4, 6, 7, 101. Another kinase able to phosphorylate S6 in vitro, the 85-90-kDa S6 kinase (re- ferred to as p85 or rsk S6 kinase) does not play a major role in phosphorylation of S6 in G1 in somatic cells [8] and does not appear to be stimulated by 1L-2 [3, 41. Some common intermediates in the signal transduction pathways triggered by IL-2, TL-3, and EPO, such as the phosphorylation and activation of Raf-1 kinase [11-141, have been defined. Different members of the JAK family of tyrosine kinases

00 14-2980/94/1llI-26(14$IO .OO + .2.5/0 0 VCH Verlagsgesellschaft mhH, D-69451 Weinheim, lYY4

Eur. J. Immunol. 1994. 24: 2664-2671 Cytokines, 70-kDa S6 kinase activity, rapamycin, and cell proliferation 2665

have recently been shown to be activated by IL-3 and EPO (for review, see [15]).We wished to investigate whether the activation of p70 S6 kinase constitutes a common element of the signal transduction pathways recruited upon cyto- kine triggering by IL-3 and EPO, as it is for IL-2, and the requirement for this kinase activity in cell proliferation.

The immunosuppressant rapamycin, a macrolide antibiot- ic, has been shown to inhibit IL-2-dependent proliferation [16,17] and appears to inhibit or delay proliferation [5,8,9, 16, 171. In a particular IL-2 dependent cell line, rapamycin was shown to inhibit entry into the first cycle but not the subsequent cycles of cell division, despite inhibition of p70 S6 kinase activation in both cases [S]. Rapamycin is structurally related to FKS06, and both drugs bind to the same family of intracellular receptors (immunophilins), termed FKBP (for review, see [18-201. While both agents bind to the same family of FKBP, they inhibit different signal transduction pathways. The complex of an FKBP bound to FK.506, but not to rapamycin, binds to and inhibits the activity of calcineurin, a calcium- and calmodulin- dependent serinekhreonine phosphatase [21, 221, and inhibits the activation of a number of cytokine genes. Conversely, rapamycin, but not FKS06, binding to an FKBP, interferes with the activation of p70 S6 kinase in several cell types, including IL-Zdependent cells [3,4], and appears to inhibit or to delay proliferation [8, 9, 16, 171. While the direct target of the rapamycin-immunophilin complex is not known, rapamycin provides a specific probe for the analysis of the activation of p70 S6 kinase. We, therefore, exploited this ability of rapamycin to inhibit p70 S6 kinase activity specifically to determine whether IL-3- or EPO-dependent proliferation was dependent upon the stimulation of the 70-kDa S6 kinase triggered by activated IL-3 or EPO receptors. Incubation of factor-dependent cell lines with IL-3 and with EPO induced p70 S6 kinase activity which was inhibitable by rapamycin. A cell line was found, however, that was able to proliferate in the presence of rapamycin, despite inhibition of p70 S6 kinase activity. Rapamycin inhibited p70 S6 kinase activity both in GO/Gl synchronized cells and in actively dividing cells, demon- strating that this effect was independent of cell cycle status. Taken together, these data suggest the existence of p70 S6 kinase-independent pathways of cell cycle progression.

2 Materials and methods

2.1 Cell culture, growth factors, and proliferation assays

The murine pro-B cell line BalF3 is dependent upon IL-3 for growth. Transfection of BalF3 with the murine EPO receptor (BaR3-EPOR) renders these cells able to proli- ferate in the presence of either murine IL-3 or EPO [23]. Proliferation of Ba/F3-EPOR cells is dependent upon the concentration of IL-3 or EPO. The EPOR was initially cloned from a cDNA library derived from the transformed murine erythroleukemia cell line MEL [24]. MEL cells express both the murine EPOR and the Friend spleen focus-forming virus (SFFV) gpSS glycoprotein; this combi- nation renders MEL cells capable of proliferation indepen- dent of the addition of growth factor. BaF3-EPOR were further transfected with the SFFVgpSS glycoprotein; Ba/F3 cells expressing both EPOR and the gp5S glycoprotein

(BalF3-EPORgp.55) are able to proliferate in the absence of either IL-3 or EPO [23].

The parental Ba/F3 cells and Ba/F3-EPOR cells were grown in RPMI 1640 supplemented with 10% FCS, 2 mM glutam- h e , 10 mM Hepes pH 7.4, and 50 p~ 2-ME (RPMI-10% FCS) supplemented with 5-20'70 IL-3-containing WEHI- 3B-conditioned medium prepared as described [2S]. BalF3- EPORgpSS cells were cultured in RPMI-10% FCS. MEL cells were cultured in DMEM supplemented with 10% FCS, 2 mM glutamine, and SO pM 2-ME (DMEM-10% FCS). All cells were passaged every 48 h. In proliferation assays, purified murine IL-3 (the kind gift of Biogen Labs, Cambridge, MA) and recombinant human EPO (R & D Systems, Minneapolis, MN) were used at concentrations that resulted in maximal proliferation at 24 h (see below) in preliminary assays (data not shown). Ba/F3-EPOR cells in the plateau phase of growth were washed three times in RPMI free of growth factors and pre-incubated at 106 celldm1 for 3 h in RPMI-10% FCS at 37°C prior to proliferation assays. Alternatively, cells in plateau phase of growth were starved of growth factor and serum by culture in RPMI 1640 with no serum [RPMI 1640 supplemented with 1% BSA, 2 mM glutamine, 10 mM Hepes pH 7.4, and SO pM 2-ME (RPMI-1% BSA)] prior to refeeding. Rapa- mycin, FKS06, cyclosporin A (CsA), or equivalent ethanol diluent (final concentration 0.1'70) was added to cells plated in 96-well plates in triplicate. Cell proliferation was assessed by (1) cell counting using trypan blue exclusion at 24 and 48 h and by (2) the incorporation of [3H] thymidine in a 4-6-h pulse following an 18-h incubation by harvesting the incorporated radioactivity with the use of a Skatron microtiter plate harvester (Skatron, Sterling,VA).

2.2 Treatment of cells with rapamycin/FK506/CsA and stimulation of Ba/F3-EPOR cells with IL-3/EPO/PMA

Ba/F3-EPOR cells in the plateau phase of growth were washed free of growth factors and incubated at S x 106 cells/ml for 2 h in RPMI-10% FCS at 37°C. Rapamycin, FKS06, CsA, different combinations, or equivalent ethanol diluent (final concentration 0.1%) at the indicated concen- trations were added and the cells were further incubated for 1 h in the same conditions. IL-3, EPO (at concentrations that rendered maximal proliferation), or PMA (SO ng/ml) were added, and the cells were further incubated at 37 "C in the presence or absence of drug. BaF3-EPORgp5S and MEL cells (5 x 106/ml) were incubated for 1 h at 37 "C with rapamycin/FK506/CsA or ethanol diluent in RPMI-10% FCS or DMEM-10% FCS, respectively, prior to assay. Similar analyses were performed on Ba/F3-EPOR, BalF3- EPORgp55 and MEL cells that had been starved of growth factor and/or serum prior to assay.

2.3 Antiphosphotyrosine immunoblotting analysis

Cells ( 107/sample) were washed, lysed, and centrifuged as previously described [26], except that the lysis buffer contained 150 mM NaCl. The protein concentration in the supernatants was determined by a colorimetric method [27]. Equivalent protein amounts (80 pg) of each superna-

2666 V. Calvo, M. Wood, C. Gjertson, T. Vik and B. E. Bierer Eur. J. Immunol. 1994. 24: 2664-2671

tant were resolved by reducing 8.5%-10% SDS-PAGE and analyzed by immunoblotting with the antiphosphotyrosine mAb 4G10 and anti-mouse IgG alkaline phosphatase- conjugated antibody (Promega, Madison,WI), as previous- ly described [26].

2.4 p70 S6 kinase and pS5 (rsk) S6 kinase immunocomplex assays

Cells (0.5 x lo7 - 2 X 107/sample) were washed with cold PBS containing 0.4 mM Na3V04,5 mM EDTA, and 10 mM sodium pyrophosphate, and cell pellets were frozen in liquid nitrogen and stored at -80°C until analysis. Cell pellets were lysed in 0.5-1 ml of lysis buffer (10 mM Hepes pH 7.8, 15 mM KC1, 1 mM EDTA, 1 mM EGTA, 10% glycerol, 20 mM [bglycerophosphate, 0.1 mM Na3V04, 1 mM DTT, 1 mM PMSF, 10 pg/ml leupeptin, 10 pg/ml pepstatin, 10 pg/ml antipain, 0.2% NP-40). After 20 min, rocking at 4"C, lysates were centrifuged (14000 x g, 15 min, 4 "C), and supernatants were used for immunopre- cipit at ion.

For the detection of p70 S6 kinase activity, equivalent amounts of protein (300-500 pg as determined by the Bradford assay [27]) were incubated with 2 p1 of a p70 S6 kinase-specific antiserum for 2 h at 4 "C.This antiserum was generated by immunizing rabbits with a C-terminal portion of the rat p70 S6 kinase in the form of a bacterially- produced glutathione S-transferase fusion protein (the kind gift of Dr. Joseph Avruch, Massachusetts General Hospital, Boston, MA). Protein A-Sepharose CL4B beads were washed in lysis buffer and 100 pl of a 10% solution were added. The mixture was incubated for 40 min at 4°C. The immunocomplexes were washed at 4 "C once with lysis buffer, twice with 10 mM Tris-HC1 pH 7.2, 1 M NaC1, 1% Tween 20, 20 mM fl-glycerophosphate, 1 mM Na3V04, once with 20 mM Tris-HCI pH 7.4, 0.1% Triton X-100, 1 mM EDTA, 1 mM EGTA, 10% glycerol, 20 mM p- glycerophosphate, 1 mM DTT, and once with kinase buffer (10 mM MOPS pH 7.4,lO mM MgCl2, 1 mM DTT). Kinase reactions were performed on the immunocomplexes by adding 30 p1 of kinase buffer containing 5 pM protein kinase inhibitor (Sigma), 100 p~ ATP with 10 pCi [ Y - ~ ~ P ] ATP and rat 40s ribosomal subunits (0.4 A260) as substrate, pre- pared as previously described [28]. The mixture was incubated for 15 min at 30°C and analyzed by reducing SDS-PAGE and autoradiography. The activity of the p85 (rsk) S6 kinase was assayed in immunocomplexes obtained with specific antisera (125, [29] or anti-mouse rsk kinase, Upstate Biotechnologies, Inc., Lake Placid, NY) as pre- viously described [30, 311.

2.5 In vivo S6 phosphorylation

Cells (107) were washed three times in phosphate-free, serum-free medium (Flow Laboratories) and resuspended in 4 ml of this medium. All subsequent incubations were performed at 37 "C. Cells were incubated for 3 h, recovered by centrifugation, resuspended in 2 ml of the same medium containing 1 mCi of 32Pi, and incubated for 2 h. Rapamycin (100 nM) or ethanol diluent (0.1%) were added, and cells were incubated for 1 h, after which 0.2 ml of dialyzed FCS were added. Cells were recovered 1 h later by centrifuga-

tion and lysed, and ribosomes were partially purified as previously described [32]. Proteins were resolved by reduc- ing SDS-PAGE, and 32P incorporation into S6 was deter- mined by autoradiography.

3 Results

3.1 The cytokines IL-3 and EPO induce activation of p70 S6 kinase activity that is inhibited by rapamycin

To analyze the intracellular signals induced by the cytokines IL-3 and EPO, we used a model system in which the EPOR had been transfected into a murine, IL-3-dependent, pro-B cell line, BaR3 (BaF3-EPOR). BaF3-EPOR is able to proliferate in the presence of either IL-3 or EPO. BaR3-EPOR cells were starved of growth factor for 3 h, after which cells were stimulated by the addition of either IL-3 or EPO for 1 h. Cells were pelleted, lysates prepared, and p70 S6 kinase activity was measured by an immuno- complex assay [33] using specific anti-p70 S6 kinase antise- rum. The addition of either purified murine IL-3 or recombinant EPO resulted in the induction of p70 S6 kinase activity (Fig. l).Thus, like IL-2, both IL-3 and EPO stimulation induced the activity of p70 S6 kinase. We and others have shown that the macrolide immunosuppressant rapamycin is able to inhibit the induction of p70, but not p85 (rsk), S6 kinase activity. We, therefore, added rapamy-

- IL-3 + Rnpamyein (nM) - Ba/F3- EPOR

BalF3- EPORgp55

MEL

Figure I. Effect of rapamycin on the 70-kDa S6 kinase activity in BalF3-EPOR, Ba/F3-EPORgp55, and MEL cells. Cells were incubated in the absence or presence of rapamycin (0.1%) for 1 h at the indicated concentrations. BalF3-EPOR cells were further incubated with IL-3 or EPO for 1 h or left untreated (NS). Samples labeled as M correspond to cells incubated in medium (M) alone in the absence of rapamycin. 70-kDa S6 kinase activity was measured by the immunocomplex assay. The region of the gel corresponding to the S6 protein is shown.

Eur. J. Irnrnunol. 1994. 24: 2664-2671 Cytokines, 70-kDa S6 kinase activity, raparnycin, and cell proliferation 2667

cin to these cells to determine if cytokine-induction of p70 S6 kinase activity was sensitive to this agent. Preincubation of BdF3-EPOR cells in rapamycin resulted in a concentra- tion-dependent inhibition of p70 S6 kinase activity which was essentially complete at 1 nM rapamycin. Rapamycin did not cause a general inhibition of activated ser- inekhreonine kinases in these cell lines, since rapamycin treatment did not inhibit the PMA-induced stimulation of the p85 (rsk) S6 kinase in this cell line (Fig. 2). The inhibitory effect on the p70 S6 kinase was specific for rapamycin, since the structurally related compound FK506 and the structurally unrelated immunosuppressant CsA were ineffective (Fig. 3). Taken together, both IL-3 and EPO were able to induce, and rapamycin was able to inhibit, p70 S6 kinase activity.

BalF3- Ba/F3- MEL EPOR EPORgp55

Raps + + + + + + + IL-3 + + PMA + + + + + +

1 2 3 4 5 6 7 8 9 1 0 I 1 12 1 3 14

Figure2. Effect of raparnycin on the p85 S6 kinase activity in Ba/F3-EPOR, BdF3-EPORgp55, and MEL cells. Cells were incubated in the absence or presence of raparnycin (100 nM) for 1 h and were further incubated with IL-3 or PMA for 10 min or left untreated as indicated. p85 S6 kinase activity was measured by the irnrnunocornplex assay. The region of the gel corresponding to the S6 protein is shown.

3.2 Cytokine-independent cell lines exhibit rapamycin-inhibitable p70 S6 kinase activity

The cDNA encoding the EPOR was initially isolated from the transformed murine erythroleukemia MEL cell line. The co-expression of SFFV gp55 renders the EPOR constitutively active in these cells. Like MEL cells, BalF3- EPOR cells co-expressing gp55 (Ba/F3-EPORgp55) are able to grow in medium alone in the absence of exogenous IL-3 or EPO. We tested the activity of p70 S6 kinase in immune complex kinase assays from both Ba/F3- EPORgp55 and MEL cells, and were able to detect p70 S6 kinase activity (Fig. l ) , which was not consistently altered by the addition of IL-3, EPO, or PMA (data not shown). This activity was inhibited by rapamycin at similar concen- trations to those used to inhibit p70 S6 kinase activity from BalF3-EPOR cells (Fig. 1). As expected, CsA and FK506 had no effect (Fig. 3). Furthermore, the effects were specific to the pathways leading to p70 S6 kinase activation as p85 rsk activity was not affected (Fig. 2). These findings suggested that p70 S6 kinase is active in growing cells, and that rapamycin inhibits this activity downstream of growth factor binding to its growth factor receptor.

3.3 The complex of rapamycin bound to FKBP mediates inhibition of p70 S6 kinase activity

Rapamycin and the structurally related macrolide com- pound FK506 both bind specifically to a family of intracel-

Rapa + + + FK506 + + CsA + +

BalF3- EPOR

BdF3- EPORgp55

MEL

Figure 3. Reversal of raparnycin effects on the 70-kDa S6 kinase activity by FK.506. Cells were incubated in the absence or presence of rapamycin (1 nM) alone or together with FK506 (1 pM) or CsA (1 p f ) for 1 h as indicated. Ba/F3-EPOR cells were further incubated with IG3 for 1 h. The 70-kDa S6 kinase activity was measured by the irnmunocomplex assay. The region of the gel corresponding to the S6 protein is shown.

lular receptors termed FKBP [18,19,34]. Common binding to these FKBP appears to be the molecular basis for the ability of FK506 to antagonize the inhibitory effects of rapamycin on IL-Zstimulated cell cycle progression [16], in that FK506, unable to inhibit growth factor-dependent stimulation, displaces rapamycin from the FKBP-immuno- suppressant complex. Therefore, excess concentrations of FK506 will reverse the actions of rapamycin [16, 351. We tested the ability of FK506 to reverse rapamycin-induced inhibition of 70-kDa S6 kinase activity compared with that of untreated cells in the three cell lines. In each case, a 1000-fold excess of FK506 was able to reverse rapamycin- induced inhibition of p70 S6 kinase (Fig. 3). These results are consistent with a model in which the complex of immunophilin bound to immunosuppressant is the media- tor of the inhibitory effects of rapamycin on p70 S6 kinase activity in these cell lines.

3.4 Rapamycin did not alter tyrosine phosphorylation of a number of polypeptides

Both IL-3 and EPO have been shown to increase tyrosine phosphorylation of a number of intracellular polypeptides [36-411, although the identity of many of the tyrosine kinases activated by IL-3 or EPO receptor stimulation remains unknown. The activation of tyrosine kinases constitutes a necessary signal for the induction of prolifer- ation in a variety of cell types. Since rapamycin inhibited p70 S6 kinase activity downstream of growth factor binding to the growth factor receptor,we wished to determine if the effects of rapamycin were exerted through inhibition of tyrosine kinase activity. Stimulation of Ba/F3-EPOR cells with IL-3 for 10 min increased the tyrosine phosphorylation of a protein of - 95 kDa, which appeared to be constitu- tively phosphorylated in BdF3-EPORgp55 cells (Fig. 4). The phorbol ester PMA induced the phosphorylation of

2668 V. Calvo, M. Wood, C. Gjertson, T. Vik and B. E . Bierer Eur. J. Immunol. 1994. 24: 2664-2671

proteins of - 85 kDa and - 42 kDa in both cell lines. The appearance of tyrosine phosphorylation of these polypep- tides was not affected by incubation of the cells with concentrations of rapamycin at least 100-fold above those needed for complete inhibition of p70 S6 kinase activity, suggesting that the agents'effects are not mediated through general inhibition of tyrosine kinase activity. In agreement with these results, rapamycin treatment did not affect the general pattern of tvrosine-phosphorylated proteins in MEL celis (Fig. 4).

120

100

80

60

40

BalF3- EPOR

Rapa + + + IL-3 + + PMA + +

1 2 3 4 5 6

228-

110-

+ 7 0-

4 4-

-b

2 8-

Ba/F3- MEL EPORgp55

+ + + + +

7 8 9 I 0 11 12

- 215

-105

-7 0

-4 3

-2 8

Figure 4. Effect of rapamycin on the tyrosine phosphorylation pattern in Ba/F3-EPOR, BalF3-EPORgp55, and MEL cells. Cells were preincubated in the absence or the presence of rapamycin (100 nM) for 1 h and treated with IL-3 or for 10 min PMA or left untreated as indicated. Lysates were analyzed by anti-phosphoty- rosine immunoblotting.The positions of molecular weight markers for the left and middle panels are indicated on the left and the positions of molecular weight markers for the right panel are indicated on the right. The arrows indicate the positions of the - %-, 85-, and 42-kDa substrates described in the text.

3.5 Rapamycin inhibits the proliferation of Ba/F3-EPOR and Ba/F3-EPORgp55, but not MEL cells

Rapamycin has been shown to inhibit the proliferation of a number of growth factor-dependent cell lines. As expected, rapamycin inhibited, in a concentrtion-dependent fashion, IL-3 and EPO-supported proliferation of Ba/F3-EPOR cells; rapamycin further inhibited the proliferation of BaF3-EPORgp55 cells grown in medium alone (Fig. 5) . Inhibition of proliferation was specific for rapamycin since neither the structurally related immunosuppressant FK506 nor the unrelated immunosuppressant CsA had any effect on proliferation of these cells (Fig. 5 and data not shown). Thymidine incorporation of the transformed MEL cells, however, was either unchanged or minimally (10-15%) affected by 100 n M rapamycin, despite complete inhibition of p70 S6 kinase activity at these and lower concentrations of drug. Parallel results were obtained when cells were counted at 24 and 48 h following rapamycin addition (data not shown).

n t .0001 .001 . 01 . 1 1 10 100 1000

Rapamycin (nM)

Figure 5. Effect of rapamycin on the proliferation of BdF3- EPOR, Ba/F3-EPORgp55, and MEL cells. Ba/F3-EPOR cells incubated with IL-3 (0) or EPO (m), BaIF3-EPORgp55 cells (0), or MEL cells (0) were treated with rapamycin at the indicated concentrations. Cell proliferation (initial concentration 5 x lo3 cells/well) was measured by ["H] thymidine incorporation and normalized to that of untreated cells, designated as maximal proliferation. The cprn corresponding to the maximal incorpora- tion were: BaE3-EPOR cells incubated with IL-3, 68740: BalF3- EPOR cells incubated with EPO, 43 678; BalF3-EPORgp55 cells incubated in medium alone, 27225; MEL cells incubated in medium alone, 31 758. ['HI thymidine incorporation of Ba/F3- EPOR cells in medium alone was 646 cpm. Standard deviations of triplicate determinations were less than 10% of maximal prolifer- ation. This assay is representative of three independent experi- ments with similar results.

The influence of cell cycle status on the ability of rapamycin to inhibit proliferation was assessed. MEL cells were compared both to Ba/F3-EPOR and to Ba/F3-EPORgp55. Cells were grown without FCS (RPMI-1% BSA) for 24 h prior to treatment with rapamycin and refeeding. Both [3H] thymidine incorporation and cell counts were deter- mined at 24,48, and 72 h (seeTable 1 and data not shown). Despite equivalent incorporation of thymidine and cell counts at the time of treatment, rapamycin inhibited, in a concentration-dependent fashion, proliferation of the Ba/F3-EPORgp55 cell line, but not of MEL cells grown similarly. Following rapamycin treatment, no p70 S6 kinase activity could be detected after serum-starvation or 24 h after refeeding, the latter a time of active cell division (data not shown). Therefore, there was no recovery of p70 S6 kinase activity in the time course of the experiments. MEL cells, therefore, do not appear to require p70 S6 kinase activity for proliferation and cell cycle progression. These data imply that the activity of the target of rapamycin, presumably an upstream activator or regulator of p70 S6 kinase activity, is not an absolute requirement for cell proliferation.

3.6 Rapamycin inhibits S6 phosphorylation in MEL cells

Rapamycin did not inhibit proliferation of MEL cells but nevertheless inhibited p70 S6 kinase activity i n these cells (Figs. 1 and 3). Although p70 S6 kinase appears to be the major physiological S6 kinase in somatic cells, the pS5 S6

Eur. J. Immunol. 1994. 24: 2664-2671 Cytokines, 70-kDa S6 kinase activity, rapamycin, and cell proliferation 2669

Table 1. Effects of rapamycin on proliferation of serum-starved ce1W A.

Barn3-EPORgp55 MEL

+ - + Rapamycin __

24 h 55400 17020 26200 38360 48 h 34330 13820 40320 66840 72 h 29 800 4860 27780 21 155

B ~ _ _ _ _ _ ~

BaE3-EPORgp55 MEL ( X 104) ( X 104)

+ - + Rapamycin -

24 h 1.6 0.44 1.8 2.67

48 h 9.7 2.7 3.1 4.0

72 h 44.0 6.2 16.9 16.4

(27.5%)’) (150%)

(27.5 Yo ) (129%)

(14.1 Yo) (97.1 Yo) ~ _ _ _ _ _ _ _ _____ ~ ~

a) Cells were cultured in RF’MI-1% BSA for 24 h prior to treatment with 100 nM rapamycin and refeeding with RPMI- 10% FCS. Cells were then cultured for the indicated times. (A) Cell proliferation (initial concentration, 2 x lo4 cells/well) was measured by [3H] thymidine incorporation (cpm). In the absence of refeeding, the cpm of untreated Ba/F3-EPORgp55 cells was 5675 cpm and of MEL cells was 3890 cpm. (B) Cell counts (initial concentration, 1 x lo4 cells/well) were deter- mined by trypan blue exclusion. The table is representative of three similar experiments.

b) Percent change from cell counts of cells that had not been treated with rapamycin.

Rapa - +

Figure6. Effect of rapamycin on the phosphorylation of the ribosomal S6 protein in vivo in MEL cells. Cells were incubated in the absence or presence of rapamycin (100 nM). Ribosomes were partially purified and the phosphorylation of the S6 protein (indicated with the arrow) was visualized on autoradiography.

kinase is nevertheless a physiological S6 kinase, at least in Xenopus luevis oocytes [7]. Rapamycin did not inhibit p85 S6 kinase in MEL, Ba/F3-EPOR, or BaF3-EPORgp55 cells (Fig. 2), as has been previously observed in several cell types [ 3 , 4, 8, 91. It was, therefore, theoretically possible that the rapamycin-resistant p85 S6 kinase was able to phosphorylate S6 in vivo in rapamycin-treated MEL cells, providing the level of S6 phosphorylation required for proliferation. MEL cells were labeled in vivo with 32Pi, treated with rapamycin or left untreated, and the phospho- rylation of S6 protein was examined (Fig. 6). Rapamycin

reduced the in vivo phosphorylation of S6 in MEL cells, which correlated with the inhibitory effect of rapamycin on p70 S6 kinase (Figs. 1 and 3) and not with the rapamycin- resistant basal activity of the p85 (rsk) S6 enzyme (Fig. 2). Taken together, these observations suggest that neither p70 S6 kinase activity nor the phosphorylation status of the ribosomal protein S6 are critical factors in the regulation of the proliferation of the transformed MEL cell line. Fur- thermore, they suggest that the relevant enzyme able to maintain phosphorylation of the S6 protein in vivo is p70, not p85, S6 kinase.

3.7 Rapamycin mediates inhibition of proliferation via binding to an FKBP

Since the anti-proliferative effect of rapamycin had been dissociated from its ability to inhibit p70 S6 kinase activity,

Bn/FJ-EPOR + IL-3

120 I I -

100

40

20

0 Ba/F3-EPOR

120 1 EPO

I

100

80

6 0

40

20

0 BalF3-EPORgp55

120 1

100

80

60

40

20

0 0 0.1

Rapamycin (nM) Figure 7. Reversal of rapamycin effects on proliferation by FK506. Cells were incubated in the absence or presence of rapamycin (0.1 n M ) alone (0) or together with FK.506 (100 nM, M) or CsA (1 VM, M). Cell proliferation was measured by [3H] thymidine incorporation and normalized to that of untreated cells, designated as maximal proliferation. The values shown correspond to the mean of the values obtained in at least three independent experiments and the error bars correspond to the mean of the standard deviations of the independent experiments.

2670 V. Calvo, M. Wood, C. Gjertson, T.Vik and B. E. Bierer Eur. J. Immunol. 1994. 24: 2664-2671

it was possible that this effect was not mediated via an association with an FKBP in the sensitive cell lines. A 1000-fold molar excess of FK506 was added to cultures incubated with 1 nM rapamycin; the excess FK506 was able to revert rapamycin-mediated inhibition of proliferation of both the Ba/F3-EPOR (grown in either IL-3 or EPO) and of Bam3-EPORgp55 (Fig. 7). This ability was specific for FK506, since the immunosuppressive CsA, which binds to a distinct family of intracellular receptors (cyclophilins), had no effect (Fig. 7).Thus, it would appear that the biological effect of rapamycin is mediated by FKBP complex forma- tion, yielding a biologically active moiety able to inhibit both p70 kinase activity and, in some cell lines, prolifera- tion.

4 Discussion

The results presented in this report demonstrate that one of the downstream events triggered by stimulation of the IL-3 or EPO receptors is activation of the p70 S6 kinase. The signal transduction pathways of a number of cytokines appear to converge rapidly after the interaction of the cytokine with its surface receptor, since the proliferation of several cell types can be sustained by a cytokine to which they were not initially responsive after transfection of the corresponding receptor. In addition, some cytokines have been shown to increase the phosphorylation of intracellular substrates. In some cases, the phosphorylation of an apparently similar polypeptide(s) appears to be increased by stimulation with several cytokines. These observations suggest that some signal transduction pathways are used by several cytokine receptors. Our data appear to suggest that ligand binding to IL-3, EPO, and IL-2 receptors converge on a common signal transmission pathway leading to the activation of the p70 S6 kinase [3, 41.

Rapamycin inhibited the proliferation of BaE3-EPOR and BaF3-EPORgp55 cells in a dose-dependent manner (IC50 - 0.01 nM).The maximal reduction of DNA synthe- sis observed was 50%-60%, suggesting that rapamycin may lengthen the G1 phase of the cell cycle in these cell lines rather than cause an irreversible block, similar to what has been observed in fibroblasts [4]. The molecular target of this inhibition is not the cytokine-receptor interaction, since the proliferative signals generated by the constitutive- ly activated receptor present in BdF3-EPORgp55 cells were also affected by rapamycin. Additionally, rapamycin did not appear to inhibit the early increases in tyrosine phosphorylation triggered by the binding of the cytokine to the receptor.

Rapamycin inhibited the activity of the p70 S6 kinase in a dose-dependent manner (ICso - 0.1 nM) but is not a gen- eral inhibitor of serine/threonine kinase in these cells, since the p85 rsk S6 kinase was unaffected. Activation of the p70 S6 kinase was less sensitive to rapamycin than proliferation (IC5o - 0.03 nM), although rapamycin was able to block completely the activation of the kinase as detected in our assay. Since rapamycin does not affect the enzymatic activity of the p70 S6 kinase itself [4, 8, 91, it appears to interfere specifically with the regulation of the activation state of the p70 S6 kinase. This regulation is achieved by phosphorylation/dephosphorylation on multiple ser- inehhreonine residues [33, 42, 431, although the pertinent

upstream kinases and phosphatases have yet to be identif- ied. Tyrosine kinases of the JAK family, which are involved in cytokine receptor signal transduction [15], may partici- pate in the regulation of the kinases and phosphatases acting on p70 S6 kinase. These kinases and phosphatases could constitute the rapamycin target directly linked to cell cycle progression and exhibit a sensitivity to rapamycin more similar to that observed for proliferation.

Inhibition of p70 S6 kinase activity by microinjection of specific antibodies blocks proliferation of fibroblasts at the GUS boundary, supporting the view that the p70 S6 kinase is a critical regulator of G1 progression in these cells [lo]. The GUS boundary is also the point at which rapamycin affects IL-2-dependent proliferation [35, 441. These data suggest that p70 S6 kinase activity might also be essential for G1 progression of cytokine-dependent cells, as it is for fibroblasts. However, as described above, rapamycin appears to cause a delay in G1 progression rather than an irreversible block, both in fibroblasts and cytokine-depen- dent cells. The discrepancy between the consequences of direct inhibition of p70 S6 kinase activity and those of rapamycin-mediated inhibition may be explained by the hypothesis that cytokine-dependent cells and fibroblasts recruit additional (rapamycin-insensitive) pathways lead- ing to proliferation.

The rapamycin target(s) in the pathway(s) leading to p70 S6 kinase activation are very likely the same molecules in IL-2, IL-3, and EPO signal transduction. Phosphatidylinositol (PI) 3-kinase activity is stimulated in response to IL-2 and EPO [45-481. Interestingly, a target of rapamycin in yeast (TOR2) contains a PI 3-kinase domain [49], suggesting that, in cytokine-dependent cells, either the known mam- malian PI 3-kinase, unidentified mammalian TOR2 homo- logs, or upstream activators of these enzymes, could be rapamycin targets. Another possibility raised by these observations is that stimulation of PI 3-kinase might be directly linked to p70 S6 kinase activation in the IL-2 and EPO signal transduction pathways.

The proliferation of MEL cells, synchronized in GOlGl by prolonged serum starvation, was not significantly altered by rapamycin, while the proliferation of equivalently syn- chronized, cytokine-independent BaF3-EPORgp55 cells was inhibited by the drug. Nevertheless, rapamycin inhi- bited the p70 S6 kinase activation both after serum starvation but prior to serum refeeding and 24 h after serum refeeding. This result implies that the effect of rapamycin is independent of cell cycle status, since active cell division is taking place 24 h after serum repletion. Taken together, these results suggest that some downstream targets of the p70 S6 kinase may be constitutively activated in MEL cells and/or other rapamycin-insensitive signal transduction pathways may substitute for the functional lack of the kinase in rapamycin-treated MEL cells. MEL cells exhibit molecular changes that contribute to their malignant phenotype, some of which are known (Spi-1 activation and p53 inactivation) [50, 511, and could be related to the insensitivity to rapamycin. The only physiologically rele- vant substrate of the 70-kDa S6 kinase identified so far is the ribosomal protein S6. S6 phosphorylation may play a regulatory role in G1 progression by increasing the efficien- cy of translation of some key proteins [6, 71. The observa- tion that rapamycin inhibits S6 phosphorylation but not cell

Eur. J. Immunol. 1994.24: 2664-2671

cycle progression in MEL cells suggests that some trans- formed cell lines may possess alterations that neutralize or bypass this regulatory mechanism. The identification of these alterations would help to understand the mechanisms involved in cytokine signal transduction in physiological conditions.

Cytokines, 70-kDa S6 kinase activity, rapamycin, and cell proliferation 2671

21 Liu, J., Farmer, J. D. J., Lane,W. S., Friedman, J.,Weissman, I.

22 Fruman, D. A., Klee, C. B., Bierer, B. E. and Burakoff, S. J.,

23 Yoshimura, A., D’Andrea, A. D. and Lodish, H. F., Proc. Natl.

24 D’Andrea, A. D., Lodish, H. F. and Wong, G. G., Cell 1989.57:

and Schreiber, S. L. , Cell 1991. 66: 807.

proc. Natl. Acad. sci.

Acad. Sci. USA 1990. 87: 4139.

277.

1992. 89: 3686.

25 D’Andrea, A. D.,Yoshimura, A. ,Youssoufian, H., Z o n , L. I., Koo, J.-W. and Lodish, H. F., Mol. Cell Biol. 1991. If: 1980.

22: 457.

We are greatiy indebted to J. Avruch for invaluable discussions and advice and for providing the p7U S6 kinase specific antiserum. We

BalF3-EPORgp55 cell lines, and S. J. Burakoff, A. Carrera, R. Erikson, and D. Fruman for stimulating discussions.

thank A. ~ ~ ~ d ~ ~ ~ f ~ ~ ~ ~ ~ ~ i d i ~ ~ the transfected BaIF3-EpOR and 26 Calvo,v, Bierer, B. E. and Vik,T. A.3 Eur. J. Immunol. 1992.

27 Bradford, M., Anal. Biochem. 1976. 72: 248. 28 Thomas, G., Gordon, J. and Rong, H., J. Biol. Chem. 1978.

Received September 28, 1993; in revised form July 11, 1994; accepted July 28, 1994.

5 References

1 Arai, K.-I., Lee, F., Miyajima, A., Miyatake, S., Arai, N. and

2 Boulay, J. L. and Paul. W. E., J. Biol. Chem. 1992.267: 20525. 3 Calvo,V., Crews, C. M. ,Vik,T. A. and Bierer, B. E., Proc. Natl.

4 Kuo, C. J., Chung, J., Fiorentino, D. F., Flanagan, W. M.,

5 Terada. N., Franklin, R. A , , Lucas, J. L., Blenis, J. and

6 Sturgil1,T.W. and Wu, J., Biochem. Biophys. Acta 1991. 1092:

7 Erikson, R. L., J. Biol. Chem. 1991. 266: 6007. 8 Chung, J., Kuo, C. J., Crabtree, G. R. and Blenis, J. , Cell 1992.

69: 1227. 9 Price, D. J., Grove, J. R., Calvo,V, Avruch, J. and Bierer, B.

E., Science 1992. 257: 973. 10 Lane, H. A., Fernandez, A., Lamb, N. J. C. and Thomas, G.,

Nature 1993. 363: 170. 11 Turner, B., Rapp, U., App, H., Greene, M., Dobashi, K. and

Reed, J., Proc. Natl. Acad. Sci. USA 1991. 88: 1227. 12 Zmuidzinas, A., Mamon, H. J., Roberts,T. M. and Smith, K.

A., Mol. Cell Biol. 1991. 11: 2724. 13 Carroll, M. l?, Clark-Lewis, I . , Rapp, U. R. and May,W. S., J.

Biol. Chem. 1990. 265: 19 812. 14 Carroll, M. F!, Spivak, J. L., McMahon, M. ,Weich, N., Rapp,

U. L. and May, W. S., J. Biol. Chem. 1991. 266: 14964. 15 Ilhe, J. N., Witthuhn, B. A., Quelle, F. W., Yamamoto, K.,

Thierfelder, W. E., Kreider, B. and Silvennoinen, O., TZ. B.S. 1994. 19: 222.

16 Bierer, B. E., Mattila, P. S., Standaert, R. F., Herzenberg, L. A., Burakoff, S. J., Crabtree, G. and Schreiber, S. L., Proc. Natl. Acad. Sci. USA 1990. 87: 9231.

17 Dumont, F. J., Staruch, M. J., Koprak, S. K., Melino, M. R. and Sigal, N. H., J. Immunol. 1990. 144: 251.

18 Sigal, N. H. andDumont, F. J.,Annu. Rev. Zmmunol. 1992.10: 519.

19 Schreiber, S. L., Cell 1992. 70: 365. 20 Fruman, D. A., Burakoff, S. J. and Bierer, B. E., FASEB J.

Yokota, T., Annu. Rev. Biochem. 1990. 59: 783.

Acad. Sci. USA 1992. 89: 7571.

Blenis, J. and Crabtree, G. R., Nature 1992. 358: 70.

Gelfand, E. W., J. Biol. Chem. 1993. 268: 12062.

350.

1994. 8: 391.

29

30

31 32

33

34 35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50 51

II

253: 1101. Alcorta, D. A., Crews, C. M., Sweet, L. J., Bankston, L., Jones, S.W. and Erikson, R. L., Mol. Cell Biol. 1989. 9: 3850. Vik,T. A., Sweet, L. J. andErikson, R. L., Proc. Natl. Acad. Sci. USA 1990. 87: 2685. Chen, R.-H. and Blenis, J., Mol. Cell Biol. 1990. 10: 3204. Blenis, J., Chung, J., Erikson, E., Alcorta, D. A. and Erikson, R. L., Cell Growth Diff. 1991. 2: 279. Price, D. J., Gunsalus, J. R. and Avruch, J., Proc. Natl. Acad. Sci. USA 1990. 87: 7944. Bierer, B. E., Curr. Opin. Hematol. 1993. 1: 149. Dumont, F. J., Melino, M. R., Staruch, M. J., Koprak, S. L., Fischer, l? A. and Sigal, N. H., J. Zmmunol. 1990. 144: 1418. Koyasu, S., Tojo, A., Miyajima, A., Akiyama,T., Kasnga, M., Urabe, A., Schreurs, J., Arai, K.-I., Takata, F. F. and Yahara, I . , EMBO J. 13: 3979. Isfort, R., Huhn, R. D., Frackelton, A. R. and Ihle, J. N., J. Biol. Chem. 1988. 263: 19203. Ferris, D. K. ,Willet-Brown, J., Martensen, W. and Farrar,W. L., Biochem. Biophys. Res. Commun. 1988. 154: 991. Morla, A. O., Shreurs, J., Miyajima, A. and Wang, J.Y. J.. Mol. Cell Biol. 1988. 88: 2214. Sorensen, P. H. B., Mui, A. L., Murthy, S. C. and Kystal, G., Blood 1989. 73: 406. Quelle, F.W. and Wojchowski, D. M., J. Biol. Chem. 1991.266: 609. Ballou, L. M., Siegmann, M. and Thomas, G . , Proc. Natl. Acad. Sci. USA 1988. 85: 7154. Mukhopadhyay, N. K., Price, D. J., Kyriakis, J. M., Pelech, S., Sanghera, J. and Avruch, J., J. Biol. Chem. 1992. 267: 3325. Morice,W. G., Brunn, G. J., Wiederrecht, G., Siekierka, J. J. and Abraham, R. T., J. Biol. Chem. 1993. 268: 3734. Remillard, B., Petrillo, R., Maslinski, W., Tsudo, M., Strom, T. B., Cantley, L. and Varticovski, L., J. Biol. Chem. 1991.266: 14 167. Merida, I . , Diez, E. and Gaulton, G. N., J. Zmmunol. 1991. 147: 2202. Augustine, J. A,, Sutor, S. L. and Abraham, R. T., Mol. Cell Biol. 1991. 11: 4431. Damen, J. E., Mui, A. L. F., Pawson, I? L. and Krystal, G., Blood 1993. 81: 3204. Kunz, J., Henriquez, R., Schneider, U., Deuter-Reinhard, M., Movva, N. R. and Hall, M. N., Cell 1993. 73: 585. Ben-David,Y. and Berstein, A., Cell 1991. 66: 831. Mowat, M., Cheng, A., Kimura, N., Bernstein, A. and Benchimol, S., Nature 1985. 314: 633.