Functionally active fusion protein of the novel composite cytokine CLC/soluble CNTF receptor Catherine Guillet 1 , Eric Lelie ` vre 1 ,He ´ le ` ne Plun-Favreau 1 , Josy Froger 1 , Marie Chabbert 1 , Jacques Hermann 1 , Amelie Benoit de Coignac 2 , Jean-Yves Bonnefoy 2 , Hugues Gascan 1 , Jean-Franc¸ois Gauchat 2 and Greg Elson 2, * 1 INSERM U564, CHU d’Angers, Angers, France; 2 Centre dı ´ Immunologie Pierre Fabre, St-Julien-en-Genevois, France The h eterodimeric c ytokine composed of the soluble ciliary neurotrophic factor receptor (sCNTFR) and the IL-6 family member cardiotrophin-like c ytokine (CLC) w as recently identified as a new ligand for gp130–leukemia inhibitory factor receptor (LIFR) complex [Plun-Favreau, H., Elson, G., Chabbert, M., Froger, J., deLapeyriere, O., Lelievre, E., Guillet, C., Hermann, J., Gauchat, J. F., Gascan, H. & Chevalier, S. (2001) EMBO J. 20, 1692– 1703]. This heterodimer shows overlapping biological properties with LIF. Although CLC contains a putative signal peptide and therefore should enter into the classical secretory pathway, the p rotein has been shown t o be retained within transfected mammalian cells, unless coex- pressed with either sCNTFR or cytokin e like factor ( CLF) [Elson, G. C., Lelievre, E., Guillet, C., Chevalier, S., Plun- Favreau, H., Froger, J., Suard, I ., de Co ignac, A. B., Delneste, Y., Bonnefoy, J. Y., Gauchat, J. F. & Gascan, H. (2000) Nat. Neurosci. 3, 867–872]. In the present study, we demonstrate t hat a fusion protein comprising CLC covalently coupled through a glycine/serine linker to sCNTFR (CC–FP) is efficiently secreted from transfected mammalian cells. CC–FP shows enhanced activities in respect to the CLC/sCNTFR native complex, on a number of cells expressing gp130 and LIFR on their surface. In addition, CC–FP is able to compete with CNTF for cell binding, indicating that b oth cytokines s hare binding epitope(s) expressed by their receptor c omplex. Analysis of the downstream signaling events revealed the recruitment by CC–FP of the signal transducer and activator of tran- scription (STAT)-3, Akt and mitoge n-activated protein (MAP) kinase pathways. The monomeric bioactive CLC/ sCNTFR fusion protein is therefore a powerful tool to study the biological role o f the recently described c ytokine CLC. Keywords: C LC; s CNTFR; fusion protein. Ciliary n eurotrophic f actor (CNTF) was named based on its ability t o maintain t he survival of parasympathetic neurons of chick ciliary ganglions [1,2]. Subsequent stud ies have revealed that CNTF also enhances the survival of sensory [3], motor [4], cerebellar and hippocampal neurons [5,6]. It can also prevent lesion-induced degeneration of motor neurons and s lows disease progression in mice with inherited neuromuscular deficits [7–9]. CNTF is also known to be a trophic factor for skeletal muscles [10,11]. CNTF belongs to a family of structurally related cytokines t hat i ncludes leukemia i nhibitory factor (LIF), interleukin-6 ( IL-6), interleukin-11 (IL-11), oncostatin M (OSM), cardiotrophin-1 (CT-1) [12–14] and cardiotrophin- like cytokine (CLC) [15,16]. These cytokines share one or both of the receptor signal transducing subunits gp130 or LIF rec eptor ( LIFR) i n t heir respe ctive rece ptor complexes [17–20]. The functional CNTF r eceptor i s a ternary complex, that in addition to gp130 and LIFR also includes a specificity-determining binding component called CNTF receptor (CNTFR) anchored to the membrane through a glycosylphosphatidylinositol motif [21–25]. Binding of the cytokine to the membrane-bound, nonsignaling a chain (CNTFR; [21]), l eads to the recruitment of t he sha red signaling subunits gp130 a nd the LIFR with the formation of the high-affinity functional receptor complex [22,23]. The subsequent s ignaling cascade implicates activation of the Janus kinase 1 ( JAK1)/STAT3 pathway [26–30]. We recently identified a second ligand for the tripartite CNTF receptor as a complex formed between the IL-6 family cytokine CLC (also known as novel neurotrophin-1/B cell s timulatory factor-3) [15,16], and the soluble t ype-I cytokine receptor CLF [31,32]. We initially observed that CLC, although containing a signal peptide, was inefficiently secreted when expres sed in mammalian cells. This secretion could be induced upon coexpression with CLF, with the two proteins forming a heterodimer (CLF/CLC). This was the first demonstration of such a secretion mechanism for a cytokine of the IL-6 family and shares certain similarities with the formation of the functional IL-12 heterodimer [33]. Like CNTF, CLF/CLC recruits cells Correspondence to H. Gascan: INSERM U564, CHU d’Angers, 4 rue Larrey, 49033 Angers, France. Fax: + 33 241 73 16 30, Tel.: + 33 241 35 47 29, E-mail: hugues.gascan@univ-angers.fr or J F. Gauchat Centre dı ´ Immunologie Pierre Fabre, 5 Avenue Napole ´ on III, 74164 St-Julien-en-Genevois, France. Fax: + 33 450 35 35 90, Tel.: + 33 450 35 35 55, E-mail: jean.francois.gauchat@pierre-fabre.com. Abbreviations: sCNTFR, soluble ciliary neurotrophic factor receptor; CLC, cardiotrophin-like cytokine; LIFR, leukemia inhibitory factor receptor; CLF, cytokine like factor; IL, interleukin; MCS, multiple cloning site; PVDF, poly(vinylidene difluoride); JAK, janus kinase. *Presen t address: NovImmune, G eneva, Switzerland. (Received 8 November 2001, accepted 20 February 2002) Eur. J. Biochem. 269, 1932–1941 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.02850.x expressing on their surface the tripartite CNTF receptor, induces the tyrosine phosphorylation of gp130, LIFR and STAT3 in neuroblastoma cells and acts as a survival factor for motor neurons cultured in vitro [32,34]. We subsequently observed that CLC could also form a secreted composite cytokine when associated with sCNTFR. Similarly to LIF, CLC/sCNTFR displays activ- ities on cells which are negative for the expression of surface- bound CNTFR, but expressing gp130 and LIFR [35]. The a ssociation of CLC w ith s CNTFR is similar t o the situation reported previously for CNTF/sCNTFR, IL-6/sIL-6R and IL-11/sIL-11R, w here composite cyto- kines implicating a soluble r eceptor alpha component in their structure display functional activities mediated through the appropriate signaling subunits [23,36,37]. A closely related situation also exists for the IL-12 and IL-23 heterodimeric cytokines, composed of an a-recep tor-like chain (p40), respectively associated to p35 or p19 [33,38]. These studies revealed an interesting degree of binding promiscuity between the IL-6 and IL-12-type ligands and their multichain receptor complexes. Previous studies have demonstrated that the addition of a flexible glycine/serine link er between the two subunits of such composite cytokines allows the expression of a single chain fusion protein retaining functional activity [38–41]. For example, a fusion protein between IL-6 and soluble IL-6R, and named ÔHyper IL-6Õ, was shown to be functionally active in cells where IL-6 alone had no effect (i.e. lacking the membrane-bound form of IL-6R) [39]. These designer molecules display an increased stability compared with their respective composite cytokines pre- sumably because the cytokine and its cognate a receptor component are covalently associated, and both components remain bound to the signaling receptor subunits for a longer period of time. In an attempt to f acilitate the func tional characterization of the n ovel neurotrophic co mplex CLC/sCNTFR, we have generated a soluble fusion protein. CLC is irreversibly associated to its cognate receptor, the sCNTFR subunit, via a 10-amino-acid glycine/serine linker. We demonstrate that the CLC/sCNTFR fu sion protein is efficiently secreted from transfected mammalian cells and is highly active on c ell types expressing gp130 and LIFR on their s urface. MATERIALS AND METHODS Reagents Human IL-2, CNTF and L IF were purch ased from R & D Systems (Minneapolis, MN, USA). T he 4G10 monoclonal anti-phosphotyrosine Ig was bought from Upstate Biotech- nology (Lake Placid, NY, USA) and the 9E10 antic-myc epitope mAb was obtained from the ATCC (Rockville, MD, U SA). The antibody raised against STAT3 was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies detecting phospho-ERK1/ERK2, phospho- STAT3 (Tyr705) and phospho-AKT were purchased from New England Biolabs (Beverly, MA, USA). The monocl- onal antibodies d irected against t he human forms of LIFR (AN-E1, IgG1), gp130 (AN-HH1, IgG2a) and CNTFR (AN-B2, AN-C2, IgG2a) were generated in t he laboratory [35]. T he 4–68 monoclo nal anti-CNTF Ig was bought from Roche diagnostics (Meylan, France). Cell cultures Ba/F3 cells modified to e xpress functional receptors for LIF, CNTF or IL-6 were a kind gift from K. J. Kallen, University of Kiel, Germany. Cells were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum and 5 ngÆmL )1 recombinant LIF, CNTF or IL-6. HepG2 hepatoma cells, KB epidermoid c arcinoma, HEK 293 cells and SK-N-GP neuroblastoma cells (ATCC, Rockville, MD, USA) were maintained in RPMI 1640 supplemented with 10% fetal bovine serum. Construction of a single chain CLC/sCNTFR fusion protein (CC–FP) The cDNA encoding a soluble form of CNTFR (sCNTFR) was amplified b y PCR using the primers 5¢-CCGGAATTC GCCAGTGGTGAAGAGATG-3¢ and 5¢-CCGCTCGAG GTCACAGATCTTCGTGGT-3¢, a nd cloned i nto the EcoRI and XhoI restriction sites of pcDNA3. The oligonu- cleotides encoding the (G 4 S) 2 flexible polypeptide 5¢-TCG AAGGCGGAGGCGGGAGCGGCGGGGGCGGAAG CGGAGGCGGGGGAAGCCTCGAGT-3¢ and 5¢-CTA GACTCGAGGCTTCCCCCGCCTCCGCTTCCGCCCC CGCCGCTCCCGCCTCCGCCT-3 ¢, w ere a nnealed and cloned into the XhoIandXbaI restriction sites of the afore mentioned pcDNA3 vector containing the sCNTFR cDNA. T he cDNA encoding a derative of CLC containing the c-myc epitope was amplified from CLC cDNA using the primers 5¢-CCGCTCGAGCTCAATCGCACAGGGGAC CC-3¢ and 5¢-CCGCTCGAGTCAGAGGTCCTCCTCG GAGA-3¢ and cloned into pcDNA3 containing the mod- ified sCNTFR cDNA. Protein purification and Western blotting HEK 293 cell line was stably transfected with pcDNA3 expression vector encoding CC–FP. Cell culture medium containing CC–FP was concentrated approximately 10-fold using Centricon-30 units (Millipore, Bedford, MA, USA) and the fusion protein subsequently purified by affinity chromatography using an anti-(c-myc) Ig affinity matrix. Bound protein was eluted with 100 m M Glycine-HCl (pH 2 .75). A neutral pH was immediately restored using 1 M Tris base. Protein concentrations were determined by SDS/PAGE and silver staining using a BSA protein standard. Western-blotting of CC–FP was performed after SDS/PAGE and transfer onto a nylon membrane using a peroxidase coupled anti-(c-myc) Ig. Gel filtration Sample containing CC–FP was fractionated on a Superose 12 size exclusion column. Fractions were then analysed by Western-blotting as described before. Column calibration was performed using standard purified proteins. Protein modeling CC–FP has been modeled from the molecular models of CLC and CNTFR. CLC was modeled f rom residues 7 to 181 by homology with human CNTF (PDB accession number 1CNT) [42] and with mouse LIF (PDB accession Ó FEBS 2002 Bioactive CLC/sCNTFR fusion protein (Eur. J. Biochem. 269) 1933 number 1LKI) [43], as described previously [35]. R esidues 1–286 of CNTFR were modeled by homology with gp130 (PDB accession numbers: 1BQU for the cytokine-binding domain of gp130 and 1I1R for the Ig-like and the CBD domains of gp130 i n the complex with v iral IL-6) [44,45]. A flexible loop including the C-terminal part of CNTFR (residues 287–316), the linker joining the t wo proteins (LEGGGGSGGGGSLE) and the N-terminal part of CLC (residues 1–6) was generated and refined by simulated annealing. Computations were carried out with the model- ing program MODELER Ò [46], as implemented in INSIGHT (MSI, San Diego, USA) on a SGI Octane workstation. The quality of the model was checked with PROFILE 3 D [47]. Tyrosine phosphorylation analysis After a 24-h serum s tarvation, cells were stimulated for 10 min in the presence of the indicated c ytokine. Cells were lysed in 10 m M Tris/HCl pH 7.6, 5 m M EDTA, 50 m M NaCl, 30 m M sodium pyrophosphate, 50 m M sodium flu- oride, 1 m M sodium orthovanadate, proteinase inhibitors (1 lgÆmL )1 pepstatin, 2 lgÆmL )1 leupeptin, 5 lgÆmL )1 aprotinin, 1 m M phenylmethanesulfonyl fluoride) and 1% NP40 or Brij 96 depending on the experiments [35]. After pelleting insoluble material and protein standardization, the supernatants were immunoprecipitated overnight. The complexes were t hen isolated w ith beads coupled to protein A,submittedtoSDS/PAGEandtransferredontoan Immobilon m embrane ( Millipore, Bedford, MA, U SA). The membranes were subsequ ently incubated with t he relevant primary antibody before being incubated with the appropriate secondary antibody lab eled with peroxidase for 60 min. The r eaction was visualized on an X-ray film using ECL reagents (Amersham, Les Ullis, France) according t o the manufacturer’s instructions. In some experiments, the membranes were stripped overnight in 0.1 M glycine-HCl, pH 2.7, and n eutralized in 1 M Tris/HCl, pH 7.6, before reblotting. Cell proliferation assays BAF GL (gp130, LIFR ), BAF G LC (gp130, LIFR, CNTFR), BAF gp130/IL-6R or TF1 cells were seeded at 5 · 10 3 cellsÆwell )1 (in 96-well plates) in RPMI 1 640 m edium supplemented with 5% fetal bovine serum containing the indicated amount of recombinant cytokine. Following a 72-h incubation period, a [ 3 H]thymidine pulse was performed for 4 h and t he incorporated radioactivity determined as described previously [48]. KB cell IL-6 production assay KB cells were plated in 96-well plates at a conce ntration o f 5 · 10 3 cells per well in culture medium containing serial dilutions of recombinant c ytokines as indicated. After 4 8 h, the supernatants were harvested, an d their IL-6 conten t determined by ELISA as described previously [32]. Gene reporter assay Transient transfection of KB cells were carried out in 24-well culture plates using t he lipid reagent Fugene 6 from Roche Diagnostics. Cells were transfected with 300 ng SIEM-luciferase reporter gene, as described p reviously [32]. Forty-eight hours a fter transfection, cells were incubated with IL-2, LIF, CLC/sCNTFR or CC–FP for an additiona l 18 h . Transfected cells were washed with ice cold NaCl/P i , and 100 lL of lysis buffer was added to the wells (0.1 M KH 2 PO 4 ,pH7.8,0.1%TritonX-100).Extractswerethen used directly to measure the luciferase activity by integrating total light emission over 10 s using a Packard Topcount luminometer (Meriden, CT, USA). Luciferase activity was normalized based on protein concentrations. FACS analysis and cytokine displacement BAF GLC and SK-N-GP cells were incubated in the presence of increasing concentrations of putative competitor (CC–FP, IL-11, IL-4) and a fixed amount of CNTF (2 ng in a20-lL final volume). After a 2-h incubation period, cells were washed and i ncubated with the 4–68 monoclonal a nti CNTF Ig, or with an IgG1 control antibody, for 30 min. After washing, cells were fu rther incubated with a phy- coerythrin-conjugated anti-(mouse IgG) I g. Fluorescence was subsequently analyzed on a FACScan flow cytometer from Becton & Dickinson (Mountain View, CA, U SA). RESULTS The bioactive designer cytokine hyper-IL-6 (H-IL-6) [39] was used as a model to generate a functional CLC/ sCNTFR complex through monocistronic expression of a CLC/sCNTFR fusion protein (hereafter noted as CC–FP). H-IL-6 is composed of a soluble form of the IL-6 receptor (IL-6R) connected to the mature IL-6 protein via a flexible polypeptide consisting of the glycine/serine linker (G 4 S) 2 . The first 16 N -terminal amino acids of IL-6 are nonhelical and therefore presumably flexible, thus contributing to the connecting loop. As CNTFR share a high level of s equence homology with IL-6R and CLC shares significant structural homology with IL-6 [15,16], we hypothesized that sCNTFR connected to CLC in a similar f ashion would also b e functional. In contrast to the sIL-6R portion of H-IL-6, the N-terminal signal peptide and Ig-like domain of t he sCNTFR precursor protein were maintained in the fusion protein with CLC in order to allow f or the secretion of the protein in mammalian cells. A c-myc epitope tag was introduced at the C-terminus of the f usion p rotein for detection and purification purposes (Fig. 1A). HEK 293 cells were stably transfected with an expres- sion vector encoding the fusion protein. CC–FP was purified from the culture supernatant to apparent homo- geneity with an a nti-(c-myc) Ig affinity column (Fig. 1B, left panel). SDS/PAGE analysis r evealed a single band displaying an ap parent m olecular weight o f 85 kDa, which was quantified based on known concentrations of BSA run on neighboring lanes. W estern blotting was e mployed in order to formally demonstrate that the detected single chain fusion protein effectively was the observed protein (Fig. 1B, right panel). To further check the integrity and folding of the protein, CC–FP was immunoprecipitated with several monoclonal anti-CNTFR Ig, submitted to SDS/PAGE and Western blotted with an anti-(c-myc) Ig. As presented in Fig. 2A, the mAbs used were able to recognize the protein, indicating that the purified CC–FP was correctly folded. 1934 C. Guillet et al. (Eur. J. Biochem. 269) Ó FEBS 2002 To determine whether the protein could self associate to generate dimers, CC–FP was submitted to a gel filtration size exclusion column. Fractions were then studied by western-blotting using an anti-(c-myc) I g. The large major- ity of CC–FP appeared in fraction 17 corresponding to a molecular mass of 60–75 kDa (Fig. 2B). This result indi- cates that CC–FP preferentially stays a s a monomer in solution. We then tried to determine a molecular model of CC–FP according to the model of both proteins. The immunoglob- ulin-like domain and the two cytokine b inding domains of CNTFR are represented at the N-terminal side, followed by a loop containing the glycine linker and, at the C-terminal end, the fou r helices of CLC (Fig. 2C). In this model, the putative sites of interaction o n CNTFR and CLC are highlighted in green and red, r espectively. The f unctional properties of CC–FP were tested in proliferation a ssays using derivatives of the IL-3-dependent Ba/F3 cell line, rendered responsive to cytokines of t he IL-6 family by transfection o f cDNAs e ncoding the appropriate receptor chains [48]. The CC–FP complex induced a robust proliferation of Ba/F3 ce lls expressing LIFR and gp130 (BAF GL), whereas the response observed in the presence of the CLC/sCNTFR composite cytokine was 10-fold weaker (Fig. 3A). T he specific activities were 5 · 10 6 UÆmg )1 and 5 · 10 5 UÆmg )1 , respectively. Higher specific activities were observed when using the B AF GLC cells as test cell line (Fig. 3B). This is likely to reflect the fact that membrane bound CNTFR is more potent in mediating CLC signaling than its soluble counterpart [34]. Similar experiments were carried out on Ba/F3 cells expressing only gp130, or gp130 and IL-6R on their surface, with no detectable signal being measured in response to either the CLC/sCNTFR composite cytokine, or CC–FP (Fig. 3C, and data not shown). This demonstrates an absolute requirement for the LIFR subunit to mediate the signaling response to CLC/sCNTFR or to CC–FP. The involvement of gp130 in CC–FP signaling was further confirmed by the inhibition of the BAF GLC proliferative response following the addition to the culture of a neutral- izing monoclonal anti-gp130 Ig (Fig. 3D). We then tested the fusion protein in a second functional assay using the TF1 cell line. This is a human erythroleu- kemia cell line known to coexpress LIFR and gp130 on its membrane, but not CNTFR [23]. T he experiments per- formed revealed a proliferative response to CC–FP similar to that observed with LIF (Fig. 4A). Surprisingly, and i n a reproducible manner, we failed to detect any functional effect of the CLC/sCNTFR composite cytokine on TF1 cells. This was interpreted a s a weaker sensitivity of the gp130–LIFR pathway in the TF1 cell line compared to other LIF-sensitive c ell lines. In agreement with t his observation, the LIF spe cific activity displayed in TF1 assay was 1.5 · 10 6 UÆmg )1 whereas it is known to reach 4 · 10 7 UÆmg )1 using mouse DA1.a cells [49]. T he CC–FP fusion protein displayed a specific activity of 1 · 10 6 UÆmg )1 in the TF1 erythroleukemia assay. The KB carcinoma cell line, which expresses gp130 a nd LIFR, h as been well characterized for its ability to produce IL-6 in response to cytokines signaling via gp130 and LIFR, such as LIF or OSM [32,50]. We therefore use d KB cells to further characterize the functional activity of CC–FP. In accordance with the results obtained in experiments using transfected Ba/F3 cells and TF1 cells, w e found CC–FP to be more potent than CLC/sCNTFR in inducing IL-6 production in KB cells (Fig. 4B). Cytokines s ignaling t hrough gp130/LIFR can usually compete, at least to some extent, for binding to the same receptor complex [51,52]. We examined whether CC–FP and CNTF could be mutually displaced from the cell membrane. CNTF binding to BAF GLC and SK-N-GP neuroblastoma cells was monitored by flow cytometry u sing a mAb recognizing CNTF (Fig. 5A,B). Increasing concen- trations of CC–FP were added with a fixed amount of CNTF. A dose-dependent competition for receptor binding was observed using CC–FP, whereas no displacement of CNTF binding was observed with either IL-11 or IL-4 in control experiments. These results show that CC–FP and CNTF share binding epitope(s) expressed b y their receptor complexes. We next studied the induction of downstream signaling components by the single chain fusion protein following receptor engagement. The ability of CC–FP to transduce a signal in cells expressing on their s urface the functional L IF receptor co mplex was subseq uently demonstrated. The role of gp130 and LIFR as signaling components for CC–FP was reinforced when analysing their tyrosine phosphoryla- tion following activation by CC–FP in HepG2 hepatoma Fig. 1. Generation of secreted CLC/sCNTFR fusion protein. (A) Schematic representation of the CC–FP fusion protein. A cDNA adaptor e ncoding a glycine/serine linker (GGGGS) 2 was p ositioned between cDNA encoding the sCNTFR protein and CLC. sp, signal peptide. (B) Detection of recombinant CC–FP purified from HEK 293 transfected cells. In the left panel, purified CC–FP was analyzed by SDS/PAGE and silve r staining of th e gel using a BSA protein stan- dard. In the right panel the presence of c-myc epitope-tagged proteins (CLC/sCNTFR and CC–FP) was detected by Western blotting using the anti-(c-myc) Ig. Ó FEBS 2002 Bioactive CLC/sCNTFR fusion protein (Eur. J. Biochem. 269) 1935 Fig. 2. Biochemical and structural characterization o f the CLC/sCNTFR fusion protein. (A) P urified CC–FP was immunoprecipitated w ith anti- CNTFR I g (AN-B2, AN-C2), anti-(c-myc) Ig or control mAb as indicated, submitted to SDS/PAGE and Western blotting as described in F ig. 1 . (B) CC–FP was submitted to a Superose 12 size exclusion column. Fractions were analysed by Western blotting using an anti-(c-myc) Ig. Column calibration was performed using standard purified proteins. (C) Ribbon model of CC–FP: the immunoglobulin-like domain and the two cytokine binding domains of CN TFR are link ed to the four h elices of CLC by a loop containing the glycine linker (cyan). The putative binding sites of CNTFR and CLC a re highlighted in gr een and red, resp ectively. Fig. 3. The CLC/sCNTFR fusion protein induces the proliferation of transfected Ba/F3 cells. BAF GL (A) and BAF GLC (B) cells were cultured in the presence of s erial dilutions of indicated re combinant cytokines. Proliferation was measured by [ 3 H]thymidine incorporation and experiments were performed in triplicate. Error bars represent the SEM. (C) BAF gp130/IL-6R cells were cultured in the presence of serial dilutions of recombinant IL-6, CLC/sCNTFR, CC–FP or IL-2, as c ontro l. (D) Transfected Ba/F3 cells were incub ated in triplicate in culture medium alone (marked as 0), o r containing 1 n gÆmL )1 CNTF or CC–FP. T he AN-HH1 Ig (black bars) or a control IgG2a Ig (grey bars) was added at a final concentration of 30 lgÆmL )1 . 1936 C. Guillet et al. (Eur. J. Biochem. 269) Ó FEBS 2002 and i n SK-N-GP neuroblastoma cells. In response to CC–FP, CLC/sCNTFR and LIF, a clear induction of tyrosine phosphorylation was detected for gp130 and LIFR (Fig. 6). A similar result was also observed when analyzing the activation levels of STAT3 in response t o CC–FP and CNTF (Fig. 7A). The transcriptional activity o f STAT3 was measured in the KB ce ll line, w hich can e asily be transfected in a transient manner. For this, cells were transfected with a reporter construct containing two STAT3 consensus binding sites located upstream of a thymidine kinase minimal p romoter [53]. Two days after transfection, cells were serum starved and stimulated for an additional 15-h with 20 ngÆmL )1 of the indicated cytokines. Whereas the CLC/sCNTFR composite cytokine very weakly stimu- lated STAT3 transcriptional activity, a twofold increase was observed with both CC–FP and LIF (Fig. 7B). These results indicate that CC–FP recruits STAT3 to a similar extent than LIF, for both s ignaling and transcriptional activation of target genes. It has been p reviously reported that t he PI3-kinase/Akt pathway could regulate gp130 signaling [34]. PI3-kinase pathway recruitment by CC–FP led to a marked increase in the tyrosine phosphorylation content of AKT (Fig. 8). Comparable results were obtained when treating the cells with CNTF or CLC/ sCNTFR. In addition to the PI3- kinase/AKT and STAT3 activation pathways, LIFR/gp130 signaling is also kn own to implic ate the GRAB2/Sos adaptators and regulate the MAP kinase pathway [54–56]. ERK1 and ERK2, involved in the MAP kinase pathway, have been shown to play important roles in mediating the Fig. 4. Proliferation of the TF1 cells and induction of IL-6 production in KB cells by CC–FP. (A) TF1 cells were cultured in the presence of serial dilutions o f p urified recombinant human LIF, CLC/sCNTFR, CC–FP, or IL-2, as a control. Proliferation was measured by [ 3 H]thymidine incorporation and e xperiments were performe d i n triplicate. (B) KB cells were exposed to serial dilutions of CLC/ sCNTFR, CC–FP, LIF or IL-2 as control. After a 48-h culture period, the supernatants were analyzed by ELISA for their IL-6 content. Experiments were performed in triplicate. Fig. 5. CC–FP and CNTF compete for receptor complex binding. BAF GLC and SK-N-GP cells were incubated with 2 ng CNTF and increasing concentrations of CC–FP, IL-11 or IL-2. Detection of CNTF binding was monitored by m easuring the mean fluorescence by flow cytometry using an a nti-CNTF Ig. Ó FEBS 2002 Bioactive CLC/sCNTFR fusion protein (Eur. J. Biochem. 269) 1937 mitogenic effects of IL-6 family members. ERK1 and ERK2 activation was determined b y measuring their tyrosine phosphorylation levels. Stimulation o f the SK-N- GP neuroblastoma cell line with C C–FP quickly increased basal values (Fig. 8). These results demonstrate the involve- ment of the PI3-kinase/AKT and MAP kinase signaling pathways in functional responses to the CC–FP fusion cytokine. DISCUSSION We have demonstrated that the fusion of CLC to the C-terminus of sCNTFR via a flexible linker leads to the generation of a bioactive fusion protein. Whereas CLC is inefficiently secreted when expressed in the absence of CLF or CNTFR [32,35], CC–FP i s e fficiently e xpressed a nd secreted in mammalian cells. Similar a pproaches have been successfully used to gener- ate a number of composite cytokines. The first described example r eported the generation of a protein consisting of IL-3fusedtoGM-CSF,whichdisplayedanincreased activity when compared to the respective i ndividual cytokin- es [57]. The discovery of the composite n ature of IL-12, encompassing a cytokine-like component (p35) associated to Fig. 8. Analysis of AKT and ERK1, ERK2 tyrosine phosphorylation induced by CC–FP. SK-N-GP cells were incubated either with or without 5 0 ngÆmL )1 of CNTF, CLC/sCNTFR and CC–FP for 10 m in. After l ysis in 1% N P-40, lysate s were s ubjected to i mmun oblot analysis with antibodies specific for ac tivated AKT (AKT-P) or recognizing activated forms o f ERK1 and ERK2 (ERK1-P and ERK2-P). Fig. 7. Analysis of STAT3 tyrosine phosphorylation and transcriptional activation induced by CC–FP. (A) CC–FP induces STAT3 tyrosine phosphorylation in SK-N-GP n euroblastoma and HepG2 cells. Fol- lowing a 10-min exposure to either NaCl/P i (marked as 0), CNTF (50 n gÆmL )1 ), CLC/sCNTFR (50 ngÆmL )1 ), or p urified CC –FP (50 n gÆmL )1 ), cells were lysed a nd subject to Western blotting using an anti-(STAT3-P) mAb. (B) Effect of CC–FP stimulation on STAT3 transcriptional activity. KB carcinoma cells were transiently trans- fected with a reporter plasmid gene (SIEM-luciferase). 48 h later, cells were treated with 20 ng ÆmL )1 of LIF, CLC/sCN TFR, CC–FP or IL-2 as a control, for an additional 18 h. Cellular extracts were prepared and used t o directly measure luciferase activity. Fig. 6. Analysis of gp130 and LIFR t yrosine phosphorylation i nduced by CC–FP. CC–FP induces gp130 and LIFR t yrosine phospho- rylation in SK-N-GP neuroblastoma and in HepG2 cells. Following a 10-min exposure to either NaCl/P i (marked as 0), LIF (50 ngÆmL )1 ), CLC/sCNTFR (50 ngÆmL )1 ), or purified CC–FP (50 ng ÆmL )1 ), cells were lysed and subject to immunoprecipitation (IP) using an anti-LIFR Ig an d Western blotting (WB). 1938 C. Guillet et al. (Eur. J. Biochem. 269) Ó FEBS 2002 a soluble receptor-like subunit (p40), opened the possibility of fusing the two components, or even adding an immuno- globulin portion to fused IL-12 to reinforce the targeting of the cytokine towards a defined cell type [41,58,59]. Designed IL-12 fusion proteins do not display any increase in their specific activity, when compared to the wild type protein. This is in part explained by the fact that p35 and p40 components are already covalently a ssociated through a disulfide bridge leading to a stable association. Many examples of cytokine receptors existing in soluble form in vivo have been reported. Almost all o f these solu ble receptors are able to interfere with the activity of their ligands. An interesting feature of the a receptor components belonging to the IL-6 family is their ability t o promote the action of their ligands. These have been described in detail for IL-6/IL-6R, CNTF/CNTFR and IL-11/IL-11R [23,36,37]. For this cytokine family the ligand–receptor interaction i s m ainly governed b y t he dissociation rate, suggesting t hat the average half-life of the cytokine–solub le receptor complex may be shorter than the time required to recruit the larger signaling receptors, gp130, LIFR or OSMR. Accordingly, fused proteins interacting with their cognate receptor show a lower off-rate leading to a stronger recruitment of the signaling subunits. As the CC–FP fusion protein represents an irreversibly bound and therefore stabilized derivative of its respective complex, it is noteworthy that CC–FP displays enhanced biological activity relativ e to CLC/sCNTFR. On cells expressing gp130 and LIFR, t he CC–FP fusion protein was shown to be 10- to 100-fold more poten t w hen compared to the unlinked composite cytokine. This result corroborates previous studies showing that single-chain fusion proteins between cytokines and their nonsign aling binding receptors exhibit enhanced functional activity with respect to their native cytokine/receptor complexes [39,60–63]. In the p resent study we analyzed the proliferative potential of CC–FP in cells displaying a hematopoietic background (TF1 and Ba/F3 cells). CC–FP also activates these cells by increasing their t ranscriptional m achinery leading to specific protein synthesis, such as acute phase protein synthesis for hepatocytes [64], or IL-6 production in the case of the KB e pidermoid carcinoma [50]. Collectively, these results support t he idea that CC–FP should be able to substitute for LIF in a large number of situations. It is worth underlining t he synergistic potential of gp130 activating cytokines together with SCF, GM-CSF and erythropoie- tin in increasing the maintenance and proliferation of CD34+CD38– or CD34+Thy1+ hematopoietic stem cells in vitro [65,66]. Therefore, the involvement of CC–FP in hematopoietic stem cell expansion is currently under investigation. CNTF promotes the differentiation and s urvival of a wide range of cell t ypes in the nervous system [1–6]. We can therefore assume t hat composite C LC-containing cytokines will display overlapping functions. A lthough CLC uses th e same functional receptor as CNTF, it differs from the latter in that it is apparently naturally secreted under nontrau- matic conditions [32,35]. Recent studies reported the possibility of expanding human central nervous system stem cells by in v itro growth [67,68]. D eveloped cultures can continuously propagate a heterogeneous population of early neural stem and/or progenitor cells. E xperiments have been carried out with neurosphere c ultures, requiring a cocktail of cytokines. Among them, L IF was shown to p lay an important role for correct culture d evelopment. The availability of a CLC/sCNTFR fusion protein using the LIF signaling receptor complex should b e of g reat interest in this context. Due to its neuroprotective effects, much investigation has been made in to the potential utility o f CNTF i n the treatment of neurodegenerative disorders such as a myo- trophic lateral sclerosis and Huntington’s disease. Indeed, promising preclinical studies have led to clinical trials with varying success [69–72]. The high level of toxicity upon systemic injection of the prote in has led to t he targeted administration of CNTF to the CNS via i ntrathecal implantation of encapsulated transfected cells. For this reason, similar studies should be performed to determine the ability of CLC/sCNTFR to convey neuroprotective effects whilst assessing its toxicity. The availability of a monomeric bioactive CC–FP fusion protein should t herefore allow the production of sufficient purified protein and facilitate the generation of stable transfected cell lines and vectors for alternative gene therapy approaches. ACKNOWLEDGEMENTS He ` le ` ne Plun-Favreau and Catherine Guillet were funded by grants from the city of Angers and the Departement du Maine et Loire, respectively. The project w as supported b y a grant from the Association Franc¸ aise contre les Myopathies. REFERENCES 1. Stockli, K.A., Lottspeich, F., Sendtner, M., Masiakowski, P., Carroll, P., Gotz, R., Lindholm, D. & Thoenen, H . (1989) Molecular cloning, expression and regional distribution of rat ciliary neurotrophic factor. Nature 342, 9 20–923. 2. Lin, L.F., Mismer, D., Lile, J.D., Armes, L.G., B utler, E.T., 3rd, Vannice, J.L. & Collins, F. (1989) Purification, cloning, and expression of ciliary neurotrophic factor (CNTF). Science 246, 1023–1025. 3. Simon, R., Thier, M ., Kruttgen, A., Rose-John, S., W eiergrabe r, O., Heinrich, P.C., Schroder, J.M. & Weis, J. (1995) Human CNTF and related cy tokines: effe cts on DRG n eurone surviv al. Neuroreport 7, 153–157. 4. Oppenheim, R.W., Prevette, D., Yin, Q.W., Collins, F. & MacDonald, J. (1991) Control of embryonic motoneuron survival in vivo by ciliary neurotrophic factor . Science 251, 1616–1618. 5. Larkfors, L., Lindsay, R.M. & Alderson, R.F. (1994) Ciliary neurotrophic factor enhances the survival of Purkinje cells in vitro. Eur. J. Neurosci. 6, 1015–1025. 6. Ip, N.Y., Li, Y.P., van de Stadt, I., Panayotatos, N ., Alderson, R.F. & Lindsay, R.M. (1991) Ciliary neurotrophic factor enhances neuronal s urvival in emb ryonic rat h ippocamp al cultures. J. Neurosci. 11, 3 124–3134. 7. Sendtner, M., Gotz, R., Holtmann, B. & Thoenen, H. (1997) Endogenous ciliary neurotroph ic factor is a lesion factor for axotomized motoneurons in adult mice. J. Neurosci. 17 , 6999–7006. 8. Sendtner, M., Schmalbruch, H., Stockli, K.A., Carroll, P., Kreutzberg, G.W. & Thoenen, H. (1992) Ciliary neurotrophic factor preve nts degeneration o f motor neurons in mouse mutant progressive motor neuronopathy. Nature 358, 502–504. 9. Mitsumoto, H., Ikeda, K., Klinkosz, B., C e darbaum, J.M., Wong, V. & Lindsay, R.M. (1994) Arrest of motor neuron disease in wobbler mice cotreated with CNTF and BDNF. Science 265 , 1107–1110. Ó FEBS 2002 Bioactive CLC/sCNTFR fusion protein (Eur. J. Biochem. 269) 1939 10. Helgren, M.E., Squinto, S.P., Davis, H.L., Parry, D.J., Boulton, T.G., Heck, C.S., Zhu, Y., Yancopoulos, G.D., Lindsay, R.M. & DiStefano, P.S. (1994) T rophic effec t of ciliary n eurotrophic facto r on denervated skeletal muscle. Cell 76, 493–504. 11. Guillet, C., Auguste, P., Mayo, W., Kreher, P. & Gascan, H. (1999) Ciliary neurotrophic factor is a regulator of muscular strength in aging. J. Neurosci. 19, 1257–1262. 12. Heinrich, P.C., Behrmann, I., Muller-Newen, G., Schaper, F. & Graeve, L. (1998) Interleukin-6-type cytokine signalling through the gp130/Jak/STAT pathway. Biochem. J. 334 , 297–314. 13. Bravo, J. & Heath, J.K. (2000) Receptor recognition by gp130 cytokines. EMBO J. 19, 2399–2411. 14. Bazan, J.F. (1991) Neuropoietic cytokines in the hematopoietic fold. Neuron 7, 197–208. 15. Shi, Y., Wang, W., Yourey, P.A., Gohari, S., Zukauskas, D., Zhang, J., Ruben, S. & Alderson, R.F. (1999) Computational EST database analysis identifies a novel member of the neuro- poietic cytokine family. Biochem. Biophys. Res. Commun. 262, 132–138. 16. Senaldi,G.,Varnum,B.C.,Sarmiento,U.,Starnes,C.,Lile,J., Scully, S., Guo, J., E lliott, G ., McNinch, J., S haklee, C.L. et al. (1999) Novel neurotrophin-1/B cell-stimulating factor-3: a cytokine of the IL-6 family. Proc. Natl Acad. Sci. USA 96, 11458–11463. 17. Hibi, M., Murakami, M., Saito, M., Hirano, T., Taga, T. & Kishimoto, T. (1990) Molecular cloning and expression of an IL-6 signal transducer, gp130. Cell 63, 1149–1157. 18. Gearing, D.P., T hut, C.J., VandeBos, T., Gimpel, S.D., Delaney,P.B.,King,J.,Price,V.,Cosman,D.&Beckmann, M.P. (1991) Leuk emia inhibitory factor receptor is structur ally related to the IL-6 signal transducer, g p130. EMBO J. 10, 2839–2848. 19. Gearing , D.P., Comeau, M.R., F riend, D.J., Gimpel, S.D., Thut, C.J., McGourty, J., Brasher, K.K., King, J.A., Gillis, S., Mosley, B. et al. (1992) The IL-6 signal transducer, gp130: an oncostatin M receptor and affinity con verter for the LIF receptor. Science 255, 1 434–1437. 20. Murakami, M ., Hibi, M., Nakagawa, N., Nakagawa, T., Yasukawa,K.,Yamanishi,K.,Taga,T.&Kishimoto,T.(1993) IL-6-induced homodimerization of gp130 and associated a ctiva- tion of a tyrosine kinase. Science 260 , 1808–1810. 21. Davis, S., A ldrich, T.H., Valenzuela, D.M., Wong, V.V., Furth, M.E., Squinto, S.P. & Yancopoulos, G.D. (1991) The r eceptor for ciliary neurotrophic factor. Science 253, 59–63. 22. Davis, S., Aldrich, T.H., Stahl, N., Pan, L., Taga, T., Kishimoto, T., Ip, N.Y. & Yancopoulos, G.D. (1993) LIFR beta and gp130 as heterodimerizing signal transducers of the tripartite CNTF receptor. Science 260, 1805–1808. 23. Davis, S., Aldrich, T.H., Ip, N.Y., Stahl, N., Scherer, S., Farruggella, T., DiSt efano, P.S., C urtis, R., Panayotatos, N., Gascan, H. et al. (1993) Released form of CNT F receptor alpha component as a soluble mediator of CNTF responses. Science 259, 1736–1739. 24. Baumann, H., Ziegler, S.F., Mosley, B ., Morella, K.K., Pajovic, S. & Gearing, D .P. ( 1993) Re constitut ion of the response to l euke mia inhibitory factor, oncostatin M, and ciliary neurotrophic factor in hepatoma cells. J. Biol. Chem. 268, 8 414–8417. 25. Gearing, D.P., Ziegler, S.F., Comeau, M.R., Friend, D ., Thoma, B., Cosman, D., Park, L. & Mosley, B. (1994) Proliferative responses and binding propertie s of hemato poietic cells trans- fected with low-affinity receptors for leukemia inhibitory factor, oncostatin M, and ciliary neurotrophic factor. Proc. N atl Acad. Sci. USA 91, 1 119–1123. 26. Stahl, N., Boulton, T.G., Farruggella, T., Ip, N.Y., Davis, S., Witthuhn, B.A., Quelle, F.W., Silvennoinen, O., Barbieri, G., Pellegrini, S. et al. (1994) Association and activation of Jak-Tyk kinases by C NTF-LIF-OSM-IL-6 beta receptor components. Science 263, 9 2–95. 27. Stahl, N., F arruggella, T.J., Boulton, T.G., Zhong, Z., Darnell, J.E. Jr & Yancopoulos, G.D. (1995) Choice o f STATs and other substrates specified by modular tyrosine-based motifs in cytokine receptors. Science 267, 1349–1353. 28. Lutticke n, C., Wegenka, U.M., Yuan, J., Buschmann, J., Schindler, C., Ziemiecki, A ., Harpur, A.G., Wilks, A.F., Yasuk- awa, K., Taga, T. et al. (1994) Association of transcription factor APRF and protein k inase Jak1 wit h the interleukin-6 signal transducer gp130. Science 263 , 89–92. 29. Narazaki, M., Witthuhn, B.A., Yoshida, K., Silvennoinen, O., Yasukawa, K., I hle, J .N., Kishimoto, T. & Taga, T. (1994) Activation of JAK2 kinase mediated by the interleukin 6 signal transducer gp130. Proc.NatlAcad.Sci.USA91, 2285– 2289. 30. Guschin, D., Rogers, N ., Briscoe, J., Witth uhn , B ., Watling, D., Horn, F., Pellegrini, S., Y asukawa, K., Heinrich, P., Stark, G.R. et a l. (1995) A major role for the protein tyrosine kinase JAK1 in the JAK/STAT signal transduction pathway in response t o in- terleukin-6. EMBO J. 14 , 1421–1429. 31. Elson, G.C., Graber, P., Losberger, C., Herren, S., Gretener, D., Menoud, L.N., Wells, T .N., Kosco-V ilbois, M.H. & Gauchat, J.F. (1998) Cytokine-like factor-1, a n ovel soluble protein, shares homology with members of the cytokine type I receptor family. J. Immunol. 161, 1 371–1379. 32. Elson, G.C., Lelievre, E., Guillet, C., Chevalier, S., Plun-Favreau, H.,Froger,J.,Suard,I.,deCoignac,A.B.,Delneste,Y.,Bonne- foy, J.Y., Gauchat, J.F. & Gascan, H. (2000) CLF associates with CLC to form a functional heteromeric ligand for the CNTF receptor complex . Nat. Neurosci. 3, 867–872. 33. Trinchieri, G . (1995) Interleukin-12: a proinflammatory cytokine with immunoregulatory functions that bridge innate resistance and antigen-specific adaptive i mmunity. Annu . Rev. Immunol. 13, 251–276. 34. Lelievre, E., P lun-Favreau, H., C hevalier, S., Froger, J., Guillet, C., Elson, G.C., Gauchat, J.F. & Gascan, H. (2001) Signaling pathways recruited b y t he c ardiotrop hin-like c ytokine/cytokine - like f actor-1 co mposite cytokine. s pecific re quirement of the membrane-bound form of ciliary neurotrophic factor receptor alpha component. J. Biol. Chem. 276, 22476–22484. 35. Plun-Favreau , H., E lson, G ., C habbe rt, M ., F roger, J ., deLapeyriere, O., Lelievre, E., Guillet, C., Hermann, J., Gauchat, J.F., G ascan, H. & C hevalier, S. (2001) T he ciliary neurotrophic factor receptor alp ha c omp onent in duces the secretion of and is required for fun ctional responses to c ardio trophin-like c ytokine. EMBO J. 20 , 1692–1703. 36. Peters, M., Muller, A.M. & Rose-John, S. (1998) Interleukin-6 and soluble interleukin-6 receptor: direct stimulation of gp130 and hematopoiesis. Blood 92 , 3495–3504. 37. Karow, J., H udson, K.R., Hall, M.A., Vernallis, A.B., Taylor, J.A., G ossler, A. & Heath, J.K. (1996) Mediation of interleukin- 11-dependent biological responses by a s oluble form of the inte r- leukin-11 r eceptor. Biochem. J. 318, 489–495. 38. Oppmann,B.,Lesley,R.,Blom,B.,Timans,J.C.,Xu,Y.,Hunte, B. & Vega, F., Yu, N., Wang, J., Singh, K. et al. (2000) Novel p19 protein engages IL-12p40 t o form a cytokine, IL-23, with biolo- gical activities similar as well as distinct from IL-12. Immunity 13, 715–725. 39. Fischer, M., Goldschmitt, J., Peschel, C., Brakenhoff, J.P., Kallen, K.J., Wollmer, A., Grotzing er, J. & R ose-John, S . (1997 ) I. A bioactive designer c yt okine for human hematopoietic progenit or cell ex pansion. Na t. Biotechnol. 15, 1 42–145. 40. Pflanz, S., T acken, I., G rotzinger, J., Jacques, Y., Minvielle, S., Dahmen, H., He inrich, P.C. & Muller-New en, G. ( 1999) A fusion protein of interleukin-11 and soluble interleukin-11 receptor acts 1940 C. Guillet et al. (Eur. J. Biochem. 269) Ó FEBS 2002 as a superagonist on cells expressing gp130. FEBS Lett. 450, 117– 122. 41. Lieschke, G.J., Rao, P.K., Gately, M.K. & Mulligan, R.C. (1997) Bioactive murine and human interleukin-12 fusion proteins which retain antitumor a ctivity in vivo. Nat. Bio technol. 15, 35–40. 42. McDonald, N .Q., P anayotatos, N. & H endrickson, W.A . (1995 ) Crystal structure of dimeric human ciliary neurotrophic factor determined by MAD phasing. EMB O J. 14, 2 689–2699. 43. Robinson , R.C., Grey, L.M., Staunton, D., Vankelecom, H., Vernallis, A.B., Moreau, J.F., Stuar t, D.I., Heath, J.K. & Jones, E.Y. (1994) The crystal structure and biological f unction o f leu- kemia inhibitory factor: implications for receptor binding. Cell 77, 1101–1116. 44. Bravo, J., Staunton, D., Heath, J.K. & Jones, E.Y. (1998) Crystal structure of a cytokine-binding region of gp 130. EMBO J. 17, 1665–1674. 45. Chow,D.,He,X.,Snow,A.L.,Rose-John,S.&Garcia,K.C. (2001) Structure of an extracellular gp130 cytokine rec eptor sig- naling complex. Sc ience 291, 2150–2155. 46. Sali, A. & B lundell, T.L. (1993) Comparative protein modelling by satisfaction of spatial r estraints. J. Mol. Biol. 234 , 779–815. 47. Luthy, R., Bowie, J.U. & Eisenberg, D. (1992) Assessment of protein models with three-dimensional profiles. Natu re 356, 8 3–85. 48. Kallen, K.J., Grotzinger, J., Lelievre, E., Vollmer, P., Aasland, D., Renne, C., M ullbe rg, J., M ye r zum Buschenfelde, K.H., Gascan , H. & Rose-John, S. ( 1999) Receptor re cognitio n sites of cytokines are organized as exchangeable modules. Transfer of the l eukemia inhibitory factor receptor-binding site from ciliary neurotrophic factor to interleukin-6. J. Biol. Chem. 274, 11859–11867. 49. Gascan, H., Godard, A., Ferenz, C., Naulet, J., Praloran, V., Peyrat, M.A., Hewick, R., Jacques, Y., Moreau, J.F. & Soulillou, J.P. (1989) Characterization and NH 2 -terminal amino a cid sequence of natural human interleukin for DA cells: leukemia inhibitory factor. Differentiation i nhibitory activity s ecreted by a T lymphoma cell line. J. Biol. Chem. 26 4 , 21509–21515. 50. Thoma,B.,Bird,T.A.,Friend,D.J.,Gearing,D.P.&Dower,S.K. (1994) Onc ostatin M and leukemia inhibitory f actor trigger overlapping and different signals through partially shared receptor complexes. J. Biol. Chem. 269 , 6215–6222. 51. Robledo, O., Auguste, P., Coupey, L., Praloran, V., Chevalier, S., Pouplard, A. & Gascan, H . (1996) B inding interactions of leuke- mia inhibitory factor and ciliary neurotrophic factor with the different subunits of their high affinity receptors. J Neurochem. 66, 1391–1399. 52. Robledo, O., Fourcin, M., Chevalier, S., Guillet, C., Auguste, P., Pouplard-Barthelaix, A., P ennica, D. & Gascan, H. (1997) Sig- naling of the cardiotrophin-1 receptor. E vidence for a third receptor component . J. Biol. C hem. 272, 4855–4863. 53. Coqueret, O. & Gascan, H. (2000) Functional interaction of STAT3 tran scription factor with the cell c ycle in hibitor p21WAF1/CIP1/SDI1. J. Biol. Chem. 275 , 18794–18800. 54. Boulton, T.G., Stahl, N. & Yancopoulos, G.D. (1994) Ciliary neurotrophic factor/leukemia inhibitory factor/interleukin 6/ oncostatin M family of c ytokines induces tyro sin e phosphorylation of a common set of proteins ov erlapping those induced by other cytokines and growth factors. J. Biol. Chem. 269, 11648–11655. 55. Kim, H. & Baumann, H. (1999) Dual signaling role of t he protein tyrosine phosphatase SHP-2 in regulating expression of acute- phase plasma proteins b y interleuk in-6 cytokin e receptors in hepatic c ells. Mol. Cell. Biol. 19, 5326–5338. 56. Takahashi-Tezuka, M., Yoshida, Y., Fukada, T., Ohtani, T., Yamanaka, Y., Nishida, K., Nakajima, K., Hibi, M. & Hirano, T. (1998) Gab1 acts as an adapter molecule linking the cytokine receptor gp130 to E RK mitogen-activated protein kinase. Mol. Cell. Biol. 18 , 4109–4117. 57. Curtis,B.M.,Williams,D.E.,Broxmeyer,H.E.,Dunn,J.,Farrah, T., Jeffery, E., Cle venger, W., de Roos, P., Martin, U., Friend, D. et al. (1991) Enhanced hematopoietic activity of a human g ranu- locyte/macrophage colony-stimulating factor-interleukin 3 f usion protein. Proc. Natl Acad. Sci. USA 88, 5809–5813. 58. Lode, H.N., Dreier, T., Xiang, R., Varki, N.M., Kang, A .S. & Reisfeld, R.A. (1998) Gene therapy with a single chain interleukin 12 fu sion protein induces T cell-dependent protective im munity in a syngeneic model of murine ne uroblastoma. Proc.NatlAcad.Sci. USA 95, 2475–2480. 59. Peng, L.S., Penichet, M .L. & Morrison, S.L. (1999) A single-chain IL-12 IgG3 a ntib ody f usion p ro tein retains antibody specificity and IL-12 bioac tivity and demonstrates antitumor activity. J. Immunol. 16 3 , 250–258. 60. Oh,J.W.,VanWagoner,N.J.,Rose-John,S.&Benveniste,E.N. (1998) Role of IL-6 and the soluble IL-6 recept or in inhibition of VCAM-1 gene expression. J. Immunol. 161, 4992–4999. 61. Peters, M., Blinn , G., Solem , F., Fischer, M., Meyer zum Bus- chenfelde, K.H. & Rose-John, S . (1998) In vivo and in vi tr o activities of the gp130-stimulating designer cytokine Hyper-IL-6. J. Immunol. 16 1 , 3575–3581. 62. Rakemann, T., N iehof, M., Kubicka, S., Fischer, M., Manns, M .P., Rose-John, S. & Trautwein, C. (1999) The designer cytokine hyper-interleukin-6 is a potent a ct ivator of STAT3-dependent gene transcription in vivo and in vitro. J. Biol. Chem. 274, 1257– 1266. 63. Igaz, P., Horvath, A., Horvath, B., Szalai, C., P allinger, E., Rajnavolgyi, E., Toth, S., Rose-John, S . & Falus, A. (2000) Soluble interleuk in-6 rec eptor (sIL -6R) makes I L-6R ne gative T cell line respond t o IL-6; it inhibits TNF production. Immunol. Lett. 71, 1 43–148. 64. Lai, C.F., Ripperger, J., Morella, K.K., Wang, Y., Gearing, D.P., Fey, G.H. & Baumann, H. (1995) Separate signaling mechanisms are involved i n t he co ntrol of STAT protein activation and g ene regulation via the interleukin 6 response element by the box 3 motif of g p130. J. Biol. C hem. 270, 14847–14850. 65. Shih, C.C., Hu, M.C., Hu, J., Medeiros, J. & Forman, S.J. (1999) Long-term ex vivo maintenance and expansion of t ransplantable human hematopoietic stem c ells. Blood 94 , 1623–1636. 66. Galy, A.H., Cen, D ., Travis, M., Chen, S. & Chen, B.P. (1995) Delineation of T-progenitor cell acti vity within the CD34+ compartment o f adult bo ne marrow. Blood 85, 2770–2778. 67. Uchida, N., Buck, D.W., He, D., Reitsma, M.J., Masek, M., Phan, T.V., Tsukamoto, A.S., G age, F.H. & Weissman, I.L. (2000) Direct isolation of human c entral nervous system stem ce lls. Proc. Natl Acad. Sci. USA 97 , 14720–14725. 68. McKay, R. (1997) Stem cells in the central nervous system. Science 276, 66–71. 69. Miller, R.G., Petajan, J.H., Bryan, W.W., Armon, C ., Barohn, R.J., Goodpasture, J.C., Hoagland, R.J., Parry, G.J., Ross, M.A. & Stromatt, S.C. (1996) A placeb o-controlled trial of recombinant human c iliary neurotrophic (rh CNTF) factor in am yotrop hic lateral sclerosis. rhCNTF ALS Study Group. Ann. Neurol. 39, 256–260. 70. Penn, R.D., Kroin, J.S., York, M.M. & Cedar baum, J.M. (1997) Intrathecal ciliary neurotrophic factor delivery for treatment of amyotrophic lateral sclerosis (phase I trial). Neurosurgery 40 , 94–100. 71. Aebischer, P., Schluep, M., Deglon, N., Joseph, J.M., Hirt, L., Heyd, B., Goddard, M., Hammang, J.P., Zurn, A.D., Kato, A.C., Regli, F. & Baetge, E.E. (1996) Intrathecal delivery of CNTF using encapsulated genetically modified xenogeneic cells in amyotroph ic lateral sclerosis patients. Nat. Med. 2, 696–699. 72. Mittoux, V., Joseph, J.M., Conde, F., Palfi, S., Dautry, C., Poyot , T., Bloch, J., Deglon, N., Ouary, S., Nimchinsky, E.A., Brouillet, E., Hof, P.R., Peschanski, M., Aebischer, P. & Hantraye, P. (2000) Restoration of cognitive and motor functions b y ciliary neuro- trophic factor in a primate mod el of Hu ntingto n’s disease. Hu m. Gene. Ther. 11, 1177–1187. Ó FEBS 2002 Bioactive CLC/sCNTFR fusion protein (Eur. J. Biochem. 269) 1941 . Functionally active fusion protein of the novel composite cytokine CLC/soluble CNTF receptor Catherine Guillet 1 , Eric Lelie ` vre 1 ,He ´ le ` ne. have demonstrated that the fusion of CLC to the C-terminus of sCNTFR via a flexible linker leads to the generation of a bioactive fusion protein. Whereas CLC