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,

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

´
on III, 74164 St-Julien-en-Genevois, France.
Fax: + 33 450 35 35 90, Tel.: + 33 450 35 35 55,
E-mail:
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

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-

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

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

,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

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

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

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,

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

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.

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

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

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-

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.

(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.,

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)