Differential response of neuronal cells to a fusion protein of ciliary
neurotrophic factor/soluble CNTF-receptor and leukemia inhibitory
factor
Pia Ma¨rz
1,
*, Suat O
¨
zbek
2,
*, Martina Fischer
3
, Nicole Voltz
4
, Uwe Otten
1
and Stefan Rose-John
4,5
1
Department of Physiology, University of Basel, Switzerland;
2
Department of Biophysical Chemistry, Biocenter, University of Basel,
Switzerland;
3
Xerion Pharmaceuticals, Martinsried, Germany;
4
Department of Medicine, Section Pathophysiology, Johannes
Gutenberg University of Mainz, Germany;
5
Department of Biochemistry, Christian Albrechts University of Kiel, Germany
Ciliary neurotrophic factor (CNTF) displays neurotrophic
activities on motor neurons and neural cell populations both

differentiation factor for a variety of neuronal and glial cells.
It has been proposed to act as a lesion factor preventing
motor neuron degeneration after injury [1] and exerting
myotrophic activity on denervated skeletal muscle [2].
CNTF belongs to the IL-6 type family of neuropoietic
cytokines that comprises interleukin-6 (IL-6), interleukin-11
(IL-11), leukemia inhibitory factor (LIF), oncostatin M,
cardiotrophin-1 (CT-1), and novel neurotrophin-1 (NNT-
1)/cardiotrophin-like cytokine (CLC) [3–7]. All IL-6 type
cytokines use a membrane spanning 130-kDa glycoprotein,
gp130, as a signal transducing receptor subunit. The
biological response to CNTF is elicited by formation of a
multimeric receptor complex [8]. CNTF first binds
to a specific glycosyl-phosphatidylinositol-anchored a unit,
CNTF receptor (CNTF-R), which is not involved in
signaling. This is followed by the recruitment of gp130
and LIF receptor (LIF-R) as signal transducing b units,
which in turn form a disulfide-linked heterodimer that
activates the JAK/STAT and the Ras/MAP kinase path-
ways [6,9]. IL-6, CNTF as well as IL-11 and presumably
CT-1 and NNT-1 act via specific membrane receptors which
together with their ligands associate with signal transducing
b subunits thereby initiating cytoplasmic signaling. Cells
that only express signal transducing but no ligand binding
subunits for these cytokines are refractory to stimulation.
An unusual feature of the IL-6 cytokine family is that the
soluble forms of the ligand binding receptor subunits
generated by one cell type in complex with their ligands can
directly stimulate the signal transducing receptor b subunits
on different cell types which lack ligand binding a subunits

cardiotrophin-like cytokine; JAK, Janus kinase; STAT, signal
transducer and activator of transcription; MAPK, mitogen activated
protein kinase; DMEM, Dulbecco’s modified Eagle’s medium.
*Note: these authors contributed equally to this work.
(Received 6 February 2002, revised 25 April 2002, accepted 3 May 2002)
Eur. J. Biochem. 269, 3023–3031 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.02977.x
Superagonistic cytokines have been designed that consist
of covalently linked cytokines and soluble receptors. The
first such molecule was Hyper-IL-6, a fusion protein in
which IL-6 and soluble IL-6-R were connected by a flexible
polypeptide linker. Hyper-IL-6 turned out to be fully active
on cells expressing gp130 at 100–1000 fold lower concen-
trations than unlinked IL-6 and sIL-6R [18]. This approach
has been adopted to obtain a superagonist of IL-11 and sIL-
11R [19].
We generated a CNTF/soluble CNTF-receptor (sCNTF-
R) fusion protein with superagonistic activity on target cells
expressing gp130 and LIF-R, but lacking membrane-bound
CNTF-R. In contrast to the existing cytokine–cytokine
receptor fusion proteins, Hyper-IL-6 and Hyper-IL-11,
which directly activate the ubiquitously expressed gp130
protein, such a protein allows more specificity due to the
restricted expression pattern of the LIF-R. While the effects
of Hyper-CNTF and LIF on BAF3/-gp130/LIF-R cells
were similar, Hyper-CNTF but not LIF induced neuronal
differentiation of rat pheochromocytoma cells (PC12).
These data point to a cell-specific difference in signaling
via the heterodimeric receptor complex of gp130 and LIF-R.
MATERIALS AND METHODS
Chemicals

streptomycin (50 lgÆmL
)1
), and 10% fetal bovine serum
at 5% CO
2
in a water saturated atmosphere. BAF/3-
gp130 cells were cultured in the presence of 10 ngÆmL
)1
Hyper-IL-6, BAF/3-gp130/LIF-R cells with 5 ngÆmL
)1
human LIF. Recombinant human IL-6 and human
CNTF were prepared as described previously [21,22].
The fusion protein hIL-6/hsIL-6R designated Hyper-IL-6
was expressed in the methylotrophic yeast Pichia pastoris
and purified to homogeneity by ion-exchange chromatog-
raphy followed by gelfiltration as described previously
[18,23]. Nerve growth factor (NGF) was isolated [24] with
modifications as described previously [25]. Recombinant
human LIF was expressed as glutathione S-transferase
(GST)-fusion protein, purified by glutathione Sepharose
4B and cleaved from GST by thrombin treatment as
described by the manufacturer (Pharmacia, Freiburg,
Germany). The fusion proteins gp130-Fc and LIF-R-Fc
were transiently expressed in COS-7 cells and purified by
protein A-Sepharose, as described previously [26,27].
Recombinant growth factor concentrations were estimated
using standard protein assays. The polyclonal anti-(phos-
pho-STAT3) Ig and anti-(phospho-p44/42 MAP kinase)
Ig were from New England Biolabs (Schwalbach,
Germany). The monoclonal anti-(CNTF-R) Ig (AN-D3)

S]cysteine
(Tran-
35
S-Label) in methionine/cysteine-free medium for
6 h. For production of Hyper-CNTF protein, transfected
cells were transferred to serum-free medium after 24 h and
supernatants were collected on day 4 post-transfection.
Immunoprecipitation
Metabolically labeled Hyper-CNTF was precipitated from
culture media using 0.5 lgÆmL
)1
gp130-Fc, 0.5 lgÆmL
)1
LIF-R-Fc or 1 lgÆmL
)1
monoclonal anti-CNTF-R Ig
(AN-D3) followed by protein A–Sepharose. Immune
complexes were analyzed by SDS/PAGE [31] and visual-
ized by fluorography using the fluorographic intensifier
solution ÔAmplifyÕ (Amersham International, Aylesbury,
UK).
Proliferation assays
BAF/gp130 and BAF/gp130/LIF-R cells were extensively
washed with NaCl/P
i
, and resuspended in cytokine free
medium at 5 · 10
3
cells per well of a 96-well plate. They
were cultured in a final volume of 100 lL with cytokines as

IgG) Ig was used (Sigma, Deisenhofen, Germany). The blot
was developed using the ECL-detection system (Amersham
International, Aylesbury, UK). The STAT3 and MAPK
phosphorylation assays were reproduced three times with
one representative experiment shown.
RESULTS
Construction of CNTF/sCNTF-R fusion protein
We engineered an expression vector encoding a CNTF/
sCNTF-R fusion protein by linking the C-terminus of
human CNTF-R to the N-terminus of human CNTF
(Fig. 1A). In principle, we followed the design of Hyper-
IL-6 [18] with two specific modifications. First, we included
the N-terminal Ig domain of the sCNTF-R, as deletion of
this region lead to reduced expression levels of recombinant
sCNTF-RDIg protein (P. Ma
¨
rz, M. Fischer & S. Rose-
John, unpublished work). This observation is in line with
recent results indicating that the Ig-like domain of the IL-6R
is important for intracellular transport of IL-6R through the
secretory pathway [32]. Secondly, we avoided the use of a
synthetic polypeptide linker in order to minimize immun-
ogenicity. Instead, the 16 C-terminal amino acids of
CNTF-R (amino acids 331–346) that are not part of the
membrane-proximal cytokine binding domain [33] and the
14 N-terminal nonhelical and presumably flexible amino
acids of CNTF (amino acids 1–12) [34] were linked by one
additional glycine residue. The resulting length of 31 amino
acids, in analogy to Hyper-IL-6 and Hyper-IL-11, is
presumably sufficient to connect both molecules and to

35
S-labeled Hyper-
CNTF protein was incubated with Fc-fusion proteins
containing the extracellular portion of gp130 or the
extracellular portion of LIF-R. Protein complexes were
precipitated with protein A-Sepharose. As can be seen
in Fig. 2B, Hyper-CNTF interacted with gp130-Fc and
LIF-R-Fc to a similar extent.
Fig. 1. Schematic representation of the fusion protein of CNTF and
sCNTF-R. (A) Construction of the fusion protein. The C-terminus of
sCNTF-R was linked via one additional glycine residue (G) to the
N-terminus of CNTF. (B) Schematic model of the Hyper-CNTF ter-
tiary structure. Ig denotes the immunoglobulin-like domain, D2 and
D3 the two cytokine-binding receptor domains.
Ó FEBS 2002 CNTF/sCNTF-R fusion protein with enhanced activity (Eur. J. Biochem. 269) 3025
Biological activity of the Hyper-CNTF fusion protein
To assess the biological activity of Hyper-CNTF, we first
investigated the proliferative response of transfected BAF/3
cells. Murine BAF/3 cells, which normally grow IL-3-
dependently, are known to proliferate in response to various
cytokines after transfection of the corresponding receptor
chains. BAF/3 cells transfected with human gp130 and/or
additional transfection of the human LIF-R were stimulated
with increasing amounts of Hyper-IL-6, Hyper-CNTF, LIF
or medium alone. Proliferation of cells was assayed by
measuring [
3
H]thymidine incorporation into DNA. As
shown in Fig. 3A, BAF/3-gp130 cells proliferate upon
stimulation with Hyper-IL-6, but absence of the LIF-R

CNTF. (A) BAF/3-gp130 cells and (B) BAF/3-gp130/LIF-R cells were
stimulated with increasing amounts of Hyper-CNTF, Hyper-IL-6, LIF
or medium alone. Proliferation of cells was assayed by measuring
[
3
H]thymidine incorporation into DNA. One representative experi-
ment is shown.
3026 P. Ma
¨
rz et al. (Eur. J. Biochem. 269) Ó FEBS 2002
STAT3 and MAPK activation by Hyper-CNTF
in transfected BAF/3 cells
Downstream signal transduction pathways were analyzed
by studying the activation level of JAK/STAT and MAP
kinase signaling components known to be mainly tyrosine
phosphorylated in response to IL-6 type cytokines [36–38].
BAF/3-gp130 cells and BAF/3-gp130/LIF-R cells were
stimulated with medium alone, 10 ngÆmL
)1
Hyper-IL-6,
20 ngÆmL
)1
Hyper-CNTF, 50 ngÆmL
)1
IL-6, 50 ngÆmL
)1
CNTF or 20 ngÆmL
)1
LIF for 10 min (Fig. 4). Cells were
lysed in Laemmli buffer and proteins were separated via

)1
Hyper-CNTF, 100 ngÆmL
)1
LIF, 100 ngÆmL
)1
NGF or 20 ngÆmL
)1
Hyper-IL-6 was analysed. As expected
from earlier studies [39,40], stimulation of the cells with
NGF or Hyper-IL-6 led to robust formation of neurites.
Surprisingly, exposure of the cells to Hyper-CNTF also
induced pronounced neuronal differentiation, whereas LIF
and CNTF (at concentrations up to 500 ngÆmL
)1
, data not
shown) did not result in significant morphological changes
(Fig. 5A). Hyper-CNTF induced neurites extending longer
than twice the diameter of the cell bodies appear within a
day, and maximal response is reached in 2 days. For direct
comparison, the amount of responsiveness was evaluated
forallfactorsat48h.AspresentedinFig.5B,Hyper-
CNTF turned out to be virtually as effective as NGF and
Hyper-IL-6 to elicit neuronal differentiation.
STAT3 and MAPK activation by Hyper-CNTF in PC12 cells
We then asked which signal transduction pathways are
involved in Hyper-CNTF-induced neurite outgrowth. In a
first experiment, PC12 cells were stimulated with medium
alone, Hyper-IL-6, NGF, Hyper-CNTF, or LIF for 10 min
(Fig. 6A). Cells were lysed in Laemmli buffer and cell lysates
Fig. 4. STAT3 and MAPK phosphorylation by Hyper-CNTF in

NGF or 20 ngÆmL
)1
Hyper-IL-6 was analyzed (magni-
fication: 300·) and (B) the extent of responsiveness was evaluated by
analysis of neurite outgrowth. Vertical bars represent S.E.M. (n ¼ 3).
Ó FEBS 2002 CNTF/sCNTF-R fusion protein with enhanced activity (Eur. J. Biochem. 269) 3027
were analyzed for STAT3 and MAPK phosphorylation as
described above. We found that stimulation with Hyper-IL-
6 led to an increase of both, STAT3 and MAPK phos-
phorylation. Consistent with other reports [41,42], a strong
activation of p42/p44 MAP kinases was observed by NGF.
Interestingly, as compared to Hyper-IL-6, treatment of the
cells with Hyper-CNTF resulted in small but significant
STAT3 phosphorylation and strong MAPK phosphoryla-
tion which was at least equal to Hyper-IL-6. In contrast,
stimulation with LIF alone had a similar effect on STAT3
phosphorylation but no effect on MAP kinase activation. As
demonstrated in Fig. 6B, the dose–response phosphoryla-
tion pattern for both STAT3 and p42/p44 MAP kinases
clearly confirmed that of the cytokines signaling through
gp130/LIF-R only Hyper-CNTF but not LIF or CNTF
(even at high concentrations) alone were able to activate the
MAPK pathway. MAP kinases and STAT3 are rapidly
activated within 10 min in response to Hyper-CNTF, the
phase of activation lasting for at least 30 min before
returning to near basal levels within 1 h (Fig. 6C). These
data are in line with the different abilities of Hyper-CNTF,
LIF and CNTF to induce neuronal differentiation in PC12
cells, as observed above. We conclude that the Hyper-
CNTF-induced neurite outgrowth is most likely mediated by

pattern of STAT3 and MAP kinases, mainly p42, were
identical for BAF/3 cells stimulated with Hyper-IL-6,
Hyper-CNTF, and LIF.
Analysis of the biological activity of Hyper-CNTF in
non-neuronal vs. neuronal cells revealed unexpected func-
tional and biochemical differences between LIF and Hyper-
CNTF activity. In contrast to LIF, Hyper-CNTF rapidly
induced neurite outgrowth and formation of a neuronal
network in PC12 cells. Looking at the signaling events, we
observed that both LIF and Hyper-CNTF induced phos-
phorylation of STAT3. However, only Hyper-CNTF has
the potential to activate MAP kinases. This finding is in
agreement with the experiments of Sterneck et al. who failed
to induce neuronal differentiation with CNTF and LIF in
PC12 cells [46,47].
How can the differential response of BAF/3 cells and
PC12 cells be explained? The phenomenon that stimulation
of the gp130/LIF-R complex by different cytokines might
result in different biological responses in neuronal cells has
already been discussed in a review [48]. It is known that
gp130 stimulation leads to the activation of multiple
signaling cascades including the STAT3 and the MAPK
Fig. 6. STAT3 and MAPK activation by Hyper-CNTF in PC12 cells.
PC12 cells were stimulated with medium alone, 20 ngÆmL
)1
Hyper-IL-
6, 100 ngÆmL
)1
NGF, 20 ngÆmL
)1

by Hyper-IL-6 showed a more profound and elongated
response as compared to IL-6 [51]. This was most likely due
to decreased internalization of Hyper-IL-6 as compared to
IL-6. We have also recently described differential effects of
IL-6 and Hyper-IL-6 on PC12 cells. Whereas PC12 cells
responded to both IL-6 and Hyper-IL-6 with an increase in
expression of growth associated protein (GAP)-43 mRNA
and protein, only Hyper-IL-6 induced neuronal differenti-
ation in these cells [39]. Intriguingly, it has been shown by
Ihara et al. 1997 [52] that gp130 mutants incapable of
activating the MAPK pathway failed to induce neurite
outgrowth. Consistently, a MAPK kinase inhibitor,
PD98059, inhibited neurite outgrowth. These results suggest
that the activation of the MAPK pathway is essential for
gp130 induced neurite outgrowth of PC12 cells whereas
STAT3 is believed to inhibit this response [52,53]. In line
with these findings, Hyper-CNTF led to a profound
activation of the MAPK pathway with little stimulation
of STAT3. We therefore conclude that upon receptor
stimulation by Hyper-CNTF and LIF in PC12 cells, the
intracellular signal transduction pathways diverge leading to
the observed differences in physiological response in neur-
onal cells. The underlying molecular mechanism might
include the recruitment of the transducing proteins through
binding of Hyper-CNTF and LIF to distinct functional
motifs in the extracellular region of the receptor, leading to
minor conformational changes in the cytoplasmic domains.
Strobl et al. were able to show that for comparable levels of
STAT1 phosphorylation by slightly different chimeric
gp130 receptors, significantly changed transcriptional

CNTF has been demonstrated to slow the progression of
motor dysfunction in wobbler mice, another animal model
for motor neuron disease [61]. These findings encouraged
the use of CNTF and related neuropoietic cytokines in
human motor disease. The interest in the neuroprotective
potential of gp130/LIF-R stimulation has been revived by
the demonstration that the CNTF-R not only complexes
with CNTF but also with the newly identified cytokine CLC
[15].
Recently it has been shown that delivery using CNTF-
releasing implants, as described by Aebischer et al. [62–64],
was efficient to treat motor neuron disease in animals. We
propose that similar implants containing recombinant
Hyper-CNTF protein could represent a more optimal way
to stimulate degenerating neuronal cells in amyotrophic
lateral sclerosis or other neurological diseases.
ACKNOWLEDGEMENTS
We thank Dr Birgit Oppmann, Dr Marc Ehlers and Dr Barbara Krebs
for the production of recombinant LIF and CNTF, Dr Thomas
Jostock for cloning of the LIF-R-Fc fusion construct and Dr Hughes
Gascan for the CNTF-R antibody. This work was supported by grants
from the Deutsche Forschungsgemeinschaft (Bonn, Germany), the
Stiftung Rheinland Pfalz fu
¨
r Innovation (Mainz, Germany) and the
Naturwissenschaftlich-Medizinisches Forschungszentrum (Mainz,
Germany) to S. R J., and from the Swiss National Foundation for
Scientific Research (Grant 3100-061571.00/1) and the Deutsche
Forschungsgemeinschaft (SFB505/B5) to U. O.
REFERENCES

ropy of cytokines and growth factors. Biochem. Biophys. Res.
Commun. 260, 303–308.
10.Taga,T.,Hibi,M.,Hirata,Y.,Yamasaki,K.,Yasukawa,K.,
Matsuda, T., Hirano, T. & Kishimoto, T. (1989) Interleukin-6
triggers the association of its receptor with a possible signal
transducer, gp130. Cell 58, 573–581.
11. Rose-John, S. & Heinrich, P.C. (1994) Soluble receptors for
cytokines and growth factors: generation and biological function.
Biochem. J. 300, 281–290.
12. Mu
¨
llberg, J., Althoff, K., Jostock, T. & Rose-John, S. (2000) The
importance of shedding of membrane proteins for cytokine biol-
ogy. Eur. Cyt. Netw. 11, 27–38.
13. Davis, S., Aldrich, T.H., Ip, N.Y., Stahl, N., Scherer, S.,
Farruggella, T., DiStefano, P.S., Curtis, R., Panayotatos, N.,
Gascan, H., Chevalier, S. & Yancopulos, G.D. (1993) Released
form of CNTF receptor alpha component as a soluble mediator of
CNTF responses. Science 259, 1736–1739.
14. Ma
¨
rz, P., Otten, U. & Rose-John, S. (1999) Neuronal activities of
IL-6 type cytokines often depend on soluble cytokine receptors.
Eur. J. Neurosci. 11, 2995–3004.
15. 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.
16. Masu,Y.,Wolf,E.,Holtmann,B.,Sendtner,M.,Brem,G.&

llberg, J., Meyer zum Bu
¨
schenfelde, K H.,
Gascan, H. & Rose-John, S. (1999) Receptor recognition sites of
cytokines are organized as exchangeable modules: transfer of the
LIFR binding site from CNTF to IL-6. J. Biol. Chem. 274, 11859–
11867.
21. van Dam, M., Mu
¨
llberg, J., Schooltink, H., Stoyan, T.,
Brakenhoff, J.P., Graeve, L., Heinrich, P.C. & Rose-John, S.
(1993) Structure-function analysis of interleukin-6 utilizing
human/murine chimeric molecules. Involvement of two separate
domains in receptor binding. J. Biol. Chem. 268, 15285–15290.
22. Kru
¨
ttgen, A., Gro
¨
tzinger,J.,Kurapkat,G.,Weis,J.,Simon,R.,
Thier, M., Schro
¨
der, M., Heinrich, P., Wollmer, A., Comeau, M.,
Mu
¨
llberg, J. & Rose-John, S. (1995) Human ciliary neurotrophic
factor: a structure–function analysis. Biochem. J. 309, 215–220.
23. Vollmer, P., Peters, M., Ehlers, M., Yagame, H., Matsuba, T.,
Kondo, M., Yasukawa, K., Bu
¨
schenfelde, K.H. & Rose-John, S.

Cloning: a Laboratory Manual, Cold Spring Harbor Laboratory
Press, Cold Spring Harbor, NY.
30. McMahan, C.J., Slack, J.L., Mosley, B., Cosman, D., Lupton,
S.D., Brunton, L.L., Grubin, C.E., Wignall, J.M., Jenkins, N.A.,
Brannan, C.I. et al. (1991) A novel IL-1 receptor, cloned from B
cells by mammalian expression, is expressed in many cell types.
EMBO J. 10, 2821–2832.
31. Laemmli, U.K. (1970) Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227, 680–685.
32. Vollmer, P., Oppmann, B., Voltz, N., Fischer, M. & Rose-John, S.
(1999) A role for the immunoglobulin-like domain of the human
IL-6 receptor: intracellular protein transport and shedding. Eur. J.
Biochem. 263, 438–446.
33. Davis, S., Aldrich, T.H., Valenzuela, D.M., Wong, V.V., Furth,
M.E., Squinto, S.P. & Yancopoulos, G.D. (1991) The receptor for
ciliary neurotrophic factor. Science 253, 59–63.
34. McDonald, N.Q., Panayotatos, N. & Hendrickson, W.A. (1995)
Crystal structure of dimeric human ciliary neurotrophic factor
determined by MAD phasing. EMBO J. 14, 2689–2699.
35. Chebath, J., Fischer, D., Kumar, A., Oh, J.W., Kolett, O.,
Lapidot,T.,Fischer,M.,Rose-John,S.,Nagler,A.,Slavin,S.&
Revel, M. (1997) Interleukin-6 receptor-interleukin-6 fusion pro-
teins with enhanced interleukin-6 type pleiotropic activities. Eur.
Cyt. Netw. 8, 359–365.
36. Fukada, T., Hibi, M., Yamanaka, Y., Takahashi Tezuka, M.,
Fujitani, Y., Yamaguchi, T., Nakajima, K. & Hirano, T. (1996)
Two signals are necessary for cell proliferation induced by a
cytokine receptor gp130: involvement of STAT3 in anti-apoptosis.
Immunity 5, 449–460.
37. Hirano,T.,Ishihara,K.&Hibi,M.(2000)RolesofSTAT3in

differentiation and for transformation of NIH3T3 cells. Cell 77,
841–852.
42. Pang, L., Sawada, T., Decker, S.J. & Saltiel, A.R. (1995) Inhibi-
tion of MAP kinase blocks the differentiation of PC-12 cells
induced by nerve growth factor. J. Biol. Chem. 270, 13585–13588.
43. Hudson, K.R., Vernallis, A.B. & Heath, J.K. (1996) Character-
ization of the receptor binding sites of human leukemia inhibitory
factorandcreationofantagonists.J. Biol. Chem. 271, 11971–
11978.
44. Vernalis, A.B., Hudson, K.R. & Heath, J.K. (1997) An antagonist
for the leukemia inhibitory factor receptor inhibits leukemia
inhibitory factor, cardiotrophin-1, ciliary neurotrophic factor, and
oncostatin M. J. Biol. Chem. 272, 26947–26952.
45. Bravo, J. & Heath, J.K. (2000) Receptor recognition by gp130
cytokines. EMBO J. 19, 2399–2411.
46. Sterneck, E., Kaplan, D.R. & Johnson, P.F. (1996) Interleukin-6
induces expression of peripherin and cooperates with Trk receptor
signaling to promote neuronal differentiation in PC12 cells.
J. Neurochem. 67, 1365–1374.
47. Wu, Y.Y. & Bradshaw, R.A. (1996) Induction of neurite out-
growth by interleukin-6 is accompanied by activation of Stat3
signaling pathway in a variant PC12 cell (E2) line. J. Biol. Chem.
271, 13023–13032.
48. Murphy, M., Dutton, R., Koblar, S., Cheema, S. & Bartlett, P.
(1997) Cytokines which signal through the LIF receptor and their
actions in the nervous system. Prog. Neurobiol. 52, 355–378.
49. Shirogane,T.,Fukada,T.,Muller,J.M.,Shima,D.T.,Hibi,M.&
Hirano, T. (1999) Synergistic roles for Pim-1 and c-Myc in
STAT3-mediated cell cycle progression and antiapoptosis.
Immunity 11, 709–719.

3251–3256.
57. Pennica, D., Arce, V., Swanson, T.A., Vejsada, R., Pollock, R.A.,
Armanini, M., Dudley, K., Phillips, H.S., Rosenthal, A., Kato,
A.C. & Henderson, C.E. (1996) Cardiotrophin-1, a cytokine pre-
sent in embryonic muscle, supports long-term survival of spinal
motoneurons. Neuron 17, 63–74.
58. Ma
¨
rz, P., Heese, K., Dimitriades-Schmutz, B., Rose-John, S. &
Otten, U. (1999) Role of interleukin-6 and soluble IL-6 receptor in
region specific induction of astrocytic differentiation and neuro-
trophin expression. Glia. 26, 191–200.
59. Scha
¨
fer, K.H., Mestres, P., Ma
¨
rz, P. & Rose-John, S. (1999) The
IL-6/sIL-6R fusion protein promotes neurite outgrowth and
neuronal survival in cultured enteric neurons. J. Interferon Cyto-
kine Res. 19, 527–532.
60. Thier, M., Ma
¨
rz, P., Otten, U., Weis, J. & Rose-John, S. (1999)
Interleukin-6 (IL-6) supports survival of sensory neurons: auto-
crine trophic effects of IL-6 and soluble IL-6 receptor and
enhanced activity of an IL-6 designer cytokine. J. Neurosci. Res.
55, 411–422.
61. Ikeda, K., Iwasaki, Y., Tagaya, N., Shiojima, T. & Kinoshita, M.
(1995) Neuroprotective effect of cholinergic differentiation factor/
leukemia inhibitory factor on wobbler murine motor neuron dis-