Portal de Periódicos da CAPES

Co-administration of Ciliary Neurotrophic Factor with Its Soluble Receptor Protects against Neuronal Death and Enhances Neurite Outgrowth

2007; Elsevier BV; Volume: 283; Issue: 10 Linguagem: Inglês

10.1074/jbc.m709065200

1083-351X

Mark A. Ozog, Geetanjalee Modha, John A. Church, Rayne Reilly, Christian C. Naus,

RNA regulation and disease

Attempts to promote neuronal survival and repair with ciliary neurotrophic factor (CNTF) have met with limited success. The variability of results obtained with CNTF may, in part, reflect the fact that some of the biological actions of the cytokine are mediated by a complex formed between CNTF and its specific receptor, CNTFRα, which exists in both membrane-bound and soluble forms. In this study, we compared the actions of CNTF alone and CNTF complexed with soluble CNTFRα (hereafter termed “Complex”) on neuronal survival and growth. Although CNTF alone produced limited effects, Complex protected against glutamate-mediated excitotoxicity via gap junction-dependent and -independent mechanisms. Further examination revealed that only Complex promoted neurite outgrowth. Differential gene expression analysis revealed that, compared with CNTF alone, Complex differentially regulates several neuroprotective and neurotrophic genes. Collectively, these findings indicate that CNTF exerts more robust effects on neuronal survival and growth when applied in combination with its soluble receptor. Attempts to promote neuronal survival and repair with ciliary neurotrophic factor (CNTF) have met with limited success. The variability of results obtained with CNTF may, in part, reflect the fact that some of the biological actions of the cytokine are mediated by a complex formed between CNTF and its specific receptor, CNTFRα, which exists in both membrane-bound and soluble forms. In this study, we compared the actions of CNTF alone and CNTF complexed with soluble CNTFRα (hereafter termed “Complex”) on neuronal survival and growth. Although CNTF alone produced limited effects, Complex protected against glutamate-mediated excitotoxicity via gap junction-dependent and -independent mechanisms. Further examination revealed that only Complex promoted neurite outgrowth. Differential gene expression analysis revealed that, compared with CNTF alone, Complex differentially regulates several neuroprotective and neurotrophic genes. Collectively, these findings indicate that CNTF exerts more robust effects on neuronal survival and growth when applied in combination with its soluble receptor. Ciliary neurotrophic factor (CNTF), 3The abbreviations used are:CNTFciliary neurotrophic factorCBXcarbenoxoloneCNTFRαciliary neurotrophic factor receptor αDIVdays in vitroGZAglycyrrhizic acidgp130glycoprotein 130IL-6interleukin-6JAK/STATJanus kinase/signal transducers and activators of transcriptionLDHlactate dehydrogenaseLIFRβleukemia inhibitory factor receptor βMAP2microtubule-associated protein 2MAPK/ERKmitogen-activated protein kinase/extracellular signal-regulated kinaseNGFnerve growth factorPBSphosphate-buffered salinePDLpoly-d-lysinePIpropidium iodidePI3Kphosphatidylinositol 3-kinasepSTAT3Tyr-705-phosphorylated STAT3RT-PCRreverse transcription-PCRSTAT3signal transducers and activators of transcription 3TUNELterminal deoxynucleotidyltransferase-mediated deoxyuridine triphosphate nick-end labelingDAPI4,6-diamidino-2-phenylindole.3The abbreviations used are:CNTFciliary neurotrophic factorCBXcarbenoxoloneCNTFRαciliary neurotrophic factor receptor αDIVdays in vitroGZAglycyrrhizic acidgp130glycoprotein 130IL-6interleukin-6JAK/STATJanus kinase/signal transducers and activators of transcriptionLDHlactate dehydrogenaseLIFRβleukemia inhibitory factor receptor βMAP2microtubule-associated protein 2MAPK/ERKmitogen-activated protein kinase/extracellular signal-regulated kinaseNGFnerve growth factorPBSphosphate-buffered salinePDLpoly-d-lysinePIpropidium iodidePI3Kphosphatidylinositol 3-kinasepSTAT3Tyr-705-phosphorylated STAT3RT-PCRreverse transcription-PCRSTAT3signal transducers and activators of transcription 3TUNELterminal deoxynucleotidyltransferase-mediated deoxyuridine triphosphate nick-end labelingDAPI4,6-diamidino-2-phenylindole. a member of interleukin (IL)-6 cytokine family, was originally identified by its ability to support the in vitro survival of parasympathetic neurons from the chick ciliary ganglion (1Adler R. Landa K.B. Manthorpe M. Varon S. Science. 1979; 204: 1434-1436Crossref PubMed Scopus (358) Google Scholar). CNTF has been termed a “brain injury” cytokine based on several observations. First, humans and mice lacking CNTF appear normal, indicating that the cytokine is not essential for normal development or maintenance (2Masu Y. Wolf E. Holtmann B. Sendtner M. Brem G. Thoenen H. Nature. 1993; 365: 27-32Crossref PubMed Scopus (531) Google Scholar). Second, the level of CNTF in astrocytes increases dramatically in response to injury (3Ip N.Y. Wiegand S.J. Morse J. Rudge J.S. Eur. J. Neurosci. 1993; 5: 25-33Crossref PubMed Scopus (208) Google Scholar, 4Lee M.Y. Deller T. Kirsch M. Frotscher M. Hofmann H.D. J. Neurosci. 1997; 17: 1137-1146Crossref PubMed Google Scholar). Third, CNTF is normally produced as a nonsecreted cytokine and is only released when the membrane integrity of astrocytes or Schwann cells becomes compromised, e.g. in response to injury (5Nieto-Sampedro M. Lewis E.R. Cotman C.W. Manthorpe M. Skaper S.D. Barbin G. Longo F.M. Varon S. Science. 1982; 217: 860-861Crossref PubMed Scopus (347) Google Scholar, 6Rudge J.S. Alderson R.F. Pasnikowski E. McClain J. Ip N.Y. Lindsay R.M. Eur. J. Neurosci. 1992; 4: 459-471Crossref PubMed Scopus (183) Google Scholar, 7Stockli K.A. Lottspeich F. Sendtner M. Masiakowski P. Carroll P. Gotz R. Lindholm D. Thoenen H. Nature. 1989; 342: 920-923Crossref PubMed Scopus (534) Google Scholar). Fourth, although the expression of the specific receptor for CNTF, termed CNTFRα, is normally restricted to the plasma membrane of neurons (8Davis S. Aldrich T.H. Valenzuela D.M. Wong V.V. Furth M.E. Squinto S.P. Yancopoulos G.D. Science. 1991; 253: 59-63Crossref PubMed Scopus (534) Google Scholar), surviving astrocytes in the penumbra of an injury also begin to express CNTFRα (9Park C.K. Ju W.K. Hofmann H.D. Kirsch M. Ki K.J. Chun M.H. Lee M.Y. Brain Res. 2000; 861: 345-353Crossref PubMed Scopus (27) Google Scholar, 10Rudge J.S. Pasnikowski E.M. Holst P. Lindsay R.M. J. Neurosci. 1995; 15: 6856-6867Crossref PubMed Google Scholar) suggesting a mechanism by which the nervous system responds to sublethal environmental stress or injury. ciliary neurotrophic factor carbenoxolone ciliary neurotrophic factor receptor α days in vitro glycyrrhizic acid glycoprotein 130 interleukin-6 Janus kinase/signal transducers and activators of transcription lactate dehydrogenase leukemia inhibitory factor receptor β microtubule-associated protein 2 mitogen-activated protein kinase/extracellular signal-regulated kinase nerve growth factor phosphate-buffered saline poly-d-lysine propidium iodide phosphatidylinositol 3-kinase Tyr-705-phosphorylated STAT3 reverse transcription-PCR signal transducers and activators of transcription 3 terminal deoxynucleotidyltransferase-mediated deoxyuridine triphosphate nick-end labeling 4,6-diamidino-2-phenylindole. ciliary neurotrophic factor carbenoxolone ciliary neurotrophic factor receptor α days in vitro glycyrrhizic acid glycoprotein 130 interleukin-6 Janus kinase/signal transducers and activators of transcription lactate dehydrogenase leukemia inhibitory factor receptor β microtubule-associated protein 2 mitogen-activated protein kinase/extracellular signal-regulated kinase nerve growth factor phosphate-buffered saline poly-d-lysine propidium iodide phosphatidylinositol 3-kinase Tyr-705-phosphorylated STAT3 reverse transcription-PCR signal transducers and activators of transcription 3 terminal deoxynucleotidyltransferase-mediated deoxyuridine triphosphate nick-end labeling 4,6-diamidino-2-phenylindole. Although the above considerations have prompted interest in CNTF as a therapeutic agent against neuronal injury and neurodegenerative diseases, it has become apparent that the administration of exogenous CNTF elicits variable and, in any case, only weak neuroprotective and neurotrophic effects in experimental animals and humans (11Bloch J. Bachoud-Levi A.C. Deglon N. Lefaucheur J.P. Winkel L. Palfi S. Nguyen J.P. Bourdet C. Gaura V. Remy P. Brugieres P. Boisse M.F. Baudic S. Cesaro P. Hantraye P. Aebischer P. Peschanski M. Hum. Gene Ther. 2004; 15: 968-975Crossref PubMed Scopus (206) Google Scholar, 12Emerich D.F. Winn S.R. Cell Transplant. 2004; 13: 253-259Crossref PubMed Scopus (22) Google Scholar, 13Ogata N. Ogata K. Imhof H.G. Yonekawa Y. Acta Neurochir. 1996; 138: 580-583Crossref PubMed Scopus (12) Google Scholar). The limited therapeutic efficacy of CNTF as a neuroprotective and neuroregenerative agent may be consequent upon the administration of the cytokine without the soluble form of its specific receptor. Although CNTFRα is normally localized to the neuronal plasma membrane, it is anchored there by a glycosylphosphatidylinositol linkage that is sensitive to proteolysis and phospholipase-C-mediated cleavage, and following injury, it is released into the extracellular space as a soluble component (14Davis S. Aldrich T.H. Ip N.Y. Stahl N. Scherer S. Farruggella T. DiStefano P.S. Curtis R. Panayotatos N. Gascan H. Science. 1993; 259: 1736-1739Crossref PubMed Scopus (328) Google Scholar, 15Ip N.Y. Yancopoulos G.D. Prog. Growth Factor Res. 1992; 4: 139-155Abstract Full Text PDF PubMed Scopus (121) Google Scholar). Indeed, soluble CNTFRα has been identified in proximity to injured tissue and in the urine of patients with amyotrophic lateral sclerosis (4Lee M.Y. Deller T. Kirsch M. Frotscher M. Hofmann H.D. J. Neurosci. 1997; 17: 1137-1146Crossref PubMed Google Scholar, 16Rudge J.S. Li Y. Pasnikowski E.M. Mattsson K. Pan L. Yancopoulos G.D. Wiegand S.J. Lindsay R.M. Ip N.Y. Eur. J. Neurosci. 1994; 6: 693-705Crossref PubMed Scopus (145) Google Scholar). Following injury-induced release, CNTF and soluble CNTFRα together form a heterodimer (hereafter referred to as “Complex”) that can interact with the ubiquitously expressed β-receptor subunits, glycoprotein 130 (gp130) and leukemia inhibitory factor receptor β (LIFRβ) (14Davis S. Aldrich T.H. Ip N.Y. Stahl N. Scherer S. Farruggella T. DiStefano P.S. Curtis R. Panayotatos N. Gascan H. Science. 1993; 259: 1736-1739Crossref PubMed Scopus (328) Google Scholar, 17Ozog M.A. Bechberger J.F. Naus C.C. Cancer Res. 2002; 62: 3544-3548PubMed Google Scholar, 18Ozog M.A. Bernier S.M. Bates D.C. Chatterjee B. Lo C.W. Naus C.C. Mol. Biol. Cell. 2004; 15: 4761-4774Crossref PubMed Scopus (45) Google Scholar). The subsequent heterodimerization of gp130 and LIFRβ activates several signaling cascades, including the Janus kinase/signal transducers and activators of transcription (JAK/STAT) pathway (19Alonzi T. Middleton G. Wyatt S. Buchman V. Betz U.A. Muller W. Musiani P. Poli V. Davies A.M. Mol. Cell. Neurosci. 2001; 18: 270-282Crossref PubMed Scopus (116) Google Scholar). In addition, we have recently found that Complex increases both the expression of the gap junctional constituent connexin43 and intercellular coupling in glia via trans-signaling (17Ozog M.A. Bechberger J.F. Naus C.C. Cancer Res. 2002; 62: 3544-3548PubMed Google Scholar, 18Ozog M.A. Bernier S.M. Bates D.C. Chatterjee B. Lo C.W. Naus C.C. Mol. Biol. Cell. 2004; 15: 4761-4774Crossref PubMed Scopus (45) Google Scholar), the term used to describe the ability of a ligand to bind a soluble receptor (α subunit), which subsequently associates with the β subunits of the receptor on cells that do not normally express the α-receptor of the ligand (20Rose-John S. Heinrich P.C. Biochem. J. 1994; 300: 281-290Crossref PubMed Scopus (684) Google Scholar). In light of the limited efficacy of CNTF alone as a neuroprotective and neurotrophic agent, and increasing evidence that connexins and gap junctions elicit neuroprotective and neurotrophic effects (21Blanc E.M. Bruce-Keller A.J. Mattson M.P. J. Neurochem. 1998; 70: 958-970Crossref PubMed Scopus (204) Google Scholar, 22Lin J.H. Yang J. Liu S. Takano T. Wang X. Gao Q. Willecke K. Nedergaard M. J. Neurosci. 2003; 23: 430-441Crossref PubMed Google Scholar, 23Nakase T. Fushiki S. Naus C.C. Stroke. 2003; 34: 1987-1993Crossref PubMed Scopus (183) Google Scholar, 24Nakase T. Sohl G. Theis M. Willecke K. Naus C.C. Am. J. Pathol. 2004; 164: 2067-2075Abstract Full Text Full Text PDF PubMed Scopus (201) Google Scholar, 25Ozog M.A. Siushansian R. Naus C.C. J. Neuropathol. Exp. Neurol. 2002; 61: 132-141Crossref PubMed Scopus (123) Google Scholar, 26Theis M. Jauch R. Zhuo L. Speidel D. Wallraff A. Doring B. Frisch C. Sohl G. Teubner B. Euwens C. Huston J. Steinhauser C. Messing A. Heinemann U. Willecke K. J. Neurosci. 2003; 23: 766-776Crossref PubMed Google Scholar), in this study we examined the effects of exogenous CNTF, applied alone or in combination with its soluble receptor CNTFRα, on excitotoxic cell death, neuronal survival, and neuronal development. Primary co-cultures of murine cortical astrocytes and neurons were prepared as described previously (25Ozog M.A. Siushansian R. Naus C.C. J. Neuropathol. Exp. Neurol. 2002; 61: 132-141Crossref PubMed Scopus (123) Google Scholar). Briefly, astrocyte cultures were established from the cortices of 1-day-old CD-1 mouse pups and plated onto poly-d-lysine-coated (50 μg/ml) 6-well plates. Culture medium was replaced every 3 days. Because reactive and newly plated (reactive-like) astrocytes express CNTFRα, we matured our astrocytes in vitro for 6 weeks to obtain quiescent cultures (18Ozog M.A. Bernier S.M. Bates D.C. Chatterjee B. Lo C.W. Naus C.C. Mol. Biol. Cell. 2004; 15: 4761-4774Crossref PubMed Scopus (45) Google Scholar). Neurons, obtained from the neocortices of embryonic CD-1 mice aged 16 days gestation, were then seeded on top of the astrocytes and allowed to settle for 2 h before replacing the medium with Medium I (54 ml of Neurobasal Medium, 36 ml of Dulbecco's modified Eagle's medium/F-12, 1 ml of B-27 supplements; Invitrogen). On day 4, 2 μm cytosine arabinoside was added during a fresh medium change for 24 h. Thereafter, two-thirds of the conditioned medium was replaced every 3 days with fresh Medium I. Co-cultures were maintained for 2 weeks prior to experiments to ensure expression of functional glutamate receptors by the neurons (27Dugan L.L. Bruno V.M. Amagasu S.M. Giffard R.G. J. Neurosci. 1995; 15: 4545-4555Crossref PubMed Google Scholar). In a manner similar to that described above, neurons were seeded directly onto either PDL (50 μg/ml)/laminin-coated (100 μg/ml) transwells (neurite outgrowth quantification assay kit; Chemicon, Temecula, CA) or PDL/laminin-coated 2-well chamber slides (BD Biosciences/Fisher) at 2.5 × 105 cells per chamber. Two h after cell plating, the medium was replaced with serum-free Medium I. Cultures maintained for 13 days in vitro (DIV) were treated with cytosine arabinoside as described above. Astroglial contamination of the cultures was <5%, as assessed by direct morphological analysis and glial fibrillary acidic protein immunoreactivity. Rat pheochromocytoma cells (PC12; ATCC, Manassas, VA) were cultured in Dulbecco's modified Eagle's medium supplemented with 10% horse serum, 5% fetal bovine serum, 10 units/ml penicillin, and 10 μg/ml streptomycin. Differentiation of the PC12 cells was induced by exposure to serum-reduced medium (1% horse serum, 1% fetal bovine serum) supplemented with NGF (50 ng/ml; Sigma) for 72 h. Prior to the application of glutamate, neuron-astrocyte co-cultures were pretreated with vehicle (PBS), CNTF (20 ng/ml; R & D Systems, Minneapolis, MN), soluble CNTFRα (200 ng/ml; R & D Systems), or Complex (20 ng/ml CNTF + 200 ng/ml CNTFRα, an ∼1:5 molar ratio to favor the association of CNTF with soluble CNTFRα) (18Ozog M.A. Bernier S.M. Bates D.C. Chatterjee B. Lo C.W. Naus C.C. Mol. Biol. Cell. 2004; 15: 4761-4774Crossref PubMed Scopus (45) Google Scholar) with a fresh medium change every 24 h for 72 h. In some experiments, 24 h following the final pretreatment, the gap junction blocker carbenoxolone (CBX; 25 μm; Sigma), its inactive analogue glycyrrhizic acid (GZA; 25 μm; Sigma), or solvent (H2O) was added to the co-cultures for 1 h as described previously (25Ozog M.A. Siushansian R. Naus C.C. J. Neuropathol. Exp. Neurol. 2002; 61: 132-141Crossref PubMed Scopus (123) Google Scholar). The co-cultures were then bathed in Earle's balanced salt solution (Invitrogen) containing glutamate (1 mm) for 3 h. For a comparative control, selected sister cultures were treated similarly but with the omission of glutamate. Cells were then maintained in fresh Medium I for 24 h and subsequently assessed for cell viability (see below). Neuron-enriched cultures were exposed to vehicle, CNTF, CNTFRα, or Complex for 3 days starting at either 2 h (hereafter termed “3 DIV” cultures) or 10 days (“10 + 3 DIV” cultures) after initial plating. A fresh medium change was performed every 24 h during treatments. The JAK/STAT, MAPK/ERK, and PI3K/Akt signaling cascades were inhibited with 50 μm AG490 (Calbiochem), 10 μm U0126 (Promega, Madison, WI), and 20 μm LY294002 (Calbiochem), respectively. The inhibitors were applied 45 min prior to and throughout the treatment of neurons with vehicle or Complex. The inhibitors were applied at concentrations commonly used by others on neuronal cells in vitro and adequately blocked the signaling pathways as assessed by substrate phosphorylation detection on control immunoblots (data not shown). To examine if CNTF, CNTFRα, or Complex induced or enhanced neuronal differentiation, undifferentiated PC12 cells were exposed to the test agent in place of NGF for 3 days or were co-treated with NGF and the test agent for 24 h, respectively. Cell survival was determined by measuring LDH content (Sigma) within the conditioned medium, the ability of the cells to exclude propidium idodide (PI), as well as the absence of cleaved caspase3 (see below) and/or TUNEL (DeadEnd Fluorometric TUNEL System; Promega). RNA Isolation—Cytoplasmic RNA was isolated from cells using TRIzol reagent (Invitrogen) and stored at -80 °C until used for microarray hybridization, real time PCR, or reverse transcription (RT)-PCR experiments. Microarray Hybridization—RNA samples were processed for microarray hybridization as described previously by Treister et al. (28Treister N.S. Richards S.M. Lombardi M.J. Rowley P. Jensen R.V. Sullivan D.A. J. Dent. Res. 2005; 84: 160-165Crossref PubMed Scopus (25) Google Scholar). Briefly, 2 μg of RNA was used to synthesize cDNA with the aid of CodeLink expression assay reagent kits (Amersham Biosciences). Target cRNA product was subsequently acquired using RNeasy kits (Qiagen), fragmented, and applied to a Codelink Mouse Uniset I microarrays (Amersham Biosciences) containing ∼10,000 mouse oligonucleotide gene probes. Positive signals on the arrays were detected with streptavidin-Alexa 647 and scanned using ScanArray Express software and a ScanArray Express HT scanner. Microarray images were analyzed using CodeLink image and data analysis software. Experiments were carried out in duplicate, with consistent results. Values presented consist of the mean ratio of normalized intensities of the different treatment regimens as described, and experiments were carried out in duplicate. The Ingenuity Pathways Knowledge Base (Ingenuity Systems, Inc., Redwood City, CA) was used to assess gene expression in the context of regulatory clusters. Real Time PCR—Real time PCR was used to verify the differential expression of selected genes in a manner similar to that described by James et al. (29James C.G. Appleton C.T. Ulici V. Underhill T.M. Beier F. Mol. Biol. Cell. 2005; 16: 5316-5333Crossref PubMed Scopus (112) Google Scholar). Briefly, specific target primers and probe sets (Taqman Assays-on-Demand, Applied Biosystems), their corresponding probes, and template RNA were combined with TaqMan (Universal PCR Mastermix, No AmpErase UNG, Applied Biosystems). Amplified target sequences were detected with the ABI Prism 7500 sequence detector (Applied Biosystems). Gene expression values were corrected using 18 S RNA as the endogenous control. RT-PCR—Isolated RNA was subjected to RT-PCR as described previously (18Ozog M.A. Bernier S.M. Bates D.C. Chatterjee B. Lo C.W. Naus C.C. Mol. Biol. Cell. 2004; 15: 4761-4774Crossref PubMed Scopus (45) Google Scholar). The RT product was amplified in the presence of primers for CNTF (30Malgrange B. Rogister B. Lefebvre P.P. Mazy-Servais C. Welcher A.A. Bonnet C. Hsu R.Y. Rigo J.M. De Water Van T.R. Moonen G. Neurochem. Res. 1998; 23: 1133-1138Crossref PubMed Scopus (36) Google Scholar), CNTFRα (30Malgrange B. Rogister B. Lefebvre P.P. Mazy-Servais C. Welcher A.A. Bonnet C. Hsu R.Y. Rigo J.M. De Water Van T.R. Moonen G. Neurochem. Res. 1998; 23: 1133-1138Crossref PubMed Scopus (36) Google Scholar), gp130 (31Zaheer A. Zhong W. Uc E.Y. Moser D.R. Lim R. Cell. Mol. Neurobiol. 1995; 15: 221-237Crossref PubMed Scopus (47) Google Scholar), and LIFRβ (32Nakashima K. Yanagisawa M. Arakawa H. Taga T. FEBS Lett. 1999; 457: 43-46Crossref PubMed Scopus (114) Google Scholar). Control experiments were performed in parallel in the absence of the reverse transcriptase enzyme and did not produce any products. Cells were fixed with paraformaldehyde, processed as described previously (18Ozog M.A. Bernier S.M. Bates D.C. Chatterjee B. Lo C.W. Naus C.C. Mol. Biol. Cell. 2004; 15: 4761-4774Crossref PubMed Scopus (45) Google Scholar), and labeled with antibodies against MAP2 and neurofilament 200 (Sigma), pSTAT3 (Tyr-705; Cell Signaling, Beverly, MA), or cleaved caspase3 (Cell Signaling), with subsequent Alexa- and fluorescein-conjugated secondary IgG antibody application (Molecular Probes Inc., Eugene, OR). Immunolabeled cells were mounted in ProLong Gold containing DAPI (Molecular Probes Inc.), viewed on a Zeiss Axioskop microscope, and analyzed using AxioVision 4.2 software. Specificity of antibody labeling was assessed by omitting the primary antibody from the labeling protocol. PC12 cells were stained with fluorescently tagged wheat germ agglutinin (Molecular Probes) to define cell limits, fixed in paraformaldehyde, and processed as described above. Neuron-enriched cultures were seeded in 6-well PDL/laminin-coated culture plates (BD Biosciences) and were maintained in culture for 24 h prior to a 15-min exposure to vehicle or Complex. Protein was harvested and processed as described previously (18Ozog M.A. Bernier S.M. Bates D.C. Chatterjee B. Lo C.W. Naus C.C. Mol. Biol. Cell. 2004; 15: 4761-4774Crossref PubMed Scopus (45) Google Scholar), electrotransferred onto a nitrocellulose membrane, and immunoblotted against pSTAT3 (Tyr-705; Cell Signaling). To normalize protein loading, bound antibodies were stripped, and the blot was probed for total STAT3 (Cell Signaling). Densitometric analyses of immunoblot signals on Kodak X-Omat x-ray film were performed using Scion Image software (Scion, Frederick, MD). Neurite outgrowth was analyzed in two ways. Indirect measurements were performed on neurons treated with agents for 3 days using the neurite outgrowth quantification assay kit (Chemicon) (33Smit M. Leng J. Klemke R.L. BioTechniques. 2003; 35: 254-256Crossref PubMed Scopus (32) Google Scholar). Direct morphometric analyses were conducted on individual MAP2-immunolabeled neurons. Because MAP2 immunoreactivity can be observed in both dendrites and axons in developing neurons (34Dehmelt L. Halpain S. J. Neurobiol. 2004; 58: 18-33Crossref PubMed Scopus (194) Google Scholar), no attempt was made to differentiate dendrites from axons; neurofilament 200 immunolabeling was employed only to confirm that axons of the neurons used for analyses were not in contact with neighboring cells. The total number of neurites was estimated by counting the tip ends of all MAP2-positive neurites elaborated by a single neuron. The cumulative length of primary neurites was the sum of the lengths of all neurites ≥8 μm long that extended directly from the soma. A branching point was defined as a point at which a neurite ≥8 μm in length extended from another neurite; the ratio of the sum of the number of branching points on a single neuron to the sum of the number of primary neurites on that neuron defined the branching ratio. Each experiment was performed on at least four different batches of PC12 cells or four different culture preparations obtained from different litters of mice. Neurite outgrowth measurements were performed on at least 16 neurons per treatment group from each culture preparation, for a total of at least 64 neurons per treatment group. For assessment of PI, TUNEL, and cleaved caspase3 labeling, at least eight randomly selected areas from each culture per treatment group were used. Data are presented as means ± S.E., and the results were analyzed by one-way analysis of variance followed by the Student-Newman-Keuls post hoc test. A p value of less than 0.05 was considered statistically significant. Complex Induces Neuroprotection in Vitro—We have reported previously that gap junctions exert a neuroprotective effect against glutamate excitotoxicity and that Complex, but not CNTF alone, up-regulates functional gap junctional coupling (18Ozog M.A. Bernier S.M. Bates D.C. Chatterjee B. Lo C.W. Naus C.C. Mol. Biol. Cell. 2004; 15: 4761-4774Crossref PubMed Scopus (45) Google Scholar, 25Ozog M.A. Siushansian R. Naus C.C. J. Neuropathol. Exp. Neurol. 2002; 61: 132-141Crossref PubMed Scopus (123) Google Scholar). Initially, therefore, we sought to determine whether pretreating neuron-astrocyte co-cultures with Complex can reduce glutamate-induced neurotoxicity and whether this effect is dependent on gap junctional communication. As glutamate-mediated cell death can proceed by both necrosis and apoptosis (35Dirnagl U. Iadecola C. Moskowitz M.A. Trends Neurosci. 1999; 22: 391-397Abstract Full Text Full Text PDF PubMed Scopus (3185) Google Scholar), the effects of Complex on neuronal viability were assessed by measuring the release of LDH into the extracellular space, the ability of cells to exclude the cationic dye PI, the presence of cleaved caspase3, and TUNEL. Under control conditions, pretreatment of neuron-astrocyte co-cultures with CNTFRα, CNTF, or Complex did not significantly alter LDH release (Fig. 1), PI staining (Fig. 2), or TUNEL (Fig. 2), compared with vehicle pretreatment alone. As expected, application of 1 mm glutamate to the co-cultures for 3 h resulted in substantial increases in LDH release (p < 0.001; Fig. 1) as well as PI staining and TUNEL (p < 0.01 in both cases; Fig. 2) at 24 h post-insult. Whereas pretreatment of co-cultures with CNTFRα alone failed to significantly alter glutamate-induced cell death (Figs. 1 and 2), pretreatment with CNTF alone significantly reduced TUNEL (p < 0.001; Fig. 2) but not LDH release (p > 0.05; Fig. 1) or PI staining (p > 0.05; Fig. 2). In contrast, Complex pretreatment significantly reduced glutamate-induced cell death as measured in all three cell viability assays (p < 0.01 or less in all cases; Figs. 1 and 2). Together, these results indicate that Complex affords greater protection against glutamate-induced cytotoxicity than either CNTF or CNTFRα alone.FIGURE 2Complex reduces both propidium iodide staining and TUNEL in neuron-astrocyte co-cultures treated with 1 mm glutamate for 3 h. A, representative photomicrographs of co-cultures pretreated for 3 days with vehicle (PBS), CNTFRα (200 ng/ml), CNTF (20 ng/ml), or Complex (CNTFRα + CNTF) following the sham insult or challenged with glutamate in the absence or presence of the gap junction blocker CBX, and subsequently stained with PI (red) and then labeled with TUNEL (green). Several cells can be identified as being labeled for both markers (overlaid in yellow). Nuclei were stained with DAPI (blue). Bars = 100 μm. A reduction in glutamate-induced TUNEL can be seen in CNTF pretreated cultures compared with vehicle or CNTFRα pretreatments in both the absence and presence of CBX. Complex further reduced PI staining and TUNEL under each condition. B, when quantified, CNTF significantly reduced TUNEL but not PI staining in glutamate-treated cultures in both the absence and presence of CBX, as compared with vehicle and CNTFRα pretreatments. Pretreatment with Complex in the absence of CBX significantly reduced glutamate-mediated PI staining compared with vehicle, CNTFRα, and CNTF pretreatments and significantly reduced TUNEL compared with vehicle and CNTFRα pretreatments. In the presence of CBX, Complex significantly reduced both PI staining and TUNEL compared with all other agents. The number of PI-stained and TUNEL-labeled cells is expressed as a percentage of total cells within the same field examined. #, p < 0.05; ^, p < 0.01; and *, p < 0.001 compared with other treatments as indicated. NS, not significant.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Functional Gap Junctions Contribute to Complex-mediated Neuroprotection—Substantial evidence supports a neuroprotective role for gap junctions and their constituent proteins, the connexins (see under “Discussion”). Previously, we reported that Complex, but not CNTF alone, enhances gap junctional communication and connexin43 expression in astrocytes (18Ozog M.A. Bernier S.M. Bates D.C. Chatterjee B. Lo C.W. Naus C.C. Mol. Biol. Cell. 2004; 15: 4761-4774Crossref PubMed Scopus (45) Google Scholar), suggesting a possible mechanism by which this agent could elicit neuroprotection. Therefore, in the next series of experiments, we applied glutamate to neuron-astrocyte co-cultures pretreated with CNTFRα, CNTF, or Complex in the absence or presence of the gap junction blocker CBX or the inactive analogue GZA. Consistent with our previous report (25Ozog M.A. Siushansian R. Naus C.C. J. Neuropathol. Exp. Neurol. 2002; 61: 132-141Crossref PubMed Scopus (123) Google Scholar), neit