SEARCH

SEARCH BY CITATION

Keywords:

  • myelin;
  • multiple sclerosis;
  • PI3K;
  • differentiation;
  • growth factors

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  9. Supporting Information

Oligodendrocytes (OLGs) produce and maintain myelin in the central nervous system (CNS). In the demyelinating autoimmune disease multiple sclerosis, OLGs are damaged and those remaining fail to fully remyelinate CNS lesions. Therefore, current therapies directed to restrain the inflammation process with approaches that protect and reconstitute oligodendrocyte density would be essential to pave the way of myelin repair. A critical signal for oligodendrocytes is insulin-like growth factor-1 (IGF-1), which promotes their development and ultimately myelin formation. PTEN inhibits the phosphoinositide 3-kinase (PI3K)/Akt signaling, a convergence downstream pathway for growth factors such as IGF-1. In this report, we temporarily inhibited PTEN activity by treating rat and human oligodendrocyte progenitors (OLPs) cultured alone or with dorsal root ganglion neurons (DRGNs) with bisperoxovanadium (phen). Our findings show that phen potentiates IGF-1 actions by increasing proliferation of OLPs in a concentration-dependent manner, and caused a sustained and time-dependent activation of the main pathways: PI3K/Akt/mammalian target of rapamycin (mTOR) and MEK/ERK. At low concentrations, IGF-1 and phen stimulated the differentiation of rat and human OLPs. Concordantly, the PTEN inhibitor together with IGF-1 robustly augmented myelin basic protein accumulation in rat newborn and human fetal OLGs co-cultured with DRGNs in a longer timeframe by promoting the elaboration of organized myelinated fibers as evidenced by confocal microscopy. Thus, our results suggest that a transient suppression of a potential barrier for myelination in combination with other therapeutic approaches including growth factors may be promising to improve the functional recovery of CNS injuries. GLIA 2013;62:64–77


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  9. Supporting Information

The process of myelination in the central nervous system (CNS) involves generation of oligodendrocyte progenitors (OLPs), their migration to specific brain regions and finally, their differentiation into oligodendrocytes (OLGs), which synthesize myelin components to elaborate myelin sheaths (Small et al., 1987). Multiple signals are required for OLG development, including growth factors, neurotransmitters and hormones, which act in concert with developmental gene expression programs (Cohen and Almazan, 1994; McMorris and Dubois-Dalcq, 1988).

The OLG-myelin-axon unit represents a unique structural and functional specialization in the CNS. Myelin not only increases an axon's diameter, which enhances conduction velocity, but also provides protection and trophic support (Wilkins et al., 2003). Disruption of these critical factors is present in the early stages of demyelinating diseases such as multiple sclerosis (MS) (Trapp et al., 1998). MS is primarily a neuroinflammatory disorder of the CNS in which focal leukocyte infiltration leads to damage of OLGs and destruction of myelin. In late neurodegenerative phases of MS, myelin repair appears limited by OLG density, which could be a product of impaired survival, proliferation, maturation, and/or myelin formation (Hanafy and Sloane, 2011). The processes leading to OLG differentiation and myelination appear to be most vulnerable to dysregulation (Franklin and Ffrench-Constant, 2008), suggesting that the current therapies aimed at limiting the inflammation process should be combined with approaches that protect and reconstitute OLG density to facilitate myelin repair.

Insulin-like growth factor (IGF-1) plays a prominent role in OLP proliferation, survival and differentiation. Specifically, in vivo and in vitro studies have demonstrated that IGF-1 signaling is required for myelin formation via its receptor, IGF1R (D'Ercole et al., 2002; Zeger et al., 2007). One of the major components of the signaling cascades stimulated by IGF-1 is phosphatidylinositol-3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTOR), a pathway involved in CNS myelination and OLP differentiation (Flores et al., 2008; Tyler et al., 2009). Using specific kinase inhibitors in OLP cultures, it has been shown that PI3K is necessary for full mitogenic response as well as for IGF-1-mediated cell proliferation, survival and protein synthesis (Bibollet-Bahena and Almazan, 2009; Cui and Almazan, 2007; Cui et al., 2005; McMorris and Dubois-Dalcq, 1988). Activated PI3K phosphorylates 4,5-biphosphate (PIP2) to generate a second messenger, phosphatidylinositol 3, 4, 5 triphosphate (PIP3), and the lipid phosphatase PTEN (phosphatase and tensin homologue deleted on chromosome 10) reverses this reaction, thus negatively regulating the pathway (Engelman et al., 2006). Phosphatases such as PTEN act as gatekeepers by terminating tyrosine kinase signal transduction pathways, including IGF-1 (Himpe and Kooijman, 2009). Recent data showed that an amplification of PI3K signaling in mice with a conditional deletion of PTEN in OLGs produced hypermyelination (Goebbels et al., 2010; Harrington et al., 2010); however, the animals develop a progressive neuropathy and myelin sheath abnormalities. Conversely, other studies suggest that perturbing PTEN can prevent neuronal cell death in ischemic brain injury and promote axonal regeneration by modulating PTEN/mTOR signaling (Park et al., 2008; Shi et al., 2010; Sun et al., 2011). Moreover, pharmacologically blocking PTEN resulted in regeneration of peripheral neurons and improved recovery in a stroke animal model (Christie et al., 2010; Mao et al., 2013). Therefore, we hypothesized that modulating PTEN actions by temporarily inhibiting its activity could ultimately enhance myelination by facilitating OLP proliferation and/or differentiation.

In this report, we investigated whether bisperoxovanadium (phen), a pharmacological inhibitor of PTEN, increases or potentiates the trophic effects of IGF-1 on OLPs and their ability to myelinate dorsal root ganglion neurons (DRGNs). Our results suggest that the PTEN inhibitor potentiates IGF-1 actions by increasing OLP proliferation in a concentration-dependent manner through a time-dependent hyperactivation of the main downstream components of PI3K/Akt/mTOR and MEK/ERK pathways. In addition, lower concentrations of phen enhance differentiation of rat newborn and human fetal OLPs, marked by induction of higher myelin basic protein (MBP) expression in the IGF-1-treated cells. Importantly, we found that a short time treatment with the PTEN inhibitor and IGF-1 increases myelination of rat DRGNs by OLGs.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  9. Supporting Information

Reagents and Supplies

The following reagents were purchased from various suppliers: Ham's F12 medium, PBS, 7.5% BSA fraction V, Leibovitz's medium (L-15), B27 supplement and penicillin/streptomycin from Invitrogen (Burlington, ON); fetal or newborn calf serum and DMEM from Wisent Inc. (St-Bruno, QC); human recombinant platelet-derived growth factor-AA (PDGFAA), basic fibroblast growth factor (bFGF) and IGF-1 from PeproTech Inc. (Rocky Hill, NJ); nerve growth factor (NGF) (2.5S) from Alomone Labs (Jerusalem, Israel); extracellular matrix, poly-d-lysine, poly-l-ornithine, human transferrin, insulin, HEPES, Triton-X-100, 1,4-dithio-treitol (DTT), mouse monoclonal anti-α-tubulin, and bicinchonic protein assay kit from Sigma-Aldrich (Oakville, ON); phen from EMD chemicals (Gibbstown, NJ); malachite green phosphatase assay kit from Echelon Biosciences (Salt Lake City, UT); protein A/G magnetic beads by Fisher (Nepean, ON); nitrocellulose membranes from Mandel Scientific (Guelph, ON, Canada); ECL Western blotting reagents from GE Healthcare Life Sciences (Baie d'Urfe, QC); A2B5 mouse monoclonal antibody from American Type Culture Collection; rabbit polyclonal Ki67 from Abcam (Toronto, ON); mouse monoclonal anti-MBP (SMI-99) and anti-neurofilament N52 antibodies from Chemicon (Temecula, CA); rabbit polyclonal anti-cyclin D1 (Santa Cruz Biotechnology, Santa Cruz, CA), mouse monoclonal anti-p27kip1 (BD Biosciences, Mississauga, ON); rabbit polyclonal antibodies to phospho-PTEN (Ser380/Thr382/383), total PTEN, phospho-Akt (Ser473), phospho-Akt (Thr308), phospho-GSK3 (glycogen synthase kinase-3 α/β) (Ser21/9), phospho-S6rp (S6 ribosomal protein) (Ser235/236), phospho-Erk1/2 (p44/42 MAPK [mitogen-activated protein kinase]) (Thr202/Tyr204), phospho-4EBP1 (Thr37/46), phospho-eIF4G (eukaryotic initiation factor 4G) (Ser1108) and phospho-eEF2 (eukaryotic elongation factor 2) (Thr56) from Cell Signalling Technology (Danvers, MA); mouse monoclonal anti-human CD140α from AbD Serotec (Kidlington, Oxford); anti-mouse IgG1 magnetic beads (Miltenyi Biotech, Auburn, CA); Alexa Fluor 488-, Cy5-, HRP-, FITC-, or Texas Red-conjugated secondary antibodies from Southern Biotechnology, Jackson Immunoresearch Laboratories (Cedarlane, Hornby, ON), BIO-RAD Canada (Mississauga, ON) or Invitrogen (Burlington, ON); DAPI and Hoechst nuclear stains from Molecular Probes Inc. (Eugene, OR). O4, O1 and mouse monoclonal antibodies to sulfatides and galactocerebroside (GC) were donations (Sommer and Schachner, 1981) and Caspr (contactin-associate protein) from David Colman. All other reagents were obtained from Fisher Scientific (Whitby, ON), or VWR (Mont-Royal, QC).

Rat Oligodendrocyte Cultures

Cultures of OLPs were prepared from the brains of newborn Sprague-Dawley rats as described previously (Almazan et al., 1993). Experiments were approved by the McGill Faculty of Medicine Animal Care Committee in accordance with the Canadian Council on Animal Care guidelines. OLPs were plated on poly-d-lysine-coated culture dishes and grown in serum free medium (SFM) consisting of a DMEM-F12 mixture (1:1), 10 mM HEPES, 0.1% bovine serum albumin (BSA), 25 µg/mL human transferrin, 30 nM triiodothyronine, 20 nM hydrocortisone, 20 nM progesterone, 10 nM biotin, 5 µg/mL insulin, 16 µg/mL putrescine, 30 nM selenium. Addition of 2.5 ng/mL of PDGFAA and bFGF maintain progenitor proliferation, while removal of these mitogens initiates their differentiation. In differentiation experiments, 3% newborn calf serum was supplemented to the SFM. Culture medium was replaced every 2 days under all experimental conditions. At the beginning of the culture period, the purity of the preparations was determined immunocytochemically using cell-type-specific antibodies. More than 95% of the cells were immunostained with monoclonal antibody A2B5 that recognizes cell surface gangliosides, a marker for OLPs in culture while less than 5% were GC+ OLGs, glial fibrillary acidic protein+ astrocytes or complement type-3+ microglia. The progressive expression of stage-specific markers was previously studied in detail (Cohen and Almazan, 1994). Cells start to express surface sulfatides (O4) on days 1 to 2 following growth factor removal, GC and MAG (myelin-associated glycoprotein) on day 2 to 3, and finally MBP on days 2 to 4.

Rat Dorsal Root Ganglion Neuron Cultures

Purified DRGNs cultures were prepared as described previously (Giasson and Mushynski 1996). Briefly, DRGNs were dissected from Sprague-Dawley rat embryos at 15 to 16 days of gestation, dissociated with trypsin, and plated onto rat tail collagen-coated dishes. The cultures were maintained with 12.5 ng/mL NGF in serum-free N1 media. The anti-mitotic cytosine-1-β-d-arabinofuranoside (1 µM) was applied for 2 pulses of 24 h at the beginning of the culture in order to rid them of proliferating Schwann cells and fibroblasts. Myelination was initiated in the third week of culture by the addition of OLPs, plated at a density of 0.7 x 105 cells/cm2. At this stage, DRGNs are morphologically mature, displaying a profuse axonal network. The medium was replenished every 3 days.

Human Fetal Oligodendrocyte Cultures

Human fetal brain tissue obtained from 15- to 18-gestational week (gw) embryos was provided by the Human Fetal Tissue Repository (Albert Einstein College of Medicine, Bronx, NY). Both Albert Einstein College of Medicine and McGill University institutional review boards approved these studies. PDGFRα+ cells were selected with a monoclonal mouse anti-human CD140α antibody followed by rat anti-mouse IgG1 magnetic beads. Purified cells were plated on coverslips coated with a matrix of lysed human fetal astrocytes, and maintained in DMEM-F12 supplemented with N1, 0.01% BSA, 1% penicillin-streptomycin, B27, triiodothyronine (2 ng/mL), PDGFAA (10 ng/mL) and bFGF (10 ng/mL).

MTT Assay

Cell viability was assessed using the MTT assay, which measures mitochondrial dehydrogenase activity. The assay detects the reduction of MTT by active mitochondria in viable cells to an insoluble formazan product. Following the removal of PDGFAA and bFGF to initiate differentiation, cells were incubated with phen (0.1, 0.5, 1, and 2 μM) in DMEM alone for 24 h. Cultures were subsequently incubated with 0.5 mg/mL MTT at 37oC for 3 h, the formazan crystals solubilized in acidified isopropanol and absorbance was measured at 595 nm with a BIO-RAD spectrophotometer.

Proliferation Assay

OLPs, grown in 24-well dishes (~1.5 x 105 cells/cm2), were incubated with 1 µCi/mL [3H]-thymidine diluted in DMEM in the absence or presence of phen (0.5, 1, 2, or 5 μM) and/or IGF-1 (20 ng/mL) for 24 h. Cells were washed three times with 5% ice-cold trichloroacetic acid and lysed in 0.2 N NaOH and 0.1% Triton-X-100. Aliquots were mixed with Ecolite liquid scintillation counting fluid and emissions recorded using a β-counter.

Immunofluorescence Staining

Immunofluorescence was performed as described (Radhakrishna and Almazan, 1994). Briefly, cultures grown on glass coverslips were incubated with A2B5, O4, or anti-GC monoclonal antibodies at 37°C, fixed with 4% paraformaldehyde in PBS, and incubated with Texas Red-, Alexa 488-, or FITC-conjugated secondary antibodies. For intracellular markers (MBP, NF, Caspr and Ki67) cells were permeabilized with 0.3% triton X-100 for 30 min before incubation with appropriate primary and secondary antibodies, followed by DAPI or Hoechst staining. Coverslips were mounted in ImmunoMount medium and imaged with Leitz fluorescence and Zeiss LSM 510 confocal microscopes (Department of Pharmacology & Therapeutics or Montreal Neurological Institute, McGill University) with PlanFluor 10, 20, 40, or 63x water/oil objectives and Zeiss LSM acquisition software. Nuclear staining images were taken with the two-photon laser. All images from co-culture experiments were taken as z-stacks and subsequent three-dimensional projections were compiled for further analysis using LSM 5 Image Browser and ImageJ softwares.

Phosphatase Activity Assay

Cells were treated with 2 μM phen for 1 h, and harvested in RIPA buffer (150 mM sodium chloride, 50 mM Tris (pH 8.0), 0.5% sodium deoxycholate, 0.1% SDS, 1% Triton X-100). The protein concentrations of samples were determined with the bicinchonic protein assay kit. Protein extracts (4 mg/mL) were added to prewashed protein A agarose beads with total PTEN antibody to immunoprecipitate PTEN, then incubated overnight at 4°C. The beads/antibody/PTEN immunocomplex was washed with PTEN reaction buffer (25 mM Tris-HCl pH 7.4, 140 mM NaCl, 2.7 mM KCl, and 10 mM DTT) and after centrifugation, PIP3 substrate was added to the immunocomplex to start the reaction. No substrate was added to the negative control. Enzyme reactions were stopped by Malachite green solution provided by the manufacturer and absorbances were measured at 620 nm. Results were compared with a standard curve and plotted as free phosphate release.

Western Blot Analysis for the Activation of PI3K/Akt/mTOR and ERK Pathways in OLPs

Rat OLPs were maintained in DMEM alone for 4 h before treatment with phen (2 μM) and/or IGF-1 (20 ng/mL) for 30, 60, 240 min or 24 h, and were harvested in RIPA buffer (150 mM sodium chloride, 50 mM Tris pH 8.0, 0.5% sodium deoxycholate, 0.1% SDS, 1% Triton X-100). Protein concentration was determined by bicinchonic assay, and 10 to 50 µg of the cellular extracts were resolved by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes by standard protocols. Blots were blocked with 5% dry milk in Tris-buffered saline with 0.1% Tween-20 and then incubated with primary antibodies. Membranes were incubated with the appropriate HRP-conjugated secondary antibodies and immunoreactive bands visualized by enhanced chemiluminescence and quantified by densitometry using ImageQuant software. To normalize for equal loading and protein transfer, the membranes were re-probed with an antibody for α-tubulin.

Drug Treatments for Differentiation of OLP and Myelination of DRGN Experiments

Phen at 2 or 0.25 μM with and without 20 ng/mL IGF-1 was applied to rat OLPs for 1 or 4 to 6 days, respectively. All treatment groups were harvested on day 6 and immunoblotted for MBP. For isolated human fetal OLPs, cultures were treated with IGF-1 (20 ng/mL) alone or combined with phen (0–500 nM) for 2 days when the cells were immunostained for GC, sulfatides, and PDGFRα.

For rat OLP/DRGN co-cultures, OLPs were allowed to attach to DRGNs for 18 to 24 h before treatment with 0.25 μM phen and/or IGF-1 (20 ng/mL) for only 2 days. Cells were maintained for 5 or 12 more days in SFM when they were immunostained with specific antibodies. On the other hand, human fetal OLP/DRGN co-cultures were treated with IGF-1 (20 ng/mL) alone or with graded concentrations of phen (0–500 nM) for 4 weeks before being harvested for immunostaining. The medium was replenished along with phen and IGF-1 every 3 days.

Data Analysis

Unless indicated otherwise, results are presented as the mean ± SEM of a minimum of three independent experiments performed in triplicate. Statistical significance was determined by unpaired t-test for two group comparisons, and one-way or two-way ANOVA followed by Tukey or Bonferroni test for multiple group comparisons through Graph Pad Prism software. Statistical differences were considered significant when P values were <0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  9. Supporting Information

PTEN Inhibitor Potentiates IGF-1 Effects on proliferation of OLPs in a Concentration-Dependent Manner

Previous work showed that PI3K/Akt/mTOR pathway is activated by IGF-1 in OLP cultures (Cui et al., 2005) as well as in mice with conditional inactivation of Pten (Harrington et al., 2010). To further investigate PTEN function in OLP development, we used the inhibitor phen, a potassium bisperoxo vanadate derivative, alone or in combination with a submaximal concentration of IGF-1 (Cui and Almazan, 2007). Mitochondrial dehydrogenase activity was assessed by MTT cleavage as an index of cell viability. Phen in increasing concentrations (0.1–1 µM) was not toxic to OLPs; and at 2 µM it increased MTT cleavage alone and potentiated IGF-1 (Fig. 1a), suggesting an effect on survival or proliferation of OLPs.

image

Figure 1. The PTEN inhibitor amplifies IGF-1 signals on proliferation of OLPs in a dose-dependent manner. Progenitors were deprived of growth factors for 4 h before a 24-h treatment with phen and/or 20 ng/mL IGF-1 in DMEM alone. (a) The MTT reduction assay was carried out to determine the viability of OLPs treated with 0.1 to 2 µM phen and/or IGF-1. (b) The proliferation of progenitors treated with 0.5 to 5 µM phen and/or IGF-1 was assessed by [3H]-thymidine incorporation, as described in Material and Methods. (c) Protein extracts from OLPs treated with 2 µM phen and/or IGF-1 were analyzed by immunoblotting for cyclin D1 and p27kip1 expression. Representative immunoblots and their quantification, normalized to non-treated control, are shown. α-Tubulin was used as a loading control. The results are expressed as the mean ± SEM of three independent experiments performed in triplicate or quadruplicate. Statistical differences compared with nontreated control: ***P < 0.001, **P < 0.005, *P < 0.05; compared with phen alone: aP < 0.001, bP < 0.005, cP < 0.05 at its corresponding concentration, or IGF-1 alone: xP < 0.001, yP < 0.005, zP < 0.05.

Download figure to PowerPoint

To determine whether phen affects OLP proliferation, [3H]-thymidine incorporation and Ki67 expression were assessed. Phen alone significantly increased OLP proliferation at 1 and 2 µM in comparison with the control (P < 0.05). IGF-1 increased [3H]-thymidine incorporation, and the effect was potentiated by phen in a concentration-dependent manner, reaching its peak at 2 µM (approximately fourfold above phen alone). At a higher concentration, phen (5 µM) was toxic (Fig. 1b). These results were substantiated by immunostaining with a Ki67 antibody, which recognizes a nuclear protein present in all active phases of the cell cycle. OLPs were co-immunostained with A2B5 and O4 antibodies, which detect surface gangliosides and sulfatides, respectively, 24 h after treatment with 20 ng/mL IGF-1 and/or phen (1-5 µM). In agreement with the [3H]-thymidine incorporation results, the PTEN inhibitor at 1 and 2 µM alone increased the number of Ki67+ cells in a subpopulation of A2B5+ and/or O4+ OLPs (Supp. Info. Fig. 1a,b). Phen most significantly amplified the effect of IGF-1 at 2 µM with a fourfold increment above IGF-1 or inhibitor alone (P < 0.001). Thus, approximately 8% of OLP nuclei were immunolabeled with Ki67 in the IGF-1+phen co-treated cultures.

Two critical molecules in OLP cell cycle progression, cyclin D1 (a regulator of G1 to S-phase transition) and p27kip1 (a family member of the cell cycle inhibitor cip/kip), were next analyzed by Western blotting. Potentiated by the PTEN inhibitor (2 µM), IGF-1 up-regulated cyclin D1 and reduced p27kip1 levels when compared with either control or phen alone (Fig. 1c), consistent with the increased proliferation.

PTEN Inhibitor Alone or in Combination with IGF-1 Activates the PI3K/Akt/mTOR and ERK Pathways in OLPs

Activated PI3K phosphorylates PIP2 to generate PIP3, and the lipid phosphatase PTEN reverses this reaction. To confirm phen as an effective inhibitor of PTEN, we assessed PTEN levels by Western blotting and its lipid phosphatase activity by quantifying the free phosphate released from PIP3 after 1 h treatment with 2 µM phen. Phen did not affect PTEN total protein level but increased its phosphorylation state as assessed with an anti-phospho-PTEN (Ser380/Thr382/383) antibody (Fig. 2a). Furthermore, phen decreased PTEN enzymatic activity in rat OLPs (Fig. 2b).

image

Figure 2. Phen alters PTEN phosphorylation status and its phosphatase activity in OLPs. Progenitors were maintained in the absence of growth factors for 4 h and treated with 2 µM phen in DMEM for 1 h when samples were harvested for immunoblotting or phosphatase activity assays. (a) Protein extracts were subjected to Western blotting using specific antibodies to phosphorylated and total PTEN, quantified and normalized to control. Membranes were reprobed for α-tubulin and used as a loading control. In (b) the phosphatase activity was determined. Total PTEN was immunoprecipitated from samples with an anti-PTEN antibody, then the substrate PIP3 was added and the amount of free phosphate released per well was determined. Samples with no substrate added served as negative controls. The results are shown as the mean ± SEM of three experiments performed in duplicate. Statistical differences were measured by comparison to control: ***P < 0.001.

Download figure to PowerPoint

Most biological actions of IGF-1 involve two downstream signaling pathways: the PI3K/Akt/mTOR and/or the MEK/ERK cascades. To investigate their engagement in rat OLPs under our experimental conditions, the activation levels of downstream effectors were investigated by Western blotting or immunocytochemistry with phospho-specific antibodies. OLPs were treated with phen at 2 µM and/or 20 ng/mL IGF-1, and harvested at different time points. Although IGF-1 was applied at a low concentration, it increased phosphorylation of most effectors tested, mainly at 60 min, decreasing by 240 min (Fig. 3 and Supp. Info. Fig. 2). Treatment of cultures with 2 µM phen alone increased p-Akt, p-S6rp and p-Erk1/2 to higher levels than IGF-1 alone at most time points, while the repressor of protein translation, 4EBP1, was uniformly phosphorylated throughout. The combination of IGF-1 and phen maximally increased p-Akt (Thr308), p-S6rp and p-Erk1/2 (Fig. 3) at 30 min as compared with the 60 min required for IGF-1, and maintained an 8-fold increase over controls at 240 min in p-S6rp, p-Erk1/2 and p-Akt (Ser473) (Fig. 3 and Supp. Info. Fig. 2). GSK3, a downstream target of Akt, was phosphorylated in a pattern that paralleled Akt activation (Supp. Info. Fig. 2). IGF-1 with phen also induced phosphorylation of the translational repressor 4EBP1 above the individual treatments at all time points tested (Fig. 3).

image

Figure 3. The PTEN inhibitor alone and in combination with IGF-1 activate the PI3K/Akt/mTOR and ERK pathways. OLPs were subjected to growth factor withdrawal for 4 h in DMEM alone, treated with 2 µM phen and/or 20 ng/mL IGF-1 for 30, 60, or 240 min. Samples were analyzed by Western blotting using phospho-specific antibodies against Akt (Thr303), S6rp (Ser235/236), Erk1/2 (Thr202/Tyr204), and 4EBP1 (Thr37/46). (a) Representative immunoblots of samples are shown. α-Tubulin was used as a loading control. Quantifications of blots from three independent experiments performed in triplicate were normalized to control and are shown graphically as mean ± SEM in (b), as fold increase. Statistical differences compared with non-treated control: ***P < 0.001, **P < 0.005, *P < 0.05; compared with phen alone: aP < 0.001, bP < 0.005, cP < 0.05, or IGF-1 alone: xP < 0.001, yP < 0.005, zP < 0.05 at its corresponding time point.

Download figure to PowerPoint

The effect of IGF-1 and/or phen on eIF4G (initiation) and eEF2 (elongation) factors involved in protein translation was next studied. IGF-1 alone increased the number of A2B5+/O4+ cells co-labeled with the p-eIF4G antibody (~35%) as compared with control (Supp. Info. Fig. 3b,e). The combination with phen further increased the number of p-eIF4G+ OLPs (~50%) (Supp. Info. Fig. 3d,e). Likewise, IGF-1 alone significantly activated eEF2 (~40%) by dephosphorylation over a 240 min treatment when compared with control (Supp. Info. Fig. 3f). Together with phen, IGF-1 highly decreased p-eEF2 levels (~80%), thereby stimulating the elongation stage of mRNA translation. Overall, these data suggest that the PTEN inhibitor potentiates the action of IGF-1 not only on its main downstream effectors in a time-dependent manner by stimulating them faster and sustaining their activation longer but also via important steps of the mRNA translation machinery.

PTEN Inhibitor in Combination with IGF-1 Enhances Expression of Earlier and Late Lineage Markers in Rat and Human Differentiating OLPs

To assess whether phen can alter differentiation of rat OLPs in the absence or presence of IGF-1, MBP protein levels were determined. Cells were treated with one pulse of 2 µM phen and/or IGF-1 (20 ng/mL) for 24 h, then the drug and IGF-1 were removed, and cells were maintained until day 6 for harvesting. In another experiment, cells were chronically treated with 0.25 µM phen for 4 or 6 days and harvested on day 6. The lower dose of phen was used in this longer-term study to minimize possible toxic effects, whereas the higher concentration (2 μM), capable of inducing prominent proliferation on OLPs, was applied in short-term experiments. Phen by itself did not alter MBP levels regardless of the regimen applied (Fig. 4). Similarly, application of IGF-1 alone did not increase MBP expression, with the exception of 6 days of continuous treatment when it moderately augmented MBP (Fig. 4a,c). On the other hand, IGF-1 in combination with 2 µM phen increased MBP (1.7-fold) in the 1-day treatment group (Fig. 4b). Interestingly, 0.25 µM PTEN inhibitor with IGF-1 increased MBP levels in both the 4-d and 6-d treatment groups (P < 0.005 and P < 0.001, respectively) as compared with controls (Fig. 4).

image

Figure 4. PTEN inhibition increases myelin basic protein expression in differentiating IGF-1-treated OLPs. Progenitors were treated with IGF-1 (20 ng/mL) and/or 2 µM phen for 24 h, or 0.25 µM phen for 4 or 6 days. All groups were harvested at day 6, followed by Western blotting for MBP. Representative immunoblots are shown in (a). α-Tubulin was used as a loading control. MBP levels normalized to control are shown in (b) and (c), expressed as the mean ± SEM of three independent experiments performed in triplicate. Statistical differences compared with non-treated control: **P < 0.005; compared with phen alone: aP < 0.001 at its corresponding concentration, or IGF-1 alone: xP < 0.001, yP < 0.005 at its corresponding time point.

Download figure to PowerPoint

Given the findings that phen in conjunction with IGF-1 enhances MBP expression in rat OLPs, we explored their impact in human OLPs. OLPs derived from 15 to 18 gw human fetal brain tissues were selected by the expression of PDGFRα, an early marker of the oligodendroglial lineage. Cultures were treated with increasing concentrations of phen (0-500 nM) with IGF-1 (20 ng/mL) for 2 days and immunostained with monoclonal antibodies against PDGFRα (Fig. 5a,d), sulfatides (Fig. 5b,d) and GC, two myelin lipids used as markers of OLP differentiation. IGF-1 alone or with phen at the lowest concentration tested (10 nM) had no effect on the number of O4+ cells in comparison to control (Fig. 5e). Higher doses of the PTEN inhibitor (50–500 nM) with IGF-1 increased the number of O4+ cells by approximately 50% over IGF-1 alone. Similarly, the number of cells immunostained for GC out of O4+ cells were elevated in a concentration-dependent manner, starting at 10 nM phen (Fig. 5f). This amounted to an 80% increase over control or IGF-1 alone. Therefore, our data suggest that phen and IGF-1 in low concentrations can exert prominent maturational effects in rat and human OLP cultures.

image

Figure 5. Phen in combination with IGF-1 promotes the differentiation of human fetal OLPs. PDGFRα-selected human fetal progenitors were treated with 20 ng/mL IGF-1 alone or together with increasing concentrations of phen (0–500 nM) for 2 days, when they were harvested and immunostained for sulfatides (with O4 mAb) and GC. The images in (a) to (d) show cells immunocytochemically labeled with an antibody against PDGFRα (a) (green) and sulfatides (b) (red). Cell nuclei were stained with DAPI (c) (blue). (d) Merged confocal images. A higher power view of a double-labeled OLP in the center of the field is shown in the inset. Scale bar = 30 µm. (e) The number of O4+ cells or (f) GC+ OLGs relative to the total population of O4+ cells was counted and expressed as a percentage of the total. A minimum of 16 fields and 1,100 PDGFRα+ cells per coverslip was analyzed. The results are expressed as the mean ± SEM of three independent experiments performed in duplicate. Statistical differences compared with IGF-1 alone: xP < 0.001, yP < 0.005, zP < 0.05. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

PTEN Inhibitor Amplifies IGF-1 Effects by Increasing MBP Expression and Ensheathment of DRGN Axons by Rat or Human OLGs

We next investigated the ability of phen and IGF-1 to induce myelination of embryonic dorsal root ganglia by rat or human OLGs. To assess whether a single exposure could increase the myelination potential of rat OLPs, the cultures were treated with 0.25 µM phen and/or 20 ng/mL IGF-1 for the first 2 days and maintained for additional 5 or 12 days before immunostaining for MBP and Caspr. Neuronal integrity was determined by co-immunolabeling for neurofilaments (NFs). In compiled z-stack confocal images taken at high and low magnification, MBP+ patches (areas where MBP forms linear segments on NF+ fibers) or individually myelinated fibers were counted in the co-cultures. We observed a significant increase in the number of MBP+ patches after IGF-1 treatment alone or in combination with phen at both time points analyzed whereas phen alone was most effective in the 7-d co-culture (Fig. 6). Importantly, when combined with phen, IGF-1 increased the number of MBP+ patches to a greater extent than IGF-1 alone at day 14 (Fig. 6d–f). In addition, the number of MBP+ patches counted at 14-d was threefold higher than at 7-d. Moreover, IGF-1 and phen positively affected myelination, which increased the approximate numbers of MBP+ fibers by twofold above control at both time points (Fig. 6d,f). In order to assess the maturity of the myelin formed, we examined the expression of Caspr, a major component of the paranodal junction and important player in axon-glia signaling (Charles et al., 2002). As anticipated, we observed clustering of this marker along neurites in 14-day co-cultures, indicating a mature state of the myelinating fibers after phen and IGF-1 treatment (Fig. 6g–j).

image

Figure 6. The PTEN inhibitor enhances myelination in rat OLG-DRGN co-cultures by potentiating IGF-1. Co-cultures treated with 0.25 µM phen alone or in combination with 20 ng/mL IGF-1 for 2 days after OLP seeding, and maintained for an additional 5 or 12 days, were immunostained for MBP (red) and NF (green or purple) and Caspr (green). Myelination was evaluated by counting MBP+ patches or individually myelinated fibers. Representative confocal images of control (a), IGF-1- (b), phen- (c), and phen+IGF-1 (d)-treated 14 day cultures are shown. In (d), a higher magnification of myelinating co-cultures is shown (inset). Scale bar = 50 µm. (e) and (f), Quantification of the number of MBP+ patches or fibers at day 7 and 14 cultures, respectively. Images of a minimum of 20 fields per treatment were taken at a lower magnification (10× objective), and at least 2,500 cells per coverslip were analyzed. Shown in (g)–(j) are representative confocal images of day 14 phen+IGF-1-treated cultures. Double-labeled fibers are shown at higher magnification in the inset. Cell nuclei were stained with DAPI (blue). Scale bar = 20 µm. The results are expressed as the mean ± SEM of three independent experiments performed in triplicate. Statistical differences compared with nontreated control: ***P < 0.001, **P < 0.005, *P < 0.05; compared with phen alone: aP < 0.001, or IGF-1 alone: xP < 0.001, zP < 0.05 at its corresponding time point. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

We next assessed whether abrogating PTEN activity by phen in the presence of IGF-1 might enhance myelination of DRGNs by human fetal OLPs. As human OLPs require longer time to differentiate and synthesize myelin than rat cells (Cui et al., 2010), PDGFRα-selected cells were co-cultured with rat DRGNs for 4 weeks and examined for the expression of GC (Fig. 7a,d), MBP (Fig. 7b,d) and NFs (Fig. 7c,d). Thus, co-cultures were treated with IGF-1 (20 ng/mL) alone or with phen (0–500 nM) every 3 days when the medium was replenished. In the presence of IGF-1 alone, a small rise in the number of GC+ (Fig. 7e) and MBP+ cells per coverslip (Fig. 7f) was observed. However, under PTEN inhibition, IGF-1 increased GC+ cells at most phen concentrations (10–200 nM), starting at 10 nM and most markedly at 100 nM, which produced a twofold increase (P < 0.001) as compared with IGF-1 alone (Fig. 7e). Consistently, these phen concentrations positively affected MBP expression as well, with maximal effect at 100 nM phen (approximately threefold that of IGF-1 alone; P < 0.001) (Fig. 7f). To determine whether the PTEN inhibitor combined with IGF-1 is also able to increase the ensheathment of axons by OLGs, the number of GC+ cells making contacts with axons per coverslip was counted. IGF-1 with phen (50–500 nM) increased the relative numbers of GC+ OLGs making contacts with axons (P < 0.001) (Fig. 7g). Hence, our data suggest that inhibition of PTEN by phen in presence of IGF-1 improved the competence of rat and human OLPs to differentiate and ensheath DRGN axons.

image

Figure 7. The PTEN inhibitor amplifies IGF-1 effects by increasing expression of MBP in human fetal OLGs and ensheathment of DRGN axons in culture. PDGFRα-selected OLP cells co-cultured with DRGNs were treated with 20 ng/mL IGF-1 alone or with graded doses of phen (0–500 nM). Drug and media were replenished every 3 days. Co-cultures were maintained for 4 weeks, harvested, and immunostained for GC (a) (green), MBP (b) (red), and NF (c) (blue). Merged confocal image is shown in (d). Scale bar = 10 µM. (e) and (f), quantification of the number of GC+ or MBP+ cells per coverslip (cvs), respectively. (g) Quantification of the percentage of ensheathing GC+ cells relative to the total population of GC+ OLGs. The results are expressed as the mean ± SEM of three independent experiments performed in duplicate. Statistical differences compared with nontreated control: **P < 0.005, *P < 0.05, or IGF-1 alone: xP < 0.001, yP < 0.005, zP < 0.05. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  9. Supporting Information

Our results demonstrate that pharmacological inhibition of PTEN potentiates the effects of IGF-1 on proliferation and differentiation of OLPs. This action may occur through a sustained, faster, time-dependent activation of the main downstream PI3K/Akt/mTOR and MEK/ERK pathways. Interestingly, these trophic effects were obtained using a concentration of IGF-1 below its half maximal effect (Bibollet-Bahena and Almazan, 2009; Cui and Almazan, 2007). Moreover, IGF-1 in combination with a substantially lower phen concentration was able to increase differentiation of OLPs and to markedly enhance myelination of DRGNs by rat and human OLGs.

While the importance of IGF-1 in OLG growth has been demonstrated in transgenic studies, the underlying mechanisms are not completely understood (Ye et al., 2002). PI3K activity is essential for IGF-1-mediated proliferation, survival and protein synthesis (Bibollet-Bahena and Almazan, 2009; Cui and Almazan, 2007). IGF-1 has been shown to induce the differentiation of human OLPs (Armstrong et al., 1992; Cui et al., 2010, 2012) and increase myelination in rat brain aggregate cultures (Mozell and McMorris, 1991). Similarly, our results show that IGF-1 alone is able not only to promote proliferation and differentiation of OLPs but also to induce myelination of DRGNs during a 7-day course regime. The effect of IGF-1 on proliferation assessed by [3H]-thymidine incorporation was potentiated at higher concentrations of phen. This was confirmed by a marked increment in the number of Ki67+ cells in A2B5+ and/or O4+ OLPs. The expression of the markers A2B5 and O4 characterizes the proliferative states of OLG development (Gard and Pfeiffer, 1990). In addition, combined treatments of IGF-1 and phen induced an increase in cyclin D1 and lower p27kip1 levels. Similarly, previous studies have demonstrated higher expression of cyclin D1 mRNA in IGF-1-stimulated OLG cell line (Ye et al., 2010) as well as an elevated proliferation of OLPs followed by p27kip1 loss (Casaccia-Bonnefil et al., 1999). PTEN is a key functional antagonist of the Akt/mTOR pathway and its removal can cause cellular proliferation and increased cell size (Fraser et al., 2004). PTEN removal has often been associated with higher levels of cell proliferation and accordingly, neoplasia (Li et al., 1997). Conversely, it has been also reported that PTEN loss of function is not tumorigenic (Stiles et al., 2006; Wijesekara et al., 2005). Surprisingly, conditional inactivation of PTEN in OLGs in mice reduced or did not alter the proliferative capacity of these cells (Goebbels et al., 2010; Harrington et al., 2010). It has been suggested that developmental balance of cell loss and proliferation may have different requirements in vivo and in vitro. As such, basal PI3K/Akt/mTOR activation may be sufficient to control and maintain the optimal number of OLGs. Thus, stimulating PI3K activity in vivo may not have the ability to overcome that regulation (Flores et al., 2008).

Our prior studies demonstrated that during cell proliferation and growth, both PI3K/Akt/mTOR and MEK/ERK signaling play essential roles in IGF-1-stimulated OLPs (Cui and Almazan, 2007). In our experiments, treatment with IGF-1 together with phen led to a time-dependent phosphorylation of Akt at Ser473 and Thr308 along with activation or inhibition of other downstream targets, namely S6rp and 4EBP1, respectively. Akt has a N-terminal pleckstrin domain, which is essential for binding lipids such as PIP3 (Manning et al., 2002). Akt binds to PIP3 at the plasma membrane, allowing 3-phosphoinositide-dependent kinase 1 (PDK1) to access and phosphorylate Thr308, leading to a partial Akt activation. This Akt modification is sufficient to activate the mTORC1 complex and, as a result, phosphorylates the translation repressor 4EBP1 and p70 S6K, which in turn, phosphorylates S6rp, promoting protein synthesis and cell proliferation (Hemmings and Restuccia, 2012). Two of the well-studied targets of the mTOR kinase are S6rp and 4EBP1. S6rp phosphorylation is widely used to monitor mTOR activity (Guertin and Sabatini 2007), while MEK/ERK is involved in IGF-1-mediated cell proliferation in different systems (Cui and Almazan, 2007; Sunayama et al., 2010). In this context, we also assessed activation of Erk1/2 and found a pattern similar to the other downstream effectors. Thus, the results suggest that the proliferative effects induced by IGF-1 in phen-pretreated OLPs are mediated through sustained and full Akt stimulation as well as S6rp and Erk1/2 activation, which are crucial downstream pathways of IGF-1 signaling.

Recent studies have demonstrated that PTEN deletion prevents neuronal cell death resulting from ischemic brain injury by stimulating mTOR pathway (Shi et al., 2010). It appears that downstream kinases mainly Akt are recruited to the plasma membrane and activated under IGF-1 stimulation and PTEN blockade. Our laboratory has shown that Akt plays a key role in mediating IGF-1-induced OLP survival (Cui et al., 2005). Here we demonstrated full activation of Akt at two sites. Moreover, GSK3, an indicator of Akt activity, was highly phosphorylated (inactivated) by phen+IGF-1 at all time points. Inhibition of GSK3β increased OLP proliferation and survival (Azim and Butt, 2011). Given that we have not specifically investigated cell survival, it may be possible that IGF-1 along with phen is also capable of increasing OLP survival.

IGF-1 or insulin can induce activation of translational regulators in a few cell types, including myotubes (Shen et al., 2005) and embryonic stem cells (Wang et al., 2001). Once phosphorylated, 4EBP1 is released from inhibiting the initiation factor eIF4E, which together with eIF4A and eIF4G are key components in promoting protein translation (Prevot et al., 2003). Downstream effectors of PI3K/Akt and MEK/ERK pathways, including mTOR and Erk, activate important regulators of translation such as eIF4G and eEF2 (Holz et al., 2005; Pyronnet et al., 1999). Phosphorylation of eEF2 inhibits its activity by preventing it from binding to the ribosome thus affecting the elongation cycle of protein synthesis. We found that IGF-1 in combination with PTEN inhibitor induced the eIF4G and eEF2 activation in addition to 4EBP1, Erk1/2 and S6rp phosphorylation, which are required for protein synthesis through IGF-1 (Bibollet-Bahena and Almazan, 2009). During myelination, OLGs require large amounts of protein synthesis even early on in differentiation to give rise to specialized membrane sheets. Higher levels of MBP, one of the major CNS myelin proteins, were detected after phen treatment followed by IGF-1 in OLGs as well as in co-cultures with DRGNs. Therefore, while not addressed in this work, it is reasonable to speculate from our data that the pharmacological inhibition of PTEN may also augment protein synthesis by amplifying the IGF-1 effects in OLPs.

Phosphatases such as PTEN are key to the termination of IGF-1 signal transduction. Once PTEN is inhibited, the PI3K pathway requires a ligand such as IGF-1 to bind a receptor with tyrosine kinase activity, and as a result, to be activated, suggesting that phen cannot trigger downstream signaling by itself. When applied alone in different concentrations and in varying time courses, phen enhanced OLP proliferation, and increased differentiation and myelination. While short-term treatments were given after growth factor withdrawal, OLPs express IGF-1 mRNA, suggesting, at least in part, an autoregulatory development (Shinar and McMorris 1995). OLPs can receive multiple signals from neurons through growth factors that modulate proliferation, survival and differentiation. It has been suggested that IGF-1 may play an important role in regulating the expression of different tyrosine kinase receptors in DRGNs through the Erk and PI3K/Akt pathways (Li et al., 2013). OLGs also express receptors for other growth factors released by neurons and astrocytes, including platelet-derived growth factor (PDGF) (Barres et al., 1993), fibroblast growth factor (FGF) (Redwine et al., 1997), NGF (Sofroniew et al., 2001), and neuregulin (Vartanian et al., 1997). Neuregulin-1 (Nrg1) signaling, for example, regulates neuronal development, migration, myelination, and synaptic maintenance by activating epidermal growth factor tyrosine kinase receptors (Liu et al., 2007). Studies in transgenic mice showed that neuronal overexpression of Nrg1 causes hypermyelination, whereas diminished Nrg1 expression induces hypomyelination (Michailov et al., 2004). Most of the pathways stimulated by these growth factors through their tyrosine kinase receptors lead ultimately to PI3K/Akt and MEK/ERK stimulation. Therefore, it suggests that basal levels of growth factors released into the milieu may have their effects potentiated since they bind those receptors where PTEN has its actions hindered by phen. It appears that in this report the pharmacological inhibition of PTEN in the absence of 20 ng/mL IGF-1 favored activation of multiple signaling pathways, which moderately influenced OLP proliferation and differentiation as well as myelination. In our co-culture system, inhibition of PTEN in partnering cells such as DRGNs may have also intensified the effects of growth factors including IGF-1 and indirectly affected OLP differentiation giving rise to an overall increase in myelination.

Conditional knockout of PTEN in mouse OLGs showed hypermyelination with myelin abnormalities and no improvements in axonal remyelination (Goebbels et al., 2010; Harrington et al., 2010), demonstrating adverse effects of chronic Pten inactivation. These may be related to the evidence that the regulation of PIP2 availability at the cell surface is coupled to MBP synthesis in OLGs, suggesting that permanent PTEN suppression may potentially affect the formation and maintenance of myelin (Musse et al., 2008; Nawaz et al., 2009). Recent evidence has suggested that the pharmacological inhibition of PTEN or its temporary knockdown in neurons promoted regenerative axonal outgrowth following severe nerve injuries (Christie et al., 2010; Park et al., 2008). Our work in conjunction with the findings in neural injury suggests that the pharmacological inhibition of PTEN may be an interesting approach to apply to animal models of CNS diseases to improve remyelination of axonal lesions as well as boost regenerative outgrowth of neurons in spinal cord injury. Furthermore, in our experiments we demonstrate that short-term inhibition of PTEN with relatively low concentrations of inhibitor was efficient to enhance myelination, an essential aspect to consider when blocking a tumor gatekeeper molecule. Approaches that encourage myelination by transiently suppressing a potential obstacle, in combination with other strategies including growth factors, may provide promising therapies to promote the functional recovery of white matter injuries and demyelinating diseases such as MS.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  9. Supporting Information

The authors thank Jun Fang and Jacynthe Laliberté for technical assistance and Dr. Debra Fulton for critical reading and editorial help with the article.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. References
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
glia22584-sup-0001-suppinfo01.tif10124KSupporting Information
glia22584-sup-0002-suppinfo02.tif2586KSupporting Information
glia22584-sup-0003-suppinfo03.tif21198KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.