Inhibition of p38 mitogen-activated protein kinase interferes with cell shape changes and gene expression associated with Schwann cell myelination

Available online at www.sciencedirect.com R

Experimental Neurology 183 (2003) 34 – 46

www.elsevier.com/locate/yexnr

Inhibition of p38 mitogen-activated protein kinase interferes with cell shape changes and gene expression associated with Schwann cell myelination Gabriela Fragoso,a Janice Robertson,b Eric Athlan,b Emily Tam,b Guillermina Almazan,a and Walter E. Mushynskib,* a

Department of Pharmacology and Therapeutics, McGill University, Montreal, Quebec, Canada b Department of Biochemistry, McGill University, Montreal, Quebec, Canada Received 14 November 2002; revised 21 January 2003; accepted 22 January 2003

Abstract In the present study we demonstrate that p38, a member of the mitogen-activated protein kinase (MAPK) family, is essential for ascorbate- and laminin-induced myelination in Schwann cell– dorsal root ganglion neuron cocultures. The inhibitory effect of the specific p38 blockers, PD 169316 and SB 203580, on ascorbate-induced myelination was exerted during the early stages (1–2 days) of ascorbate treatment. Inhibition of p38 was further shown to prevent the alignment of Schwann cells along axons in laminin-treated cocultures. The addition of laminin to Schwann cell– dorsal root ganglion neuron cocultures stimulated phosphorylation of p38, thereby demonstrating a link between laminin-induced myelination and p38 activation. Similarly, the small heat shock protein, Hsp27, which is phosphorylated by MAPKAPK2, a downstream substrate of p38, was phosphorylated in response to the addition of laminin to the cocultures. The p38 inhibitors did not affect the proliferation or survival of Schwann cells in the cocultures as assessed by BrdU incorporation and total cell counts. However, p38 inhibition interfered with an early stage in myelination, thereby preventing ascorbate-induced increases in the levels of mRNAs encoding MBP, MAG, and P0 and reducing laminin deposition. These results indicate that activation of p38 by a signaling pathway(s) involving laminin and appropriate integrin receptor(s) is required for the alignment of Schwann cells with axons that precedes myelination. © 2003 Elsevier Science (USA). All rights reserved. Keywords: DRG; HSP27; Integrins; Laminin; Myelin; Myelin basic protein; Myelinating cultures; p38 MAPK; Schwann cells; Sensory neurons

Introduction Peripheral nerve myelination proceeds through several discernible stages including the early establishment of a one-to-one relationship between a Schwann cell (SC)1 and a

* Corresponding author. Department of Biochemistry, McGill University, 3655 Promenade Sir William Osler, Room 914, Montreal, Quebec H3G 1Y6, Canada. E-mail address: [email protected] (W.E. Mushynski). 1 Abbreviations used: DRGN, dorsal root ganglion neuron; ECM, extracellular matrix; ERK, extracellular signal regulated kinase; Hsp27, small (27 kDa) heat shock protein; MAPK, mitogen-activated protein kinase; MAPKAPK, mitogen-activated protein kinase-activated protein kinase; MEK, MAPK/ERK kinase; MBP, myelin basic protein; NGF, nerve growth factor; SC, Schwann cell.

large caliber axon, growth of the inner membrane around the axon, and compaction of the spiral lamellae (Bunge et al., 1982; Bunge, 1993; Mirsky and Jessen, 1996). Experiments with primary cell cultures have shown that in addition to its contact with an axon destined for myelination, a SC must adhere to a basal lamina for myelination to be initiated (Carey and Todd, 1987; Eldridge et al., 1989; FernandezValle et al., 1993). Hence, myelination in SC and dorsal root ganglion neuron (DRGN) cocultures requires the addition of ascorbate to the culture medium to allow synthesis of collagen IV, which serves as a scaffold for the deposition of other constituents of the basal lamina (Eldridge et al., 1987; Woodley et al., 1983). Laminin appears to be a critical component of this basal lamina, as soluble laminin (but not collagen IV) supports

0014-4886/03/$ – see front matter © 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0014-4886(03)00101-8

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

myelination when added to culture medium lacking ascorbate (Eldridge et al., 1989; Podratz et al., 2001). Indeed, mutations in the human and mouse ␣-chain of the laminin-2 subtype are associated with peripheral nerve dysmyelination (Shorer et al., 1995; Uziyel et al., 2000; Xu et al., 1994). The laminin likely exerts this effect by interacting with ␤1 integrin-containing receptor(s), since treatment of SC/ DRGN cocultures with function-blocking ␤1 integrin antibody blocks myelination (Fernandez-Valle et al., 1994) and conditional disruption of ␤1 integrin in SCs has demonstrated its requirement in the first steps of axonal ensheathment (Feltri et al., 2002). However, some SCs in ␤1 integrin-null mutant mice form normal myelin indicating that other integrin receptors are involved in myelination (Feltri et al., 2002). Several studies have shown that the main laminin receptors in myelinating SCs are dystroglycan and the integrins, ␣6␤1 and ␣6␤4 (Einheber et al., 1993; Feltri et al., 1994; Previtali et al., 2001; Yamada et al., 1996). The SC shape changes and alignment with axons that precede myelination can be induced by exogenous extracellular matrix (ECM) (Carey et al., 1986) and are blocked by disrupting actin polymerization with cytochalasin D (Fernandez-Valle et al., 1997). The implication of laminin, ␤1integrin-containing receptor(s), and the actin cytoskeleton in early stages of myelination suggests the involvement of signaling pathways known to modulate myelin gene expression and actin dynamics. This likely includes integrinlinked signaling pathways involving focal adhesion kinase, Fyn kinase, paxillin, and the Rho GTPases, Rac and Cdc42 (Chen et al., 2000; Hall, 1998; Schoenwaelder and Burridge, 1999; Terashima et al., 2001). Possible downstream elements of pathways activated by Rac and Cdc42 include p21-activated kinase (Pak) and members of the mitogenactivated protein kinase (MAPK) family, the extracellular signal regulated kinases, c-Jun N-terminal kinases, and p38 kinases (Ivaska et al., 2002; Ono and Han, 2000; Teramoto et al., 1996; Vojtek and Cooper, 1995). The present report documents our finding that specific inhibition of p38 in SC/DRGN cocultures blocked an early stage in myelination induced by ascorbate, by soluble laminin, or by exogenous ECM. Inhibition of p38 was shown to prevent the morphological reorganization of SCs that precedes myelination, resulting in blockage of myelin gene expression.

Materials and methods Materials Dulbecco’s modified Eagle’s medium, phosphate-buffered saline (PBS), 7.5% bovine serum albumin (BSA) fraction V, and penicillin/streptomycin mix were from Gibco/ BRL (Burlington, ON); nerve growth factor (2.5S) was purchased from Prince Laboratories (Toronto, ON, Canada). Extracellular matrix gel, laminin peptide (Cys-laminin A chain 2091-2108), Triton X-100, and monoclonal NFH

35

antibody N52 were from Sigma–Aldrich (Oakville, ON), while PD 169319, SB 202474, SB 203580, and U0126 were purchased from Calbiochem (San Diego, CA). Radiolabeled [␥-32P]CTP was provided by Mandel (Guelph, ON). Antibromodeoxyuridine (anti-BrdU) IgG1 mAb was purchased from ICN Canada Ltd. (Montreal, QC), while anti-myelin basic protein (MBP) was from Sternberger Monoclonals (Lutherville, MD). Phosphospecific p38 (Thr 180 and Tyr 182) antibody was from New England Biolabs (Mississauga, ON), and 4,6-diamidino-2-phenylindole dihydrochloride (DAPI) was from Polysciences Inc. (Warrington, PA). Immobilon-P membrane was from Millipore (Mississauga, ON) while the ECL Western Blotting Detection kit was from New England Nuclear (Oakville, ON). Secondary antibodies used for immunofluorescence were purchased from Southern Biotechnology (Birmingham, AL) or Jackson Immunoresearch Laboratories (West Grove, PA). All other reagents were obtained from Fisher (Montreal, QC) or VWR (Mississauga, ON). Cell culture Purified DRGN/SC cocultures were prepared using methods described previously (Giasson and Mushynski, 1996). DRGs were obtained from Sprague–Dawley rat embryos at 15–16 days gestation, dissociated with trypsin, and plated onto rat tail collagen-coated wells. The cultures were maintained in serum-free N1 medium until their use after 21 days in culture. Myelination was then initiated by the addition of ascorbate (50 ␮g/ml), which was replenished every 2 days with fresh culture medium. In defined medium induction of myelination by ascorbate did not occur prior to 17–19 days in culture. Western blot analysis After treatment, cells grown in 24-well culture plates were washed in PBS and harvested in 2% SDS, 62.5 mM Tris–HCl (pH 6.8) and placed in a boiling water bath for 5 min. Protein content in the cell lysates was determined using the bicinchoninic acid assay, after which aliquots with the appropriate amount of protein were mixed with 1/3 volume of 3 ⫻ SDS sample buffer (6% SDS, 30% glycerol, 15% ␤-mercaptoethanol, 188 mM Tris–HCl, pH 6.8, 0.03% bromphenol blue). Samples were resolved by SDS polyacrylamide gel electrophoresis, on a 6 or 12% polyacrylamide gel (Laemmli, 1970), and transferred to Immobilon-P membrane. The membrane blots were blocked for 1 h with 5% dry milk in Tris-buffered saline containing 0.1% Tween 20 and then incubated with primary antibody. Then the blots were incubated with horseradish-peroxidase-conjugated secondary antibody and developed by chemiluminescence using an ECL Western Blotting Detection kit. The signals were quantified with a Master Scan Interpretative densitometer. To normalize for equal loading and protein transfer, the membranes were stripped and reprobed with an antibody for ␤-actin.

36

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

Fig. 1. Transmission electron micrograph of a transverse section of a myelinating SC/DRGN coculture 10 days after addition of ascorbate. Arrows indicate compact myelin sheaths surrounding axons (asterisk). The insert in the upper left corner shows compact myelin with multiple lamellae at higher magnification. The arrowhead points to basal lamina. Scale bar ⫽ 1.5 ␮m.

Microscopy Cocultures plated on 12-mm glass coverslips were fixed with 4% paraformaldehyde in PBS, pH 7.4, for 20 min at room temperature and then with methanol for 5 min at ⫺20°C. Afterward the cells were washed three times with PBS and blocked for 15 min in PBS containing 0.2% BSA, 5% goat serum, 5% rabbit serum, and 0.2% Triton X-100. Monoclonal anti-MBP (IgG2b) and anti-NF (N52, IgG1) or polyclonal rabbit anti-laminin antibodies were diluted in the same solution and applied for 45 min at room temperature. The secondary goat anti-mouse IgG2b-TxR and goat antimouse IgG1– or donkey anti-rabbit–FITC conjugates were

applied for 20 min at room temperature. Cellular nuclei were examined by using the DAPI dye. The coverslips were mounted on glass slides and examined under a Leitz Diaplan epifluorescent microscope or a confocal microscope. Myelinating cell cultures were fixed, embedded, and sectioned for electron microscopy as previously described (Therien and Mushynski, 1976). Cell proliferation (2-bromodeoxyuridine incorporation) In order to determine the proportion of SCs undergoing mitosis, BrdU was added to culture medium at a final concentration of 10 ␮M for 24 h. Paraformaldehyde pre-

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

fixed cells growing on coverslips were immersed in 70% ethanol at room temperature for 20 min followed by denaturation with 1 M HCl and neutralization with 0.1 M sodium tetraborate, pH 9. Cells were immunostained with antiBrdU (1:40) followed by goat anti-mouse IgG1 conjugated to FITC. In all experiments, cell nuclei were stained for 10 min with 500 ng/ml DAPI to determine the total number of cells per microscopic field. Coverslips were mounted in Immuno-mount and cell numbers for two separate experiments were determined by examining under oil immersion at 40⫻, 3– 6 fields for each of three coverslips per culture. The number of BrdU-positive cells is expressed as a percentage of the total number of DAPI⫹ nuclei. Northern blot analysis Total RNA was extracted from the cultured cells as previously reported (Cohen et al., 1996). RNA pellets were resuspended in 50 ␮l of 50% formamide/2.2 M formaldehyde/20 mM N-morpholinopropanesulfonic acid, denatured for 30 min at 65°C, and resolved by electrophoresis on a 1.3% agarose–formaldehyde gel. RNA was transferred to a nylon membrane (Hybond-N, Amersham) using standard Northern blot procedures. The cDNA probes for rat MBP (Mentaberry et al., 1986), MAG (Salzer et al., 1987), P0 (Lemke and Axel, 1985), and ␤-tubulin (Farmer et al., 1986) were labeled with [␣-32P]dCTP using a random primer kit from Pharmacia to a specific activity of 108 cpm/␮g DNA. Membranes were hybridized at 65°C for 24 h with 106 cpm of cDNA per ml of hybridization solution (0.5 M sodium phosphate buffer, pH 7, 7% SDS, 1 mM EDTA, and 0.5% BSA). Membranes were washed for 1 h at 65°C in 0.1⫻ standard saline citrate, 0.1% SDS. Membranes were then exposed to Kodak XAR-5 AR film for 24 h at ⫺80°C with an intensifying screen. To standardize for equal RNA loading and transfer, the membranes were stripped and reprobed with a ␤-tubulin cDNA.

Results Characterization of myelinating Schwann cell/neuron cocultures SCs cocultured with rat embryonic DRGN proliferated rapidly on contract with axons so that 3 weeks after plating, the cultures were fully populated and proliferation of SCs had slowed down. At this time myelination was initiated by the addition of ascorbate to the cocultures. The SCs then began to elongate and ensheath the axons over a period of days, and robust myelination was evident after 10 days of ascorbate treatment. The electron micrograph in Fig. 1 shows a transverse section of a representative coculture containing several myelinated axons, with other SCs engulfing axons at the premyelinating stage. The higher magnification insert in Fig. 1 shows a typical basal lamina, indi-

37

Fig. 2. Time course for stimulation of laminin deposition and MBP expression in ascorbate-treated SC/DRGN cocultures. Three-week-old SC/ DRGN cocultures exposed to 50 ␮g/ml ascorbate in N1 medium from 1 to 10 days were harvested at the same time and subjected to immunoblot analysis. Equal amounts of protein (30 ␮g) were loaded in each lane Top: Representative immunoblot probed with laminin-antibody. Middle: All isoforms of MBP increased during ascorbate-induced SC differentiation. Bottom: Actin levels in different lanes to monitor sample loading. The first lane, indicated by a “C,” is from control cells grown in the absence of ascorbate and harvested at the same time as the other samples. The number above each lane indicates the days of ascorbate treatment.

cated by an arrowhead, as well as multiple myelin lamellae in the cross-section of a myelinated axon. Changes in expression of MBP and laminin, a component of the basal lamina, were monitored by Western blot analysis of ascorbate-treated SC/DRGN cocultures over a 10-day period. The Western blots in Fig. 2 show a correlation between ascorbate-induced synthesis of laminin and MBP. After 2 days of ascorbate treatment there was an evident increase in laminin content, which reached a maximum at 6 days. Increased laminin synthesis (or deposition) was paralleled by increased expression of MBP, since all isoforms of MBP were evident after 2 days of ascorbate treatment with the maximal increase being seen at 9 to 10 days of treatment, 3– 4 days later than for laminin. This suggests that laminin deposition precedes myelin formation. Peripheral myelination is blocked by p38 inhibitors Inhibitors of p38 were tested to determine their effect on SC/DRGN cocultures during the myelination phase. Fig. 3 shows micrographs of myelinating cocultures double immunostained with anti-MBP (Texas Red, Figs. 3A, C, and E) and anti-laminin antibody (FITC, Figs. 3B, D, and F). Micrographs of the positive, ascorbate-treated control show extensive MBP (Fig. 3A) and laminin expression (Fig. 3B). On the other hand, cocultures treated with ascorbate plus PD 169316 (Figs. 3C and D) showed a ⬃90% reduction in MBP expression based on measurement of myelinated internodes. Laminin organization took on a granular appearance (Fig. 3D) compared to the smooth staining in untreated cultures (Fig. 3B). The MEK1-specific inhibitor, U0126, also reduced MBP expression, by about 75% (Fig. 3E), but did not modify laminin distribution (Fig. 3F). The effect of PD 169316 on neuronal integrity was tested by treating cocultures for 10 days with ascorbate and the p38 blocker, followed by immunostaining with N52 monoclonal antibody, which is specific for the heavy neurofilament subunit. The

38

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

39

Fig. 4. Inhibition of p38 reduces MBP synthesis without decreasing neurofilament expression. Neurofilament expression (A, B) and myelination (C, D) visualized in SC/DRGN cocultures treated with ascorbate for 10 days in the absence (A, C) or presence of 5 ␮M PD 169316 (B, D). Cells were stained with N52/FITC (A, B) and anti-MBP/Texas Red (C, D). The arrowhead points to DRGN cell bodies stained with anti-neurofilament antibody. The dull red color of DRGN cell bodies in (D) is a staining artifact.

micrographs in Fig. 4 show that neurofilament staining was not altered by the p38 inhibitor (Fig. 4B) upon comparison with the culture treated with ascorbate only (Fig. 4A), while MBP expression was reduced (Fig. 4D) compared to the positive control (Fig. 4C). Note that the dull red staining of neuronal cell bodies by Texas Red in Fig. 4D is a background artifact and does not represent labeling of MBP. The effect of p38 inhibition is not due to cell death or altered proliferation rate Confluent SC/DRGN cocultures were treated with ascorbate and different MAPK inhibitors from 1 to 10 days. Immunocytochemical detection of BrdU incorporation using a specific monoclonal antibody (Fig. 5B) showed that

the increase in SC proliferation that accompanied ascorbate treatment was inhibited by U0126 whereas PD 169316 had little or no effect. Quantification of cell number following DAPI staining (Fig. 5C) showed a marked decrease in total number of cell nuclei as well as a greater incidence of nuclear condensation or fragmentation indicative of apoptosis in cultures treated with 10 ␮M U0126, compared to those treated with 5 ␮M SB 203580 or PD 169316. Treatment of cocultures with 10 ␮M U0126 resulted in a 45–50% reduction in the number of SCs after 5 to 10 days. These results indicate that inhibition of peripheral myelination by p38 inhibitors was not due to a cytotoxic effect or a decrease in the numbers of SCs. This conclusion has been verified in preliminary studies involving TUNEL assay of cocultures treated with 5 or 10 ␮M SB 203580 (results not shown).

Fig. 3. Reduced MBP expression and altered laminin deposition induced by p38 inhibitors. SC/DRGN cocultures were grown on rat tail collagen for 3 weeks in N1 medium followed by 10 days in N1 medium plus 50 ␮g/ml ascorbate with or without p38 or MEK1 inhibitors. Double staining was with anti-MBP/Texas Red (A, C, E) and anti-laminin/FITC (B, D, F). Micrographs taken by confocal microscopy show the induction of myelination by ascorbate in the presence or absence of MAPK inhibitors. Note the myelinated internodes in (A) correlating with laminin deposition in (B) in cultures treated with ascorbate only. C and D show the inhibitory effects of the p38 inhibitor, PD 169316 (5 ␮M), on MBP expression (C) and laminin deposition (D) in ascorbate-treated cultures. On the other hand, U0126 (10 ␮M), a MEK1-specific inhibitor, reduced MBP expression (E) without altering the pattern of laminin deposition (F). The bar in F is equal to 50 ␮m. Quantification of MBP expression was achieved by counting myelinated internodes in a total of six different microscope fields for each condition. With the ascorbate-treated sample set at 100% (A), relative expression of MBP in the culture treated with ascorbate plus PD 169316 was 5.7 ⫾ 1.1% (C) while the value for the culture treated with U0126 was 25.1 ⫾ 3% (E).

40

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

Fig. 5. The effect of p38 inhibitors on MBP expression is not due to changes in SC proliferation or survival. A: Micrographs of DAPI-stained cocultures with control (a), 50 ␮g/ml ascorbate plus 5 ␮M PD 169316 (b), ascorbate plus 5 ␮M SB 203580 (c), and ascorbate plus 10 ␮M U0126 (d). The arrowheads point to fragmented or condensed nuclei. B: Histogram showing the effects of MAPK inhibitors on SC proliferation. BrdU-positive cell counting was done after 1 and 5 days treatment with MAPK inhibitors. SC proliferation was inhibited at 1 and 5 days by 10 ␮M U0126, but not by SB 203580 or PD 169316. C: Histogram showing total cell counts with DAPI staining after 5 and 10 days of treatment with MAPK inhibitors. Both graphs show the mean ⫾ SEM of three separate experiments. **P ⬍ 0.01 compared with VC.

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

41

the myelination process. In view of the known involvement of p38 in actin filament dynamics (Ono and Han, 2000) and actin cytoskeleton rearrangement in early myelination (Fernandez-Valle et al., 1997), we set out to determine whether p38 inhibition affects the extensive morphological reorganization of Schwann cells that precedes myelination induced by treating cocultures with ECM or laminin (Carey et al., 1986; Eldridge et al., 1989). Exogenous extracellular matrix and laminin peptide induce MBP synthesis We first verified that myelination in our cocultures could be induced by treatment with ECM or laminin. SC/DRGN

Fig. 6. Inhibition of p38 reduces MBP, MAG, and P0 mRNA levels. Identification of mRNA encoding myelin proteins in SC/DRGN cocultures by Northern blot analysis after 2 to 10 days of treatment with ascorbate (VC) (50 ␮g/ml) in the presence or absence of the p38 inhibitor, PD 169316. SC/DRGN cocultures were maintained in N1 medium for 3 weeks after which ascorbate was added with or without PD 169316. The series of three samples for each interval from 2 to 10 days shows the control (no ascorbate), ascorbate-treated, and ascorbate-treated plus PD 169316 (5 ␮M) as indicated at the bottom of each lane. A more intense signal for each mRNA was seen in cultures treated with ascorbate and this effect was reduced in the presence of PD 169316. Tubulin mRNA levels remained constant under all conditions.

Ascorbate-induced myelin gene expression and laminin synthesis are reduced by p38 inhibition The effect of p38 inhibition on steady-state levels of mRNAs encoding myelin proteins was determined by purifying total RNA from SC/DRGN cocultures treated with ascorbate for 2 to 10 days in the presence and absence of PD 169316. Northern blot analysis showed that MBP, myelinassociated glycoprotein (MAG), and P0 mRNA levels were increased by ascorbate treatment, but this effect was blocked at all stages by 5 ␮M PD 169316 (Fig. 6). The ascorbate-induced increase in MBP is dramatically reduced by the p38-specific inhibitors, PD 169316 and SB 203580, but not by the inactive analog, SB 202474 (Fig. 7A). This effect also involves a less pronounced reduction in laminin expression. In contrast, the MEK1 inhibitor, U0126, reduced MBP synthesis, but had no significant effect on laminin (Fig. 7A). MBP expression was reduced in a timedependent manner by PD 169316, which completely blocked ascorbate-induced MBP synthesis when cocultures were treated simultaneously with the p38 blocker for the first 6 days of the 10-day interval and partially blocked MBP synthesis with 4 days of treatment (Fig. 7B, left). It is noteworthy that PD169316 was effective in blocking MBP expression only when it was present at the start of ascorbate-induced SC differentiation, since its addition 2–3 days later had little or no effect (Fig. 7B, right). This indicates that inhibition of p38 affects an early, essential stage(s) in

Fig. 7. Effect of p38 inhibitors on ascorbate-induced MBP synthesis. (A) SC/DRGN cocultures after 3 weeks in culture were treated with ascorbate (VC) with or without the MEK1 inhibitor, U0126 (10 ␮M), or the p38 inhibitors SB 203580 (5 ␮M) or PD 169316 (5 ␮M) or the inactive negative control agent SB 202474 (5 ␮M). After 10 days treatment, cells were harvested and subjected to immunoblot analysis with MBP antibody. The lane labeled “C” is the control without ascorbate harvested at day 10 of treatment. The numbers under the individual lanes of the laminin blot denote the densitometric units detected in the corresponding laminin panel. (B) Left: SC/DRGN cocultures in medium containing 50 ␮g/ml ascorbate were exposed to 5 ␮M PD 169316 for different time periods (4, 6, 8, and 10 days) beginning at day zero. Right: Samples from cells treated with ascorbate for 10 days and with 5 ␮M PD 169316 beginning at day 1 (1–10), day 2 (2–10), day 3 (3–10), or day 4 (4 –10). All samples were harvested at the same time (day 10) and subjected to immunoblot analysis of MBP. ␤-Actin levels (bottom) in the different samples remained constant.

42

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

(Fig. 9C), which was added to the cultures at the same time as laminin. Addition of ascorbate to the cultures produced a more gradual change in SC morphology, which was less evident than that of laminin after 2 days of treatment (Fig. 9D). Laminin increases p38 and Hsp27 phosphorylation

Fig. 8. A: Laminin peptide and extracellular matrix (ECM)-induced MBP expression in SC/DRGN cocultures. Cells maintained for 3 weeks in N1 medium (0.5 ml/well) were treated with different concentrations of laminin peptide (Cys-laminin A chain; 1–30 ␮g/well, top) or ECM (0.6 –20 ␮g/ well, middle) in medium lacking ascorbate. Cells were harvested after 10 days of treatment and subjected to immunoblot analysis with 30 ␮g of protein loaded in each lane. The number at the top of each lane is the amount of laminin peptide or ECM protein added to 0.5 ml of medium in each well. (Bottom) MBP expression induced by addition of 10 ␮g/well of ECM protein was markedly reduced by 5 ␮M PD 169316 (PD169), while 10 ␮M U0126 had a much weaker effect. Duplicate 30 ␮g samples of protein were loaded in each case. C is the control in N1 medium with no additives.

cocultures were analyzed by Western blotting after being exposed for 10 days to different concentrations of laminin peptide (Fig. 8, top) or to ECM (Fig. 8, middle). Both laminin (not shown) and laminin peptide as well as ECM were able to induce MBP synthesis in a concentrationdependent manner, although ECM was the more potent of the two. The appearance of only one MBP species in samples from laminin peptide-treated cultures is due to lower levels of MBP induction so that only the more prominent isoform is detected. ECM-induced MBP synthesis was significantly reduced by 5 ␮M PD 169316 but not by U0126 (Fig. 8, bottom), indicating that myelination events initiated through interaction between a component(s) of the ECM, likely laminin, and integrins occurred downstream of p38. P38 inhibitors block laminin-induced changes in Schwann cell morphology It has been shown that addition of soluble laminin to cocultures induces SCs to assume a more elongated morphology and to align with axons (Carey et al., 1986; Eldridge et al., 1989). In order to determine whether these morphological and organizational changes occurring during early myelination could be affected by p38 inhibition, we treated SC/DRGN cocultures with laminin alone or in the presence of SB 203580. As shown in Fig. 9, laminin-induced SC elongation was evident after 2 days of treatment (Fig. 9B). This effect was blocked by 5 ␮M SB 203580

p38 phosphorylates the downstream kinase, MAPKAPK2/3 (Stokoe et al., 1992), which can subsequently phosphorylate the small heat shock protein, Hsp27. Hsp27 is involved in the modulation of actin filament assembly and stability (Lavoie et al., 1995). SC/DRGN cocultures were exposed to exogenous laminin and levels of active (phosphorylated) p38 and phosphorylated Hsp27 were measured by Western blot analysis with phosphoepitope-specific antibodies. Laminin at a concentration of 50 ␮g/ml increased phosphorylated p38 and Hsp27 levels for a period greater than 1 h (Fig. 10A). Note in Fig. 10B that the laminininduced increase in Hsp27 phosphorylation is reduced by 5 ␮M SB 203580.

Discussion In this study, we have shown that p38 plays a fundamental role in PNS myelination. Experimental evidence was obtained using p38-specific inhibitors, PD169316 and SB203580, which reduced steady-state levels of mRNA for the myelin proteins MAG, MBP, and P0 and blocked myelination induced by ascorbate or ECM. These p38 inhibitors were found to reduce laminin content and alter the arrangement of laminin in PNS cultures without affecting the viability or proliferation of SCs. Addition of soluble laminin to the cocultures, previously shown to initiate myelination, caused activation of p38 and phosphorylation of Hsp27 while p38 inhibitors blocked early laminin-induced SC morphological changes and myelination promoted by added ECM in an ascorbate-free environment. These results suggest that early myelination events initiated by interaction of laminin with SC receptors, presumably integrins, occur downstream of p38 and involve changes in cytoskeletal dynamics and lead to SC–axon association required for the induction of genes encoding myelin components. The evidence we have presented implicating p38 in SC myelination adds to the growing list of functions involving this important group of MAPKs in mammalian cells, which includes stress-induced signaling and myogenic and neuronal differentiation (Hansen et al., 2000; Ono and Han, 2000). Of the four known p38 kinases, only p38␣ and p38␤ are inhibited by SB 203580 (Ono and Han, 2000), hence our reference to p38 excludes p38␥ and p38␦. In our study, treatment of DRGN/SC cocultures with p38 inhibitors did not alter neuronal morphology or neurofilament expression in contrast to their effects on myelination. The reduced myelination caused by p38 inhibition was not due to cell death at the inhibitor concentration that was used. However,

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

43

Fig. 9. SB 203580 impedes laminin-induced SC morphological changes. Phase contrast micrographs of SC/DRGN treated with 5 ␮g/ml laminin plus or minus 5 ␮M SB 203580 or ascorbate for 2 days. A: control; B: cells treated with laminin; C: cells treated with laminin plus SB 203580; D: cells treated with 50 ␮g/ml ascorbate. The arrowheads point to SCs while arrows point to sensory neurons. Note in B the alignment of spindle-shaped SCs along axonal neurites.

we observed that higher concentrations of SB 203580 (20 ␮M) were somewhat toxic to SCs (not shown). In addition, inhibition of p38 did not block proliferation of SCs following addition of ascorbate to the cultures, further providing support for the selective effect of the inhibitors on differentiation of SCs leading to myelination of DRGN. Others have shown that p38 inhibition does not alter SC proliferation in medium containing fetal calf serum, while decreasing the effect of endothelin-1 on both [3H]-thymidine incorporation and cell shape changes (Berti-Mattera et al., 2001). The inhibition of MEK1, the immediate upstream activator of ERK1/2, with U0126 (at 10 ␮M) in SC/DRGN cocultures also caused a reduction in the formation of myelinated internodes, as demonstrated by immunofluorescence microscopy and Western blot analysis using MBP as a marker. However, MEK1 inhibition also had a strong anti-mitotic effect and reduced SC numbers. This observation contrasts with the report of Maurel and Salzer (2000), showing that U0126 did not cause cell death in SC/DRGN cocultures. This discrepancy may be due to our use of culture medium that did not contain serum, which may have

a protective effect against this inhibitor. In addition, the observed anti-mitotic effect of U0126 contributes to the reduction in SC number shown in Fig. 5B. The role of p38 in myelination appears to be twofold, involving both the actin cytoskeleton and transcription, since blockage of laminin-induced elongation and axonal alignment of SCs by specific chemical inhibitors of p38 prevents the expression of genes encoding myelin proteins. The relevance to myelination of the laminin-induced increase in p38 phosphorylation (Fig. 10) is that it complements our finding that myelination initiated by adding ECM gel to SC/DRGN cocultures is blocked by inhibition of p38. Laminin is also involved in SC–axon interaction, as previously demonstrated in studies with a mouse strain (dy/dy) that has reduced levels of laminin-2 (Uziyel et al., 2000). This animal model of merosin-deficient congenital muscular dystrophy (Tome et al., 1994) harbors mutations in the gene encoding the ␣2 laminin chain and presents a dysmyelinating peripheral neuropathy. Interactions between laminin-2, located in the PNS basal lamina, and SC ␤1 integrin-containing receptors appear to

44

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

Fig. 10. Laminin-induced phosphorylation of p38 and Hsp27. A, SC/ DRGN cocultures were incubated with 50 ␮g/ml of laminin in N1 medium for 5 min to 5 h and subjected to Western blot analysis. A significant time-dependent increase in p38 and Hsp27 phosphorylation was obtained over a period greater than 1 h. Equal amounts of protein (30 ␮g) were loaded in each lane. B, SC/DRGN cocultures were incubated for 30 min in N1 medium alone (control) or N1 containing 50 ␮g/ml laminin (Lam) or N1 containing 50 ␮g/ml laminin plus 5 ␮M SB 203580 (Lam/SB). Equal amounts of protein (25 ␮g) were loaded in each lane and ␤-actin staining provides an index of protein loading.

be of critical importance in early signaling between axons and SCs to initiate myelination. The increase in p38 phosphorylation in cocultures treated with laminin concurs with previous reports showing the involvement of ␤1 integrincontaining receptor(s) in early stages of SC myelination (Feltri et al., 2002; Fernandez-Valle et al., 1994; Podratz et al., 2001) as well as their role in activation of p38 (Ivaska et al., 1999). Signaling between laminin-activated ␤1 integrin receptor and p38 may be mediated by Rho family GTPases, Cdc42, and/or Rac1 (Price, et al., 1998), MAPK kinase kinases such as PAK and ACK (Yang, et al., 1999), and the MAPK kinases, MKK3/4/6 (Ono and Han, 2000). The role of Rho family GTPases in regulating the actin cytoskeleton is well known (Hall, 1998) and these signaling components are found in peripheral nerve (Terashima et al., 2001) where they may play a role in myelination. Rho GTPases are activated by guanine nucleotide exchange factors which can themselves be regulated by specific lipids such as phosphatidylinositol-3,4,5-trisphosphate, produced by phosphatidylinositol 3-kinase (PI3K), following activation of membrane receptors, including integrins (Giancotti and Ruoslahti, 1999; Hall, 1998). A recent report has shown that PI3K activity is essential for proliferation and survival of SCs induced by neuregulins and/or other factors produced by axons and also plays a role in the early myelination events (Maurel and Salzer, 2000). Of relevance to this study, IGF-1, which causes PI3K activation in SCs (Delaney et al., 1999), has been shown to promote myelination of peripheral sensory axons (Russell et al., 2000).

Several downstream targets of p38 are known (Ono and Han, 2000), including the first identified substrate, MAPKAPK 2, which phosphorylates Hsp27 (Stokoe et al., 1992). Indeed, addition of laminin to SC/DRGN cocultures caused an increase in Hsp27 phosphorylation concomitant with p38 activation (Fig. 10A), and this was inhibited by SB203580 (Fig. 10B). Phosphorylation/dephosphorylation of Hsp27 modulates actin filament dynamics (Lavoie et al., 1995; Piotrowicz and Levin, 1997) and smooth muscle cell migration (Hedges et al., 1999). Our demonstration that p38 inhibitors block the early laminin-induced changes in SC morphology may well reflect the concomitant inhibition of Hsp27 phosphorylation since disruption of actin polymerization by cytochalasin D also inhibits myelination in SC/ DRGN cocultures (Fernandez-Valle et al., 1997). In addition to preventing the functional alignment of SCs with axons, the negative effect of p38 inhibition on myelin gene expression could in part be due to the known role of p38 signaling in modulating transcription, including that of collagen genes (Ivaska et al., 1999). MAPKAPK 2 phosphorylates transcription factors such as CREB, and ATF1 (Tan et al., 1996) and p38 itself phosphorylates several transcription factors (Ono and Han, 2000). It is interesting to note that p38 inhibition (this study) and disruption of the actin cytoskeleton (Fernandez-Valle et al., 1997) both inhibit transcription of genes encoding myelin proteins. Identifying p38-related transcription factors involved in regulating myelin-specific gene expression and elucidating the role of F-actin in this process will provide major new insights into mechanisms underlying myelination.

Acknowledgments This work was supported by grants from the MS Society of Canada and CIHR (W.M. and G.A.), as well as the American ALS Association and the Motor Neuron Disease Association, UK (J.R.).

References Berti-Mattera, L.N., Harwalkar, S., Hughes, B., Wilkins, P.L., Almhanna, K., 2001. Proliferative and morphological effects of endothelins in Schwann cells: roles of p38 mitogen-activated protein kinase and Ca(2⫹)-independent phospholipase A2. J. Neurochem. 79, 1136 – 1148. Bunge, M.B., Williams, A.K., Wood, P.M., 1982. Neuron–Schwann cell interaction in basal lamina formation. Dev. Biol. 92, 449 – 460. Bunge, R.P., 1993. Expanding roles for the Schwann cell: ensheathment, myelination, trophism and regeneration. Curr. Opin. Neurobiol. 3, 805– 809. Carey, D.J., Todd, M.S., Rafferty, C.M., 1986. Schwann cell myelination: induction by exogenous basement membrane-like extracellular matrix. J. Cell Biol. 102, 2254 – 63. Carey, D.J., Todd, M.S., 1987. Schwann cell myelination in a chemically defined medium: demonstration of a requirement for additives that promote Schwann cell extracellular matrix formation. Brain Res. 429, 95–102.

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46 Chen, L.M., Bailey, D., Fernandez-Valle, C., 2000. Association of beta 1 integrin with focal adhesion kinase and paxillin in differentiating Schwann cells. J. Neurosci. 20, 3776 –3784. Cohen, R.I., Molina-Holgado, E., Almazan, G., 1996. Carbachol stimulates c-fos expression and proliferation in oligodendrocyte progenitors. Brain Res. Mol. Brain Res. 43, 193–201. Delaney, C.L., Cheng, H.L., Feldman, E.L., 1999. Insulin-like growth factor-I prevents caspase-mediated apoptosis in Schwann cells. J. Neurobiol. 41, 540 –548. Einheber, S., Milner, T.A., Giancotti, F., Salzer, J.L., 1993. Axonal regulation of Schwann cell integrin expression suggests a role for alpha 6 beta 4 in myelination. J. Cell Biol. 123, 1223–1236. Eldridge, C.F., Bunge, M.B., Bunge, R.P., 1989. Differentiation of axonrelated Schwann cells in vitro: II. Control of myelin formation by basal lamina. J. Neurosci. 9, 625– 638. Eldridge, C.F., Bunge, M.B., Bunge, R.P., Wood, P.M., 1987. Differentiation of axon-related Schwann cells in vitro. I. Ascorbic acid regulates basal lamina assembly and myelin formation. J. Cell Biol. 105, 1023– 1034. Farmer, S.R., Robinson, G.S., Mbangkollo, D., Bond, J.F., Knight, G.B., Fenton, M.J., Berkowitz, E.M., 1986. Differential expression of the beta-tubulin multigene family during rat brain development. Ann. N. Y. Acad. Sci. 466, 41–50. Feltri, M.L., Graus Porta, D., Previtali, S.C., Nodari, A., Migliavacca, B., Cassetti, A., Littlewood-Evans, A., Reichardt, L.F., Messing, A., Quattrini, A., Mueller, U., Wrabetz, L., 2002. Conditional disruption of beta 1 integrin in Schwann cells impedes interactions with axons. J. Cell Biol. 156, 199 –209. Feltri, M.L., Scherer, S.S., Nemni, R., Kamholz, J., Vogelbacker, H., Scott, M.O., Canal, N., Quaranta, V., Wrabetz, L., 1994. Beta 4 integrin expression in myelinating Schwann cells is polarized, developmentally regulated and axonally dependent. Development 120, 1287– 1301. Fernandez-Valle, C., Fregien, N., Wood, P.M., Bunge, M.B., 1993. Expression of the protein zero myelin gene in axon-related Schwann cells is linked to basal lamina formation. Development 119, 867– 880. Fernandez-Valle, C., Gorman, D., Gomez, A.M., Bunge, M.B., 1997. Actin plays a role in both changes in cell shape and gene-expression associated with Schwann cell myelination. J. Neurosci. 17, 241–250. Fernandez-Valle, C., Gwynn, L., Wood, P.M., Carbonetto, S., Bunge, M.B., 1994. Anti-beta 1 integrin antibody inhibits Schwann cell myelination. J. Neurobiol. 25, 1207–1226. Giancotti, F.G., Ruoslahti, E., 1999. Integrin signaling. Science 285, 1028 – 1032. Giasson, B.I., Mushynski, W.E., 1996. Aberrant stress-induced phosphorylation of perikaryal neurofilaments. J. Biol. Chem. 271, 30404 –30409. Hall, A., 1998. Rho GTPases and the actin cytoskeleton. Science 279, 509 –514. Hansen, T.O., Rehfeld, J.F., Nielsen, F.C., 2000. Cyclic AMP-induced neuronal differentiation via activation of p38 mitogen-activated protein kinase. J. Neurochem. 75, 1870 –1877. Hedges, J.C., Dechert, M.A., Yamboliev, I.A., Martin, J.L., Hickey, E., Weber, L.A., Gerthoffer, W.T., 1999. A role for p38(MAPK)/HSP27 pathway in smooth muscle cell migration. J. Biol. Chem. 274, 24211– 24219. Ivaska, J., Nissinen, L., Immonen, N., Eriksson, J.E., Kahari, V.M., Heino, J., 2002. Integrin alpha 2 beta 1 promotes activation of protein phosphatase 2A and dephosphorylation of Akt and glycogen synthase kinase 3 beta. Mol. Cell. Biol. 22, 1352–1359. Ivaska, J., Reunanen, H., Westermarck, J., Koivisto, L., Kahari, V.M., Heino, J., 1999. Integrin alpha2beta1 mediates isoform-specific activation of p38 and upregulation of collagen gene transcription by a mechanism involving the alpha2 cytoplasmic tail. J. Cell. Biol. 147, 401– 416. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680 – 685.

45

Lavoie, J.N., Lambert, H., Hickey, E., Weber, L.A., Landry, J., 1995. Modulation of cellular thermoresistance and actin filament stability accompanies phosphorylation-induced changes in the oligomeric structure of heat shock protein 27. Mol. Cell. Biol. 15, 505–516. Lemke, G., Axel, R., 1985. Isolation and sequence of a cDNA encoding the major structural protein of peripheral myelin. Cell 40, 501–508. Maurel, P., Salzer, J.L., 2000. Axonal regulation of Schwann cell proliferation and survival and the initial events of myelination requires PI3-kinase activity. J. Neurosci. 20, 4635– 4645. Mentaberry, A., Adesnik, M., Atchison, M., Norgard, E.M., Alvarez, F., Sabatini, D.D., Colman, D.R., 1986. Small basic proteins of myelin from central and peripheral nervous systems are encoded by the same gene. Proc. Natl. Acad. Sci. USA 83, 1111–1114. Mirsky, R., Jessen, K.R., 1996. Schwann cell development, differentiation and myelination. Curr. Opin. Neurobiol. 6, 89 –96. Ono, K., Han, J., 2000. The p38 signal transduction pathway: activation and function. Cell Signal. 12, 1–13. Piotrowicz, R.S., Levin, E.G., 1997. Basolateral membrane-associated 27kDa heat shock protein and microfilament polymerization. J. Biol. Chem. 272, 25920 –25927. Podratz, J.L., Rodriguez, E., Windebank, A.J., 2001. Role of the extracellular matrix in myelination of peripheral nerve. Glia 35, 35– 40. Previtali, S.C., Feltri, M.L., Archelos, J.J., Quattrini, A., Wrabetz, L., Hartung, H., 2001. Role of integrins in the peripheral nervous system. Prog. Neurobiol. 64, 35– 49. Price, L.S., Leng, J., Schwartz, M.A., Bokoch, G.M., 1998. Activation of Rac and Cdc42 by integrins mediates cell spreading. Mol. Biol. Cell. 9, 1863–1871. Russell, J.W., Cheng, H.L., Golovoy, D., 2000. Insulin-like growth factor-I promotes myelination of peripheral sensory axons. J. Neuropathol. Exp. Neurol. 59, 575–584. Salzer, J.L., Holmes, W.P., Colman, D.R., 1987. The amino acid sequences of the myelin-associated glycoproteins: homology to the immunoglobulin gene superfamily. J. Cell Biol. 104, 957–965. Schoenwaelder, S.M., Burridge, K., 1999. Evidence for a calpeptin-sensitive protein-tyrosine phosphatase upstream of the small GTPase Rho. A novel role for the calpain inhibitor calpeptin in the inhibition of proteintyrosine phosphatases. J. Biol. Chem. 274, 14359 –14367. Shorer, Z., Philpot, J., Muntoni, F., Sewry, C., Dubowitz, V., 1995. Demyelinating peripheral neuropathy in merosin-deficient congenital muscular dystrophy. J. Child. Neurol. 10, 472– 475. Stokoe, D., Engel, K., Campbell, D.G., Cohen, P., Gaestel, M., 1992. Identification of MAPKAP kinase 2 as a major enzyme responsible for the phosphorylation of the small mammalian heat shock proteins. FEBS Lett. 313, 307–313. Tan, Y., Rouse, J., Zhang, A., Cariati, S., Cohen, P., Comb, M.J., 1996. FGF and stress regulate CREB and ATF-1 via a pathway involving p38 MAP kinase and MAPKAP kinase-2. EMBO J. 15, 4629 – 4642. Teramoto, H., Coso, O.A., Miyata, H., Igishi, T., Miki, T., Gutkind, J.S., 1996. Signaling from the small GTP-binding proteins Rac1 and Cdc42 to the c-Jun N-terminal kinase/stress-activated protein kinase pathway. A role for mixed lineage kinase 3/protein-tyrosine kinase 1, a novel member of the mixed lineage kinase family. J. Biol. Chem. 271, 27225–27228. Terashima, T., Yasuda, H., Terada, M., Kogawa, S., Maeda, K., Haneda, M., Kashiwagi, A., Kikkawa, R., 2001. Expression of Rho-family GTPases (Rac, cdc42, RhoA) and their association with p-21 activated kinase in adult rat peripheral nerve. J. Neurochem. 77, 986 –993. Therien, H.M., Mushynski, W.E., 1976. Isolation of synaptic junctional complexes of high structural integrity from rat brain. J. Cell Biol. 71, 807– 822. Tome, F.M., Evangelista, T., Leclerc, A., Sunada, Y., Manole, E., Estournet, B., Barois, A., Campbell, K.P., Fardeau, M., 1994. Congenital muscular dystrophy with merosin deficiency. C. R. Acad. Sci. III 317, 351–357.

46

G. Fragoso et al. / Experimental Neurology 183 (2003) 34 – 46

Uziyel, Y., Hall, S., Cohen, J., 2000. Influence of laminin-2 on Schwann cell–axon interactions. Glia 32, 109 –121. Vojtek, A.B., Cooper, J.A., 1995. Rho family members: activators of MAP kinase cascades. Cell 82, 527–529. Woodley, D., Sauder, D., Talley, M.J., Silver, M., Grotendorst, G., Qwarnstrom, E., 1983. Localization of basement membrane components after dermal– epidermal junction separation. J. Invest. Dermatol. 81, 149 – 153. Xu, H., Wu, X.R., Wewer, U.M., Engvall, E., 1994. Murine muscular

dystrophy caused by a mutation in the laminin alpha 2 (Lama2) gene. Nat. Genet. 8, 297–302. Yamada, H., Denzer, A.J., Hori, H., Tanaka, T., Anderson, L.V., Fujita, S., Fukuta-Ohi, H., Shimizu, T., Ruegg, M.A., Matsumura, K., 1996. Dystroglycan is a dual receptor for agrin and laminin-2 in Schwann cell membrane. J. Biol. Chem. 271, 23418 –23423. Yang, W., Lin, Q., Guan, J.L., Cerione, R.A., 1999. Activation of the Cdc42-associated tyrosine kinase-2 (ACK-2) by cell adhesion via integrin betal. J. Biol. Chem. 274, 8524 – 8530.