Stress proteins in oligodendrocytes: differential effects of heat shock and oxidative stress

Authors


Address correspondence and reprint requests to Prof. Dr Christiane Richter-Landsberg, Department of Biology, Molecular Neurobiology, University of Oldenburg, POB 2503, D-2611 Oldenburg, Germany. E-mail: christiane.richter.landsberg@university

Abstract

Heat shock proteins (HSP) or stress proteins serve as biomarkers to identify the contribution of stress situations underlying the pathogenesis of degenerative diseases of the CNS. We have analyzed by immunoblot technique the constitutive and inducible occurance of stress proteins in cultured rat brain oligodendrocytes subjected to heat shock or oxidative stress exerted by hydrogen peroxide, or a combination of both. The data demonstrate that oligodendrocytes constitutively express HSP32, HSP60 and the cognate form of the HSP70 family of proteins, HSC70. After heat shock, HSP25, αB-crystallin and HSP70 were up-regulated, while after oxidative stress the specific induction of HSP32 and αB-crystallin was observed. HSP32 represents heme oxygenase 1 (HO-1), a small stress protein with enzymatic activity involved in the oxidative degradation of heme which participates in iron metabolism. The presence of the iron chelators phenanthroline or deferoxamine (DFO), which previously has been shown to protect oligodendrocytes from oxidative stress-induced onset of apoptosis, caused a marked stimulation of HSP32 without affecting HSP70. This indicates that DFO possibly exerts its protective role by directly influencing the antioxidant capacity of HO-1. In summary, HSP in oligodendrocytes are differentially stimulated by heat stress and oxidative stress. Heme oxygenase-1 has been linked to inflammatory processes and oxidative stress, its specific up-regulation after oxidative stress in oligodendrocytes suggests that it is an ideal candidate to investigate the involvement of oxidative stress in demyelinating diseases.

Abbreviations used
DFO

deferoxamine

HO

heme oxygenase

HS

heat stress

HSP

heat shock proteins

MBP

myelin basic protein

MS

multiple sclerosis

MTT

3-4,5-dimethyl-2-yl]-2,5-diphenyl-tetrazoliumbromide

OS

oxidative stress

PCD

programmed cell death

PDTC

pyrrolidine dithiocarbamate

PLL

poly-l-lysine.

Heat shock proteins (HSP) or stress proteins play a role in normal CNS development and function, and are enhanced after traumatic injury of the brain and during neurodegenerative diseases (Brown 1994; Marcucilli and Miller 1994). Their involvement in autoimmune diseases, such as multiple sclerosis (MS), has been suggested (Birnbaum 1995; Brosnan et al. 1996; Van Noort et al. 1995). Stress proteins are induced by a variety of stress situations, including hyperthermia, viral infection, ischemia, anoxia and oxidative stress, and many of them are also constitutively expressed (Birnbaum 1995). Constitutively expressed HSP function as molecular chaperones and participate in protein synthesis, protein folding, transport and translocalization processes.

Heat shock proteins are classified according to their molecular weight, and in the nervous system the 27, 60 and 70 kDa families have been mostly studied. The small 25–28 kDa HSP is constitutively expressed in many cell types, is regulated by phosphorylation through serine protein kinases and shows sequence homologies to the αB-crystallin (20 kDa) of the lens, which is stress-inducible and has chaperone-like properties (Boelens and DeJong 1995). In the nervous system, αB-crystallin is expressed in oligodendrocytes in MS lesions but not in unaffected areas, as revealed by immunohistochemical analysis (Van Noort et al. 1995). Increased immunoreactivity has been observed in astrocytes in a variety of neurodegenerative diseases and its accumulation might be related to inclusion body formation (Head and Goldman 2000). HSP60 is associated with mitochondria and participates in the folding and assembly of transported proteins into the mitochondrion (Itoh et al. 1995; Brosnan et al. 1996). The HSP70 family comprises multiple members. The constitutive cytosolic form is known as HSC70 or cognate HSP70, has a molecular weight of 73 kDa, and is only moderately inducible. The inducible form HSP70 (72 kDa) is not constitutively expressed and strongly induced following neural trauma, including tissue injury and ischemia (Brown 1994; Nowak et al. 1994). HSP32, also known as heme oxygenase 1 (HO-1), is a small stress protein and belongs to the heme oxygenase (HO) family of proteins which catalyzes the oxidative degradation of heme to biliverdin, which is subsequently converted to bilirubin, and equimolar amounts of carbon monoxide and iron. Heme and iron may increase the formation of reactive oxygen intermediates, and hence exacerbate intracellular oxidative stress (Schipper 2000). On the other hand, biliverdin and bilirubin have free radical scavenging capabilities and can act as potent antioxidants. Heme oxygenase-2 is constitutively expressed, while HO-1 is synthesized in response to heat shock, heme and oxidative stress (Elbirt and Bonkovsky 1999; Sharp et al. 1999). Heme oxygenase-1 up-regulation in a number of neurodegenerative diseases, e.g. Alzheimer's and Parkinson's disease, indicates chronic oxidative stress. However, it remains controversial whether increased protein levels are cytoprotective or lead to further cell damage (Schipper 2000).

Stress proteins serve as biomarkers to identify the contribution of stress situations underlying the pathogenesis of degenerative diseases of the CNS. Oxidative stress has been implicated in the pathogenesis of MS, and specifically, HSP32/HO-1 might indicate oxidative stress and is involved in antioxidant defence mechanisms, but so far has not been investigated in myelinating cells. To further characterize the stress responses in oligodendrocytes and possible consequences for demyelinating diseases, we have analyzed the constitutive and inducible occurrance of stress proteins in cultured rat brain oligodendrocytes using a panel of antibodies directed against HSP/HSC70, HSP60, HSP25, αB-crystallin and HSP32/HO-1. The data demonstrate that oligodendrocytes differentially up-regulate HSPs after heat shock and oxidative stress, and that oxidative stress, but not heat stress, leads to the induction of HSP32/HO-1. Hence, HSP32/HO-1 might represent a useful marker to further elucidate the possible involvement of oxidative stress in inflammatory demyelinating diseases.

Materials and methods

Materials

Cell culture media were from Gibco/BRL (Grand Island, NY, USA). The 3-[4,5-dimethyl-2-yl]-2,5-diphenyl-tetrazoliumbromide (MTT) was from Pharmacia (Piscataway, NJ, USA); deferoxamine (DFO), Phenanthroline and PDTC were from Sigma (St Louis, MO, USA). Antibodies against heat shock proteins were from StressGen (Victoria, BC, Canada): HSP/HSC70 (SPA-820), HSP70 (SPA-810), HSP60 (SPA-806), HSP32 (SPA-895), HSP25 (SPA-801), αB-crystallin (SPA-223). Monoclonal anti-α-tubulin antibody was from Sigma (T-9026). Polyclonal anti-myelin basic protein antibody was a generous gift of Dr A. McMorris (Wistar Institute, Philadelphia, USA; McMorris et al. 1981). HRP-conjugated anti-mouse IgG was from Amersham (Freiburg, Germany) and anti-rabbit IgG from Biorad (Munich, Germany).

Oligodendrocyte cultures

Oligodendroglial cultures were prepared from the brains of 1-day-old Wistar rats as described before (Richter-Landsberg and Vollgraf 1998). Briefly, precursor cells, growing on the astrocytic cell layer were separated by vigorous shaking (for 16 h at 240 r.p.m.) and taken off (McCarthy and DeVellis 1989). To obtain mature oligodendrocytes, precursors were replated on poly-l-lysine (PLL)-coated culture dishes (3 × 106 cells/10-cm dish) and kept for 7 days in serum-free DMEM to which insulin- (5 µg/mL) transferrin- (5 µg/mL) sodium selenite (25 ng/mL) supplement (Boehringer, Mannheim, Germany) was added. All cells were kept at 37°C and 10% CO2. Growth medium was changed twice a week.

Heat shock treatment

Growth medium was exchanged and cultures were kept in the incubator (10% CO2) for 6–8 h before the experiments were started. Then culture dishes were sealed with parafilm and immersed for 30 min in a water bath at the indicated temperatures. Thereafter cells were put into the incubator for recovery as indicated. Control cells were sealed for 30 min, but remained in the incubator.

MTT-Viability assay

To assess the cytotoxic potential of the compounds the MTT (Tetrazolium) assay was carried out as described before (Richter-Landsberg and Vollgraf 1998). Briefly, oligodendrocyte precursor cells were prepared as described above and plated on PLL-coated 96-microwell cell culture plates (2 × 104 cells per well) then grown for 7 days. Thereafter, the growth medium was removed and fresh medium (100 µL/well) was added, and cells were stressed and incubated for the indicated times. 10 µL of MTT solution (5 mg/mL in PBS) were added to the wells, containing 100 µL medium, and the plates were incubated for 4 h. Thereafter 100 µL of a solubilization solution (10% sodium dodecyl sulfate in 0.01 mol/L HCl) was added and incubated overnight to dissolve the water-insoluble formazan salt. Quantification was then carried out with an ELISA reader at 595 nm using a 655-nm filter as a reference. Data are expressed as percentage of the untreated controls, and values represent the mean ± SD of eight microwells each of three independent experiments (n = 24).

Assay for DNA fragmentation

Total genomic DNA was isolated from cell pellets of oligodendrocytes (3–6 × 106 cells, for each experimental condition). DNA samples (5 µg) and a 1-kb DNA-ladder standard (Gibco, Rockville, MD, USA) were separated on 1.5% agarose gels, visualized by ethidium bromide staining, and photographed.

Immunoblot analysis

Cellular monolayers of control and stressed cells were washed with PBS once, scraped off in sample buffer containing 1% SDS and boiled for 10 min. Protein contents in the samples were determined according to Neuhoff et al. (1979). For immunoblotting, total cellular extracts (10 µg protein per lane) were separated by one-dimensional SDS–PAGE using 7.5% (HSP/HSC70, HSP70, HSP60), 10% (HSP32, α-tubulin) or 12.5% (HSP25, αB-crystallin, MBP) polyacrylamide gels, and transferred to nitrocellulose membranes (Schleicher and Schuell, Dassel, Germany; 0.45 µm) according to Towbin et al. (1979). The blots were washed and incubated with the individual antibodies, followed by HRP-conjugated anti-mouse (1: 2000) or anti-rabbit (1 : 5000) IgG, and visualized by the enhanced chemiluminescence (ECL) procedure as described by the manufacturer (Amersham, Braunschweig, Germany). The following antibodies were used (dilutions are given in brackets): anti-HSP/HSC70 (1 : 1000), anti-HSP70 (1 : 1000), anti-HSP60 (1 : 1000), anti-HSP32/HO-1 (1 : 1000), anti-HSP25 (1 : 500), anti-αB-crystallin (1 : 500), anti-α-tubulin (1 : 1000) and anti-MBP (1 : 1000).

Quantitative evaluation of the immunoblots was carried out by densitometric scanning and Image Quant software (Molecular Dynamics, Sunnyvale, CA, USA).

Results

Oligodendrocyte cell extracts were prepared from control and stress-induced cells and assessed for the presence of constitutively and stress-induced HSP. Antibodies directed against the following stress proteins were applied: HSP25, αB-crystallin, HSP32/HO-1, HSP60, HSP70 and HSC70. Two antibodies against the HSP70 family were used, one recognizing only the inducible isoform (HSP70) and the second recognizing both the constitutive and inducible forms (HSC/HSP70).

Constitutive HSP in oligodendrocytes during in vitro differentiation

Under unstressed conditions only HSP32, HSP60 and the constitutive form (HSC70) of the HSP70 family was detectable (Fig. 1). All three stress proteins were continuously expressed during in vitro differentiation of oligodendrocytes from precursor cells (0–24 h after plating) to mature oligodendrocytes (7 days in vitro), which were characterized by the presence of all four myelin basic protein (MBP) isoforms (Fig. 1, lower panel).

Figure 1.

Developmental regulation of constitutively expressed heat shock proteins in oligodendrocytes. Cell lysates were prepared from oligodendrocytes at the indicated times in vitro and subjected to immunoblot analysis using antibodies against HSC70, HSP60, HSP32, α-tubulin or MBP, as indicated by the arrows on the right. HSP70, HSP25 and αB-crystallin were not detectable under these unstressed conditions.

Stress responses in mature oligodendrocytes

To investigate stress responses experiments were carried out with 7-day-old cultured oligodendrocytes, which are characterized by their numerous cellular processes (Fig. 2a). We have shown before that these cells express myelin markers and mRNAs encoding the major myelin proteins, such as myelin basic protein, proteolipid protein and the myelin associated glycoproteins MAG and MOG (Richter-Landsberg and Vollgraf 1998; Richter-Landsberg and Gorath 1999).

Figure 2.

Effect of stress on oligodendrocyte morphology. Oligodendrocytes (7 div) were subjected to heat shock or oxidative stress (100 µm H2O2, 30 min) and photographed after a 24-h recovery period. Hoffmann modulation contrast images are shown: control (a), oxidative stress (b), heat shock (44°C, 30 min; c), heat shock (46°C, 30 min; d). Bar represents 50 µm.

Heat-stress

Cells were exposed to heat stress (HS) by incubation at 44°C for 30 min, and allowed to recover for 0–72 h. Immunoblot analysis revealed that HS caused the induction of HSP70, HSP25 and αB-crystallin, which were not detectable in control cultures (Fig. 3a). Additionally, the level of HSP60 was slightly enhanced by HS, while the levels of HSC70 and HSP32 remained unchanged. No changes in the appearance of MBP were observed (Fig. 3a). Quantitative evaluation of the time course showed that the three inducible proteins were maximally expressed after an 18-h recovery period and still present after 3 days (Fig. 3b). This treatment did not cause cellular damage as observed by microscopy (Fig. 2c). HSP expression was temperature dependent and hyperthermia of 44–45°C was most effective (Fig. 3c), however, at temperatures above 44°C cytotoxic effects were prominent (see Figs 2d and 7c).

Figure 3.

Heat shock response in oligodendrocytes. Oligodendrocytes (7 div) were subjected to a heat shock (44°C, 30 min) and cell lysates were prepared during a 72-h recovery period (37°C) at the indicated times (a and b). (a) The immunoblot depicts that HSP70, HSP25 and αB-crystallin are heat-inducible. Arrow on the right indicate the bands corresponding to the individual proteins. Note that MBP levels are not affected. (b) Quantitative evaluation of heat-inducible proteins in oligodendrocytes during the 72-h recovery period. Results represent the mean of three independent experiments. The values of the inducible proteins after an 18-h recovery period were set as 100%. (c) Temperature-dependent induction of HSPs in oligodendrocytes. Cells were subjected to a 30 min heat shock at the indicated temperatures and cell lysates were prepared after a 24-h recovery period and analyzed. Co, untreated control.

Figure 7.

MTT-survival assay of oligodendrocytes (7 div) exposed to single or double stress situtation. (a) Oligodendrocytes were exposed to an initial heat stress followed by an oxidative stress. OS: cells were treated with H2O2 (50–200 µm, 30 min) and MTT assay was carried out after a 24-h recovery period. HS→OS: Cells were subjected to heat stress (44°C, 30 min) and after an 18-h recovery period to H2O2 at the indicated concentrations for 30 min. After an additional recovery period of 24 h MTT-assay was carried out. Note that a heat stress prior to an oxidative stress does not exert protective effects. (b) Oligodendrocytes were exposed to two oxidative stresses. OS: cells were treated with H2O2 at the indicated concentrations and MTT assay was carried out after 24 h. OS→OS: cells were treated with H2O2 (50 µm, 30 min) and after a 18-h recovery period again with H2O2 at the indicated concentrations. After an additional recovery period of 24 h MTT-assay was carried out. Note that a mild oxidative stress prior to a second one exerts protective effects. (c) Oligodendrocytes were exposed to two heat shocks. HS: cells were exposed to heat shock (44°C, 30 min) and MTT-assay was carried out after 24 h. HS→HS: cells were exposed to heat shock (44°C, 30 min) and after an 18-h recovery period to a second heat shock either at 44°C or 46°C for 30 min. After an additional recovery period of 24 h, MTT-assay was carried out. Note that a heat shock (44°C) prior to a second one (46°C) exerts protective effects. Co, untreated cells left at 37°C for the entire period of time.

Oxidative stress

To exert oxidative stress, oligodendrocytes were exposed to hydrogen peroxide (H2O2) at various concentrations for 30 min, cell extracts were prepared after a 24-h recovery period and analyzed. This treatment caused an increase in αB-crystallin and a strong induction of HSP32, without affecting the other stress proteins or MBP (Fig. 4a). Under these conditions cell morphology was disturbed and membranous beads and apoptotic bodies were seen (Fig. 2b). In comparison to the effect of heat stress, αB-crystallin was only slightly induced (Fig. 5). Quantitative evaluation further demonstrated that αB-crystallin and HSP32 were maximally induced at a H2O2 concentration of 100–150 µm(Fig. 4b). Time course analysis showed that maximal elevation of both proteins occurred after 18–24 h recovery period (results not shown). Hence, HSP70 in oligodendrocytes was not inducible by oxidative stress alone.

Figure 4.

Induction of HSP after oxidative stress. (a) Oligodendrocytes (7 div) were subjected to H2O2 (50–200 µm) for 30 min. After a 24-h recovery period, cell lysates were prepared and subjected to immunblot analysis. Experiments were carried out at least three times with cells prepared from independent litters. The blots show that specifically HSP32 and αB-crystallin are induced. (b) Quantitative evaluation of HSP after oxidative stress. The results represent the mean of three independent experiments. The value of HSP at 100 µm H2O2 was set as 100%. Co, untreated control.

Figure 5.

Comparison of HSP induction after heat stress and oxidative stress.Oligodendrocytes were subjected to heat shock (44°C, 30 min) or oxidative stress (100 µm H2O2, 30 min) and cell lysates were prepared at the indicated time of recovery and analyzed. Co, untreated control.

Effects of double stress

Heat shock treatment followed by a secondary stress has been reported to exert protective effects and diminish a secondary stress response, e.g. exposure of cells to a first milder heat shock enabled the cells to withstand temperatures that would otherwise be lethal, or to withstand other toxic insults, such as chemical stress, oxidative stress or ischemia (for review see Sherman and Goldberg 2001). Here we have investigated if oxidative stress followed by heat stress alters the stress responses observed by either stress alone. Figure 6(a) indicates that when cells were treated with H2O2 (30 min; 100 µm; 3 h recovery time) and then subjected to a heat shock (44°C, 30 min; 20.5 h recovery), αB-crystallin and specifically HSP70 were induced to a greater extent than by either stress alone. In contrast thereto, HSP60 induction was at the same level as observed with HS alone and HSP25 was reduced (Fig. 6a). Figures 6(b), (d) and 6(c) depict the quantitative evaluation of the double stress situation in comparison to MBP and α-tubulin. In this and also several other experiments heat stress resulted in an up-regulation of α-tubulin, which might be due to tubulin aggregation and a decrease in tubulin proteolysis (Goldbaum and Richter-Landsberg, unpublished). The level of MBP, however, remained constant at all experimental conditions. The extent of the synergistic activation of HSP70 was concentration-dependent (Fig. 6d), and decreased at shorter periods of recovery time after the first oxidative stress (not shown). When cells were subjected to two heat shocks, separated by a 3-h recovery period, no further activation of HSP70 in comparison to a single heat shock was observed (not shown). As HSP70 could not be induced by oxidative stress alone, the synergistic effect of the double stress situation is very interesting, and suggests that cells might be protected by this situation.

Figure 6.

Effect of a double stress on oligodendrocytes. (a) Oligodendrocytes (7 div) were subjected to H2O2 (OS) 100 µm for 30 min or heat stress (HS) (44°C, 30 min) and analyzed by immunoblot after a 24-h recovery period. Double stress (DS) was exerted by treating cells with H2O2 (100 µm, 30 min; or as indicated in d) first, and after a recovery period of 3 h a secondary heat stress (44°C, 30 min) was applied. Cells were incubated for further 20.5 h to recover and then subjected to immunoblot analysis. The individual proteins analyzed are indicated by arrows on the right. (b) Quantitative evaluation of the constitutively expressed HSPs in comparison to α-tubulin of the experiment in (a). The levels of the proteins in the untreated control was set as 100%. (c) Quantitative evaluation of the inducible HSPs in comparison to MBP of the experiment in (a). The levels of the proteins after heat shock was set as 100%. (d) HSP/HSC70 expression after pre-treatment with various concentrations of H2O2 (HP 50, 100, 150 µm, respectively). Cells were subjected to H2O2 (HP) as indicated or double stress (HP + HS) as described in (a). Control (Co); oxidative stress (OS); heat stress (HS); double stress (DS).

Cell survival and DNA fragmentation

To test this hypothesis, we have analyzed whether cell survival is enhanced after double stress situations. Cell viability was assessed using the MTT assay which indicates mitochondrial activity. This assay was chosen, as the washing procedures of the neutral red viability assay caused the detachment of oligodendrocytes from the multiwell plates. The data demonstrate that HS cannot protect oligodendrocytes from the cytotoxic effects of a secondary oxidative stress (Fig. 7a), and similarly, an oxidative stress prior to HS is not protective (not shown). In contrast, cell viability was enhanced, when cells were subjected to oxidative stress twice (Fig. 7b) and also, when two consecutive heat shocks were applied (Fig. 7c). The latter effect was only observable when the initial heat shock (44°C) was followed by a secondary heat shock at a higher temperature (46°C) which otherwise had severe toxic consequences (Fig. 2d).

As we have shown before, concentrations of H2O2 which exerted half maximal cytotoxicity (100 µm) led to the onset and execution of the apoptotic programme (Richter-Landsberg and Vollgraf 1998). To investigate whether the conditions leading to a synergistic increase in HSP70 (Fig. 6) might rescue the cells from the onset of apoptosis, DNA was isolated from control and stressed cells and analyzed by agarose gel electrophoresis. The data demonstrate that HS and oxidative stress alone in oligodendrocytes led to DNA fragmentation (Fig. 8). That HS (44°C, 30 min, 24 h recovery) alone led to the onset of apoptosis was rather surprising, as cell morphology did not seem to be disturbed (Fig. 2). On the other hand, MTT assay indicated a 20% decrease in viability after HS. Also, when cells were subjected to a double stress, i.e. oxidative stress first and then to heat stress, as described in Fig. 6, the onset of apoptosis could not be prevented (Fig. 8). These data demonstrate that the synergistic up-regulation of HSP70, which was observed after a double stress situation (oxidative stress followed by a heat stress) (Fig. 6), was not efficient to rescue cells from cytotoxic effects or the onset of programmed cell death (PCD).

Figure 8.

DNA-fragmentation after stress in oligodendrocytes. Oligodendrocytes (7 div) were treated as indicated and DNA was isolated and separated on 1.5% agarose gels, stained with ethidium bromide, and visualized under UV light. 1, untreated control; 2, cells were incubated in the presence of staurosporine (200 nm) for 24 h; 3, cells were subjected to oxidative stress (H2O2, 100 µm, 30 min; 24 h recovery); 4, cells were subjected to heat shock (44°C, 30 min, 24 h recovery); 5, cells were subjected to oxidative stress (H2O2, 100 µm, 30 min; 3 h recovery) followed by a heat shock (44°C,30 min, 20.5 h recovery); 6, cells were subjected to heat shock (44°C, 30 min, 3 h recovery) followed by an oxidative stress (H2O2, 100 µm, 30 min; 20.5 h recovery). M, 1kb DNA-ladder (Gibco). Note that under these conditions cells were not rescued from the onset of apoptosis. The experiment was repeated several times with nearly identical results.

Effects of the iron chelator deferoxamine (DFO) on HSP expression

We have shown previously that oxidative stress-induced DNA fragmentation could be prevented by pre-incubation of oligodendrocytes with the iron chelator DFO. Also, DFO protected oligodendrocytes from H2O2-induced cytotoxic effects and suppressed the formation of free radicals, as assessed by dichlorofluorescein measurements (Vollgraf et al. 1999). Here we have tested whether the H2O2-induced up-regulation of HSP32 could be influenced by pre-incubating oligodendrocytes with DFO. Additionally we have applied another iron chelator, namely phenanthroline. While both iron chelators did not affect the levels of HSP/HSC70 and MBP, incubation with DFO (1 mm) or phenanthroline (100 µm) alone led to a marked induction of HSP32 (Fig. 9a). A pre-incubation with DFO (6 h) or phenanthroline (30 min) and the continuous presence during the stress and the 24 h recovery period, did not prevent or augment H2O2-induced HSP32 stimulation (Figs 9b and c). In contrast thereto, the treatment with the radical scavenger pyrrolidine dithiocarbamate (PDTC) 100 µm; 30 min pre-incubation inhibited the stimulation of HSP32 (Fig. 9d). Deferoxamine and PDTC were applied in aquaeous solution, and phenanthroline in ethanol which alone did not cause an up-regulation of HSPs. Interestingly, after the treatment with DFO or phenanthroline, HSP32 immunoreactivity was detected as a closely migrating double band (Fig. 9b). Treatment with alkaline phosphatase did not change this pattern (not shown).

Figure 9.

Effect of iron chelators on HSP 32 induction. (a) Oligodendrocytes were pre-incubated with DFO (1 mm, 6 h) or phenanthroline (PA) 100 µm, 30 min and then subjected to oxidative stress (H2O2, 100 µm, 30 min). After a 24-h recovery period (with DFO and PA) cells were extracted and analyzed by immunoblot procedure. Arrows indicate the position of the individual proteins which are named on the right. Co, untreated control; OS, H2O2 (100 µm, 30 min, 24 h recovery); DFO, cells were incubated with deferoxamine for 30 h; OS + DFO, cells were pre-incubated with DFO (6 h) and then subjected to oxidative stress; PA, cells were incubated with phenanthroline for 24 h; OS + PA, cells were pre-incubated with phenanthroline and subjected to oxidative stress. (b) Effect of DFO and PA on H2O2-induced HSP32. Cells were treated as in (a), but oxidative stress was exerted either with 50 or 150 µm of H2O2 (30 min, 24 h recovery). (c) Quantitative evaluation of 9b. (d) Effect of the radical scavenger PDTC on HSP32 induction. Cells were pre-incubated with PDTC (100 µm, 30 min) and PDTC was present throughout the oxidative stress (OS: H2O2, 100 µm; 30 min) and the 24 h recovery period.

Discussion

The autoinflammatory disease multiple sclerosis leads to the selective destruction of the myelin sheaths and oligodendroglia cell death. The mechanisms underlying oligodendrocyte destruction or regeneration potential have not yet been clarified. A number of stress proteins have been identified in MS brain tissue and plaques, pointing to both beneficial and harmful effects during the pathogenesis of the disease (Brosnan et al. 1996). The HSP70/HSC70 and HSP27 accumulation in MS lesions was observed by Aquino et al. (1997). Bajramovic et al. (1997) investigated the expression of αB-crystallin at different stages of MS lesion development. αB-crystallin, which can act as an immunodominant CNS myelin antigen to human T-cells (Van Noort et al. 1995), was expressed in active lesions at the earliest stages of lesional development in a subpopulation of oligodendrocytes, while in inactive lesions a tenfold reduction in αB-crystallin expressing oligodendrocytes occurred. Both HSP70 and HSP60 can elicit strong immunological reactions and are targets for autoimmune attacks (Brosnan et al. 1996). Elevated HSP60 expression was described as a prominent feature of acute MS lesions and at the height of the disease, astrocytes and oligodendrocytes displayed both mitochondrial and cytosolic immunoreactivity (Raine et al. 1996). The prolonged activity in surviving, structurally intact oligodendrocytes has been attributed to a protective mechanism of HSP60 (Raine et al. 1996). On the other hand, no correlation was found between the expression of HSP60 and αB-crystallin in MS lesions, suggesting different regulatory pathways for either HSP (Bajramovic et al. 1997). The small heat shock protein HO-1 has been linked to oxidative damage and inflammatory processes (Schipper 2000). So far, no data are available on the presence of HSP32/HO-1 in MS lesions. However, the induction of HO-1 in glia after traumatic brain injury was described (Fukuda et al. 1996) and a recent study indicates that oligodendroblasts are preferentially injured by a short hypoxic-ischemic insult and degenerated changes were coincident with the induction of HO-1 (Jelinski et al. 1999).

Stress induced up-regulation of HSP may provide protection and after a recovery period in the absence of stress stimuli, cells might resume their normal functions. However, this depends on the level of stress, if it is too high, the presence of HSP fails to protect the cells. HSP synthesis might stop and cells die by the initiation of PCD or, after an even higher stress stimulus, necrotic cell death becomes prominent (Sharp et al. 1999). Heat shocked cells acquire thermotolerance, show a higher resistance to environmental stress and may become more resistant to cytotoxic stimuli which otherwise induce apoptosis (Samali and Cotter 1996).

To investigate the cell type specific effects and possible contribution of stress situation and HSP to dismyelinating disorders, we have analyzed the response of cultured rat brain oligodendrocytes to heat shock and oxidative stress, and a combination of both. The data demonstrate that oligodendrocytes during in vitro development constitutively express HSP32, HSP60 and HSC70. After heat shock mainly HSP25, αB-crystallin and HSP70 were up-regulated, and H2O2 -induced oxidative stress caused the specific induction of HSP32 and αB-crystallin, without affecting the other stress proteins. Furthermore, the application of double stress situations implies that the timely sequence of the individual stressors is important for the survival capacity, and only if the same stress signals are applied sequentially, cell survival might be improved. Although a synergistic activation of HSP70 was observed when cells were first treated with H2O2 and then subjected to a heat shock, the survival capacity was not enhanced. Also, heat shock treatment did not protect the cells from a second oxidative insult, and led to the onset of the apoptotic programme. αB-crystallin was up-regulated in oligodendrocytes both after HS and oxidative stress, however, to a much lower extent by OS than by HS (Fig. 5). Hence, HSP32/HO-1 induction in oligodendrocytes may reflect a very sensitive measure to oxidative stress-induced insults.

Oxidative stress contributes to a number of acute and chronic neurodegenerative diseases (Halliwell 1992; Leist and Nicotera 1998). It also causes oligodendroglia cell death and might occur in inflammatory lesions of demyelinating diseases (Boccaccio and Steinman 1996; Lucchinetti et al. 1996; Ludwin 1997; Smith et al. 1999). Oligodendrocytes are specifically sensitive to oxidative stress (Smith et al. 1999), which has been attributed to their low level of antioxidant defence mechanisms and their high iron content (Gelman 1995; Connor and Menzies 1996; Thorburne and Juurlink 1996). Reduced iron released from the intracellular stores can be pathogenic, as it might mediate the formation of free radicals, in particular the damaging hydroxyl radical from peroxides via the Fenton reaction (Halliwell 1992). We have recently shown that mature oligodendrocytes in culture are highly susceptible to oxidative damage exerted by H2O2 and respond by the onset of programmed cell death (Richter-Landsberg and Vollgraf 1998). H2O2-induced onset and execution of PCD in oligodendrocytes is accompanied by the induction and nuclear translocation of the transcription factors AP-1 and NF-κB, which both have been discussed as regulators of cell death and survival (Vollgraf et al. 1999). DNA-fragmentation could be prevented by the antioxidants PDTC and vitamin E, the nuclease inhibitor aurintricarboxylic acid, and by pre-incubation with the iron chelator DFO. DFO protected oligodendrocytes from H2O2-induced cytotoxic effects as assessed by MTT-assay, and suppressed the formation of free radicals (Vollgraf et al. 1999).

Deferoxamine is a well known therapeutic substance used in the treatment of chronic iron overload and might be effective in protecting against oxidative stress (for review see Galey 1997). Deferoxamine has been shown to suppress experimental allergic encephalomyelitis (Pedchenko and LeVine 1998), an animal model of multiple sclerosis, and thus might be useful in the treatment of dismyelinating disorders. Interestingly, in the present communication we can show that in oligodendrocytes DFO caused a marked stimulation of HSP32/HO-1, without affecting HSP70. The same effect was observed after treatment with the iron chelator phenanthroline. Thus, DFO which protects oligodendrocytes from oxidative stress via induction of early response genes, and prevents the onset of apoptosis (Vollgraf et al. 1999), might exert a protective role by directly influencing the antioxidant capacity of heme oxygenase 1. On the other hand, the antioxidant PDTC which is a potent inhibitor of NF-κB activation (Schreck et al. 1992), did not stimulate HSP32/HO-1 in oligodendrocytes and attenuated the H2O2-induced enhancement. Hence, PDTC protects oligodendrocytes from oxidative stress-induced PCD by blocking free radical formation (Vollgraf et al. 1999) and through mechanisms not requiring HSP32 up-regulation. This is in contrast to recent reports where PDTC was observed to induce HO-1 transcription through the activation of AP-1 in rat aortic vascular smooth muscle (Hartsfield et al. 1998) and endothelial cells (Stuhlmeier 2000), indicating cell-type specific regulatory mechanisms of cell death and survival.

HO-1 seems to be required for iron reutilization (Poss and Tonegawa 1997), and growing evidence supports a role for HO-1 in protecting cells from oxidative stress and the brain against blood- and hemoglobin-mediated injury. A prolonged expression of HO-1 in humans after brain injury has been observed (Beschorner et al. 2000) and the association of HO-1 with brain lesions (Dwyer et al. 1996) might indicate the involvement of oxidative stress and point to a long-lasting stress contributing to the pathogenesis of the various diseases. Neurons overexpressing HO-1 were protected against glutamate-mediated oxidative stress and the accumulation of ROS was attenuated (Chen et al. 2000). The findings that DFO protects oligodendrocytes and concomitantly induces HO-1 further point to a protective role of HSP32/HO-1. On the other hand, HeLa cells overexpressing HO-1 consumed heme faster than control cells and showed an increased level of free chelatable iron (Kvam et al. 2000) which rendered them more susceptible to oxidative stress. Hence, if the stress response it too strong and HO-1 induction passes a certain threshold, it might not be protective, but contribute to the onset of cell death. This is crucial during pathological situations and has to be taken into consideration during therapeutic strategies.

To summarize, HSP in oligodendrocytes are differentially expressed by HS and oxidative stress. The time and sequence of the differential stress events determine the individual stress responses which either promote survival and regeneration, or degenerative processes and cell death. As oligodendrocytes specifically up-regulate HO-1 after oxidative stress, the presence of HO-1 in MS lesions could indicate that oxidative stress plays an important role in MS pathogenesis.

Acknowledgements

We thank Angelika Spanjer for expert technical assistance with the cell cultures. This work was supported by the Deutsche Forschungsgemeinschaft (DFG, Ri 384/11–1).

Ancillary