• free radical;
  • glia;
  • hemoglobin;
  • iron;
  • signal transduction;
  • toxicity


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

Hemin is present in intracranial hematomas in high micromolar concentrations and is a potent, lipophilic oxidant. Growing evidence suggests that heme-mediated injury may contribute to the pathogenesis of CNS hemorrhage. Extracellular signal-regulated kinases (ERKs) are activated by oxidants in some cell types, and may alter cellular vulnerability to oxidative stress. In this study, the effect of hemin on ERK activation was investigated in cultured murine cortical astrocytes, and the consequence of this activation on cell viability was quantified. Hemin was rapidly taken up by astrocytes, and generated reactive oxygen species (ROS) within 30 min. Increased immunoreactivity of dually phosphorylated ERK1/2 was observed in hemin-treated cultures at 30–120 min, without change in total ERK. Surprisingly, ERK activation was not attenuated by concomitant treatment with antioxidants (U74500A or 1,10-phenanthroline) at concentrations that blocked ROS generation. Cell death commenced after 2 h of hemin exposure and was reduced by antioxidants and by the caspase inhibitor Z-VAD-FMK. Cytotoxicity was also attenuated by MEK inhibition with PD98059 or U0126 at concentrations that were sufficient to prevent ERK activation. Whereas the effect of Z-VAD-FMK on cell survival was transient, the effect of MEK inhibitors was long-lasting. MEK inhibitors had no effect on cellular hemin uptake or subsequent ROS generation. The present results suggest that hemin activates ERK in astrocytes via a mechanism that is independent of ROS generation. This activation sensitizes astrocytes to hemin-mediated oxidative injury.

Abbreviations used

2′,7′-dichlorofluorescin diacetate


dimethyl sulfoxide


extracellular signal-regulated kinase


lactate dehydrogenase


MAP kinase kinase


minimal essential medium


phenylmethylsulfonyl fluoride


polyvinylidene difluoride


reactive oxygen species


sodium dodecyl sulfate


Tris-buffered saline.

Extracellular stimuli are communicated to the cell interior by cascades of kinase reactions that phosphorylate effector enzymes and induce immediate early genes. Although signal transduction pathways are often activated by ligand-dependent events, a growing body of evidence suggests that they also may be directly activated by oxidants (Herrlich and Böhmer 1999; Allen and Tresini 2000). This phenomenon has been most fully characterized for the extracellular signal-regulated kinase (ERK) pathway, which can be oxidatively activated via at least two mechanisms. First, hydrogen peroxide, nitric oxide, and metals may act directly on Ras to promote guanine nucleotide exchange (Lander et al. 1995). Activated Ras then serves as a scaffold that allows the serial phosphorylation of Raf, MEK 1/2, and ERK 1/2 (Morrison and Cutler 1997; Herrlich and Böhmer 1999; Allen and Tresini 2000). Second, oxidants can indirectly up-regulate signal transduction by inhibiting the inactivation of kinases. Tyrosine phosphatases have a redox-sensitive sulfhydryl group that is essential for catalytic function (Monteiro et al. 1991; Allen and Tresini 2000). If oxidized to a disulfide or sulfenic acid derivative, enzyme activity is lost, and the normal balance of kinase and phosphatase reactions is disrupted.

Although redox regulation of signal transduction pathways has been the subject of a number of studies, the effect of ERK1/2 on oxidative injury remains in dispute. Its activation by neurotrophins and other growth factors prevents apoptosis (Xia et al. 1995; Fukunaga and Miyamoto 1998), leading to the general assumption that it is beneficial to stressed cells. Consistent with this idea, blocking the ERK pathway with the MEK inhibitor PD98059 potentiated the toxicity of hydrogen peroxide and hypoxanthine/xanthine oxidase on HeLa cells, cardiac myocytes, and renal epithelial cells (Aikawa et al. 1997; Wang et al. 1998; di Mari et al. 1999), and enhanced ischemia/reperfusion injury in isolated perfused heart (Yue et al. 2000). However, other studies that used cells of neural origin have yielded opposite results. Both PD98059 and the more potent MEK inhibitor U0126 attenuated the toxicity of hydrogen peroxide and glutamate in primary neuronal cultures and in cell lines (Bhat and Zhang 1999; Satoh et al. 2000; Stanciu et al. 2000). In vivo, brain ERK activation was observed after ischemia and hypoglycemia (Hu and Wieloch 1994; Kurihara et al. 1994); MEK inhibitors decreased lesion volume and improved behavioral outcome (Alessandrini et al. 1999; Namura et al. 2000; Wang et al. 2000). These observations suggest that the effect of ERK1/2 on cell survival may vary considerably with the cell type, and that the pathway to oxidative death in the CNS may surprisingly involve these kinases.

Release of heme proteins from their usual compartments is a likely source of oxidative stress after an acute CNS insult. The necrotic component of a traumatic or ischemic injury produces mitochondrial and cell membrane disruption and indiscriminate cytochrome release into the extracellular space (Kroemer et al. 1998). However, heme decompartmentalization is most extreme when injury is complicated by hemorrhage. Extravascular erythrocytes lyse over hours (Hua et al. 2000), and released hemoglobin oxidizes to methemoglobin in a predictable fashion that can be detected with magnetic resonance imaging (Hosada et al. 1987). Oxidation decreases the affinity of the heme moiety for the globin chain, increasing the probability of hemin release (Vandegriff 1995). Although the toxicity of hemin on CNS cells has not been investigated in a quantitative fashion, it is lethal to a variety of other cell types at low micromolar concentrations (Balla et al. 1992; Bhoite-Solomon et al. 1993; Braverman et al. 1995; da Silva et al. 1996). Potentially cytotoxic quantities are likely present in intracranial hematomas. Seven days after injection of autologous subarachnoid blood, Letarte et al. (1993) observed a mean free hemin concentration of 390 µm in the clot. This observation and toxicity data provide the rational basis for the hypothesis that hemin may contribute to cellular injury after CNS hemorrhage.

Little is known of the effect of hemin on ERK activation in CNS cells. Moreover, the effect of ERK on cellular vulnerability to heme-mediated oxidative stress has not been investigated. In addition to its mechanistic importance, the availability of selective MEK inhibitors suggests that such information may have therapeutic implications. We therefore utilized a cell culture model of hemin toxicity to test the following hypotheses: (i) hemin activates ERK1/2 in astrocytes; and (ii) inhibiting ERK activation protects cells from hemin-mediated oxidative injury.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

Glial cultures

Cultures were prepared from 1 to 3 day postnatal Swiss–Webster mice using minor modifications of methods described by Choi et al. (1987). Cortices were dissected free, minced with forceps, and placed in medium containing 0.075% acetylated trypsin at 37°C for 1 h. Tissue was then collected by low-speed centrifugation for 5 min, and was dissociated by trituration through a flamed Pasteur pipette in plating medium containing Eagle's minimal essential medium (MEM), 10% fetal bovine serum (HyClone, Logan, UT, USA), 10% heat inactivated horse serum (HyClone), glutamine (2 mm), glucose (21 mm) and epidermal growth factor (10 ng/mL). The cell suspension was diluted with additional plating medium and was plated on 24 well plates (Greiner Labortechnik, purchased from Bellco Inc., Vineland, NJ, USA) at a density of 1 hemisphere/plate. Cultures were incubated in a humidified 5% CO2 atmosphere at 37°C, and were fed weekly with medium that contained MEM, glucose (23 mm), glutamine (2 mm), and equine serum (10%). Cultures were confluent by 3 weeks, with > 95% of cells immunoreactive for glial fibrillary acidic protein.

Hemin exposure

At 24–48 h prior to exposure, growth medium was replaced with MEM containing 10 mm glucose and 0.5% equine serum in order to induce a quiescent state (Neary et al. 1999). Hemin and other reagents were added from concentrated stock solutions. In experiments that tested the effect of lipid-soluble compounds (PD98059, U0126, phenanthroline, Z-VAD-FMK), an equal volume of dimethyl sulfoxide (DMSO) vehicle was added to cultures exposed to hemin alone. After addition of drugs or hemin, cultures were gently agitated and returned to the incubator. In experiments in which exposure exceeded 5 h, drugs were replenished at 4 h intervals to a maximum of three doses.


Culture medium was aspirated, and the monolayer was immediately lysed in 75 µL of buffer containing 50 mm Tris-HCl, 100 mm NaCl, 50 mm NaF, 5 mm EDTA, 40 mmβ-glycerophosphate, 1% Triton X-100, 0.1% sodium dodecyl sulfate (SDS), 200 µm NaVO3, 1 µm phenylmethylsulfonyl fluoride (PMSF), 1 mg/mL leupeptin, and 1 µm pepstatin A. The protein content of samples was determined via the BCA reagent technique (Pierce, Rockford, IL, USA); the accuracy of this method was confirmed by staining gels with Coomassie blue. An equal amount of 2× sample buffer (0.5 m Tris-HCl, 10% glycerol, 10% SDS, 0.5% bromophenol blue, 5% 2-mercaptoethanol) was then added, and samples were heated to boiling for 5 min. Protein samples (6 µg/lane) were loaded onto 12% SDS–polyacrylamide gel (Ready Gel, Bio-Rad Laboratories, Hercules, CA, USA), and gels were run at 120 V for 60 min. Protein was then transferred to a polyvinylidene difluoride (PVDF) membrane (Osmonics, Inc., Minnetonka, MN, USA), which was subsequently incubated for 1 h in 5% non-fat milk in phosphate-buffered saline (PBS)/0.1% Tween-20. The membrane was then incubated overnight in primary antibody followed by secondary antibody conjugated to horse radish peroxidase. Immunoreactive proteins were detected via chemiluminescence (ECL, Amersham Pharmacia Biotech, Piscataway, NJ, USA).


Cultures were washed in cold Tris-buffered saline and fixed in cold 4% paraformaldehyde for 30 min. After washing, cultures were serially exposed to: 0.25% Triton X-100 for 10 min, 10% normal goat serum in Tris-buffered saline (TBS) for 15 min, and primary antibody raised against dually phosphorylated ERK 1/2 (Promega, Madison, WI, USA) at room temperature with continuous shaking for 2 h. After TBS wash, cultures were exposed to biotinylated secondary antibody (goat anti-rabbit IgG, 1 : 100, Vector Laboratories, Burlingame, CA, USA) for 30 min, and then were processed with the Vectastain Elite ABC kit, rabbit IgG (Vector Laboratories). Diaminobenzidine tetrahydrochloride (0.05%) in 0.01% hydrogen peroxide was used as chromagen.

Quantification of reactive oxygen species (ROS)

ROS formation was quantified by measuring fluorescence intensity of cultures after loading cells with 2′,7′-dichlorofluorescin diacetate (DCFH-DA) (Molecular Probes, Eugene, OR, USA) (Reynolds and Hastings 1995). DCFH-DA is a permeable ester that is de-esterified within cells to yield the ionized free acid (DCFH), which accumulates intracellularly. DCFH is non-fluorescent until it is oxidized by peroxides and the hydroxyl radical to 2′,7′-dichlorofluorescein. At indicated intervals, cultures were washed into minimal essential medium (MEM) containing 10 mm glucose; 20 µm DCFH-DA was then added, and cultures were returned to the incubator for 15 min. Cultures were then washed into a HEPES-buffered salt solution containing (in mm): NaCl, 120; KCl, 5.4; MgCl2, 0.8; CaCl2, 1.8; HEPES, 20; glucose, 5.5 (pH 7.4). One 100× field from the center of each culture was imaged with a Nikon Diaphot epifluorescence microscope and DAGE/MTI 10 bit cooled video camera attached to a Macintosh computer with VG-5 Scientific PCI frame grabber. A standard FITC filter was used along with a neutral density filter to minimize photo-oxidation of the dye. Images were immediately captured, and fluorescence intensity was determined with Scanalytics IPLab software. Background fluorescence from cultures subjected to sham wash and dye incubation only was subtracted from each recording. Using this procedure, the autofluorescence of culture plates was minimal.

Cellular hemin content

Culture medium was completely aspirated, and cells were washed twice with 0.75 mL PBS. After PBS aspiration, cells were immediately lysed in concentrated formic acid. Absorbance of the lysate at 398 nm was then determined with a microtiter plate reader, and was compared with a hemin in formic acid standard curve (Motterlini et al. 1995). Protein concentration was determined on sister cultures from the same plating using the BCA reagent (Pierce). Hemin content was expressed as nmol/mg protein.

Assessment of injury

In all cytotoxicity experiments, glial injury was estimated by examination of cultures under phase contrast microscopy, or under fluorescence microsopy after staining with propidium iodide, which stains the nuclei of cells with disrupted membranes. Cell death was quantified by measurement of lactate dehydrogenase (LDH) activity in the culture medium. In these cultures, LDH activity is an accurate marker of cell death that correlates well with cell counts (Koh and Choi 1988). Stock solutions of sterile 100 mm potassium phosphate buffer and 27.2 mm sodium pyruvate in potassium phosphate buffer (pH 7.4) were prepared in advance and stored at 4°C until use. A solution of NADH (3 mg/10 mL phosphate buffer) was prepared immediately prior to use. A 25-µL sample of medium was removed from each culture and was placed in a 96 well assay plate. Ten minutes after addition of 125 µL phosphate buffer and 100 µL NADH solution into each well, 25 µL pyruvate was rapidly added using an Eppendorf repeater pipette with eight-well adapter. The absorbance of the reaction mixture at 340 nm was determined at 6 s intervals for 2 min, using a kinetic plate reader (Molecular Devices). The LDH signal produced by 100% cell death was determined by lysing cells with 0.1% Triton X-100. The low LDH activity in the medium of sister cultures subjected to sham wash only was subtracted from the values obtained from treated cultures, to yield the signal specifically associated with hemin toxicity.

In other experiments, nuclear morphology was assessed after staining with Hoechst 33258. Cultures were washed with Tris-buffered saline and fixed with 4% paraformaldehyde for 30 min. After membranes were permeabilized with 0.25% Triton X-100 for 15 min, cultures were exposed to Hoechst 33258 (20 µg/mL) for 30 min, and then were visualized with an ultraviolet filter.

Statistical analysis

Differences between treatment groups were assessed with the Tukey–Kramer's test or Dunn's multiple comparisons test.


MEM and glutamine were purchased from Gibco (Life Technologies Inc, Grand Island, NY, USA), and serum was purchased from Hyclone Inc. (Logan, UT, USA). Z-VAD-FMK and polyclonal antibody to activated ERK1/2 were purchased from Promega (Madison, WI, USA), and polyclonal antibody to total ERK was provided by Dr Yizheng Wang. PD98059 was purchased from Calbiochem (San Diego, CA, USA) or Sigma (St Louis, MO, USA), U0126 was purchased from Promega, and 2′,7′-dichlorofluorescin diacetate was purchased from Molecular Probes (Eugene, OR, USA). Other reagents were purchased from Sigma (St Louis, MO, USA).


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

Characterization of hemin toxicity

In initial experiments, the toxicity of hemin was quantified in this culture system. A significant increase in cellular hemin content was detected within 10 min of exposure that progressed over the subsequent 80 min (Table 1). Later timepoints were not tested since by 2 h cells tended to slough with repeated washes. ROS production was detected with DCFH-DA within 30 min and continued at 90 min (Fig. 1). Astrocyte death as measured by LDH release was minimal after a 2-h exposure to 30 µm hemin, but peaked rapidly thereafter (Fig. 2a). With a 5-h exposure, cell injury increased between 10 and 100 µm(Fig. 2b). At the latter concentration, about three-quarters of culture LDH had been released into the medium; higher concentrations were not feasible due to the limited solubility of hemin in aqueous media at neutral pH.

Table 1.  Hemin is rapidly taken up by astrocytes; MEK inhibitors do not prevent this uptake
 Time of treatment
 10 min30 min90 min
  1. Cellular hemin content (nmol/mg protein) in cultures (n = 11–13/condition) treated with 30 µm hemin alone or with PD98050 (30 µm) or U0126 (10 µm) for indicated intervals. Hemin was not detected in control cultures that were exposed to culture medium only.

Hemin (30 µm)3.61 ± 0.354.86 ± 0.337.65 ± 0.66
+ PD98059 (30 µm)3.76 ± 0.395.90 ± 0.527.24 ± 0.81
+ U0126 (10 µm)4.07 ± 0.366.25 ± 0.547.86 ± 0.79

Figure 1. Hemin generates reactive oxygen species in astrocytes; antioxidants but not MEK inhibitors block this effect. Cultures (n = 15–17/condition) were pretreated with U74500A (20 µm), 1,10-phenanthroline (PHE, 100 µm), PD98059 (30 µm), U0126 (10 µm), or DMSO vehicle only for 3 h. Hemin (30 µm) was then added. At indicated intervals, medium was changed; cells were then loaded with DCFH-DA and imaged. Mean background fluorescence intensity from sister cultures subjected to sham wash and dye incubation only was subtracted from each value (0.0007 ± 0.0005 at 0.5 h and 0.005 ± 0.002 and 1.5 h). **p < 0.01, ***p < 0.001 versus cultures treated with hemin alone, Dunn's multiple comparisons test.

Download figure to PowerPoint


Figure 2. Characterization of hemin toxicity. Mean medium LDH (± SEM) in cultures exposed to: (a) 30 µm hemin for indicated intervals; (b) indicated concentrations of hemin for 5 h; (c) 30 µm hemin with 100 µm 1,10-phenanthroline (PHE) or 20 µm U74500A for 5 h, after a 3-h pretreatment with these antioxidants or DMSO vehicle (1%) alone. LDH values are expressed as both units/mL and as a percentage of the LDH released in sister cultures in which cells had been lysed by exposure to 0.1% Triton X-100 for the duration of the experiment (% of total). The low LDH values in sister cultures subjected to sham wash alone were subtracted from all values to yield the LDH signal due to hemin toxicity.

Download figure to PowerPoint

Unlike inorganic iron, hemin is lipophilic, and so will more readily gain access to the hydrophobic interior of the cell membrane. Since very reactive species such as the hydroxyl radical will likely react with and oxidize target molecules in proximity to their formation (Samuni et al. 1983), lipophilic antioxidants were used to test the hypothesis that hemin toxicity in this system is oxidative. 1,10-phenanthroline is a lipid-soluble iron chelator (Halliwell and Gutteridge 1989); U74500A is a 21-aminosteroid that inhibits lipid peroxidation chain reactions (Braughler et al. 1987). Both compounds markedly reduced astrocyte injury (Fig. 2c).

Recent evidence suggests that caspase activation may contribute to heme-mediated cell injury (Meguro et al. 2001). Their role in this model was therefore evaluated by treating cultures with the general caspase inhibitor Z-VAD-FMK. At a concentration that is sufficient to inhibit caspases (Okragly and O'Brien 1999), this compound significantly reduced cell death at 4 h; however, this effect was overcome by increasing the exposure duration to 8 h (Fig. 3). Inhibition of protein synthesis with cycloheximide had no significant effect on cell death at either time point.


Figure 3. Effect of caspase or protein synthesis inhibition on hemin toxicity. Cultures were treated with 30 µm hemin for indicated intervals, alone or with 30 µm Z-VAD-FMK or 1 µm cycloheximide (CHX). Cultures were treated again with Z-VAD-FMK and CHX at 4 h. ***p < 0.001, Tukey–Kramer's test.

Download figure to PowerPoint

Hemin activates ERK in astrocytes

Ultraviolet light, hydrogen peroxide, and metals cause ERK1/2 activation in a variety of cell types (Herrlich and Böhmer 1999; Allen and Tresini 2000). In order to determine if hemin-mediated oxidative stress had a similar effect on astrocytes, proteins were separated by SDS gel electrophoresis and were subjected to western blot analysis, using an antibody that is specific for dually phosphorylated ERK 1/2 (Promega, Madison, WI, USA). More prominent bands were apparent within 30 min of hemin exposure (Fig. 4a), with a greater increase of active-ERK2. Immunoreactivity was maximal at 2 h and had returned to baseline by 4 h. Total ERK expression was not significantly changed until 4 h (Fig. 4b), when cell death was widespread. ERK activation in response to hemin was also detected at the cellular level via immunocytochemistry (Fig. 5).


Figure 4. Hemin activates glial ERK. (a) Cultures were treated with hemin (100 µm) in MEM with 0.5% equine serum or medium alone for indicated intervals. Proteins were then subjected to western blot analysis, using an anti-active ERK1/2 antibody (Promega). (b) as in (a), using antibody to total ERK1/2.

Download figure to PowerPoint


Figure 5. Cultures were stained with antibody to active ERK 1/2, 1.5 h after the addition of an equal volume (9 µL) of: (a) culture medium only, or (b) hemin stock solution to give a final concentration of 30 µm in the culture; increased immunoreactivity is apparent. Scale bar = 100 µm.

Download figure to PowerPoint

Two stucturally distinct MEK inhibitors are currently available. PD98059 is a flavone that inhibits Raf activation of MEK1 with an IC50 of 2–7 µm. However, it is a much weaker inhibitor of MEK2, with an IC50 of 50 µm (Alessi et al. 1995). U0126 is a newer, more potent inhibitor of both MEK1 and MEK2 (Favata et al. 1998). Pretreatment with either compound for 3 h attenuated ERK activation by hemin at concentrations that have been reported to be effective in cultured cells (Aikawa et al. 1997; Bhat and Zhang 1999; Stanciu et al. 2000) (Fig. 6).


Figure 6. Effect of MEK inhibitors and antioxidants on activation of ERK1/2 by hemin. Cultures were exposed to hemin (100 µm) alone, with PD98059 (PD, 30 µm), U0126 (10 µm), 1,10-phenanthroline (PHE, 100 µm), or U74500A (20 µm), or to phenanthroline or U74500A alone for 2 h. Cultures were pretreated with drugs or DMSO vehicle for 3 h prior to addition of hemin. Immunoblots were stained with anti-active ERK1/2 antibody.

Download figure to PowerPoint

We hypothesized that if hemin activated ERK1/2 by generating oxidative stress, antioxidants would prevent activation. Cultures were therefore pretreated with cytoprotective concentrations of 1,10-phenanthroline or U74500A for 3 h, and then hemin was added for 2 h. Surprisingly, these agents failed to prevent ERK activation (Fig. 6).

MEK inhibitors attenuate hemin toxicity

To test the hypothesis that ERK activation potentiated hemin toxicity, the effect of MEK inhibitors on cell injury was determined. In cultures treated with 30 µm hemin plus DMSO vehicle, 49.4 ± 2.28% of culture LDH was released at 3.5 h. This cell death was largely prevented by either PD98059 or U0126 (Figs 7 and 8). Increasing the exposure duration to 7 or 9 h resulted in a slightly more severe injury. However, the absolute cytoprotective effect that was provided by MEK inhibitors was not changed. Glial injury produced by 24–48 h exposure to 10 µm hemin was also attenuated (Table 2).


Figure 7. MEK inhibitors attenuate hemin toxicity. Mean medium LDH (± SEM) in cultures that were pretreated with 30 µm PD98059 (PD), 10 µm U0126, or DMSO vehicle alone for 3 h, followed by addition of 30 µm hemin for indicated intervals. Cultures were treated again with MEK inhibitors at 4-h intervals. ***p < 0.001 for cultures treated with hemin plus PD98059 or U0126 versus cultures treated with hemin alone, Tukey–Kramer's test.

Download figure to PowerPoint


Figure 8. The MEK inhibitor PD98059 protects astrocytes from hemin toxicity. Phase contrast (a, c and e) and fluorescence photomicrographs of the same fields (b, d and f) after propidium iodide staining of astrocyte cultures that were exposed for 4 h to: (a and b) experimental medium alone; cells are viable and exclude propidium iodide; (c and d) 30 µm hemin; many cells have degenerated; they do not exclude propidium, and it stains their nuclei; (e and f) 30 µm hemin plus 30 µm PD98059; most cells are viable. Scale bar = 100 µm.

Download figure to PowerPoint

Table 2.  MEK inhibition provides sustained protection
 Time of exposure
 24 h48 h
  1. Medium LDH activity (units/mL ± SEM) after continuous exposure to 10 µm hemin alone or with 10 µm U0126 for 24–48 h. Total culture LDH was determined by measuring medium activity after lysing all cells with 0.1% Triton X-100. ***p < 0.001 versus hemin alone, Tukey–Kramer's test.

Hemin208.1 ± 12.9200.9 ± 13.2
Hemin + U012681.7 ± 5.7***109.0 ± 9.71***
Sham wash36.1 ± 2.440.4 ± 3.9
100% cell lysis313.7 ± 20.61313.7 ± 20.61

In order to determine if MEK inhibitors attenuated necrosis or apoptosis, cultures were stained with Hoechst 33258 after exposure to hemin alone or with PD98059 or U0126, and nuclear morphology was assessed. Consistent with the weak and transient effect of caspase inhibition on cell injury, condensed and fragmented chromatin suggestive of apoptosis was observed in less than 1% of cells exposed to hemin alone for 6 h. MEK inhibitors increased the percentage of fragmented nuclei (Table 3, Fig. 9).

Table 3.  Effect of hemin treatment on percentage of cells with apoptotic nuclei
Condition% apoptotic nuclei
  1. Cultures were stained with Hoechst 33258, after treatment with hemin alone or with inhibitors for 6 h. Percentage of nuclei that were condensed and fragmented was then determined. **p < 0.01, ***p < 0.001 versus sham-washed cultures, Tukey–Kramer's test.

Sham wash0.31 ± 0.12
Hemin 30 µm0.66 ± 0.23
Hemin + PD980592.08 ± 0.53***
Hemin + U 01261.68 ± 0.30**
Hemin + Z-VAD-FMK0.19 ± 0.12

Figure 9. Nuclear morphology of cultures exposed to hemin alone or with MEK or caspase inhibitors. Cultures were stained with Hoechst 33258 after 6 h exposure to: (a) sham wash only; (b) 30 µm hemin; (c) 30 mm hemin with 30 µm PD98059; (d) 30 mm hemin with 30 µm Z-VAD-FMK. Scale bar = 50 µm.

Download figure to PowerPoint

Since hemin is a highly reactive moiety, we tested the hypothesis that the effects of PD98059 and U0126 were due to an interaction of these compounds with hemin in the medium that inhibited cellular hemin uptake. At 10, 30, and 90 min after exposure to hemin with either inhibitor, cells were lysed in formic acid, and the absorbance of the lysate at 398 nm was determined. Neither compound attenuated hemin uptake (Table 1).

Effect of MEK inhibitors on free radical production

In order to determine if MEK inhibitors prevented the generation of reactive oxygen species by hemin, their effect on DCFH fluorescence was assessed after 30 and 90 min exposures. Compared with cultures that were treated with hemin plus vehicle alone, neither PD98059 nor U0126 significantly altered fluorescence intensity at either time point (Figs 1 and 10). Not surprisingly, the fluorescence signal was markedly attenuated by both U74500A and phenanthroline.


Figure 10. Reactive oxygen species production by hemin is attenuated by U74500A but not by PD98059. Phase contrast (a, c and e) and fluorescence photomicrographs of the same fields after DCFH-DA staining (b, d and f) in cultures treated for 1.5 h with: (a and b) 30 µm hemin; many cells fluoresce, indicating ROS production; (c and d) 30 µm hemin with 30 µm PD98059; (e and f) 30 µm hemin with 20 µm U74500A. Scale bar = 100 µm.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

The present study serves two ends. First, we have characterized a model of hemin toxicity in glial cultures that may be relevant to hemorrhagic CNS injury. Hemin was rapidly taken up by astrocytes and generated reactive oxygen species, as detected by DCFH fluorescence, within 30 min. By 3 h, morphologic and biochemical evidence of cell injury was apparent after exposure to a hemin concentration that is less than 10% of that found in experimental hematomas (Letarte et al. 1993). The iron-dependent, oxidative nature of this toxicity was demonstrated pharmacologically. Two lipid-soluble antioxidants, U74500A and 1,10-phenanthroline, blocked both ROS generation and cell death. The general caspase inhibitor Z-VAD-FMK was also somewhat protective, but its effect was transient and was lost by 8 h. These results suggest that hemin may initiate an apoptotic cell death program in astrocytes. However, necrosis appears to be the predominant type of cell death in this model.

Second, the effect of extracellular signal-regulated kinase activation on glial vulnerability to hemin was investigated. An increase in dually phosphorylated ERK1/2 protein was detected 30 min after adding hemin to cultures; immunoreactivity intensified at subsequent timepoints that preceded cell death (60–120 min). In this culture system, ERK activation by hemin is apparently deleterious, since preventing it with selective MEK inhibitors provides sustained cytoprotection.

Although MEK inhibitors prevented both ERK activation and hemin toxicity, they did not alter ROS generation caused by hemin. In contrast, antioxidants, U74500A and 1,10-phenanthroline, blocked ROS generation and cell death, but had no effect on ERK activation. These observations suggest that ERK pathway activation by hemin is in parallel to free radical formation in this model. PD98059 and U0126 act on the former process without affecting the latter, in agreement with results of Stanciu et al. (2000) when neuronal ERK was activated by glutamate. Furthermore, although oxidative stress per se may activate Ras/ERK (Lander et al. 1995), ERK activation by hemin does not require ROS generation. The toxicity of hemin in this system appears to be mediated by the synergistic action of mechanisms that are dependent on both free radical formation and ERK activation. Blockade of either is sufficient to prevent cell death.

The precise molecular mechanism by which hemin activates ERK in astrocytes has not been defined. Hemin is a highly reactive moiety that has a critical regulatory role in numerous metabolic pathways that use or sense oxygen (Padmanaban et al. 1989). In addition, non-toxic concentrations stimulate cell growth and differentiation, and regulate the synthesis, transport, and degradation of hemoproteins. The effect of hemin on the expression and activity of kinases that are involved in signal transduction cascades in CNS cells has not been intensively investigated. In erythroid cells, a 4-h exposure is sufficient to increase the expression of a wide variety of genes, including two that are directly involved in Ras signaling (Zhu et al. 1999). Further investigation of the regulatory role of hemin in CNS cells seems warranted. The present data suggest that either it or one of its metabolites activates at least one component of the Ras/ERK pathway that is upstream of MEK.

The potentiation of hemin toxicity by activated ERK provides further evidence that it may be deleterious to acutely stressed CNS cells, and stands in sharp contrast to the protection it provides to serum-deprived PC12 cells (Xia et al. 1995). Although the mechanisms by which ERK produces such discordant results have not been elucidated, the duration of its activation has been hypothesized to be a critical determinant of its effect (Park and Koh 1999). Transient ERK phosphorylation by growth factors, with return to basal activity within minutes, has been associated with cellular proliferation and survival (Marshall 1995). However, in injury models in which ERK was deleterious, activation persisted for hours (Rundén et al. 1998; Park and Koh 1999; Stanciu et al. 2000; Kulich and Chu 2001). The results of this study are consistent with these observations. Hemin activated ERK for at least 1.5 h; at later timepoints cell injury and membrane lysis were widespread, resulting in loss or degradation of ERK protein.

The downstream events that mediate the deleterious effect of sustained ERK activation remain undefined. Park and Koh (1999) reported that exogenous zinc produced sustained ERK activation in cortical neurons, resulting in induction of the immediate early gene egr-1. Treatment with either PD98059 or an antisense oligonucleotide to egr-1 attenuated zinc toxicity. However, the gene product that increased neuronal vulnerability to zinc was not identified. In addition to its effects on gene expression, a variety of cellular enzymes are activated by ERK1/2. Two enzyme classes that are ERK substrates and that may contribute to acute CNS injury are the phospholipases A2 and the calpains (Geijsen et al. 2000; Glading et al. 2000). Prolonged phospholipase A2 activation may compromise membrane integrity, and may also increase oxidative stress and the inflammatory response via the release and metabolism of arachidonic acid (Farooqui et al. 1997). Calpains are proteases that target cytoskeletal and regulatory proteins (Melloni and Pontremoli 1989). Inhibition of both phospholipases A2 and calpains has been beneficial in experimental models of CNS ischemia and trauma (Bartus et al. 1995; Bazan et al. 1995; Farooqui et al. 1997; Kampfl et al. 1997). Their effects on heme-mediated injury are currently under investigation in our laboratory.

The relevance of these results to CNS hemorrhage remains to be determined. Although ERK activation has been observed after ischemia and hypoglycemia in vivo (Hu and Wieloch 1994; Kurihara et al. 1994), the effect of hemorrhage and trauma on this phenomenon has not been reported. The present data suggest that hemin release from a hematoma may be sufficient to activate ERK, and that this activation may potentiate oxidative injury to astrocytes. It is noteworthy that excitotoxicity has been implicated in the pathogenesis of hemorrhagic neuronal injury (Chen et al. 1991), and that MEK inhibitors have been reported to attenuate glutamate neurotoxicity (Satoh et al. 2000; Stanciu et al. 2000). These compounds may therefore have a beneficial effect on the oxidative, excitotoxic, and perhaps inflammatory components of hemorrhagic CNS injury, and appear to be suitable candidates for further in vivo testing.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  • Aikawa R., Komuro I., Yamazaki T., Zou Y., Kudoh S., Tanaka M., Shiojima I., Hiroi Y., Yazaki Y. (1997) Oxidative stress activates extracellular signal-regulated kinases through Src and Ras in cultured cardiac myocytes of neonatal rats. J. Clin. Invest. 100, 18131821.
  • Alessandrini A., Namura S., Moskowitz M. A., Bonventre J. V. (1999) MEK1 protein kinase inhibition protects against damage resulting from focal cerebral ischemia. Proc. Natl Acad. Sci. USA 96, 1286612869.DOI: 10.1073/pnas.96.22.12866
  • Alessi D. R., Cuenda A., Cohen P., Dudley D. T., Saltiel A. R. (1995) PD 098059 is a specific inhibitor of the activation of mitogen-activated protein kinase kinase in vitro and in vivo. J. Biol. Chem. 270, 2748927494.
  • Allen R. G. & Tresini M. (2000) Oxidative stress and gene regulation. Free Rad. Biol. Med. 28, 463499.
  • Balla G., Jacob H. S., Balla J., Rosenberg M., Nath K., Apple F., Eaton J. W., Vercellotti G. M. (1992) Ferritin: a cytoprotective strategem of endothelium. J. Biol. Chem. 267, 1814818153.
  • Bartus R. T., Elliott P. J., Hayward N. J., Dean R. L., Harbeson S., Straub J. A., Li Z., Powers J. C. (1995) Calpain as a novel target for treating acute neurodegenerative disorders. Neurol. Res. 17, 249258.
  • Bazan N. G., Rodriguez de Turco E. B., Allan G. (1995) Mediators of injury in neurotrauma: intracellular signal transduction and gene expression. J. Neurotrauma 12, 791814.
  • Bhat N. R. & Zhang P. (1999) Hydrogen peroxide activation of multiple mitogen-activated protein kinases in an oligodendrocyte cell line: role of extracellular signal-regulated kinase in hydrogen peroxide-induced cell death. J. Neurochem. 72, 112119.
  • Bhoite-Solomon V., Kessler-Icekson G., Shaklai N. (1993) Myocyte injury by hemin. In Vitro Cell. Dev. Biol. 29A, 636642.
  • Braughler J. M., Pregenzer J. F., Chase R. L., Duncan L. A., Jacobsen E. J., McCall J. M. (1987) Novel 21-amino steroids as potent inhibitors of iron-dependent lipid peroxidation. J. Biol. Chem. 262, 1043810440.
  • Braverman S., Helson C., Helson L. (1995) Hemin toxicity in a human epithelioid sarcoma cell line. Anticancer Res. 15, 19631967.
  • Chen M. H., Bullock R., Graham D. I., Miller J. D., McCulloch J. (1991) Ischemic neuronal damage after acute subdural hematoma in the rat: Effects of pretreatment with a glutamate antagonist. J. Neurosurg. 74, 944950.
  • Choi D. W., Maulucci-Gedde M., Kriegstein A. R. (1987) Glutamate neurotoxicity in cortical cell culture. J. Neurosci. 7, 357368.
  • Farooqui A. A., Yang H. C., Horrocks L. (1997) Involvement of phospholipase A2 in neurodegeneration. Neurochem. Int. 30, 517522.
  • Favata M. F., Horiuchi K. Y., Manos E. J., Daulerio A. J., Stradley D. A., Feeser W. S., Van Dyk D. E., Pitts W. J., Earl R. A., Hobbs F., Copeland R. A., Magolda R. L., Scherle P. A., Trzaskos J. M. (1998) Identification of a novel inhibitor of mitogen-activated protein kinase kinase. J. Biol. Chem. 273, 1862318632.
  • Fukunaga K. & Miyamoto E. (1998) Role of MAP kinase in neurons. Mol. Neurobiol. 16, 7995.
  • Geijsen N., Dijkers P. F., Lammers J. J., Koenderman L., Coffer P. J. (2000) Cytokine-mediated cPLA(2) phosphorylation is regulated by multiple MAPK family members. FEBS Lett. 471, 8388.
  • Glading A., Chang P., Lauffenburger D. A., Wells A. (2000) Epidermal growth factor activation of calpain is required for fibroblast motility and occurs via an ERK/MAP kinase signaling pathway. J. Biol. Chem. 275, 23902398.DOI: 10.1074/jbc.275.4.2390
  • Halliwell B. & Gutteridge J. M. C. (1989) Free Radicals in Biology and Medicine. Oxford University Press, Oxford.
  • Herrlich P. & Böhmer F. D. (1999) Redox regulation of signal transduction in mammalian cells. Biochem. Pharmacol. 59, 3541.DOI: 10.1016/s0006-2952(99)00298-1
  • Hosada K., Tamaki N., Masamura M., Matsumoto S., Maeda F. (1987) Magnetic resonance images of chronic subdural hematomas. J. Neurosurg. 67, 677683.
  • Hu B. R. & Wieloch T. (1994) Tyrosine phosphorylation and activation of mitogen-activated protein kinase in the rat brain following transient cerebral ischemia. J. Neurochem. 62, 13571367.
  • Hua Y., Xi G., Keep R. F., Hoff J. T. (2000) Complement activation in the brain after experimental intracerebral hemorrhage. J. Neurosurg. 92, 10161022.
  • Kampfl A., Posmantur R. M., Zhao X., Schmutzhard E., Clifton G. L., Hayes R. L. (1997) Mechanism of calpain proteolysis following traumatic brain injury: implications for pathology and therapy: a review and update. J. Neurotrauma 14, 121134.
  • Koh J. Y. & Choi D. W. (1988) Vulnerability of cultured cortical neurons to damage by excitotoxins: Differential susceptibility of neurons containing NADPH-diaphorase. J. Neurosci. 8, 21532163.
  • Kroemer G., Dallaporta B., Resche-Rigon M. (1998) The mitochondrial death/life regulator in apoptosis and necrosis. Ann. Rev. Physiol. 60, 619642.
  • Kulich S. M. & Chu C. T. (2001) Sustained extracellular signal-regulated kinase activation by 6-hydroxydopamine: implications for Parkinson's disease. J. Neurochem. 77, 10581066.
  • Kurihara J., Hu B. R., Wieloch T. (1994) Changes in the tyrosine phosphorylation of mitogen-activated protein kinase in the rat hippocampus during and following severe hypoglycemia. J. Neurochem. 63, 23462348.
  • Lander H. M., Ogiste J. S., Teng K. K., Novogrodsky A. (1995) p21ras as a common signaling target of reactive free radicals and cellular redox stress. J. Biol. Chem. 270, 2119521198.
  • Letarte P. B., Lieberman K., Nagatani K., Haworth R. A., Odell G. B., Duff T. A. (1993) Hemin: levels in experimental subarachnoid hematoma and effects on dissociated vascular smooth muscle cells. J. Neurosurg. 79, 252255.
  • Di Mari J. F., Davis R., Safirstein R. L. (1999) MAPK activation determines renal epithelial cell survival during oxidative injury. Am. J. Physiol. 277, F195F203.
  • Marshall C. J. (1995) Specificity of receptor tyrosine kinase signaling: transient versus sustained extracellular signal-regulated kinase activation. Cell 80, 179185.
  • Meguro T., Chen B., Lancon J., Zhang J. H. (2001) Oxyhemoglobin induces caspase-mediated cell death in cerebral endothelial cells. J. Neurochem. 77, 11281135.
  • Melloni E. & Pontremoli, S. (1989) The calpains. Trends Neurosci. 12, 438444.
  • Monteiro H. P., Ivaschenko Y., Fischer R., Stern A. (1991) Inhibition of protein tyrosine phosphatase activity by diamide is reversed by epidermal growth factor. FEBS Lett. 295, 146148.
  • Morrison D. K. & Cutler R. E. (1997) The complexity of Raf-1 regulation. Curr. Opin. Cell Biol. 9, 174179.
  • Motterlini R., Foresti R., Vandegriff K., Intaglietta M., Winslow R. M. (1995) Oxidative-stress response in vascular endothelial cells exposed to acellular hemoglobin solutions. Am. J. Physiol. 269, H648H655.
  • Namura S., Nagata I., Kikuchi H., Bonventre J. V., Alessandrini A. (2000) Intravenous administration of U0126 protects brain against focal ischemia and reperfusion. Soc. Neurosci. Abstr. 26, 774.
  • Neary J. T., Kang Y., Bu Y., Akong K., Peters C. M. (1999) Mitogenic signalling by ATP/P2Y purinergic receptors in astrocytes: involvement of a calcium-independent protein kinase C, extracellular signal-regulated protein kinase pathway distinct from the phosphatidylinositol-specific phospholipase C/calcium pathway. J. Neurosci. 19, 42114220.
  • Okragly A. & O'Brien M. (1999) Death check assay systems for monitoring apoptosis. Promega Notes 72, 36.
  • Padmanaban G., Venkateswar V., Rangarajan P. N. (1989) Haem as a multifunctional regulator. Trends Biochem. Sci. 14, 492496.
  • Park J. A. & Koh J. Y. (1999) Induction of an immediate early gene egr-1 by zinc through extracellular signal-regulated kinase activation in cortical culture: its role in zinc-induced neuronal death. J. Neurochem. 73, 450456.
  • Reynolds I. J. & Hastings T. (1995) Glutamate induces the production of reactive oxygen species in cultured forebrain neurons following NMDA receptor activation. J. Neurosci. 15, 33183327.
  • Rundén E., Seglen P. O., Haug F. M., Ottersen O. P., Wieloch T., Shamloo M., Laake J. H. (1998) Regional selective neuronal degeneration after protein phosphatase inhibition in hippocampal slice cultures: evidence for a MAP kinase-dependent mechanism. J. Neurosci. 18, 72967305.
  • Samuni A., Aronivitch J., Godinger G., Chevion M., Czapski G. (1983) On the cytotoxicity of vitamin C and metal ions. A site-specific Fenton mechanism. Eur. J. Biochem. 137, 119124.
  • Satoh T., Nakatsuka D., Watanabe Y., Nagata I., Kikuchi H., Namura S. (2000) Neuroprotection by MAPK/ERK kinase inhibition with U0126 against oxidative stress in a mouse neuronal cell line and rat primary cultured cortical neurons. Neurosci. Lett. 288, 163166.
  • Da Silva J. L., Morishita T., Escalante B., Staudinger R., Drummond G., Goligorsky M. S., Lutton J. D., Abraham N. G. (1996) Dual role of heme oxygenase in epithelial cell injury: contrasting effects of short-term and long-term exposure to oxidant stress. J. Lab. Clin. Med. 128, 290296.
  • Stanciu M., Wang Y., Kentor R., Burke N., Watkins S., Kress G., Reynolds I., Klann E., Angiolieri M. R., Johnson J. W., DeFranco D. B. (2000) Persistent activation of ERK contributes to glutamate-induced oxidative toxicity in a neuronal cell line and primary cortical ′neuron cultures. J. Biol. Chem. 275, 1220012206.DOI: 10.1074/jbc.275.16.12200
  • Vandegriff K. D. (1995) Stability and toxicity of hemoglobin solutions, in Blood Substitutes: Physiological Basis of Efficacy (WinslowR. M. Vandegriff K. D. and. Intaglietta M., eds), pp. 107131. Birkhäuser, Boston.
  • Wang X., Matindale J. L., Liu Y., Holbrook N. J. (1998) The cellular response to oxidative stress: influences of mitogen-activated protein kinase signalling pathways on cell survival. Biochem. J. 333, 291300.
  • Wang Z. Q., Ye. W., Yang G. Y. (2000) Inhibition of MAP kinase kinase (MEK) reduces brain injury in mouse focal cerebral ischemia. Soc. Neurosci. Abstr. 26, 774.
  • Xia Z., Dickens M., Raingeaud J., Davis R. J., Greenberg M. E. (1995) Opposing effect of ERK and JNK-p38 MAP kinases on apoptosis. Science 270, 13261331.
  • Yue T. L., Wang C., Gu J. L., Ma X. L., Kumar S., Lee J. C., Feuerstein G. Z., Thomas H., Maleeff B., Ohlstein E. H. (2000) Inhibition of extracellular signal-regulated kinase enhances ischemia/reoxygenation-induced apoptosis in cultured cardiac myocytes and exaggerates reperfusion injury in isolated perfused heart. Circ. Res. 86, 692699.
  • Zhu Y., Hon T., Zhang L. (1999) Heme initiates changes in the expression of a wide array of genes during the early erythroid differentiation stage. Biochem. Biophys. Res. Comm. 258, 8793.DOI: 10.1006/bbrc.1999.0586