SEARCH

SEARCH BY CITATION

Keywords:

  • heat shock protein;
  • neuron;
  • oxidative stress;
  • proteasome;
  • SH-SY5Y.

Abstract

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

Recent studies have demonstrated that inhibition of the proteasome, an enzyme responsible for the majority of intracellular proteolysis, may contribute to the toxicity associated with oxidative stress. In the present study we demonstrate that exposure to oxidative injury (paraquat, H2O2, FeSO4) induces a rapid increase in reactive oxygen species (ROS), loss of mitochondrial membrane potential, inhibition of proteasome activity, and induction of cell death in neural SH-SY5Y cells. Application of proteasome inhibitors (MG115, epoxomycin) mimicked the effects of oxidative stressors on mitochondrial membrane potential and cell viability, and increased vulnerability to oxidative injury. Neural SH-SY5Y cells stably transfected with human HDJ-1, a member of the heat shock protein family, were more resistant to the cytotoxicity associated with oxidative stressors. Cells expressing increased levels of HDJ-1 displayed similar degrees of ROS formation following oxidative stressors, but demonstrated a greater preservation of mitochondrial function and proteasomal activity following oxidative injury. Cells transfected with HDJ-1 were also more resistant to the toxicity associated with proteasome inhibitor application. These data support a possible role for proteasome inhibition in the toxicity of oxidative stress, and suggest heat shock proteins may confer resistance to oxidative stress, by preserving proteasome function and attenuating the toxicity of proteasome inhibition.

Abbreviations used
AD

Alzheimer's disease

BSA

bovine serum albumin

CNS

central nervous system

DCF

2,7-dichlorofluorescin

HSP

heat shock protein

IRI

ischemia-reperfusion injury

JC-1

5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolycarbocyanine iodide

ROS

reactive oxygen species.

Cells within the CNS are continually exposed to reactive oxygen species (ROS), and must therefore systematically respond to ROS mediated damage, in order to prevent oxidative injury. Elevated levels of oxidative damage are evident in numerous neurodegenerative disorders, including Alzheimer's disease (AD) (Markesbery 1997) and ischemia-reperfusion injury (IRI) (Chan 1996), possibly contributing to the neuronal degeneration observed in those conditions (Facchinetti et al. 1998). Although it is likely that the mechanism whereby oxidative stress induces cell death is likely multifactoral, the identification of which biochemical alterations are disrupted, and those responsible for mediating oxidative stress toxicity, has not been fully elucidated.

The proteasome is a large intracellular protease that is responsible for mediating the majority of intracellular proteolysis (Rock et al. 1994; Goldberg et al. 1997; Tanaka 1998), including the degradation of most oxidized proteins (Grune and Davies 1997; Grune et al. 1997). Recent studies indicate that proteasome activity is inhibited following exposure to oxidative stress (Reinheckel et al. 1998; Okada et al. 1999), and is inhibited in both AD (Keller et al. 2000a; Lopez Salon et al. 2000) and experimental models of IRI (Keller et al. 2000c). Pharmacological inhibition of the proteasome is sufficient to cause cell death in primary neuron cultures (Boutillier et al. 1999; Canu et al. 2000; Keller and Markesbery 2000; Pasquini et al. 2000; Qui et al. 2000) and neuronal cell lines (Lopes et al. 1997; Keller et al. 2000b), suggesting that proteasome inhibition may play a role in the neurotoxicity associated with oxidative stress. Although these data indicate a possible role for proteasome inhibition in neuronal oxidative stress, direct evidence for oxidative injury inducing proteasome inhibition in neuronal cells, and a direct demonstration for proteasome inhibition playing a causal or contributory role in oxidative stress-associated neurotoxicity, has not been demonstrated.

Cells of the CNS can rapidly increase the intracellular levels of heat shock protein (Hsp) family members in response to a wide variety of environmental stimuli, including oxidative stress (Omar and Pappolla 1993; Ohtsuka and Hata 2000; Ohtsuka and Suzuki 2000). Recent studies have demonstrated that increased expression of Hsp's may suppress the toxicity of oxidative stress (Liu et al. 1997; Beucamp et al. 1998), while decreased Hsp expression results in increased levels of oxidative stress-induced toxicity (Nakano et al. 1997; Yu et al. 1999). However, the exact mechanism by which Hsps confer resistance to oxidative injury has not been determined. In the present study we examined the possible neuroprotective effects of the molecular chaperone HDJ-1, which is a member of the Hsp40 family.

In the present study, we sought to determine the role of proteasome inhibition in oxidative stress toxicity, and determine if the neuroprotective effect of Hsp expression is mediated in part by attenuating the toxicity associated with proteasome inhibition. Together, these data demonstrate a possible role for proteasome inhibition in oxidative stress toxicity, and suggest that Hsp-conferred neuroprotection is associated with a preservation of proteasome function, and resistance to the toxicity associated with proteasome inhibition.

Materials and methods

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

Materials

The SH-SY5Y cells were obtained from the American Tissue Culture Collection (Manassas, VA, USA), while all culturing supplies and transfection materials were obtained from Gibco Life Technologies (Gaithersburg, MD, USA). Western blot reagents were obtained from Amersham Life Science (Arlington Heights, IL, USA). The fluorescent ROS indicator DCF, and mitochondrial membrane potential indicator JC-1, were obtained from Molecular Probes (Eugene, OR, USA). Proteasome inhibitors were obtained from Calbiochem (La Jolla, CA, USA). Antibodies were obtained from StressGen (Victoria, BC, Canada). The proteasome substrates were obtained from Bachem (Torrance, CA, USA), and all other supplies were obtained from Sigma Chemical (St Louis, MO, USA).

Generation of stably transfected HDJ-1 cells

Neural SH-SY5Y cells were transfected with equimolar amounts of pcDNA3-HDJ-1, or pcDNA3 lacking insert, via Lipofectamine-Plus reagent. The full length human HDJ-1 expression vector (pcDNA3-HDJ-1), was a generous gift of Dr Henry Paulson (University of Iowa, IA, USA). Stably transfected cells were selected by incubation in selection medium (MEM containing 10% heat inactivated fetal bovine serum, 1% penicillin–streptomycin, 500 µg/mL G418). Populations of stably transfected vector and HDJ-1 transfected cells were isolated, characterized and remained in selection medium during all cell passages. For all experimentation cells were placed in selection medium lacking G418 overnight, and remained in this medium throughout the experimentation. Cells of fewer than 15 passages were utilized for studies described in this report. Western blot analysis revealed that there was no difference between empty vector transfected, and HDJ-1 transfected cells, with regards to the expression of Hsp27, Hsp32, or α-and β-subunits of the 20S proteasome (J.N. Keller, unpublished observations).

Analysis of cell survival, ROS and mitochondrial membrane potential

Cell survival was determined by analysis of trypan blue exclusion, as described previously (Keller et al. 2000b). Determinations of ROS levels was accomplished using fluorescent indicator DCF, as described previously (Yu et al. 1999). Mitochondrial membrane potential was determined by analysis of JC-1 as described previously (Keller et al. 1998).

Analysis of proteasome activity

The chymotrypsin-like activity of the proteasome activity was determined by measuring the rates of chymotrypsin-like and postglutamyl peptidase proteasome activities as described previously (Keller et al. 2000a,b,c). Briefly, following experimental treatment, cultures were placed on ice and collected in homogenization buffer [10 mm Tris–HCl (pH 7.8), 0.5 mm dithiothreitol, 5 mm ATP, and 5 mm MgCl2]. Protein determinations were made on the resulting lysate, and equal amounts of protein (200 µg/500 µL) incubated with 5 µL of chymotrypsin-like (suc-Leu-Leu-Val-Tyr-MCA; 5 mm) or postglutamyl peptidase (suc-Leu-Leu-Glu-MCA; 5 mm) substrate. As described in previous studies (Keller et al. 2000b), background fluorescence was determined by incubating lysate with 50 µm lactacystin, 30 min prior to the addition of fluorogenic substrate, which inhibits approximately 95–100% of proteasome activity. The amount of 7-amido-4-methylcoumarin (MCA) liberation was quantified using a fluorescent plate reader, using a standard of free MCA as standard.

Western blot analysis

The levels of HDJ-1 (1 : 1000), HDJ-2 (1 : 500), Hsc70 (1 : 1000), and Hsp90 (1 : 1000) was determined by western blot analysis as described previously (Keller et al. 1998, 2000a,b). Briefly, equal amounts of protein (50 µg) were separated on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), transferred to nitrocellulose and developed using the indicated antibody and an enhanced chemiluminescent kit.

Statistical analysis

Determinations of statistical analysis were made by anova using Scheffe's post hoc analysis.

Results

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

Oxidative stress induces loss of proteasome activity and viability

To determine the possible contribution of proteasome inhibition to oxidative stress-induced neural cell death, SH-SY5Y cells were exposed to oxidative injuries and analyzed for cellular viability and proteasome activity. Following exposure to a variety of oxidative stressors (paraquat, H2O2, FeSO4) there was a dose- and time-dependent loss of mitochondrial membrane potential and induction of cell death (Fig. 1). Oxidative stressors induced an inhibition of chymotrypsin-like proteasome activity, that occurred prior to cell death, and was evident as early as 1 h post-treatment (Fig. 2). Post-glutamyl-peptidase activity of the proteasome was also inhibited following application of oxidative stressors (data not shown), consistent with oxidative stressors inducing a gross inhibition of multiple proteasomal proteolytic activities. Application of proteasome inhibitors (MG115, epoxomycin) mimicked the effects of oxidative stressors, disrupting mitochondrial homeostasis and inducing cell death (Fig. 3). Application of 1, 10 and 25 µm MG115 for 3 h was observed to inhibit 43%, 59% and 78% of chymotrypsin-like proteasome activity in neural SH-SY5Y cells (J. N. Keller, unpublished observations). In addition to directly inducing cell death, application of proteasome inhibitors increased the vulnerability of neural SH-SY5Y cells to the toxicity associated with oxidative stressors (Fig. 4).

image

Figure 1. Oxidative stress induces a loss of mitochondrial membrane potential and loss of cellular viability in neural SH-SY5Y cells. (a) Following exposure to paraquat [(●) 20, (○) 200 µm], H2O2[(▪) 50, (□) 500 µm], or FeSO4[(▴) 50, (▵) 500 µm] mitochondrial membrane potential was determined using the fluorescent indicator JC-1. (b) Following exposure to paraquat, H2O2, or FeSO4 cellular viability was assessed 12 and 24 h following treatment using trypan blue exclusion. Data are expressed as the mean and SEM of results from at least six separate cultures from at least two separate experiments. *p < 0.05 compared with control; **p < 0.05 compared with cultures treated with lower concentrations of same treatment.

Download figure to PowerPoint

image

Figure 2. Oxidative stress impairs proteasome activity. Following exposure to paraquat [(●) 20, (○) 200 µm], H2O2[(▪) 50, (□) 500 µm], or FeSO4[(▴) 50, (▵) 500 µm] chymotrypsin-like proteasome activity was determined by analysis of suc-Leu-Leu-Val-Tyr-MCA cleavage. The basal levels of chymotrypsin-like proteasome activity ranged from 3.4 to 4.1 nmol × mg−1 × min −1. Data are expressed as the mean and SEM of results from at least six separate cultures from at least two separate experiments. *p < 0.05 compared with control; **p < 0.05 compared with cultures treated with lower concentrations of same treatment.

Download figure to PowerPoint

image

Figure 3. Proteasome inhibitors induce a loss of mitochondrial membrane potential and cellular viability. (a) Following exposure to the proteasome inhibitors MG115 [(●) 1, (▪) 10, (◆) 25 µm] or epoxomycin [(○) 1, (□) 10, (◊) 25 µm] mitochondrial membrane potential was determined using the fluorescent indicator JC-1. (b) Following exposure to MG115 or epoxomycin cellular viability was assessed 12 and 24 h following treatment using trypan blue exclusion. Data are expressed as the mean and SEM of results from at least six separate cultures from at least two separate experiments. *p < 0.05 compared with control.

Download figure to PowerPoint

image

Figure 4. Proteasome inhibitors increase vulnerability to oxidative stressors. (a) Neural SH-SY5Y cells were exposed to the proteasome inhibitor epoxomycin [(▪) 1 µm], paraquat [(●) 20 µm], H2O2[(▴) 50 µm], or paraquat (○) and H2O2 (▵) in combination with epoxomycin, and analyzed for mitochondrial membrane potential (a) or cellular viability (b). Data are expressed as the mean and SEM of results from at least six separate cultures from at least two separate experiments. *p < 0.05 compared with control; **p < 0.05 compared with cultures treated with paraquat or H2O2 alone.

Download figure to PowerPoint

Characterization of neural cells stably transfected with HDJ-1

Because previous studies have demonstrated that increased Hsp expression may confer resistance to oxidative stress-induced cell death, we sought to determine the effect of increased Hsp expression on proteasome activity, using neural SH-SY5Y cultures stably transfected with the human Hsp, HDJ-1. Two populations of HDJ-1 transfected SH-SY5Y cells were chosen based on their ∼8–10-fold higher levels of HDJ-1 expression, compared with lines transfected with empty vector (Fig. 5). Cells expressing increased levels of HDJ-1 did not exhibit increased levels of HDJ-2, Hsp90, or Hsc70 compared with cells transfected with empty vector (Fig. 5). Although HDJ-1-transfected cells displayed an altered morphology (Fig. 5), no difference in antioxidant enzyme levels (MnSOD, Cu/Zn SOD, catalase), proteasome activity, or proliferation rate was observed between HDJ-1-transfected, and empty vector-transfected cells (data not shown).

image

Figure 5. Characterization of HDJ-1-transfected cells. Cells transfected with vector alone (a) or HDJ-1 (b) were analyzed by 20 × phase microscopy. Cells transfected with vector alone have normal morphology, while HDJ-1-transfected cells display a large number of neuritic processes. Cell lysates (50 µg), from two populations of vector and HDJ-1-transfected cells, were analyzed by western blot analysis for the expression of HDJ-1, HDJ-2, Hsp90, or Hsc70. Data is representative of results from at least three separate experiments.

Download figure to PowerPoint

Ectopic expression of HDJ-1 confers resistance to the toxicity of oxidative stress

Cells transfected with HDJ-1 were more resistant to all oxidative injuries (Fig. 6), with HDJ-1-transfected cells demonstrating higher levels of mitochondrial membrane potential, proteasome activity and cell survival following application of oxidative stressors (Fig. 6). No difference in the level of neuroprotection was observed between the two populations of HDJ-1-transfected cells (data not shown). Interestingly, no difference in ROS levels were observed between HDJ-1- and empty vector-transfected cells, following exposure to oxidative injury (Fig. 7). In addition to being more resistant to the toxicity of oxidative stressors, cells transfected with HDJ-1 were more resistant to the toxicity associated with proteasome inhibitor application (Fig. 8).

image

Figure 6. Cells transfected with HDJ-1 are more resistant to the toxicity of oxidative stressors. Neural SH-SY5Y cells transfected with vector alone (vector, solid) or HDJ-1 (shaded) were analyzed for mitochondrial membrane potential (a) or chymotrypsin-like proteasome activity (b) 6 h following application of paraquat (200 µm), H2O2 (500 µm), or FeSO4 (500 µm). (c) Vector and HDJ-1 cells were analyzed for cell viability following 24 h incubation with paraquat (200 µm), H2O2 (500 µm), or FeSO4 (500 µm). The basal levels of chymotrypsin-like proteasome activity ranged from 3.3 to 4.0 nmol × mg−1 × min−1. Data are expressed as the mean and SEM of results from at least six separate cultures from at least two separate experiments. *p < 0.05 compared with control; **p < 0.05 compared with cultures transfected with vector alone.

Download figure to PowerPoint

image

Figure 7. No difference in ROS levels is observed between cells transfected with HDJ-1 (shaded) or vector alone (solid). Neural SH-SY5Y cells were analyzed for ROS levels, using the fluorescent ROS indicator DCF, 6 h following application of paraquat (200 µm), H2O2 (500 µm), or FeSO4 (500 µm). Data are expressed as the mean and SEM of results from at least six separate cultures from at least two separate experiments. *p < 0.05 compared with control.

Download figure to PowerPoint

image

Figure 8. Cells transfected with HDJ-1 are more resistant to the toxicity of proteasome inhibitors. (a) Neural SH-SY5Y cells transfected with vector alone (vector, solid) or HDJ-1 (shaded) were analyzed for mitochondrial membrane potential 6 h following application of the proteasome inhibitors MG115 (10, 25 µm) or epoxomycin (10, 25 µm). (b) Vector and HDJ-1 cells were analyzed for cell viability following 24 h incubation with the proteasome inhibitors MG115 (10, 25 µm) or epoxomycin (10, 25 µm). Data are expressed as the mean and SEM of results from at least six separate cultures from at least two separate experiments. *p < 0.05 compared with control; **p < 0.05 compared with cultures transfected with vector alone.

Download figure to PowerPoint

Discussion

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

The data presented in this study indicate a possible role for proteasome inhibition in oxidative stress-induced neurotoxicity. A causal role for proteasome inhibition as a mediator of neurotoxicity is supported by the fact that: (1) proteasome inhibition occurred rapidly following application of oxidative stressors, preceding cell death; (2) application of proteasome inhibitors alone was sufficient to induce cell death; (3) application of proteasome inhibitors increased the toxicity of oxidative stressors. Previous studies have suggested that the accumulation of oxidized, damaged and aggregated proteins following oxidative injury may be the result of proteasome inhibition (Grune et al. 1995; Grune and Davies 1997; Okada et al. 1999). These previous reports suggested that, while proteasome activity is harnessed towards the degradation of proteins modified during oxidative stress, the activity of the proteasome becomes less efficient, or possibly inhibited, resulting in the accumulation of oxidized, aggregated, or damaged proteins. It is possible that proteasome activity may become inhibited as the result of the oxidized or aggregated protein substrates themselves. For example, exposure to excessively oxidized or aggregated proteins, potently inhibits proteasome activity in vitro and in vivo (Friguet and Szweda 1997; Reinheckel et al. 1998; Okada et al. 1999). Consistent with a role for oxidative stress-modified proteins modulating the activity of the proteasome, in the present study we demonstrated that cells expressing increased levels of the Hsp (HDJ-1), had preserved proteasome activity following application of oxidative stressors. The ability of increased levels of HDJ-1 to attenuate proteasome inhibition did not appear to be due to a decrease in ROS levels, or altered levels of proteasome subunits. However, these data do not rule out the possibility that increased levels of HDJ-1 suppress the levels of non-DCF detectable ROS. Previous studies have demonstrated that Hsp play important roles in protein trafficking, protein folding and inhibiting or reversing protein aggregation (Omar and Pappolla 1993; Ohtsuka and Hata 2000; Ohtsuka and Suzuki 2000). It is therefore possible that increased Hsp levels preserve proteasome activity following oxidative injury, by delaying or reversing the deleterious effects of oxidative stress on intracellular proteins.

Data from the present study suggest that the beneficial effects of elevated Hsp expression to neural survival may be mediated, at least in part, via their amelioration of proteasome inhibition-associated toxicity. Previous studies have suggested that Hsps may play important roles in regulating proteasome activity via their roles in recruitment of substrates to the proteasome, or through beneficial Hsp–proteasome interactions (Ciechanover et al. 2000). For example, studies have demonstrated co-immunopreciptiation or co-localization of the proteasome with Hsps (Cummings et al. 1998; Luders et al. 2000), and cellular reconstitution experiments have demonstrated a requirement for some Hsps for proper proteasome activity (Conconi et al. 1996, 1998; Luders et al. 2000). Our data suggest that HDJ-1 may perform important chaperone functions that aid considerably in maintaining proteasome activity, particularly during oxidative injury. Although no previous work has identified a role for HDJ-1 in the suppression of oxidative stress-induced toxicity, previous studies have demonstrated that increased expression of HDJ proteins confers neuroprotection to the neurotoxicity associated with polyglutamine expansion (Chai et al. 1999; Jana et al. 2000; Kobayashi et al. 2000). It is interesting to note that the beneficial effects of increased HDJ expression in previous studies has been suggested to mediated, in part, by the attenuation of protein aggregation (Chai et al. 1999; Jana et al. 2000; Muchowski et al. 2000). Other evidence supporting an essential role for HDJ-1 in mediating neuron survival comes from antisense studies, in which antisense mediated down regulation of HDJ-1 in neural SH-SY5Y cells resulted in cell death (data not shown).

In the present study, proteasome inhibitors induced cell death in neural SH-SY5Y cells, and increased the toxicity of oxidative stressors. Although the mechanism by which proteasome inhibitors induced cell death, and increased vulnerability to oxidative stress is likely multifactoral, it is interesting to note the effects of proteasome inhibition and oxidative stress on mitochondrial membrane potential. At present it is unclear as to how proteasome inhibition may cause the loss of mitochondrial membrane potential. Based on the ability of increased HDJ-1 expression to attenuate proteasome inhibition-induced loss of mitochondrial membrane potential, it is likely that protein aggregation or deleterious protein–protein interactions contribute to the loss of mitochondrial membrane potential. Disruption of mitochondrial function is believed to play a contributing role to neuron death in a number of neurodegenerative conditions associated with oxidative stress (Chan 1996; Facchinetti et al. 1998; Keller et al. 1998). Data from the present study suggest that proteasome inhibition may therefore possibly contribute to the disruption of mitochondrial membrane potential in conditions associated with oxidative stress.

Although the data presented in the current study are in agreement with numerous reports (Boutillier et al. 1999; Canu et al. 2000; Keller and Markesbery 2000; Pasquini et al. 2000; Qui et al. 2000), indicating a role for proteasome inhibition in the neurodegenerative process, it is important to point out that proteasome inhibitor application has been demonstrated to attenuate neuronal death in some experimental paradigms (Sadoul et al. 1996; Favit et al. 2000). Although it is unclear how proteasome inhibitors mediate their neuroprotective effect, it is possible that proteasome inhibitors induce a beneficial form of heat shock in these experimental paradigms. For example, application of proteasome inhibitors has been demonstrated to be a potent inducer of Hsp family members in some cell types (Goldberg et al. 1997; Lee and Goldberg 1998). Alternatively, cells resistant to proteasome inhibitor toxicity may have higher levels of lysosomal proteases, and are therefore more efficient at shunting intracellular proteins to these proteases following proteasome inhibitor application, and thus prevent or delay the deleterious accumulation of toxic proteins. Recent studies also suggest that the proteasome may play a role in the regulation of some forms of NFkB activation (Goldberg et al. 1997; Tanaka 1998), which can have anti- or pro-apoptotic effects, depending on cell type (Mattson et al. 2000).

At present, the processes responsible for regulating the activity of the proteasome are poorly defined. Data in the present study provide in vitro evidence that non-proteasomal proteins, in particular Hsps, may play important roles in regulating proteasome activity. Because proteasome activity appears important to maintaining neuronal homeostasis and neuronal survival, data in the present study suggest that upregulation of Hsp family members may be of therapeutic benefit, via their amelioration of proteasome inhibition and proteasome inhibition-associated toxicity.

Acknowledgements

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

This work was supported by grants from the American Heart Association (Scientist Development Award), American Health Assistance Foundation, and Huntington's Disease Society of America (Donnellan Family Fund) (JNK). The authors thank Dr Sonia Carlson for the use of the fluorescent plate reader, Dr H. Paulson for supplying the HDJ-1 plasmid, and thank E. Dimayuga and F. F. Huang for technical assistance.

References

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