|Home | About | Journals | Submit | Contact Us | Français|
Reactive oxygen species (ROS) are implicated in 6-hydroxydopamine (6-OHDA) injury to catecholaminergic neurons; however, the mechanism(s) are unclear. In addition to ROS generated during autoxidation, 6-OHDA may initiate secondary cellular sources of ROS that contribute to toxicity. Using a neuronal cell line, we found that catalytic metalloporphyrin antioxidants conferred protection if added 1 hour after exposure to 6-OHDA, whereas the hydrogen peroxide scavenger catalase failed to protect if added more than 15 min after 6-OHDA. There was a temporal correspondence between loss of protection and loss of the ability of the antioxidant to inhibit 6-OHDA-induced ERK phosphorylation. Time course studies of aconitase inactivation, as an indicator of intracellular superoxide, and MitoSOX red, a mitochondria targeted ROS indicator, demonstrate early intracellular ROS followed by a delayed phase of mitochondrial ROS production, associated with phosphorylation of a mitochondrial pool of ERK. Furthermore, upon initiation of mitochondrial ROS and ERK activation, 6-OHDA-injured cells became refractory to rescue by metalloporphyrin antioxidants. Together with previous studies showing that inhibition of the ERK pathway confers protection from 6-OHDA toxicity, and that phosphorylated ERK accumulates in mitochondria of degenerating human Parkinson’s disease neurons, these studies implicate mitochondrial ERK activation in Parkinsonian oxidative neuronal injury.
Parkinson’s disease (PD) is a common age-related neurodegenerative disease characterized by selective neuronal cell death within several regions of the brain (Reviewed in ). Degeneration of the dopaminergic neurons of the nigral-striatal projection accounts for many of the major symptoms. Although the molecular mechanisms leading to neuronal cell death in PD remain unclear, studies of post-mortem human tissues, genetic studies of familial PD, and toxin/pesticide based models of PD suggest common pathways involving oxidative stress, mitochondrial dysfunction, disrupted protein turnover, and altered kinase signaling in dopaminergic neuron degeneration [2-9].
6-hydroxydopamine (6-OHDA), a redox cycling dopamine analog , is an oxidative neurotoxin that causes a Parkinsonian pattern of neuronal loss in rodents following intrastriatal injection [11, 12]. Although used as an exogenous neurotoxin in this model, 6-OHDA can be formed from dopamine in vivo, and elevated levels have been detected in body fluids of patients with PD . Notably, 6-OHDA injury recapitulates several features of degenerating neurons observed in human PD tissues. These include proteasome inhibition, α-synuclein aggregation, oxidation and nitration of proteins, increased protein ubiquitination, cleaved caspase 3 expression, glutathione depletion, and cytoplasmic accumulation of activated signaling proteins [14-21]. A better understanding of 6-OHDA mediated neurotoxicity could lend important insights into injury and degeneration pathways shared among different causes of dopaminergic neuron degeneration.
The mechanisms by which 6-OHDA elicits its neurotoxic effects have yet to be fully elucidated, although studies implicate a role for oxidative mediators [22, 23]. 6-OHDA metabolism generates a series of ROS at physiologic pH including hydrogen peroxide, para-quinone, and superoxide (Reviewed in ). The role of these oxidative species in 6-OHDA toxicity and their intracellular sites of action remain ill defined.
We have previously shown that catalase and metalloporphyrin antioxidants that are capable of affecting intracellular compartments conferred protection against cytotoxicity in 6-OHDA treated B65 cells . Furthermore, activation of the ERK signaling pathway contributes to 6-OHDA toxicity [25, 26]. Moreover, mitochondrial, but not extracellular, superoxide dismutase protects against delayed retrograde substantia nigra cell death following intrastriatal injection of 6-OHDA . Taking into account the temporal and spatial considerations in this model, we hypothesize that 6-OHDA elicits a secondary wave of mitochondrial ROS associated with neurotoxic ERK activation. We found that catalase and metalloporphyrins exhibit different kinetics of protection against 6-OHDA-mediated cytotoxicity that correlate with the ability of the antioxidant to inhibit sustained ERK phosphorylation. In addition, 6-OHDA elicits two phases of intracellular superoxide production associated with increased mitochondrial ROS production and phosphorylation of ERK in mitochondrial fractions. The cells become refractory to metalloporphyrin antioxidant protection upon initiation of mitochondrial ROS and ERK phosphorylation, implicating these intracellular events in 6-OHDA neurotoxicity.
Chemical reagents (except where specified) were purchased from Sigma, St. Louis, MO
Stock solutions of 6-OHDA were prepared in either MilliQ water (Millipore, Billerica, MA) or in in MilliQ water containing 0.05% (w/v) ascorbate. Spectral scanning of 6-OHDA quinone preparations revealed a peak absorbance at 480 nm (data not shown). The test samples were monitored for autoxidation at 37°C by following the increase in absorbance at 480 nm at 1-minute intervals using a SpectraMax M2 plate reader (Molecular Devices, Sunnyvale, CA), with values normalized to the A 480 for each well obtained prior to addition of 6-OHDA. For experiments performed in the presence of catalytic antioxidants, antioxidants were added to D10 15 minutes prior to the addition of 6-OHDA. For experiments performed in the presence of culture media, a 20× 6-OHDA stock dissolved in either water or 0.05% ascorbate was diluted to produce a final concentration of 500 μM 6-OHDA.
Cell injury was determined using 2 independent methods (LDH release and MTS assays) as described previously . In experiments assessing the effect of [5,10,15,20-tetrakis(1-ethylpyridinium-2-yl)porphyrinato] manganese(III) pentachloride (AEOL 10113, gift of Incara Pharmaceuticals, Research Triangle Park, NC), manganese (III) tetrakis (4-benzoic acid) porphyrin (MnTBAP, Alexis Biochemicals, San Diego, CA), and beef liver catalase (Roche Molecular Biochemicals, Indianapolis, IN, 260,000 U/ml), reagents were diluted in D10 and added 30 minutes prior to the addition of 6-OHDA or at different intervals following 6-OHDA (delayed administration time course). Heat-inactivation of catalase (5 minutes, 100 degrees) resulted in >90% loss of activity as confirmed by 2 assays for catalase activity: Amplex red catalase assay kit (Molecular Probes, Eugene, OR) or by monitoring the disappearance of hydrogen peroxide at 240 nm . To confirm that loss of protective effects observed with time was not due to loss of catalytic activity, aged catalytic antioxidants (37 °C, 5% CO2 for 2 h and 4 h for catalase and metalloporhyrins respectively) were added to untreated cells that were subsequently exposed to fresh 6-OHDA.
Cell lysis and electrophoresis was performed as previously described [24, 25]. Following electrophoresis, proteins were transferred to Immobilon-P (Millipore, Bedford, MA) and probed for phospho-ERK (Cell Signaling, Beverly, MA) and total ERK (Upstate Biotechnology, Lake Placid, NY) as previously described . Densitometric analysis of blots was performed using the electrophoresis documentation and analysis system 120 (Kodak, Rochester, NY). Data was expressed as the ratio of P-ERK/T-ERK band intensities as previously described . For mitochondrial preparations, immunoblots were also probed with lactate dehydrogenase (Santa Cruz Biotechnology, Santa Cruz, CA) and SOD2 (Upstate Biotechnology, Lake Placid, NY) to confirm the purity of the preparations.
Immunofluorescence for phospho-ERK was performed as described previously with slight modifications . B65 cells were grown on cover slips and exposed to 6-OHDA for 6 h. Following fixation with 3% paraformaldehyde, slips were probed simultaneously with phospho-ERK (1:4000) and SOD2 (1:50) followed by Cy3 conjugated donkey anti-mouse IgG (1:500, Jackson ImmunoReasearch, West Grove, PA) and Alexa fluro488 conjugated donkey anti-rabbit (1:500, Invitrogen, Carlsbad, CA). The immunofluorescence was visualized using a Zeiss Axioplan 2 confocal imaging system. An Ar 488/HeNEL 543 laser provided the excitation light beam, and a neutral density filter (T 0.01) was used to uniformly attenuate the intensity of laser light. The excitation light (488 and 543 nm) was passed through a dichroic beam splitter (HFT 488/543/635) that allows the capture of images in both green and red channels simultaneously. Co-localization was confirmed using Z-sectioning and orthogonal image analysis with Zeiss LSM 510 software (Version 3.0). Specificity of staining for P-ERK and SOD2 double fluorescence was verified using nonimmune rabbit and mouse sera and electron microscopy as described previously .
Following exposure to either vehicle (0.0025% (w/v) ascorbate final) or 500 μM 6-OHDA, cells were washed with DPBS, scraped, pelleted (7000 × g, 5 min, 4 °C), snap frozen, and stored at -80 °C until assayed. For activity determination, cell pellets were thawed, resuspended in sonication buffer (50 mM Tris, 2mM sodium citrate, 0.6 mM manganese chloride, pH 7.4 ), and sonicated. Post-sonication supernatants (2000 × g, 4 min, RT) were assayed in triplicate for aconitase activity in a linear velocity using the BIOXYTECH Aconitase-340 assay (OxisResearch, Portland, OR). Data is reported as total aconitase activity per mg of protein, which was determined using the Coomassie Plus protein assay (Pierce, Rockford, IL).
Cells were plated at 300,000 cells per well 6 well plate and grown for 48 hours at which time cells were treated with either vehicle (0.0025% ascorbate) or 6-OHDA (500 μM) for the indicated time. For experiments using the MEK 1/2 inhibitors UO126 and PD 98059 (Calbiochem, San Diego, CA), cells were incubated in D10 media containing no additives, UO126 (final concentration 10 μM), PD 98059 (final concentration 50 μM), or an equivalent amount of MEK 1/2 inhbitor vehicle (DMSO) 30 minutes prior to the addition of 6-OHDA or ascorbate. Cells were washed with warm Hanks buffered saline solution (HBSS) then incubated at 37 °C in the dark for 10 min in HBSS containing 5 μM mitoSOX (Invitrogen, Carlsbad, CA). Following the removal of excess mitoSOX reagent, quantitative measurements of fluorescence were determined using the SpectraMax M2 fluorescent plate reader (Molecular Devices, Sunnyvale, CA) using the well scan function with excitation wavelength of 510 nm, emission wavelength of 580 nm with a 570 nm cutoff filter. Data were reported as the mean of 9 separate regions +/- SEM.
B65 cells grown on 2-100 mm2 plates were treated with media change, media change plus vehicle (0.0025% ascorbate), or media change plus 500 μM 6-OHDA for 4 h and harvested for mitochondrial purification using the mitochondria isolation kit for cultured cells per manufacturer instructions (Pierce, Rockford, IL). Isolated mitochondrial pellets were resuspended in mitochondrial lysis buffer [31, 32], protein concentration of the mitochondrial and cytoplasmic fractions were determined as above, and equal amounts of protein was subjected to electrophoresis and immunoblot analysis as described above.
Statistical significance was determined using ANOVA with post-hoc multiple comparison testing by Fisher’s least significant difference (LSD). A p-value of less than 0.05 was regarded as significant.
Under physiologic conditions, 6-OHDA is capable of undergoing oxidation to produce several reactive oxygen species (ROS) as well as quinones [2, 10]. Stock solutions of 6-OHDA prepared in water showed a low rate of quinone formation at 37 °C, but 6-OHDA prepared in 0.05% (w/v) ascorbate was stable for 2 h (Figure 1A). However, in the presence of culture media, 6-OHDA rapidly oxidized to its quinone form by 15 min, and there were no differences between 6-OHDA prepared in dH2O and 6-OHDA prepared in ascorbate vehicle (Figure 1B). The presence or absence of B65 cells also did not significantly alter the autoxidation rate of 6-OHDA in media (not illustrated).
We have previously shown that increased levels of extracellular SOD has no effect on 6-OHDA toxicity, while pre-treatment with catalase and metalloporphyrin antioxidants conferred protection . Metallporphyrins are small molecules that possess catalytic antioxidant activity. These compounds have been demonstrated to be protective in a variety of in vitro models of oxidative stress (Reviewed in ). Unlike large proteins such as SOD and catalase, these compounds provide a mechanism to deliver intracellular catalytic antioxidant activity . Although these compounds are known to exhibit high levels of activity catalyzing dismutation of superoxide, they also show effects on hydrogen peroxide, peroxynitrite, and lipid peroxidation. Thus, the potential role of intracellular superoxide in 6-OHDA toxicity was still unclear.
To further study the mechanism(s) of protection against 6-OHDA toxicity, we examined the dose dependence of antioxidant protection. Preincubation of cells with 30 U/ml or greater amounts of catalase conferred significant protection from cell injury as determined by altered metabolism of MTS (Figure 2A) and LDH release (Figure 2B). This protective effect was abolished by heat-treating catalase, a procedure that resulted in a greater than 99% decrease in in vitro catalase activity (see Figure 4B). Both of the metalloporphyrin compounds also exhibited a dose-dependent protection from cell injury (Figures 3A and 3B). Since high concentrations of metalloporhyrins exhibit toxicity (Figure 3A, ), subsequent studies utilized 50 μM and 100 μM final concentrations of AEOL 10113 and MnTBAP respectively.
As potentially toxic ROS are derived from 6-OHDA autoxidation, a potential explanation for the protective effects of catalytic antioxidants in our model system is that they may affect autoxidation of 6-OHDA. To assess the effects of catalytic antioxidants on 6-OHDA autoxidation, the kinetics of quinone formation was analyzed in the presence of protective doses of catalase and MnTBAP. The rate of 6-OHDA autoxidation upon addition to cell culture media was unaffected by the presence of catalase or MnTBAP (Table 1). This is consistent with studies showing that superoxide is not essential for hydroquinone autoxidation at higher quinone concentrations . These findings suggest that the protective effect elicited by these catalytic antioxidants is independent of their ability to alter the kinetics of 6-OHDA autoxidation.
Another method to confirm that direct effects on 6-OHDA autoxidation did not account for the protective effect is to add the antioxidants at different time points after completion of 6-OHDA autoxidation. In light of the kinetics of autoxidation (Figure 1) and the delayed onset of intracellular ROS production in response to 6-OHDA that has been reported in other model systems , catalytic antioxidants were added at different intervals after the addition of 6-OHDA, and assessed for their ability to attenuate 6-OHDA toxicity. Protection was consistently achieved only if catalase was added within 15 minutes after the addition of 6-OHDA (Figures 4A and 4B). The loss of protective effect with time was not simply due to loss of activity with time, as the addition of aged catalase was able to protect against 6-OHDA-mediated cell injury when added to untreated B65 cells prior to addition of fresh 6-OHDA (data not shown).
Similar to the observed protective effect with catalase, both AEOL 10113 and MnTBAP were able to attenuate cytotoxicity when added after the addition of 6-OHDA. However, the metalloporhyrins were able to confer protection up to 1 h after the addition of 6-OHDA (Figures 5A and 5B), well after the completion of 6-OHDA autoxidation (Figure 1). Again, the loss of protective effect with time was not simply due to loss of activity with time, as the addition of aged mimetic was able to protect against 6-OHDA-mediated cell injury when added to untreated B65 cells prior to addition of fresh 6-OHDA (data not shown).
The observation that cells are still capable of being rescued from cytotoxicity up 1 hour after the addition of 6-OHDA and that this occurs well after the completion of 6-OHDA autoxidation (Figure 1) suggested a delayed component of oxidative cytotoxicity. While some studies have demonstrated a rapid (within minutes) increase in intracellular ROS in response to 6-OHDA using 2′,7′-dichlorofluorescein based assays , others have noted a delayed production in intracellular ROS production in response to 6-OHDA treatment of primary neuronal cultures using the relatively superoxide selective dihydroethidium . As both metalloporphyrin antioxidants have been shown to efficiently partition into mitochondrial fractions of neuronal cells , we studied the kinetics of both total intracellular and mitochondrial superoxide production in 6-OHDA-injured B65 cells.
Aconitase activity is a sensitive indicator of intracellular superoxide production . Time course studies revealed a significant decrease in aconitase activity compared to vehicle that occurred rapidly within the period of 6-OHDA autoxidation (Figure 6A). Analysis of aconitase activity at later time points revealed a further decrease with time that was significantly lower than that observed during the period of 6-OHDA autoxidation.
The observation that there was a second significant decrease in aconitase activity occurring well after completion of autoxidation (Figures (Figures11 and and6A),6A), suggested that there may be a delayed intracellular source of superoxide production. Based upon the recent report that mitochondrial superoxide dismutase mediates neuroprotection in a mouse model of 6-OHDA neurotoxicity , we hypothesized that 6-OHDA can induce a delayed phase of mitochondrial ROS. MitoSOX red, a mitochondrial targeted ROS indicator, was used to assess the ROS status of mitochondria. Time course studies consistently demonstrated a significant increase in MitoSOX fluorescence at 90 minutes and beyond (Figure 6B), suggesting that 6-OHDA elicits delayed mitochondrial superoxide production.
We previously have demonstrated that 6-OHDA causes a sustained activation of ERK that contributes to cytotoxicity . Based upon the findings that delayed addition of catalase and metalloporphyrins are capable of inhibiting 6-OHDA cytotoxicity, we examined the time course of inhibition of 6-OHDA-mediated sustained ERK phosphorylation by catalase and metalloporphyrins. Addition of catalase, AEOL 10113, and MnTBAP was able to attenuate 6-OHDA-induced sustained ERK activation (Figures 7A-7F). When added at time points that resulted in protection from cytotoxicity, metalloporphyrins consistently resulted in at least a 50% reduction in the level of p42 ERK phosphorylation as compared to vehicle control (Figures 7D & 7F). Likewise, when added at time points that resulted in protection from cytotoxicity, catalase consistently attenuated 6-OHDA-induced sustained ERK phosphorylation. (Figure 7B).
Recently it has been appreciated that a number of protein kinases, including members of the MAPK family, can be specifically targeted to the mitochondria where they act to modulate important mitochondrial functions (Reviewed in ). Activated ERK has been identified in mitochondria in central nervous system tissue , including diseased neurons in patients with Parkinson’s and Lewy Body Diseases . We wished to characterize the effect of 6-OHDA on mitochondrial ERK. Subcellular fractionation studies revealed that while the majority of activated ERK was found in the cytosolic fractions (Figure 8A, lane C), treatment of B65 cells with 6-OHDA resulted in significant phosphorylation of ERK within mitochondrial fractions (Figure 8A, lanes M). The activated ERK may be derived from localized phosphorylation of a resident pool of mitochondrial ERK, as there were no gross changes in the levels of total ERK present in mitochondrial fractions of untreated, vehicle treated, or 6-OHDA treated cells. Confocal microscopy demonstrated that a subset of activated ERK was present in mitochondria as confirmed by co-localization with SOD2, the mitochondrial isoform of SOD (Figure 8B).
To further define the relationship between ERK activation and the delayed phase of 6-OHDA-induced intracellular oxidant production, we utilized two chemical inhibitors of MEK 1/2, the kinases that phosphorylate and activate ERK Both MEK 1/2 inhibitors, UO126 and PD 98059, were capable of inhibiting 6-OHDA-induced mitochondrial ERK activation (Figure 9, inset), indicating that mitochondrial ERK activation requires MEK 1/2. Conversely, neither UO126 nor PD 98059 were able to attenuate the delayed 6-OHDA-induced increase in MitoSOX fluorescence (Figure 9), suggesting that delayed mitochondrial superoxide production elicited by 6-OHDA is not dependent upon MEK 1/2 activity or mitochondrial ERK activation. Taken together with the ability of mitochondrially distributed antioxidants to inhibit ERK phosphorylation (Figure 7), these data suggest that mitochondrial ROS function upstream of mitochondrial ERK activation.
While oxidative mediators have been implicated in 6-OHDA toxicity, 6-OHDA metabolism is capable of generating a series of ROS at physiologic pH , and the role of these oxidative species in 6-OHDA toxicity remains ill defined. Previously we have demonstrated that 6-OHDA is cytotoxic to B65 cells and that cytotoxicity is associated with sustained ERK phosphorylation . In addition, extracellular application of either catalase or metalloporphyrins, but not of superoxide dismutase, was able to protect from cytotoxicity in this cell culture model . The current studies indicate two phases of ROS generation, occurring concurrent with 6-OHDA autoxidation and followed by delayed onset of mitochondrial ROS production. Interestingly, this delayed mitochondrial component is temporally correlated with the onset of ERK phosphorylation by 2 h , and metalloporphyrin antioxidants added at this time can no longer prevent ERK phosphorylation and cell death. Furthermore, there is mitochondrial ERK activation in response to 6-OHDA.
As demonstrated in Figure 1, 6-OHDA undergoes rapid oxidation in cell culture medium. Moreover, catalase data indicates that a component of cytotoxicity results from the generation of hydrogen peroxide, findings that are consistent with previous reports from other laboratories using 6-OHDA with PC12 cells [40, 41]. In these model systems extracellular catalase exhibits similar kinetics of protection  and, as in our previously reported observations in B65 cells , extracellular SOD fails to protect against cytotoxicity [40, 41]. The potential role of intracellular superoxide in mediating cytotoxicity in these model systems was not addressed, and data from other model systems suggest that dopaminergic neurons may be particularly vulnerable to extracellular ROS compared to other cells . Our current data indicates that there are both extracellular and intracellular sources of ROS that are responsible for 6-OHDA toxicity. Moreover, delayed protection assays suggest that catalase is not able to protect the cells upon completion of 6-OHDA autoxidation whereas metalloporphyrin antioxidants conferred protection when added after the completion of 6-OHDA autoxidation, but prior to the onset of mitochondrial superoxide production.
Several classes of small molecular antioxidant mimetics have been shown to protect against CNS injuries, including pesticide-induced DA neuron degeneration [34, 43-45]. A comparison of the time dependence of protection from 6-OHDA toxicity of catalase and metalloporphyrins demonstrated that metalloporphyrins can be added 45 minutes later than catalase and still afford protection. Metalloporphyrins differ from catalase in several ways including possessing SOD activity and having the ability to permeate cells . The observed protective effect of metalloporphyrins and the absence of protection of extracellularly applied SOD  raise the possibility that compartmentalization of ROS may be important in mediating both cytotoxicity and sustained ERK activation. The results from this study confirm the role of a delayed phase of superoxide production in mediating 6-OHDA toxicity, findings consistent with those observed with 6-OHDA in primary neuronal cultures . Interestingly, in this study, the peak of intracellular superoxide production, as detected by dihydroethidium staining, occurred at 1 hour after the addition of 6-OHDA, the latest time at which metalloporphyrin addition was protective.
While these results suggest a role for intracellular superoxide, alternative interpretations must be considered based on the reported properties of metalloporphyrins, which include their ability to scavenge peroxynitrite . Also, metalloporphyrins possess low levels of catalase activity and have been shown to protect against H2O2 toxicity . Since extracellular and intracellular hydrogen peroxide are in dynamic equilibrium, extracellular catalase can lower intracellular levels of hydrogen peroxide . Thus it is possible that metalloporphyrins protect by providing a source of intracellular catalase activity. However, when assessed using in vitro activity assays, the catalase activity of the metalloporphyrins is similar to that of heat-inactivated catalase, findings that support a role for another mediator, such as intracellular superoxide, in mediating cytotoxicity and sustained ERK activation.
Based upon numerous lines of experimental evidence, mitochondrial dysfunction has been proposed to play an important role in Parkinson’s disease (Reviewed in ). MPTP and rotenone are two toxins used to recapitulate Parkinsonian injury in experimental model systems . Both neurotoxins are capable of inhibiting mitochondrial electron transport chain complex I, which exhibits decreased activity in PD patient material [51-53]. While 6-OHDA is capable of inhibiting complex I in isolated mitochondria [23, 54], the ability of 6-OHDA to induce mitochondrial dysfunction in intact cells is less clear [36, 55-57]. Our results suggest that mitochondria serve as an important source of intracellular superoxide production in 6-OHDA toxicity, which is supported by studies showing that MnSOD transgenic mice are protected from delayed 6-OHDA-induced retrograde neuron cell death . While the molecular mechanisms leading to delayed mitochondrial superoxide production is currently unclear, a variety of potential mechanisms may be involved.
6-OHDA-induced toxicity in B65 cells bears some similarities to glutamate-induced non-apoptotic programmed cell death. Schubert and coworkers have demonstrated that glutamate-induced cytotoxicity in HT22 cells exhibits a secondary phase of ROS generation, and prevention of this by inhibitors of transcription and translation as well as inhibitors of mitochondrial complex III confers protection from cytotoxicity . Similarly, Luetjens et al have reported a delayed ROS production in response to glutamate toxicity in primary hippocampal neurons which can be prevented by delayed addition of MnTBAP . In this system, this delayed phase of ROS production is preceded by mitochondrial cytochrome c release and can be prevented by mitochondrial complex III inhibition.
Recently, it has become established that ERK activation can contribute neurotoxicity, particularly in the context of oxidative insults (Reviewed in ). Depending on the injury, different sources of ROS generation lead to redox-activation of the ERK signaling pathway [60-64]. The present study demonstrates that the sustained ERK activation and cytotoxicity induced by the redox active molecule 6-OHDA is mediated by catalase- and metalloporphyrin-sensitive events. Given the similar redox cycling of dopamine and other catecholamines , abnormal ERK activation via a redox sensitive mechanism may contribute to neuronal cell death in PD. In support of this hypothesis, we observed a correlation between the ability of catalase and metalloporphyrins to protect B65 cells from 6-OHDA toxicity and the ability to inhibit sustained ERK activation. Furthermore, increased levels of phosphorylated ERK and increased ERK activity are observed in substantia nigra tissue from patients with PD and/or dementia with Lewy bodies as compared to age-matched controls . Finally, inhibitors of the MEK 1/2 kinases that activate ERK, protects neuronal cells from 6-OHDA toxicity  and from MPP+ toxicity [66, 67]. These observations suggest that understanding the oxidative mechanisms involved in and the identifying members of the signal transduction cascade that are affected by 6-OHDA to yield sustained ERK activation in B65 cells will be important for understanding the molecular pathogenesis of neuronal loss in PD and related neurodegenerative diseases.
The temporal and/or spatial pattern of signaling molecules in the cell is known to influence cellular responses to stimuli. Such a paradigm is known to exist with regards to ERK signaling in PC12 cells in response to different growth factors (Reviewed in ). Treatment of PC12 cells with epidermal growth factor (EGF) results in a transient activation of ERK whereas nerve growth factor (NGF) treatment results in a sustained ERK activation . In addition, this sustained ERK activation is associated with nuclear translocation and cellular differentiation, while transient EGF-mediated ERK activation remains predominantly cytosolic and is associated with a proliferative response [69-71]. In pathologic conditions, different cell fates can also be mediated by different temporal and spatial patterns of kinase activation, with nuclear localization promoting neuroprotective signaling responses not observed during cytoplasmic accumulation of activated phosphoproteins . In recent years, there has been intensifying interest in the intersection between kinase signaling pathways and mitochondria [38, 72], particularly with the recognition that kinases and other proteins implicated genetically in parkinsonian neurodegeneration localize to mitochondria [73, 74]. In addition to altered subcellular distributions of signaling proteins observed in PD neurons [20, 21, 29] normal mitochondrial functions for phosphoproteins traditionally thought of as nuclear regulators have begun to be elucidated . Poderosos and colleagues have demonstrated the presence of mitochondrial ERK in the matrix, intermembrane space, and outer membrane of the rat CNS . Interestingly, the levels of mitochondrial ERK peaked in the late in utero and early post natal periods, suggesting mitochondrial ERK may play a particularly important role in the developing nervous system.
The function of mitochondrial ERK is currently unclear. The results of our study suggest that mitochondrial ERK activation is not required to elicit 6-OHDA-induced delayed mitochondrial ROS generation, as concentrations of MEK 1/2 inhibitors capable of complexly inhibiting mitochondrial ERK activation did not attenuate the increased MitoSOX fluorescence elicited by 6-OHDA (Figure 9). These results are consistent with reports from other laboratories where ERK activation lies downstream of mitochondrially-induced ROS [76-79], although , these studies did not examine ERK activation within mitochondria. Nowak et al have suggested that mitochondrially activated ERK suppresses mitochondrial respiration in oxidatively damaged tert-butylhydroperoxide treated renal epithelial cells . Interestingly, while decreased mitochondrial respiration and ATP production were reversed by inhibition of the MEK/ERK pathway, these inhibitors did not restore mitochondrial aconitase activity .As mitochondrial aconitase inactivation is a sensitive indicator for mitochondrial superoxide, activation of the MEK/ERK pathway in this study was downstream of mitochondrial oxidative stress, but upstream of other aspects of mitochondrial dysfunction.
Catecholamine metabolism also generates reactive species which may differ from those generated by tert-butylhydroperoxide . Thus, it is possible that the delayed component of 6-OHDA-mediated mitochondrial dysfunction may involve additional metabolites such as dopamine quinone, a compound which has been previously shown to be capable of eliciting mitochondrial dysfunction in purified brain mitochondria . Coupled with the observation that phosphorylated ERK is present in mitochondria of dopaminergic neurons of patients dying from Parkinson and Lewy Body Diseases , the identification of the cellular targets of mitochondrial ERK activated during this delayed phase of 6-OHDA-induced mitochondrial ROS may yield important insight as to the role played by activated mitochondrial ERK in mediating mitochondrial dysfunction during oxidative neuronal injury.
In summary, the catalase and metalloporphyrin antioxidants exhibit different kinetics of protection from 6-OHDA in B65 cells, exhibiting a temporal correspondence between loss of protection and loss of the ability of the antioxidant to inhibit sustained ERK phosphorylation. Based upon time course studies of aconitase inactivation and MitoSOX red fluorescence, this difference suggests that catalase protects only during early phases of toxin injury, while metalloporphyrin antioxidants, which penetrate mitochondrial fractions , may act to prevent a delayed phase of mitochondrial ROS production important for 6-OHDA-mediated ERK activation. Subcellular fractionation studies further revealed that 6-OHDA induced phosphorylation of mitochondrial ERK. These results implicate a role for localized redox-activation of ERK in 6-OHDA cytotoxicity.
We would like to thank Amy Sartori, Charlotte Diges, Prajakta Sonalker, and Jianhui Zhu for technical assistance. We would like to thank Incara Pharmaceuticals (Research Triangle Park, NC) for providing AEOL 11013. This work was supported by a Veterans Administration Advanced Research Career Development Award (SMK), the National Institutes of Health NS40817, NS053777, AG026389 (CTC), NS045748 (MP), and the University of Pittsburgh Pathology Post-doctoral Research Training Program (SMK).
This work was supported by a Veterans Administration Advanced Research Career Development Award (SMK), the National Institutes of Health NS40817, NS053777, AG026389 (CTC), and NS045748 (MP), and the University of Pittsburgh Pathology Post-doctoral Research Training Program (SMK).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.