Oxidative Stress Inhibits Calpain Activity in Situ *

In this study, the effects of oxidative stress on calpain-mediated proteolysis and calpain I autolysis in situ were examined. Calpain activity was stimulated in SH-SY5Y human neuroblastoma cells with the calcium ionophore, ionomycin. Calpain-mediated proteolysis of the membrane-permeable fluorescent substrateN-succinyl-l-leucyl-l-leucyl-l-valyl-l-tyrosine-7-amido-4-methylcoumarin, as well as the endogenous protein substrates microtubule-associated protein 2, tau and spectrin, was measured. Oxidative stress, induced by addition of either doxorubicin or 2-mercaptopyridineN-oxide, resulted in a significant decrease in the extent of ionophore-stimulated calpain activity of both the fluorescent compound and the endogenous substrates compared with control, normoxic conditions. Addition of glutathione ethyl ester, as well as other antioxidants, resulted in the retention/recovery of calpain activity, indicating that oxidation-induced calpain inactivation was preventable/reversible. The rate of autolytic conversion of the large subunit of calpain I from 80 to 78 to 76 kDa was decreased during oxidative stress; however, the extent of calpain autolysis was not altered. These data indicate that oxidative stress may reversibly inactivate calpain I in vivo.

Calpains are a family of calcium-dependent thiol proteases that require both calcium (1) and a reduced environment (2) for activity. These proteolytic enzymes have been postulated to play a role in many physiological processes (3)(4)(5)(6) and disease states (7)(8)(9)(10)(11)(12). Calpains I and II are ubiquitously expressed, whereas the remaining isoforms are predominantly musclespecific (13,14). Although homologous, calpains I and II require different concentrations of calcium for activity in vitro. Calpain II requires 200 -1000 M calcium (15), and calpain I requires 3-50 M calcium (15) for half-maximal activity. Although such high calcium concentrations have been demonstrated in the presynaptic terminals of neurons (16) and under pathological conditions (17), it seems clear that physiological calcium concentrations (100 -800 nM (18)) are sufficient to activate calpain I (19 -22). The focus of this study was calpain I because it is present in neurons and has been postulated to play a role in neuronal death associated with ischemia (7,8) and certain neurodegenerative disorders (9 -12).
Calpain I is a heterodimer composed of a unique 80-kDa catalytic subunit and a 30-kDa, regulatory subunit common to both calpain I and II. Calpain undergoes calcium-dependent autolysis, a multi-step, self-proteolytic event that removes a number of amino acids from the N terminus of each subunit resulting in increased proteolytic activity and increased sensi-tivity to calcium (23). Autolysis results in the conversion of the 80-kDa subunit to a 76-kDa form through a 78-kDa intermediate (24) which occurs prior to the conversion of the 30-kDa subunit regulatory to 18 kDa (25). Although still controversial, there is substantial evidence to indicate that non-autolyzed form of calpain I is active under physiological conditions (10,21,26). The disparate findings concerning the proteolytic activity of the native 80-kDa form of calpain I may be due to the process of autolysis being modulated by more than calcium concentration alone (e.g. the presence of phospholipids, calpastatin, or other unknown factors) (23). Additionally, it is also unclear as to the specific role of the 30-kDa non-catalytic subunit in modulating calpain autolysis (27,28).
Calpain activity can be modulated by redox state, characterized by decreased activity, in vitro, in the presence of oxidant (2). The reason for this is likely due to the mechanism by which calpain I hydrolyzes a peptide bond. For proteolysis to proceed, a transfer of electrons between the specific cysteine, histidine, and arginine residues within the catalytic triad of the active site (29) must occur. This requires that these amino acid residues be maintained in a properly charged state that can be modulated by their local microenvironment. Of the amino acids in the catalytic triad, it is the cysteine residue that is the most susceptible to oxidative inactivation (30) and is therefore likely to play a significant role in decreased calpain I activity upon exposure to an oxidizing environment. The effects of an oxidizing environment on calpain I activity are important to understand because oxidative stress has been linked to pathological states in which calpain I has been suggested to play a role, including Alzheimer's disease (31,32) and ischemia (33,34).
In the present study, both calpain I autolysis and proteolytic activity were examined in situ, under control and oxidatively stressed conditions. These results indicate that calpain I activity stimulated by increasing intracellular calcium can be inhibited by oxidizing conditions. Additionally, oxidative inactivation was prevented by incubation with membrane-permeable antioxidants. Finally, oxidative stress decreased the rate of autolytic conversion of calpain I from 80 to 78 to 76 kDa, but the extent was unaffected.
Cell Culture-All calpain assays were carried out at 37°C in a humidified 5% CO 2 incubator. SH-SY5Y human neuroblastoma cells were plated on 60-mm Corning/Costar dishes and differentiated with 20 M retinoic acid for 6 -8 days in RPMI 1640 media supplemented with 10% horse serum, 5% fetal clone II, 1 mM glutamine, and 100 units/ml penicillin/streptomycin. Cells were at approximately 70 -80% confluence in a monolayer at the time of experimentation.
Free Calcium Determination-Free calcium concentrations were determined using cells attached to poly-L-lysine-coated glass coverslips (Corning) with the calcium indicator dye, fura-2. Prior to experimentation, cells were loaded with 5 M acetoxymethyl ester form of fura-2. Coverslips were placed in an imaging chamber (Warner Instrument Co.) and mounted in a heater platform on the stage of a Nikon Diaphot. The cells were maintained at 37°C in a Ringer solution for the duration of the experiments. Fura-2 was excited at alternating wavelengths of 340-and 380-nm using a 75-watt xenon light source and a filter wheel (Ionoptix system, Ionoptix Corp., Milton, MA). Emitted wavelengths passed through a 510-nm filter cube set before detection by an enhanced CCD camera. Data were stored and processed using IonWizard software. Calibration of fura-2 fluorescence was performed per standard protocols where the ratio (R) of emitted signals at 340-and 380-nm excitation wavelengths provided an index of calcium concentration, which was estimated according to the equation: [Ca 2ϩ ] free ϭ K d ⅐ S f /S b ⅐ (R/R min )/(R max /R), where K d is the effective dissociation constant of fura-2, R min and R max are the 340-to 380-nm ratios, and S f and S b are fluorescence values in the absence and presence of saturating calcium, respectively (35). The K d value of fura-2 for calcium was estimated to be 240 nM (35).
Calpain Proteolysis and Autolysis Assays-Calpain activity was measured by immunoblot analysis of endogenous substrates or proteolytic cleavage of the fluorescent calpain substrate Suc-LLVY-AMC under control and oxidizing conditions. To oxidatively stress the cells, doxorubicin was added to a final concentration of 5 M for 24 h before ionomycin treatment (36), while FF1 was used at 1 mM immediately prior to ionomycin addition. Doxorubicin, an anthracycline antibiotic, has been shown to generate superoxide radicals through redox autoxidation of the quinone group that acts on the mitochondrion (37). Production of free radicals by FF1 occurs, in the presence of UV light, resulting from photodecomposition of the compound that generates hydroxyl radicals (38). The antioxidants GSH (5 mM), Trolox (100 M), or propyl gallate (5 M) were added at the same time as oxidant (either 24 h or immediately prior to ionomycin treatment). After 24 h, cells were rinsed once in serum-free RPMI 1640 media containing the appropriate treatment to maintain experimental conditions, and then 3 ml of serum-free media were added, also containing the treatment compounds. Cells were treated with 2 M ionomycin for the times indicated, rinsed once with phosphate-buffered saline, and collected in 2ϫ Laemmli stop buffer without dye. Samples were immediately heated to 100°C for 10 min, sonicated, and protein concentrations determined by the Lowry method, after acid precipitation of the protein. Bromphenol blue was added to the samples, and 10 or 20 g of total protein were separated on SDS-polyacrylamide gels (4 -12.5% for MAP-2 and 6.5% for calpain I, 10% for tau, 8% for calpastatin, and 5% spectrin), transferred to nitrocellulose (39), and immunoblotted with a monoclonal antibody to either MAP-2 (AP14, a gift from Dr. L. Binder), calpain I (a gift from Dr. J. Elce), tau (5A6 (40); Tau-5, a gift from Dr. L. Binder (41), combination), calpastatin (a gift from Dr. J. Glass), or a polyclonal antibody to spectrin (Ab992, Chemicon). After incubation with the primary antibody, the blots were incubated with the appropriate horseradish peroxidase-conjugated secondary antibody and developed with 3,3-diaminobenzidine in the presence of hydrogen peroxide or by enhanced chemiluminescence (Amersham Pharmacia Biotech) per manufacturer's instructions. The resulting immunoblots were quantitated using a Bio-Rad imaging densitometer (model GS-670). Data were evaluated using analysis of variance, and values were considered significantly different when p Ͻ 0.05.
In the case of experiments utilizing the fluorescence compound, SH-SY5Y cells were similarly prepared and treated on 24-well Corning/ Costar plates. Prior to addition of ionomycin, cells were loaded with 80 M Suc-LLVY-AMC (42) for 20 min at 37°C in a humidified 5% CO 2 incubator. Proteolysis of the fluorescent probe was monitored using a fluorescent plate reading system (Bio-Tek FL500) with filter settings of 360 Ϯ 20 nm for excitation and 470 Ϯ 20 nm for emission. Fluorescent readings were made every 20 min for up to 1 h. Cells were replaced in the incubator between readings to maintain both temperature and pH for optimal cell viability.
Free Radical Measurements-The hydrogen peroxide-sensitive fluorescent compound DCFH 2 -DA was used to indicate levels of oxidative stress induced by either doxorubicin or FF1. DCFH 2 -DA is a well established indicator of intracellular hydrogen peroxide formation which, after intracellular deacylation, remains non-fluorescent but is transformed into the fluorescent form (dichlorofluorescein, DCF) via oxidation (43). Cells were incubated with DCFH 2 -DA 10 min after the addition of ionomycin for all treatment conditions. Fluorescent readings were then taken every 20 min for up to 60 min. Fluorescence was detected with the FL500 (Bio-Tek) fluorescence plate reader with filter wheels set at 485 Ϯ 20 nm and 530 Ϯ 25 nm for excitation and emission, respectively.
Cell Viability Assay-The release of the intracellular enzyme, lactate dehydrogenase, into the media was used as a quantitative measure of cell viability. Adherent SH-SY5Y cells on 24-well Corning/Costar dishes were differentiated, rinsed, treated, and stimulated with ionomycin as described above. The measurement of lactate dehydrogenase release was performed as described previously (44).

Determination of Free Radical Production Capacity of Various Treatment
Paradigms-To estimate the relative amount of free radical production in response to doxorubicin or FF1, the formation of the fluorescent oxidized product, DCF from DCFH 2 -DA, was measured ( Fig. 1). Treatment of SH-SY5Y cells with FF1 resulted in a significant increase in peroxide production as indicated by an increase in DCF production. Although doxorubicin has been clearly demonstrated to be a potent inducer of oxidative stress (45), treatment with doxorubicin did not significantly increase DCF production. This is probably due to the fact that DCFH 2 -DA is a selective indicator of hydrogen peroxide formation (46,47) and does not react well with superoxide anions which are the predominant free radical generated in response to doxorubicin treatment (37,46). When the antioxidants propyl gallate, Trolox, or GSH were included with either FF1 or doxorubicin, a significant decrease in DCF production was observed. Antioxidants alone also significantly reduced the hydrogen peroxide concentration to below the level observed in control cells.
In Situ Proteolysis of the Fluorescent Compound Suc-LLVY-AMC during Oxidative Stress-The fluorescent calpain-selective substrate peptide Suc-LLVY-AMC (42) was used to measure calpain activity in SH-SY5Y cells. As in studies by other investigators (48 -50) calpain activity was stimulated through the use of a calcium ionophore. Cells were treated with 2 M ionomycin, and the increase in fluorescence measured at 20min intervals for up to 60 min. Ionomycin increased the intracellular calcium concentration from the 50 -80 nM basal concentration to 700 -800 nM. Maximal calcium concentration was reached after 8 -9 min of exposure and remained elevated for the duration of the experiments (data not shown). Cells remained adherent over the time course, with no loss of viability as indicated by lack of increase in lactate dehydrogenase release (data not shown). Fig. 2 shows the increase in Suc-LLVY-AMC degradation, indicated by increased fluorescence, in response to the various treatments relative to conditions in the presence of ionophore alone (control) after 40 min. Proteolysis of the fluorescent substrate was significantly inhibited by the presence of either doxorubicin or FF1 to approximately 35% of ionomycin treatment alone. When cells were treated with FF1 in the presence of GSH, all oxidant-induced inhibition of calpain activity was prevented. GSH also significantly ameliorated the inhibition of calpain activity in response to doxorubicin. In addition, Trolox or propyl gallate significantly reduced the extent of calpain inactivation by the oxidants but to lesser extent than GSH. GSH treatment significantly increased ionomycin-stimulated calpain proteolysis of the fluorescent probe compared with ionomycin treatment under normoxic condi-tions. Because the fluorescent probe Suc-LLVY-AMC is selective but not totally specific for calpain-mediated proteolysis (2,42,51), experiments were carried out using the calpain-selective protease inhibitors Z-LLY-CHN 2 (42,52,53) or calpain inhibitor I (54) to determine the amount of Suc-LLVY-AMC proteolysis which is not calpain-dependent. Addition of either 25 M Z-LLY-CHN 2 (Fig. 2) or 20 M calpain inhibitor I (data not shown) inhibited ionomycin-stimulated proteolysis of the fluorescent compound by greater than 90%, suggesting that the majority of the increase in fluorescence in response to ionomycin is due to calpain. Because these two inhibitors may also inhibit cathepsin L (55,56), additional experiments were carried out to eliminate the possibility that cathepsin L was contributing to proteolysis of Suc-LLVY-AMC. Inhibition of cathepsin L activity with either of the cathepsin L-specific inhibitors CBZ-Phe-Gly-NHO-Bz (20 M) or CBZ-Phe-Gly-NHO-Bz-pOMe (20 M) (57) resulted in no decrease in ionophore-stimulated proteolysis of the fluorescent compound. In addition, treatment of SH-SY5Y cells with other non-calpain inhibitors (phenylmethanesulfonyl fluoride (0.1 mM), pepstatin (1 g/ml), or Ac-Tyr-Val-Ala-Asp-CMK (caspase inhibitor 2, 1 g/ml)) also did not result in inhibition of ionomycin-stimu- lated proteolysis of Suc-LLVY-AMC (data not shown), further suggesting that the observed proteolytic activity is due to calpain activation.
Effects of Oxidative Stress on Calpain-mediated Proteolysis of Endogenous Substrates-The effects of oxidative stress on the proteolysis MAP-2 by calpain were examined. MAP-2 has been identified previously both in vitro (58) and in situ (8), as a highly sensitive calpain substrate. SH-SY5Y cells were incubated with 2 M ionomycin for 10 min, and the extent of MAP-2 degradation was quantitated by immunoblot analysis. Fig. 3 shows a representative immunoblot and quantitative analysis of ionophore-stimulated MAP-2 degradation. The presence of either doxorubicin or FF1 resulted in significant reduction in MAP-2 degradation, similar to the inhibition observed with the fluorescent compound (Fig. 2). Also, in good agreement with the fluorescent data was the finding that oxidation-induced inhibition of MAP-2 degradation was prevented by addition of the antioxidant GSH. GSH treatment alone of ionomycin-stimulated cells, however, had no significant effect on MAP-2 proteolysis compared with ionomycin treatment under normoxic conditions (Fig. 3).
The second endogenous substrate that was studied was the microtubule-associated protein tau. Tau has been demonstrated to be an excellent substrate of calpain both in vitro (59,60) and in situ (61). Fig. 4 shows a representative immunoblot and quantitative analysis of tau degradation after the cells were treated for 15 min with ionomycin. Consistent with the previous findings, calpain degradation of tau was significantly inhibited by oxidative stress with both FF1 and doxorubicin. The inhibition of calpain, by either oxidative stress inducer, was similar to that observed with 25 M Z-LLY-CHN 2 .
The last endogenous substrate studied was spectrin, another well established calpain substrate in vitro (58) and in vivo (62), although less sensitive than MAP-2 (8). Calpain cleavage of spectrin results in the production of specific breakdown products at molecular masses between 150 and 155 kDa (63), which have been used extensively as a specific indicator of calpain activity (62, 64 -68). Fig. 5 shows a representative immunoblot and the quantitative analysis of spectrin breakdown product formation after 30 min exposure to 2 M ionomycin under conditions of oxidative stress and control. Data are expressed as percent of breakdown product formed in response to each treatment compared with ionomycin treatment, under normoxic conditions. As observed with the fluorescent probe and the other endogenous substrates, oxidative stress decreased the proteolytic processing of spectrin. Interestingly, oxidative stress inhibited the formation of spectrin breakdown products to the same extent as the calpain inhibitor, Z-LLY-CHN 2 (25 M).
Calpastatin was unaffected by oxidative stress at any of the time points studied as detected by immunoblot analysis (data not shown).
Effects of Oxidative Stress on Calpain I Autolysis-Previously it has been shown that in vitro, oxidative stress inhibits The presence of the antioxidant GSH prevented the oxidant-induced inhibition of calpain activity resulting in proteolytic degradation of MAP-2 similar to control conditions. Inset shows representative immunoblots (10 g of total protein per lane, probed with AP14) of the various treatments. Asterisks indicate significant difference between ionomycin alone under normoxic conditions and the noted treatment (p Ͻ 0.05, n ϭ 3-6). calpain proteolytic activity but does not inhibit calpain autolysis (2). Therefore, the effects of oxidative stress on ionomycinstimulated calpain autolysis in SH-SY5Y cells were examined. The extent of autolysis was determined by the amount of autolyzed calpain I (78/76 kDa) present compared with the amount of intact calpain I (80 kDa) at different time points after ionomycin treatment in the absence or presence of oxidant, GSH, or Z-LLY-CHN 2 . Autolytic conversion of calpain I was analyzed by immunoblotting with a calpain I antibody that recognizes intact as well as autolyzed forms of calpain I. Incubation of SH-SY5Y cells with 2 M ionomycin for 10 min in the absence of oxidant or GSH resulted in substantial conversion of the intact 80-kDa calpain to the 78-kDa form with very little formation of the 76-kDa autolytic product (Fig. 6A). Both oxidants appeared to reduce the rate of calpain autolysis but did not completely inhibit autolytic conversion. The presence of GSH alone appeared to facilitate calpain autolysis and prevent any inhibition of calpain autolysis by FF1. However, after 15 min of incubation with ionomycin (Fig. 6B), the autolytic state of calpain for controls and FF1-treated cells was not significantly different, and by 30 min the autolytic state of calpain was almost identical for all treatments, including treatment with the calpain inhibitor Z-LLY-CHN 2 (Fig. 6C), suggesting that oxidative stress alters the rate but not the extent of autolytic conversion. These results demonstrate that calpain I autolyzes to the same extent in the presence of oxidant as it does under normoxic conditions. DISCUSSION The fact that calpain requires calcium for activity has been unequivocally demonstrated (1,6,23), and numerous studies have focused on the regulatory role of calcium in calpain activation (15, 23, 69 -71). However, calpain also requires a reducing environment to be active, and therefore, the redox potential of the cell is likely to play a critical role in regulating calpain activity. The importance of understanding redox regulation of calpain is exemplified by many studies implicating calpain in several pathogenic conditions characterized by oxidative stress (7,8,31). Previously we reported that hydrogen peroxide reversibly inhibits calpain in vitro (2). In the present study we have extended our previous findings by examining the effects of oxidative stress on calpain activity in situ.
Calpain is a thiol protease that requires calcium binding for activation. In the absence of calcium, calpain remains inactive due to intramolecular interactions between the proteolytic domain and other regions on the large subunit (72,73). Upon the binding of calcium, a conformational change is induced affecting the active site in such a way that allows proteolysis to proceed. In addition, the extent of calcium binding modulates the rate of calpain proteolysis through regulation of the autolytic state. Numerous studies have demonstrated that autolytic conversion is calcium-dependent and that certain concentrations of calcium are probably required for autolytic conversion to occur (15,74). These different forms of calpain have varying degrees of calpain activity, with 80 Ͻ Ͻ 78 Х 76 kDa (75). Calcium, however, is not the only factor involved in modulation of calpain activity (23,75). Redox state is also an important factor involved in the modulation of calpain activity. Indeed, many intracellular events have been shown to be influenced by redox state. These include the regulation of transcriptional activators such as p53, NF-B, and AP1 (see Ref. 76 for review). Regulation of these transcription factors has been shown to be due to oxidation-related changes in a thiol-containing portion of a specific protein. For example, when the cysteine residue of the p50 subunit of NF-B was mutated to a serine, the effects of oxidation were abolished (77), indicating both the sensitivity and specificity of cysteine residues to oxidative attack.
In addition to transcriptional proteins, studies have shown that oxidation affects other cysteine-containing proteins. For example, it has been shown that the thiol protease papain is reversibly inhibited by hydrogen peroxide (78). It has also been demonstrated that peroxynitrite, another free radical, modulates the binding of iron regulatory proteins through oxidation of crucial cysteine residues. (79). In fact, it is thought that free radicals, and more specifically, reactive oxygen species, play an important role in mediating certain cellular responses (80,81). The present study demonstrates that the redox state of the cell may play a role in modulating the activity of the thiol protease calpain.
Similar to previous in vitro studies (2), oxidative stress leads to decreased calpain proteolytic activity in situ. Inhibition of calpain by oxidative stress caused by either doxorubicin or FF1 was also shown to be preventable by addition of antioxidant compounds such as GSH. The current findings have ramifica- tions in the interpretation of calpain activity measurements under certain conditions. In many cases, alterations in calpain activity in response to specific treatments or conditions have been measured by one of two methods which do not necessarily allow for the effects of oxidative stress to be considered. One method involves evaluating the extent of calpain autolysis as an indicator of calpain activity (7,9). Although autolytic conversion is a good gauge for determining whether calpain has been activated, it does not necessarily indicate proteolytic activity. For example, in vitro studies examining the ability of various calpain inhibitors to prevent autolytic conversion demonstrated that no calpain inhibitor, including the endogenous inhibitor calpastatin, was capable of preventing autolytic conversion of intact 80 to 78 kDa, although extremely high concentrations of inhibitors were able to prevent complete conversion to 76 kDa (82,83). Interestingly, the same is true in situ, as Z-LLY-CHN 2 did not block calpain autolysis, although it did potently inhibit proteolytic activity in ionomycin-stimulated SH-SY5Y cells. Additionally, oxidative stress inhibited calpain proteolytic activity with little impact on autolytic conversion (see Figs. 5 and 6C), similar to what was observed in vitro (2). However, because the rate of conversion is apparently slowed in the in situ paradigm, it cannot be completely ruled out that the extent of autolytic conversion may in some situations be indicative of activity.
Because calpain contains a single active site, it is important to understand how autolysis and proteolysis are affected differently by oxidation. An examination of other studies utilizing calpain active site-directed inhibitors may provide a possible answer. In previous in vitro studies (82,83), it was shown that calpain inhibitors did not bind to the active site of calpain in the absence of calcium, indicating that the active site is buried within the protein structure and inaccessible to these compounds. This suggests that until the active site is exposed to the surrounding environment, it will remain unaffected by "extramolecular" factors. Therefore, because calpain inhibitors, including calpastatin, block proteolytic activity but not auto-lytic conversion (82,83), it can be hypothesized that calpain autolysis is an intramolecular event in which the active site remains relatively protected from the surroundings, and hence protected from oxidative attack. In contrast, proteolytic processing is an intermolecular event resulting in exposure of the active site of calpain to any inhibitors that may be present, including oxidants or free radicals.
Calpain activity has also been assessed through in vitro assay of tissue homogenates with exogenous substrates (84 -86). This method is useful when examining total potential calpain activity but may be of limited use for determining activity levels at a particular point in time for two reasons. First, high concentrations of calcium are usually added, resulting in the supranormal activation of calpain through complete autolysis. Second, a thiol-reducing agent such as dithiothreitol is added that affects the redox state, resulting in recovery of calpain activity that may have otherwise been inhibited by oxidizing conditions within the cell.
Understanding the effect of oxidative stress on calpain activity may provide useful information regarding the role of calpain in ischemia/hypoxia. It has been demonstrated that calpain is responsible in part for the neuronal degeneration observed following ischemic insult (33,87,88) and is dependent on the presence of oxygen (89). In several animal models, addition of membrane-permeable calpain inhibitors resulted in decreasing the amount of neuronal damage as a consequence of ischemia. Interestingly, studies have demonstrated that the application of these inhibitors must occur within a specific "therapeutic window" to be effective (33,90). This window ranges from before the initiation of ischemic/hypoxic insult to several hours after reperfusion. The present study suggests that this "window" may result from the activity state of calpain in response to the redox potential of the cell. It can be postulated that during the insult, calcium and free radical production increases over time resulting in autolytic conversion of calpain but reduced proteolytic activity. Administration of the calpain inhibitors would be effective in reducing calpain-medi-FIG. 6. The effects of oxidant and/or antioxidants on the calcium-mediated autolysis of calpain I in SH-SY5Y cells. The effects of oxidative stress on ionomycin-stimulated calpain I autolysis was examined after 10 min (A), 15 min (B), and 30 min (C). Samples from experiments observing endogenous protein substrate degradation by calpain were also examined for changes in calpain autolytic conversion at the given time point. Immunoblot analysis shows that treatments that increase oxidative stress reduce, but do not block, autolytic conversion of calpain I. The antioxidant GSH had the reverse effect and facilitated autolytic conversion of calpain, as well as prevent autolytic inhibition by FotoFen-ton1 (FF1). Interestingly, the overall ability of calpain to autolyze was not affected as autolytic conversion was not significantly different between FF1 and control by 15 min (B), and by 30 min (C) all treatments, including treatment with 25 M N-CBZ-L-leucyl-L-leucyl-L-tyrosine, diazomethyl ketone (Z-LLY-CHN 2 ), showed nearly identical autolytic conversion to the 78-and 76-kDa forms. Samples used for determining the state of calpain autolysis were the same as were used to blot for the endogenous substrates at their respective time points. ated neuronal damage until the cellular environment becomes sufficiently reduced to allow substantial calpain-mediated proteolytic activity. After the redox potential has been restored to allow proteolysis to occur, rapid degradation occurs because calpain has been fully autolyzed to the form with the highest activity, which subsequently leads to cell damage.
The present study demonstrates that oxidative stress results in the inhibition of calpain activity in situ. In both the previous in vitro study (2) and the present in situ study, the extent of calpain autolysis was unaffected by oxidative stress. However, in contrast to the in vitro study (2), oxidation of calpain in situ resulted in an apparent reduced rate of autolytic conversion. The disparity in these findings suggest, in agreement with other studies, that calpain activity and autolysis are regulated by more than calcium alone (23). In vitro evaluation of the effects of oxidant on calpain activity and autolysis only described the relationship between calcium, calpain, substrate, and oxidant. However, in the present in situ study, additional factors, known and unknown, are involved in modulating calpain autolytic and proteolytic processes. Based on the findings of the present study as well as the previous one (2), it appears that oxidant may modulate calpain-mediated proteolysis through direct oxidation of the active site cysteine. The mechanism for modulating calpain autolysis, however, appears to be more complex but may include such factors as oxidation-induced interference with interactions of other associated proteins, membrane association, the proposed "activator" (22) or other unknown factors that play a role in autolytic conversion. Further studies are required to elucidate the mechanisms that modulate calpain autolysis in situ.