Activity of Ubiquitin-dependent Pathway in Response to Oxidative Stress

Relations between the ubiquitin pathway and cellular stress have been noted, but data regarding responses of the ubiquitin pathway to oxidative stress are scanty. This paper documents the response of this pathway to oxidative stress in lens cells. A brief exposure of lens epithelial cells to physiologically relevant levels of H2O2 induces a transient increase in activity of the ubiquitin-dependent pathway. Ubiquitin conjugation activity was maximal and increased 3.5–9.2-fold over the activity noted in untreated cells by 4 h after removal of H2O2. By 24 h after removal of H2O2, ubiquitin conjugation activity returned to the level noted in untreated cells. In parallel to the changes in ubiquitin conjugation activity, the activity of ubiquitin-activating enzyme (E1), as determined by thiol ester formation, increased 2–6.7-fold during recovery from oxidation. Addition of exogenous E1 resulted in an increase in ubiquitin conjugation activity and in the levels of ubiquitin carrier protein (E2)-ubiquitin thiol esters in both the untreated cells and the H2O2-treated cells. These data suggest that E1 is the rate-limiting enzyme in the ubiquitin conjugation process and that the increases in ubiquitin conjugation activity which are induced upon recovery from oxidation are primarily due to increased E1 activity. The oxidation- and recovery-induced up-regulation of E1 activity is primarily due to post-synthetic events. Substrate availability and up-regulation of E2 activities also appear to be related to the enhancement in ubiquitinylation upon recovery from oxidative stress. The oxidation-induced increases in ubiquitin conjugation activity were associated with an increase in intracellular proteolysis, suggesting that the transient increase in ubiquitinylation noted upon recovery from oxidative stress may play a role in removal of damaged proteins from the cells.

Relations between the ubiquitin pathway and cellular stress have been noted, but data regarding responses of the ubiquitin pathway to oxidative stress are scanty.

This paper documents the response of this pathway to oxidative stress in lens cells. A brief exposure of lens epithelial cells to physiologically relevant levels of H 2 O 2 induces a transient increase in activity of the ubiquitindependent pathway. Ubiquitin conjugation activity was maximal and increased 3.5-9.2-fold over the activity noted in untreated cells by 4 h after removal of H 2 O 2 . By 24 h after removal of H 2 O 2 , ubiquitin conjugation activity returned to the level noted in untreated cells. In parallel to the changes in ubiquitin conjugation activity, the activity of ubiquitin-activating enzyme (E1), as determined by thiol ester formation, increased 2-6.7-fold during recovery from oxidation. Addition of exogenous E1 resulted in an increase in ubiquitin conjugation activity and in the levels of ubiquitin carrier protein (E2)ubiquitin thiol esters in both the untreated cells and the H 2 O 2 -treated cells.
These data suggest that E1 is the rate-limiting enzyme in the ubiquitin conjugation process and that the increases in ubiquitin conjugation activity which are induced upon recovery from oxidation are primarily due to increased E1 activity. The oxidation-and recovery-induced up-regulation of E1 activity is primarily due to post-synthetic events. Substrate availability and up-regulation of E2 activities also appear to be related to the enhancement in ubiquitinylation upon recovery from oxidative stress. The oxidationinduced increases in ubiquitin conjugation activity were associated with an increase in intracellular proteolysis, suggesting that the transient increase in ubiquitinylation noted upon recovery from oxidative stress may play a role in removal of damaged proteins from the cells.
In the lens, epithelial cells are the first line of defense against stress and play crucial roles in maintenance of the entire organ. Oxidation is a major stress to the lens due to its constant exposure to light and oxidants (1)(2)(3)(4). H 2 O 2 in the lens (100 -300 M) has been detected using different methods (5,6). In addition, H 2 O 2 was detected in the aqueous humor (the fluid from which the lens receives its nutriture) of normal (0.03 mM) or cataract patients (0.08 -0.19 mM) by various laboratories (7)(8)(9). Oxidation-induced damage to lens epithelial cells includes bulk protein oxidation, inactivation of some key enzymes, DNA breaks, and lipid peroxidation (3, 10 -14). Lens cells also have antioxidant systems and repair mechanisms to ameliorate oxidative insult.
The ubiquitin-dependent pathway is one of the putative repair mechanisms. It participates in DNA repair and selective removal of damaged or obsolete proteins (15)(16)(17)(18)(19)(20)(21). In yeast, the ubiquitin-dependent pathway is required to withstand oxidative stress (22) and heavy metal toxicity (23). In addition, the ubiquitin-dependent pathway also plays roles in regulation of diverse cellular processes, including signal transduction (24 -27), cell cycle control (28,29), differentiation (30), malignant transformation (31,32), and apoptosis (33). The hallmark of the ubiquitin-dependent pathway is the covalent attachment of ubiquitin to proteins to form ubiquitin-protein conjugates in a process termed ubiquitinylation. The best known role of ubiquitinylation is selectively targeting proteins for degradation, but ubiquitinylation of some proteins such as calmodulin, histones H2A and H2B, actin, and some membrane receptors serves a regulatory function without targeting them for cytosolic degradation (34).
Previous studies showed that the activity of ubiquitinylation in lens epithelial cells in culture was inhibited upon exposure to 1 mM H 2 O 2 and that the activity of ubiquitinylation increased during recovery from oxidative stress (19). Recent work in this laboratory indicated that mild oxidative stress enhanced ubiquitinylation in cultured lenses (35). However, no information regarding the steps that control the response of the ubiquitinylation to oxidation was available.
Multiple enzymes are involved in the process of ubiquitinylation. In this process, ubiquitin is first activated by ubiquitinactivating enzyme (E1) 1 via formation of a thiol ester bond with E1. The activated ubiquitin is then passed to a ubiquitin carrier protein (E2), to which ubiquitin is also linked via a thiol ester bond. The activated ubiquitin is then either directly linked to substrates or is linked to substrates via a ubiquitin ligase (E3). Several E2s and E3s have been identified. The diversity of E2s and E3s is responsible for the substrate specificity of ubiquitinylation. In this work, we determined the effect of a physiologically relevant level of H 2 O 2 on ubiquitinylation and studied changes in the activities of E1 and E2s in response to oxidative stress. We also investigated mechanisms of the increased E1 activity during recovery from oxidative stress. (Tris-base), dithiothreitol, creatine phosphate,  creatine phosphokinase, ATP, 2-deoxyglucose, Coomassie Blue R-250,  and chloramine T were obtained from Sigma. Acrylamide, N,NЈ-methylene-bisacrylamide, N,N,NЈ,NЈ-tetramethylenediamine, 2-mercaptoethanol, sodium dodecyl sulfate, glycine, and protein molecular mass standards were purchased from Bio-Rad. Hexokinase was from Worthington, and magnesium chloride was from Fisher. Na 125 I and 125 Iprotein A were obtained from NEN Life Science Products. Anti-ubiquitin antibody and anti-E1 antibody were produced in this laboratory in New Zealand White rabbits by injection of SDS-denatured ubiquitin conjugated to ␥-globulin or synthetic peptides of E1 conjugated to ovalbumin.

Materials-Trizma
Exposure of Lens Epithelial Cells to Oxidative Stress-Bovine lens epithelial cells (BLEC) were cultured in Dulbecco's modified Eagle's medium containing 10% (v/v) fetal bovine serum, penicillin (100 units/ ml), streptomycin (100 g/ml), and amphotericin (250 ng/ml). The cells were maintained at 37°C in 95% air and 5% CO 2 . Exposure to oxidative stress was performed in a serum-and phenol red-free medium containing 0.1 mM H 2 O 2 for 30 min. The cells were collected immediately or were cultured in H 2 O 2 -free medium to allow them to recover from oxidative stress. Control cells were treated exactly as the exposed cells except that H 2 O 2 was not included in the medium. The viability of the cells after exposure to H 2 O 2 was monitored by exclusion of trypan blue and 3-(4,5-dimethyl thiazol-2-yl)-2,5-diphenyl tetrazolium bromide staining. The levels of reduced glutathione were determined as described previously (19). The levels of ATP in the cells were monitored using the bioluminescent somatic cell assay kit (Sigma) according to the manufacturer's instructions.
Determination of the Endogenous Levels of Free Ubiquitin, Ubiquitin Conjugates, and E1-BLEC were harvested immediately after 30 min exposure to 0.1 mM H 2 O 2 or at 1, 2, and 4 h after removal of H 2 O 2 and homogenized in 50 mM Tris-HCl, pH 7.6, containing a mixture of inhibitors of proteinase and isopeptidase (5 mM EDTA, 10 M hemin, 1 mM 4-(2-aminoethyl) benzene sulfonyl fluoride, 1 mM E-64, and 2 g/ml aprotinin, and 10 mM iodoacetamide). Following SDS-PAGE separation and transfer to nitrocellulose, the blots were probed with an affinity purified polyclonal antibody to ubiquitin or antibody to E1, followed by incubation with 125 I-protein A. The ubiquitin, ubiquitin conjugates, and E1 were detected by autoradiography and quantified by image analysis.
Determination of Ubiquitin Conjugation Activity-BLEC were harvested and homogenized in 50 mM Tris, 1 mM DTT, pH 7.6. Ubiquitin conjugation activity in the BLEC supernatant was quantified as the ability to catalyze the formation of conjugates between endogenous protein substrates and exogenous 125 I-labeled ubiquitin. This assay was done in a final volume of 50 l, containing 50 mM Tris-HCl, pH 7.6, 5 mM MgCl 2 , 1 mM DTT, 2 mM AMP-PNP (36), 2 g of 125 I-ubiquitin (Ϸ10 6 cpm), 1 M ubiquitin aldehyde, and 30 l of BLEC supernatant (10 mg of protein/ml). The reaction was started with addition of 30 l of BLEC supernatant. Following incubation at 37°C for 20 min, the reaction was stopped by addition of 50 l of 2 ϫ Laemmli buffer. After boiling at 100°C for 10 min, proteins in 20 l of the mixture were separated by a 15% SDS-PAGE. For a negative control, a parallel experiment was done in which AMP-PNP was replaced with 4.5 units of hexokinase and 12 mM 2-deoxyglucose. After drying the gel, it was exposed to film for 2-4 days. The level of ubiquitin conjugates was quantified by densitometry of the autoradiogram.
Determination of the Activities of Enzymes Involved in Ubiquitin Conjugation-BLEC were homogenized in 50 mM Tris-HCl, 1 mM DTT, pH 7.6. The activities of ubiquitin activating enzyme (E1) and ubiquitin conjugating enzymes (E2s) were determined using their property to form thiol esters with ubiquitin (37)(38)(39). This assay was done in a final volume of 50 l, containing 50 mM Tris-HCl, pH 7.6, 5 mM MgCl 2 , 1 mM DTT, 2 mM AMP-PNP, 1 g of 125 I-ubiquitin (Ϸ10 6 cpm), and 30 l of supernatant of lens cells (10 mg of protein/ml). The reaction was started with addition of 30 l of BLEC supernatant. Following incubation at 37°C for 5 min, the reaction was stopped by the addition of 50 l of 2 ϫ Laemmli buffer or thiol ester assay buffer (50 mM Tris, 4% SDS, 8 M urea, pH 6.8). After standing at room temperature for 20 min, proteins in 20 l of the mixture were separated by SDS-PAGE on 15% gels. The gels were exposed to film after drying. The activities of E1 and E2s were quantified by determining the density of the bands that disappeared on the autoradiogram after reduction with ␤-mercaptoethanol. The mass of the ubiquitin thiol ester of E1 is about 120 kDa (the mass of a subunit of E1 is about 110 kDa, and the mass of ubiquitin is 8.5 kDa). The masses of E2-ubiquitin thiol esters are ϳ8.5 kDa higher than the masses of the corresponding E2s (38). The concentrations of 125 I-ubiquitin in both the thiol ester and conjugation assays were saturating and were 15-30-fold higher than the concentrations of endogenous unlabeled ubiquitin (40). Thus, competition between the endogenous unlabeled ubiquitin and the 125 I-ubiquitin to form thiol esters or conjugates was negligible.

Changes of Redox Status and ATP Levels in BLEC Upon
Oxidative Stress and Recovery-In contrast with prior work (19), BLEC in this study were exposed to a physiological level (0.1 mM) of peroxide. To assess the effect of this level of H 2 O 2 on cultured BLEC, the levels of reduced glutathione (GSH), an indicator of cellular redox status (42), were determined during and after H 2 O 2 treatment. The level of GSH decreased 46% after 5 min of treatment with 0.1 mM H 2 O 2 , but by 15 min of treatment, the levels of GSH returned to the pretreatment level and remained constant thereafter. The rapid restitution of GSH confirms that BLEC have an active system to restore intracellular redox status (42) and that this protective apparatus is not permanently altered by the H 2 O 2 exposure. These data also indicate that cells treated with 0.1 mM H 2 O 2 need less time to restore the intracellular level of GSH than the cells treated with 1 mM H 2 O 2 , since the level of intracellular GSH in the cells treated with 1 mM H 2 O 2 was still 50% lower after 15 min of treatment (19).
Levels of ATP, another indicator of oxidative stress (43)(44)(45), were determined after exposure to 0.1 mM H 2 O 2 (Table I). After 30 min of stress, the level of ATP in the cells decreased ϳ50%. One hour after recovery, the level of ATP in the H 2 O 2 -treated cells was still ϳ30% lower than the levels observed in control cells. But by 4 h of recovery, the level of ATP in the H 2 O 2treated cells returned to the level found in untreated cells. The total restitution of GSH and ATP levels together with trypan blue staining and 3-(4,5-dimethyl thiazol-2-yl)-2,5-diphenyl tetrazolium bromide staining data, which indicate that more than 95% of the cells survived 24 h after treatment, indicate that the BLEC can withstand this level of H 2 O 2 .
Changes in Endogenous Ubiquitin-Protein Conjugates in Oxidatively Stressed BLEC-The molecular masses of endogenous ubiquitin conjugates in BLEC varied from 16 to Ͼ200 kDa ( Fig.  1). More than 50% of the conjugates had high molecular masses (Ͼ200 kDa). When BLEC were treated with 0.1 mM H 2 O 2 for 30 min, the level of high molecular weight (HMW) endogenous ubiquitin conjugates decreased 20 -40% as compared with the -treated cells were 10 -20% higher than the levels in untreated cells (Fig. 1, lanes 3-5). A low molecular mass ubiquitin conjugate (45 kDa) also increased in a time-dependent manner during recovery from oxidative damage (Fig. 1). These data indicate that oxidative stress transiently decreases the levels of endogenous ubiquitin conjugates and that recovery from oxidative stress is associated with increased levels of ubiquitin conjugates.
De Novo Ubiquitin Conjugation Activity in Oxidatively Stressed Cells-It was possible that the changes in levels of endogenous HMW-ubiquitin conjugates in oxidatively challenged BLEC were due to an alteration in ubiquitin conjugation activity. To test this possibility we determined ubiquitin conjugation activity in the BLEC by incubating BLEC supernatant with exogenous 125 I-labeled ubiquitin and 2 mM AMP-PNP (an ATP analog that supports ubiquitinylation but does not support protein degradation). As shown in Fig. 2, the BLEC supernatant has the ability to form ubiquitin conjugates (in an ATP-dependent fashion, data not shown) using endogenous substrates and exogenous ubiquitin. The sizes of the de novo formed conjugates varied from 15 to Ͼ112 kDa. In contrast with decreases in 1) de novo ubiquitinylation previously noted when lens cells were treated with 1 mM H 2 O 2 (19) and 2) the level of endogenous ubiquitin conjugates at this time point (Fig.  1, lane 2), ubiquitin conjugation activity in BLEC which were treated with 0.1 mM H 2 O 2 increased 30 -280% as compared with untreated cells (Fig. 2, lane 2). The extent of the increase varies from experiment to experiment, but the trend is constant. Accordingly, we report the range of the increases of three independent experiments hereafter (Table II). Reasons for the discrepancy between decreased levels of endogenous ubiquitin conjugates and the increased de novo ubiquitin conjugation activity upon H 2 O 2 exposure are not clear at this time. Although it appeared that even in the stressed cells ATP concentrations (Ϸ50 M) vastly exceed the K m (0.5 M) for reticulocyte E1 (46), the oxidation-induced decrease in endogenous ubiquitin conjugates may involve a temporary decrease in the level of ATP in the cells (Table I). It is also possible that damage to ubiquitin is related to the oxidation-induced decrease in the level of endogenous ubiquitin conjugates. Both ATP and ubiquitin are supplied in the de novo ubiquitin conjugation assay.
As with the increases in levels of endogenous ubiquitin conjugates during recovery from H 2 O 2 exposure, there were further increases in ubiquitin conjugation activity (Fig. 2, lanes  3-5). The ubiquitin conjugation activity was maximal (250 -820% increase as compared with the untreated cells) by 4 h of recovery (Fig. 2, lane 5). By 24 h of recovery, the ubiquitin conjugation activity returned to the level of untreated cells (Fig. 3B). These data suggest that the ubiquitin conjugation activity in BLEC is transiently up-regulated during, and upon recovery from, this mild oxidative stress.
Changes in the Activities of E1 and E2s in the Lens Epithelial . BLEC were exposed to 0.1 mM H 2 O 2 for 30 min, and then the cells were cultured in a normal medium to allow the cells recovery from oxidative stress. The BLEC were harvested at the time points indicated, and the ubiquitin conjugation activities were determined by incubation of the supernatant of BLEC with 125 I-labeled ubiquitin. Equal amounts of protein (300 g/assay) were used in each assay. The amount of the de novo ubiquitin conjugates represents ubiquitin conjugation activity. Lane 1, control; lane 2, cells exposed to H 2 O 2 for 30 min; lanes 3-5, cells exposed to 0.1 mM H 2 O 2 for 30 min and then recovered for 1, 2, and 4 h, respectively. Ub-con, ubiquitin conjugates. Cells upon Oxidative Stress and during Recovery-To determine if the H 2 O 2 -induced changes in ubiquitin conjugation activity are associated with changes in the activities of E1 or E2s, these activities in the cell extracts were determined using a thiol ester assay (37)(38)(39)47). This assay is based on the known formation of thiol esters between ubiquitin and E1 and E2s in the ubiquitin conjugation process; thus, the levels of thiol ester reflect the activities of E1 and E2s (39). Consistent with the H 2 O 2 -induced 30 -280% increases in de novo ubiquitin conjugation activity (Fig. 2, compare lane 2 with lane 1, Table  II), E1 activity increased 40 -210% after 30 min of exposure to H 2 O 2 (Fig. 3, A-C, compare lane 2 with lane 1). In addition, there were significant further increases in E1 activities during recovery (Fig. 3A, compare lanes 3-5 with lane 1). By 4 h of recovery, the E1 activity in H 2 O 2 -treated cells was maximal and increased 100 -570% as compared with untreated cells. Consistent with the changes in ubiquitin conjugation activity, the E1 activity returned to the level noted in untreated cells by 24 h of recovery (Fig. 3B, compare lane 6 with lane 1). Unlike E1 activity which increased upon oxidation, the levels of E2-ubiquitin thiol esters appeared to remain unchanged during oxidation (Fig. 3, A-C, compare lane 2 with lane 1). However, like E1 activity, the levels of two E2-ubiquitin thiol esters increased 110 -540% during recovery from oxidative damage and were also maximal at 4 h of recovery (Fig. 3, A and  B). The masses of these two E2s are 16 and 22 kDa, and their thiol esters with ubiquitin are 24.5 and 30.5 kDa, respectively. Since the formation of E2-ubiquitin thiol esters requires E1 activity and the changes in the levels of these two E2-ubiquitin thiol esters paralleled the changes in E1 activity during recovery, it appeared plausible that the oxidation/recovery-induced increases in levels of E2 thiol esters were primarily caused by the increased E1 activity.
To examine this possibility, saturating levels of purified E1 were added to the thiol ester assay. Addition of exogenous E1 to the assay resulted in 540 -1600% increases in the levels of E2-ubiquitin thiol esters in the unstressed cells (Fig. 3C, compare lane 4 versus lane 1; Table II). Addition of E1 to preparations from oxidatively stressed cells or to cells that were stressed and allowed to recover also resulted in substantial increases in levels of E2-ubiquitin thiol esters (Fig. 3C, lanes 5 and 6 versus lanes 2 and lane 3; Table II). The enhancement of E2-thiol esters upon addition of exogenous E1 indicates that E2 activities are not fully realized in the absence of exogenous E1 in BLEC and that E1 activity is rate-limiting for formation of E2-thiol esters in these cells.
In contrast with the E1-unsupplemented assays, in the presence of saturating levels of E1, E2 activities decreased 20% upon 30 min exposure to H 2 O 2 and increased Ӎ40% after 4 h of recovery in comparison with unexposed cells (Fig. 3C, compare  lanes 5 and 6 with 4). These results indicate that E2s are partially inactivated during exposure to H 2 O 2 and, like E1, are also up-regulated during recovery from oxidative stress.
The robust oxidation/recovery-induced increase in E1 activity and the observation that levels of de novo formed ubiquitin conjugates were uniformly associated with E1 activity suggested that the oxidation-induced increases in de novo conjugation ( Fig. 2; Fig. 3A, lanes 2-5 versus lane 1; Fig. 3C, lane 3 versus lane 1) are primarily caused by increased E1 activity. Corroboration of this conclusion is obtained since addition of exogenous E1 to the preparations from unexposed cells also resulted in large increases (280 -650%) in de novo formed ubiquitin conjugates (Fig. 3C, lane 4 versus lane 1; Table II). Taken together, these data indicate that E1 is rate-limiting in the ubiquitin conjugation process in the BLEC.
The data shown in lane 6 versus lane 4 (Fig. 3C) also show that substrate availability for ubiquitin conjugation also increased in the H 2 O 2 -exposed cells. Even in the presence of saturating levels of E1, there is an approximately 2-fold enhancement in de novo ubiquitin conjugates in peroxide-treated and recovered cells as compared with untreated cells. Support for this notion is obtained since the oxidation/recovery-induced 250 -820% increases in the levels of de novo formed ubiquitin conjugates are higher than the 100 -570% increase in E1 activity (Fig. 3C, lane 3 versus lane 1; Table II). The notion that oxidation results in enhanced levels of substrates for ubiquitinylation is further supported by the observation of more ubiquitin conjugates in lane 3 than in lane 4 (Fig. 3C). This is explained as follows: if oxidation only enhanced E1 activity, the amount of de novo formed ubiquitin conjugates observed in E1-supplemented preparations (Fig. 3C, lane 4) would be maximal and equal to or more than the levels of ubiquitin conjugates formed in oxidized, but E1-unsupplemented, cell preparations (lane 3). Since the level of ubiquitin conjugates shown in the E1-unsupplemented cells is greater (lane 3 versus lane 4), this implies that in addition to enhanced E1 activity, oxidation is associated with enhanced levels of substrates for ubiquitinylation. Thus, both increased substrate availability and enhanced E1 activity appear to contribute to the oxidationinduced increase in de novo formed ubiquitin conjugates.
Changes in the Levels of Protein and mRNA for E1 in BLEC in Response to Exposure to H 2 O 2 -To elucidate the mechanism of the dramatic increase in the activity of E1 during recovery from oxidative damage, we determined the levels of protein and mRNA for E1 in BLEC upon oxidation and during recovery. Two bands were detected with this anti-E1 antibody when the lens epithelial cell extracts were resolved by 6% gel. These proteins had apparent masses of 110 and 117 kDa, respectively (Fig. 4A). These proteins are indistinguishable when resolved by 12% gel (data not shown). Formation of two complementary thiol esters confirms that these proteins are isoforms of E1 (Fig.  4B). Thus, two isoforms of E1 were present in the lens epithelial cells, as has been reported in other cell types (48,49).
In the untreated cells, the levels of the two forms of E1 are comparable (Fig. 4A, lane 1), and there appeared to be little change in the levels of both forms of E1 upon exposure to H 2 O 2 (Fig. 4A, lane 2). However, after 1-2 h of recovery, both forms of E1 increased 30 -50% (Fig. 4A, lanes 3 and 4). By 4 h of recovery, the level of the 117-kDa E1 was 70 -100% higher than in the untreated cells, but the level of the 110-kDa E1 returned to the level of untreated cells. The extent of the increase in the protein levels of E1 is much smaller than the extent of the increase in the activity of E1 during recovery from oxidative stress. This indicates that the increased level of E1 protein can only partially account for the increased E1 activity during recovery from oxidative stress. Thus, post-synthetic modification of E1 appears to be involved in regulation of E1 activity in response to oxidative stress. It is interesting to note that while protein levels of the two isoforms of E1 appear to be different after 4 h of recovery, the abilities to form thiol ester are comparable. This suggests that either the specific activity of the two isoforms of E1 is changed or the immunoreactivity of these enzymes is altered during recovery.
Northern hybridization analysis showed that there is a single band of mRNA for E1 with a molecular size of about 4.3 kilobase pairs. The level of mRNA for E1 in BLEC remained constant upon oxidation and recovery (Fig. 4C). This indicates that the increases in the level of E1 protein and in E1 activity during recovery from oxidative damage are not due to increased transcription. The increase in the level of E1 protein without an increase in the level of mRNA suggests that either the translation of E1 increased or the stability of E1 increased during recovery from oxidative stress.
Changes in Proteolytic Capabilities in BLEC upon Oxidative Stress and during Recovery-To determine if the oxidationinduced alterations in ubiquitin conjugation activity are associated with rates of intracellular proteolysis, we monitored intracellular proteolysis in BLEC upon H 2 O 2 exposure and during recovery from oxidative stress by pulse-chase analysis. Consistent with previous observations, the rates of proteolysis declined with increasing time of chase as substrates were consumed (19). In the BLEC treated with H 2 O 2 for 30 min, the rate of proteolysis was not significantly different from that observed in control cells (Fig. 5). During recovery, the proteolysis rate in The increase in rates of intracellular proteolysis during recovery from oxidative stress may result from increased ubiquitin conjugation activity and/or increased levels of substrates, such as oxidatively damaged proteins. To further test if the oxidation-induced increases in ubiquitin conjugation activity has a role in proteolysis, we determined the proteolytic capacity in cell-free experiments using labeled ␤-lactoglobulin as a substrate. In control BLEC extracts, 13.7% labeled ␤-lactoglobulin was degraded in 2 h (Fig. 6). Forty-nine percent of the degradation was ATP-dependent. In extracts from BLEC that were allowed to recover for 4 h after H 2 O 2 treatment, ATP-dependent degradation of ␤-lactoglobulin increased 14% as compared with unchallenged BLEC. Previous studies demonstrated that amajorityoftheATP-dependentproteolysisinBLECisubiquitindependent (50). It appears that the increased ATP-dependent proteolysis in BLEC during recovery from oxidative stress (Figs. 5 and 6) is related to the increased ubiquitin conjugation activities.
Whereas recovery from oxidation is associated with increases in ubiquitin conjugation activity, in the levels of endogenous ubiquitin conjugates and in proteolytic capability, the recovery associated increase in de novo ubiquitin conjugation activity (250 -820%) was much greater than the increase in the levels of endogenous ubiquitin conjugates (10 -20%) and the increase in proteolytic activity (10 -20%). This indicates that the rates of intracellular proteolysis are proportional to the levels of endogenous ubiquitin conjugates rather than to the ubiquitin conjugation activity. The steady state levels of endogenous ubiquitin are defined by their relative rates of formation versus disassembly and degradation. The dramatic increase in de novo ubiquitin conjugation activity with limited increases in the levels of endogenous ubiquitin conjugates and limited increase in proteolysis suggests that the rate of disassembly of ubiquitin conjugates in the cells is also up-regulated during recovery from oxidation. However, the possibility cannot be excluded that some of the proteins that are conjugated to ubiquitin are not destined for degradation.

DISCUSSION
The ubiquitin-dependent pathway plays important roles in various cellular processes including responses to stress. Shortly after the discovery of the pathway (51,52), it was shown that some oxidized proteins are selectively degraded by this proteolytic pathway (18,53,54). Previously we demonstrated that 1) lens and retina tissues have ubiquitin-dependent proteolytic pathways (19,35,40,50,(55)(56); 2) oxidized lens proteins are selective substrates for the pathway (18); 3) efficiency of the pathway is related to cellular redox status (57); and 4) that lenses and lens epithelial cells in culture mount a robust ubiquitinylation response to oxidative stress (19,35 6. Changes in proteolytic capability in extracts of BLEC during recovery from oxidative stress. Proteolytic capacity was determined using 125 Ilabeled ␤-lactoglobulin as substrate. The amounts of 125 I-labeled ␤-lactoglobulin hydrolyzed into acid-soluble fragments in 2 h represent the proteolytic capacity. Control, BLEC without exposure to oxidative stress; oxidized, BLEC treated with 0.1 mM H 2 O 2 for 30 min and were allowed to recover in a normal medium for 4 h. Data presented are mean Ϯ S.D. (n ϭ 3). * indicates that the difference between control and H 2 O 2 -treated cells is statistically significant (p Ͻ 0.05). The ATP-dependent degradation is the difference between the degradation determined in the presence of ATP (total degradation) and in the absence of ATP (ATP-independent degradation). ubiquitinylation process are up-regulated in response to stress in these various experimental systems. The data in Figs. 1-3 and Table II show that mild oxidative stress and recovery from the stress is associated with a transient up-regulation of the ubiquitin conjugation activity. This is consistent with data from oxidatively stressed whole lenses and with data obtained during recovery from harsher stress in lens epithelial cells in culture (19,35). Furthermore, the data from experiments with added E1 (Fig. 3C) indicate that ubiquitin-activating enzyme (E1) is rate-limiting for ubiquitinylation in BLEC and that the oxidation-induced increases in ubiquitinylation result primarily from increased E1 activity. In addition, up-regulation of E1 during the mild oxidation and recovery reveals latent E2 activity that is not observed until the E1 is up-regulated or exogenous E1 is added to the system. The latter experiments also suggest that in contrast with the up-regulation of E1, E2 activity may be attenuated upon mild oxidative stress.
Two isoforms of E1 have been identified in the lens epithelial cells. Both isoforms have activity, and the data from Fig. 4 indicate that the specific activity of both isoforms appears to increase to a similar extent upon mild oxidative stress. However, upon recovery, the specific activity of the 110-kDa E1 increases even more. Together with data that indicate substantially greater increase in activity than in protein levels of E1, these data suggest that post-synthetic modification, or release of an inhibitor or presence of an activator, is involved in regulation of the activity of E1 during recovery from oxidation.
One candidate for such modification is phosphorylation (58 -60) since phosphorylation of E1 in vitro results in a stimulation of the activity of E1 (61). This would be consistent with the well established oxidative stress-induced increases in Ca 2ϩ -dependent kinase activity and with increases in the level of phosphorylated proteins (62)(63)(64). The relationship between increased activity of E1 during recovery from oxidative stress and phosphorylation is under study.
There is only a single mRNA for E1 in the lens epithelial cells. As suggested in HeLa cells, it is possible that the two isoforms of E1 are products of alternative translation of the single mRNA (49). Despite variations in levels of E1 proteins, the level of mRNA for E1 in the cells remained constant upon oxidative stress and during recovery. This suggests that the increases in E1 protein in response to oxidative stress are due to post-synthetic stabilization; however, we cannot rule out altered rates of translation.
Possible roles of increased ubiquitin conjugation activity during recovery from oxidative stress are to restore normal conditions within cells following oxidative stress. We previously hypothesized that one role for the ubiquitin pathway is to remove damaged and cytotoxic proteins (3,40). In support of this hypothesis, data in Figs. 5 and 6 showed that the intracellular proteolysis in these cells increased during recovery from oxidation, and data in Fig. 3C showed that more substrates are available for ubiquitinylation in the cells upon oxidative stress and recovery. This hypothesis is consistent with a recent study that showed that the ability of old lenses to mount a ubiquitinylation response to oxidative stress decreased coincidentally with the accumulation of damaged proteins (35,65,66). In addition to degradation of damaged proteins, the ubiquitin-dependent pathway is also involved in controlling the levels of several key regulatory proteins, such as IB (27,67,68) and p53 (31, 32, 69 -71). The increase in ubiquitin conjugation activity in response to oxidation may be involved in the degradation of IB and the consequent activation of NF-B (27, 67, 68, 72, 73).
In this work we demonstrated that ubiquitin conjugation activity increased up to 250 -820% during recovery from oxi-dative stress, but the proteolytic activity only increased 10 -20%. This suggests that ubiquitinylation has functions besides protein degradation. Support for non-proteolytic roles for ubiquitinylation are observations that ubiquitinylation of calmodulin, histones H2A and H2B, actin, and certain membrane receptors serve regulatory functions without targeting them for cytosolic degradation (34). However, the exact physiological role of the increased ubiquitin conjugation activity during recovery from oxidative stress remains to be elucidated.
In summary, this work, together with previous studies (19,35), demonstrates that the activity of the ubiquitin-dependent pathway increases during recovery from oxidative stress both in whole lenses and in cultured epithelial cells. E1 is the rate-limiting enzyme for ubiquitinylation in the lens cells and the oxidation-and recovery-induced increases in ubiquitin conjugation activity are primarily due to the increase in E1 activity. Dramatic increases in ubiquitin conjugation activity with more limited increases in proteolytic activity in the oxidatively stressed cells suggest that the increased ubiquitin conjugation activity in response to oxidative stress may have functions in addition to proteolysis.