Redox regulation of ubiquitin-conjugating enzymes: mechanistic insights using the thiol-specific oxidant diamide

^a Laboratory for Nutrition and Vision Research and Antioxidants Research Laboratory, JMUSDA-HNRCA at Tufts University, Boston, Massachusetts 02111, USA

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The ubiquitin–proteasome pathway (UPP) regulates critical cell processes, including the cell cycle, cytokine-induced gene expression, differentiation, and cell death. Recently we demonstrated that this pathway responds to oxidative stress in mammalian cells and proposed that activities of ubiquitin-activating enzyme (E1) and ubiquitin-conjugating enzymes (E2s) are regulated by cellular redox status (i.e., GSSG:GSH ratio). To test this hypothesis, we altered the GSSG:GSH ratio in retinal pigment epithelial cells with the thiol-specific oxidant, diamide, and assessed activities of the UPP. Treatment of cells with diamide resulted in a dose-dependent increase in the GSSG:GSH ratio resulting from loss of GSH and a coincident increase in GSSG. Increases in the GSSG:GSH ratio from 0.02 in untreated cells to 0.5 in diamide-treated cells were accompanied by dose-dependent reductions in the levels of endogenous Ub-protein conjugates, endogenous E1~ubiquitin thiol esters, and de novo ubiquitin-conjugating activity. As determined by the ability to form E1-ubiquitin and E2s-ubiquitin thiol esters, E1 and E2s were both inhibited by elevated GSSG:GSH ratios. Inhibition of E1 was associated with the formation of E1-protein mixed disulfides. Activities of E1 and E2s gradually recovered to preoxidation levels, coincident with gradual recovery of the GSSG:GSH ratio. These data support S-thiolation/dethiolation as a mechanism regulating E1 and E2 activities in response to oxidant insult. Ubiquitin-dependent proteolytic capacity was regulated by the GSSG:GSH ratio in a manner consistent with altered ubiquitin-conjugating activity. However, ubiquitin-independent proteolysis was unaffected by changes in the GSSG:GSH ratio. Potential adaptive and pathological consequences of redox regulation of UPP activities are discussed.—Obin, M., Shang, F., Gong, X., Handelman, G., Blumberg, J., Taylor, A. Redox regulation of ubiquitin-conjugating enzymes: mechanistic insights using the thiol-specific oxidant diamide. FASEB J. 12, 561–569 (1998)

Key Words: ubiquitin–proteasome pathway • E1 • E2 • glutathione • diamide • ubiquitin-activating enzyme

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
THE UBIQUITIN–PROTEASOME pathway (UPP)² regulates critical cell processes including the cell cycle, cytokine-induced gene expression, differentiation, cell death, and the stress response (reviewed in refs 1–4). The hallmark of UPP activity is the covalent attachment of ubiquitin (Ub) to proteins (ubiquitinylation) to form Ub-protein conjugates. In this process, ubiquitin is first activated by the Ub-activating enzyme (E1) via formation of a thiol ester bond. Subsequently, the ubiquitin is linked to one of numerous ubiquitin carrier proteins (E2s), again via thiol ester linkage. Then the ubiquitin is ligated to the protein substrate either directly or after transfer of the ubiquitin to a thiol group on a ubiquitin ligase (E3s) (2–4). The best-recognized role for ubiquitinylation is selective targeting of proteins for rapid degradation by the 26 S proteasome, but evidence of important nonproteolytic functions for this process is rapidly emerging (5–7).

Despite seminal roles for the UPP in essential cell processes, including responses to oxidative stress (8), the regulation of UPP activity is poorly understood. We recently demonstrated that UPP activity is regulated in response to oxidative stress (9–11). We observed a rapid loss of endogenous Ub-protein conjugates and down-regulation of Ub-conjugating activity in cells exposed to oxidative challenge with H₂O₂ (9, 11). Down-regulation of Ub-conjugating activity in response to H₂O₂ reflects inhibited activities of Ub-activating enzymes (E1A/B) and ubiquitin-conjugating enzymes (E2s) (10). Inhibition of E1 and E2 activities is coincident with loss of the sulfhydryl reductant glutathione (GSH) and an increase in the levels of its oxidation product, GSSG (9). Reestablishment of preoxidation levels of GSH and GSSG is coincident with recovery of E1 and E2 activities and Ub-conjugating activity (9, 11). In conjunction with the presence of active-site thiols in all members of the E1/E2 (and E3) enzyme families (12–14) and the particular sensitivity of cysteine residues to oxidation (15), these observations led us to propose a model in which inhibition and recovery of E1 and E2 activities after exposure to H₂O₂ reflected S-thiolation/dethiolation of E1 and E2s in response to altered GSSG:GSH ratios (9).

H₂O₂ can be transformed in vivo to ·OH and superoxide. These reactive oxygen species can damage various protein side chains (reviewed in ref 15). To confirm that regulation of E1/E2 activities after H₂O₂ treatment is thiol dependent, it is useful to determine whether oxidants that react only with sulfhydryls can elicit UPP responses similar to those observed during H₂O₂-induced stress. To that end, we used the thiol-specific oxidant diamide [diazenedicarboxylic acid bis (N,N-dimethylamide)] in this study. Diamide rapidly enters and reacts within cells to reversibly perturb the balance of redox-active thiols and their disulfide forms (9, 16). Diamide preferentially oxidizes low molecular weight thiols (e.g., GSH) as opposed to protein thiols. GSH is the major nonprotein thiol in cells (1–10 mM); the addition of stoichiometric amounts of diamide to cells preferentially oxidizes GSH to GSSG. We observed that incubation of human retinal pigment epithelial cells (RPE) with diamide was associated with transient dose-dependent increases in the cellular GSSG:GSH ratio and with decreases in the levels of protein sulfhydryls (PSHs). These alterations in the cellular redox state were coincident with dose-dependent reductions in 1) levels of endogenous Ub-protein conjugates and endogenous E1~Ub thiol esters, 2) the ability to form Ub-protein conjugates and E1~Ub and E2s~Ub thiol esters de novo, and 3) the capacity to degrade protein substrates via the UPP. Reestablishment of the GSSG:GSH ratio during cellular recovery from diamide treatment was associated with restoration of these measures of Ub-conjugating activity. These results support the hypothesis that transient, reversible modulation of UPP activities after oxidative stress is regulated by cellular thiol redox status.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Materials
Materials for protein electrophoresis and nitrocellulose (0.2 µm) were obtained from Bio Rad laboratories (Hercules, Calif.). Polyvinylidene difluoride (PVDF) membrane was purchased from Millipore (Bedford, Mass.). Coomassie Plus protein assay reagent was purchased from Pierce (Rockford, Ill.). Na¹²⁵I and ¹²⁵I-labeled protein A were from DuPont NEN (Boston, Mass.). The ECL chemiluminescence kit was from Amersham Life Sciences (Arlington Heights, Ill.). High-performance liquid chromatography (HPLC) grade phosphoric acid was purchased from Fisher Scientific; MG132 was from Calbiochem (La Jolla, Calif.). All other materials were purchased from Sigma (St. Louis, Mo.) and were of the highest grade available. Ubiquitin and ß-lactoglobulin were iodinated by reaction with chloramine T as described (17).

Preparation of antisera against ubiquitin and E1
Preimmune serum and an antiserum that specifically recognizes conjugated Ub were obtained and characterized as described (9). Rabbit antiserum recognizing the higher mass form (E1A, 117 kDa) of mammalian E1 (18) was raised against a synthetic peptide comprised of methionine-1 to cysteine-26 (19). The antiserum recognizes both E1A and the E1A~Ub thiol ester under denaturing conditions. IgG was purified by affinity chromatography on a protein A column.

Exposure of RPE cells to diamide
Human SV40-immortalized RPE cells (repository #AG06095, Coriell Institute, Camden, N.J.) were cultured in 100 mm² tissue culture plates as described (9). For the experiments, subconfluent cultures that had been fed 4–12 h earlier were washed twice with phosphate-buffered saline (PBS) and then incubated (37°C) for various periods of time with 8 ml of PBS (sham treatment) or with PBS containing 10–500 µM diamide. Total diamide per treatment (0.08–4.0 µm) was 2- to 100-fold greater than the average GSH content per plate (40 nmol). PBS was then aspirated and the cells were harvested in one of the following: 1) 200 µl of ice-cold 10% phosphoric acid containing 1 mM bathophenanthrolinesulfonic acid and 50 nmol of the internal standard, -glutamylglutamate (-Glu-Glu); these samples were used to determine levels of GSH, GSSG, and PSH; 2) 200 µl of lysis buffer [5 mM Tris-HCl, 4% sodium dodecyl sulfate (SDS), and 50 mM N-ethylmaleimide], followed immediately by boiling (10 min) and centrifugation (13,000xg, 10 min); the supernatants were immunoblotted to determine levels of endogenous Ub-protein conjugates and E1~Ub and E2s~Ub thiol esters; or 3) ice-cold 50 mM Tris-HCl, pH 7.8, followed by sonication (two bursts of 4 s) and centrifugation (4500 g, 10 min) to obtain supernatant; these supernatants were used for assays of E1 and E2 activities (thiol ester assays) and for proteolysis assays. In some experiments, diamide-treated cells were washed twice with PBS after removal of diamide, refed with medium, and incubated for an additional 2 h to allow `recovery' from diamide treatment. These cultures were collected as described above after aspiration of the medium and washing with PBS. All samples were stored at -80°C until use.

Levels of GSH, GSSG, and PSHs
GSH and GSSG were quantified as the dinitrophenyl derivatives by HPLC, with -Glu-Glu as internal standard, using the procedure of Fariss et al. (20) with modifications. Details of this modified procedure will be reported elsewhere. Quantitation was based on integration of the internal standard. The sensitivity of the method allows us to reliably quantify 250 pmol of GSH or GSSG in our samples. Total glutathione equivalents were calculated as [GSH] + 2 x [GSSG]. Levels of protein thiols were determined by reaction with Ellman's reagent (21). Protein levels were determined by modified Lowry's using the Protein Assay Kit (Sigma) after neutralization and resolubilization of acid-precipitated protein pellets.

Detection of endogenous Ub-protein conjugates and of E1~Ub and E2_25K~Ub thiol esters by immunoblotting
Procedures for the preparation, electrophoresis, and immunoblotting of RPE samples are detailed elsewhere (9). Briefly, RPE lysates were boiled in either SDS-PAGE (SDS-polyacrylamide gel electrophoresis) sample buffer containing 5% (final concentration) 2-mercaptoethanol or in thiol ester sample buffer containing SDS but lacking reductant. Samples prepared with thiol ester sample buffer retain Ub thiol esters of E1 and E2s. After SDS-PAGE (15% or 6% gels), proteins were transferred to either nitrocellulose or PVDF membranes. These were probed with anti-sera against Ub and E1. Specific binding was detected either with ¹²⁵I-labeled protein A, followed by autoradiography, or with the ECL kit (Amersham). Quantification was by scanning densitometry.

Formation of ¹²⁵I-labeled Ub-protein conjugates and ¹²⁵I-labeled Ub thiol esters of E1 and E2s
The assays were done in 25 µl containing 50 mM Tris-HCl, pH 7.8, 100–150 µg RPE supernatant, 1 mM ATP, an ATP-regenerating system, and 0.5–2.0 µg ¹²⁵I-labeled Ub (2–4x10⁵ cpm) (22). Some assays contained supplemental E1 (a generous gift from Dr. Martin Rechsteiner). Unless stated otherwise, assay mixtures did not contain dithiothreitol. Reactions were initiated with the addition of RPE supernatant. Assays were incubated at 37°C for the times indicated in the figure legends and terminated by boiling with SDS-PAGE sample buffer or thiol ester sample buffer. After SDS-PAGE (15 or 6% gels), gels were stained, dried, and subjected to autoradiography. Adducts of RPE proteins and radiolabeled Ub were quantified by densitometry. Functional E1 and E2 enzymes were distinguished from Ub-protein conjugates by comparing autoradiograms of samples electrophoresed in the presence or absence of 2-mercaptoethanol.

Proteolysis assays
Ub-dependent degradation of ¹²⁵I-labeled ß-lactoglobulin (5x10⁵ cpm/µg) by RPE supernatant was assayed as previously described (17). ATP-dependent proteolysis was determined to be exclusively via the UPP, since all ATP-dependent proteolysis was blocked by the proteasome inhibitor MG132 (80 µM final concentration) (22).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
To determine whether oxidation of cellular GSH results in reduced levels of endogenous Ub-protein conjugates, we exposed RPE cells to 0–500 µM diamide for 5 min and quantitated levels of GSH, GSSG, protein thiols, and endogenous Ub-protein conjugates. GSH levels decreased in a dose-dependent manner in cells exposed to 10 µM diamide, with a 90% loss observed in cells exposed to 250 µM (Fig 1 A). The loss of GSH was coincident with increased levels of its oxidation product GSSG and an even greater increase in the GSSG:GSH ratio. In sham-treated cells, the GSSG:GSH ratio was 0.02. This ratio increased by 25-fold to 0.5 in cells exposed to 50 µM diamide and attained values of more than 1.5 (75-fold) in cells exposed to >100 µM diamide (Fig 1A). Loss of protein thiols (a measure of protein-disulfide formation) was detectable in cells exposed to 250 µM diamide ( Fig. 1A). Despite significant oxidation of GSH in RPE cells exposed to 50 and 100 µM diamide, reducing equivalents remained adequate to protect PSHs from the formation of disulfides.

View larger version (28K): [in this window] [in a new window]   Figure 1. Dose-dependent effect of diamide (5 min) on levels of GSH, GSSG, and protein thiols (A) and of endogenous Ub-protein conjugates (B). GSH and GSSG values represent the average of duplicate samples, with no two replicates differing by more than 10%. Levels of protein thiols were determined in triplicate. Ub-protein conjugates were quantified in triplicate from immunoblots of RPE cell lysates probed with antiserum against Ub. All data are expressed as mean ± SEM.

Diminished levels of GSH in cells exposed to 50–250 µM diamide were accompanied by dose-dependent reductions in levels of endogenous Ub-protein conjugates, with an approximately 45% loss in cells exposed to 250 µM diamide (Fig 1B). Levels of Ub-protein conjugates tended to be greater in cells treated with 500 µM diamide than in cells treated with 250 µM diamide (Fig 1B; see Discussion).

To determine whether reductions in endogenous Ub-protein conjugates in diamide-treated cells were due to inhibition of Ub-conjugating enzyme activities, we assessed the relative activity of Ub-activating enzyme E1A in RPE cells exposed to 0–500 µM diamide for 10 min ( Fig. 2). Activity was assessed as the proportion of E1A found as the Ub-thiol ester. The E1A~Ub thiol ester (E1A~Ub, 128 kDa) is distinguished from native E1A (120 kDa) by its slightly retarded electrophoretic migration. The thiol ester linkage with Ub is confirmed by destruction in the presence of 2-mercaptoethanol (not shown). In sham-treated cells and cells exposed to 10 µM ~100 µM diamide, levels of E1A~Ub were twofold greater than the levels of native E1A, reflecting high levels of E1A catalytic activity (Fig 2, lanes 1–4). In cells exposed to 250 µM diamide, however, the levels of E1A~Ub were less than or equal to levels of native E1A ( Fig. 2, lanes 5 and 6). The losses were coincident with the formation of higher mass (135–160 kDa) immunoreactive E1A species ( Fig. 2, lanes 5 and 6, labeled with an asterisk). These higher mass E1A-containing species are likely to be disulfide-linked hetero-oligomers, since E1 was quantitatively recovered after treatment of samples with 2-mercaptoethanol (data not shown). The hetero-oligomers were presumably formed with E1A and smaller proteins, since they migrated well ahead of the predicted M_r for E1 oligomers (>200 kDa) on SDS gels. Loss of E1A~Ub in these experiments is therefore likely to reflect formation of disulfide bonds. This conclusion is consistent with the observation that in cells exposed to 50 and 100 µM diamide, the increase in the GSSG:GSH ratio was not accompanied by a detectable loss of PSHs (Fig 1A) and was insufficient to significantly inhibit E1~Ub formation ( Fig. 2, lane 3). Rather, loss of E1~Ub occurred only in the presence of diamide concentrations that induced a detectable loss of PSHs (i.e., 250 µM, 500 µM) ( Fig. 1A and Fig 2, lanes 5 and 6).

View larger version (22K): [in this window] [in a new window]   Figure 2. Loss of E1A~Ub thiol ester in RPE cells exposed to diamide. E1A and E1A~Ub thiol ester were identified on immunoblots by using rabbit anti-serum specific for the E1A isoform (E1A) and its Ub thiol ester (E1A-Ub). Detection was by ECL. Lane 1, sham-treated; Lanes 2–6, cells treated with 10–500 µM diamide. *Unidentified diamide-induced aggregates; ?, unidentified immunoreactive band. One of two representative experiments is presented.

Although reduced levels of endogenous E1~Ub thiol esters are mechanistically consistent with reduced levels of endogenous Ub-protein conjugates in cells exposed to >100 µM diamide, our inability to to detect the loss of E1 activity prevented us from ascribing the diminished levels of endogenous Ub-protein conjugates in cells exposed to 50 and 100 µM diamide to an inhibition of E1 activity ( Fig. 1B). To pursue this question, we compared rates of de novo synthesis of E1~¹²⁵I-Ub and E2s~¹²⁵I-Ub in ATP-supplemented supernatants from RPE cells that were exposed to 0–500 µM diamide (10 min). E1~¹²⁵I-Ub was identified on autoradiograms with an apparent mass of 130 kDa, and 5 E2s~¹²⁵I-Ub species were resolved with apparent masses between 22 and 40 kDa ( Fig. 3). Each radiolabeled moiety was fully reducible by 2-mercaptoethanol (not shown), thereby confirming its thiol ester linkage with ¹²⁵I-Ub.

View larger version (67K): [in this window] [in a new window]   Figure 3. Inhibitory effect of diamide on de novo synthesis of E1~125I-Ub and E2s~125I-Ub thiol esters and 125I-Ub-protein conjugates. Supernatant from RPE cells exposed to various concentrations of diamide were incubated (5 min) with ATP and 125I-labeled Ub. Reaction mixtures were then resolved on nonreducing SDS-PAGE, followed by autoradiography. E1-Ub, comigrating E1A~125I-Ub and E1B~ 125I-Ub thiol esters; E2s-Ub, E2~125I-Ub thiol esters. Molecular mass markers (kDa) are at left. The intensely radiolabeled bands at the top of the gel (>200 kDa) are comprised almost entirely of 125I-Ub-protein conjugates, based on their resistance to 2-mercaptoethanol. The intensely radiolabeled 17 kDa band is the 125I-Ub dimer.

Exposure of RPE cells to 10 and 50 µM diamide affected neither the levels of E1~¹²⁵I-Ub and E2s~¹²⁵I-Ub synthesized de novo nor levels of high mass ¹²⁵I-Ub-protein conjugates ( Fig. 3; compare lanes 2 and 3 with lane 1). In contrast, levels of E1~¹²⁵I-Ub and E2s~¹²⁵I-Ub were reduced by 16 and 32%, respectively, in supernatants derived from cells exposed to 100 µM diamide ( Fig. 3; compare lane 4 with lanes 1–3). Diminished E2 activities in cells exposed to 100 µM diamide appeared to be functionally significant, since they were associated with 40% reductions in the synthesis of high mass ¹²⁵I-Ub-protein conjugates ( Fig. 3; compare lane 4 with lanes 1–3). At diamide concentrations >100 µM, formation of E1~¹²⁵I-Ub and E2s~¹²⁵I-Ub was further attenuated ( Fig. 3; compare lanes 5 and 6 with lane 4), consistent with data relating levels of endogenous E1~Ub thiol esters and diamide concentration ( Fig. 2).

Because E2 activity requires activation of Ub by E1, reduced E2 activities in diamide-treated cells could be due entirely to reduced E1 activity. Subsequent experiments were designed to determine whether E2s were directly inhibited in diamide-treated cells. These experiments compared rates of formation of E2~¹²⁵I-Ub and ¹²⁵I-Ub-protein conjugates in supernatants from sham- and diamide-treated cells (250 µM diamide, 10 min) in the presence and absence of saturating levels of exogenous E1 (300 ng/µ l). If reduced rates of E2s~¹²⁵I-Ub formation reflected only inhibition of E1 activity, E1-supplementation should increase levels of E2s~¹²⁵I-Ub in supernatants of diamide-treated cells (10). In the absence of exogenous E1, supernatants of diamide-treated cells generated reduced levels of E1~¹²⁵I-Ub, E2s~¹²⁵I-Ub, and ¹²⁵I-Ub-protein conjugates vs. sham-treated cells ( Fig. 4; compare lanes 1 and 3). The addition of exogenous E1 resulted in a threefold increase in levels of E1~¹²⁵I-Ub ( Fig. 4; compare lane 4 vs. 3). However, this increase in E1 activity did not result in a detectable increase in E2s~¹²⁵I-Ub ( Fig. 4; compare lane 4 vs. 3). Consequently, levels of E2s~¹²⁵I-Ub and high mass ¹²⁵I-Ub-protein conjugates remained lower in E1-supplemented supernatant from diamide-treated cells than did levels generated by supernatant from sham-treated cells in the absence of exogenous E1 ( Fig. 4; compare lane 4 vs. 1). This occurred even though E1 activity was greater in supernatants from diamide-treated cells ( Fig. 4; compare lane 4 vs. 1). These results demonstrate that at least some capacity for formation of E2~Ub thiol esters and Ub-protein conjugates is compromised in diamide-treated cells independent of the inhibition of E1 activity.

View larger version (75K): [in this window] [in a new window]   Figure 4. E1 supplementation does not rescue E2 activity in RPE cells treated with diamide (250 µM, 10 min). Supernatants from sham-treated and diamide-treated cells were incubated as in Fig. 3 (but including 0.5 mM dithriothreitol) in the presence or absence of saturating amounts of purified E1 (300 ng/µl). Reaction mixtures were resolved on nonreducing SDS-PAGE and 125I-containing moieties were detected by autoradiography. E1-Ub, comigrating E1A~125I-Ub and E1B~ 125I-Ub thiol esters; E2s-Ub, E2~125I-Ub thiol esters. Other radiolabeled bands are 125I-Ub-protein conjugates, based on their resistance to 2-mercaptoethanol.

If, as hypothesized, E1 and E2 activities are regulated within cells by thiol redox status, E1 and E2 capacities to form thiol esters with Ub should be inhibited when the cellular GSSG:GSH ratio is elevated and should recover when GSSG:GSH ratio has returned to preoxidation levels. To test this prediction, we exposed RPE cells to 250 µM diamide for varying lengths of time and correlated the GSSG:GSH ratio with measures of Ub-conjugating activity either immediately after exposure to diamide or during cellular recovery after removal of diamide.

Temporal dynamics of the GSSG:GSH ratio response to 250 µM diamide are presented in Table 1. In sham-treated cells, the GSSG:GSH ratio was 0.02, reflecting an average GSH concentration 50 times greater than the concentration of GSSG. Within 2 min of exposure to diamide, the GSSG:GSH ratio increased by 170-fold due to the combined effects of a 10-fold loss of GSH and a >15-fold increase in GSSG. There was also a 25% loss of total glutathione relative to levels measured in sham-treated cells. This reduction in nonprotein thiols is likely to reflect protein S-thiolation, since it was associated with loss of PSH ( Fig. 1). It may also reflect extracellular secretion of GSSG (23). After 5 min of treatment, GSH, GSSG, and the GSSG:GSH ratio were essentially unchanged from the 2 min value ( Table 1). Evidence of recovery from thiol oxidation due to diamide was detected by 10 min of treatment. Specifically, the GSSG:GSH ratio decreased sevenfold—from 2.8 to 0.4—due to a threefold increase in GSH and a comparable decrease in GSSG. However, the GSSG:GSH ratio was still elevated >20-fold relative to control values and total glutathione remained depleted. By 30 min of treatment, cells exhibited an additional twofold decrease in GSSG but no significant increase in GSH. Consequently, the GSSG:GSH ratio decreased by twofold, but was still eight times greater than in control levels. Levels of total glutathione, although elevated over levels measured after 10 min, were still reduced relative to control values. The time-dependent recovery of the GSSG:GSH ratio does not reflect consumption of diamide, because treatment medium that was preincubated with RPE cells for up to 25 min was able to fully diminish GSH levels when transferred to fresh RPE cell cultures and incubated for an additional 5 min (data not shown). After removal of diamide and 2 h of recovery in a complete medium, levels of GSH, GSSG, and total glutathione in the cells were comparable to levels measured in sham-treated cells, and the GSSG:GSH ratio was completely reestablished to preoxidation levels (0.02).

The supernatants were also assayed for their ability to generate E1~¹²⁵I-Ub, E2s~¹²⁵I-Ub thiol esters, and ¹²⁵I-Ub-protein conjugates. Autoradiograms revealed that coincident with maximal increases in the GSSG:GSH ratio ( Table 1), the capacity to generate E1~¹²⁵I-Ub and E2s~¹²⁵I-Ub was compromised by 50% within 5 min of exposure to diamide (Fig 5B; compare lanes 1 and 2). Analysis of E1~¹²⁵I-Ub on lower percentage gels indicated that thiol esters of both E1 isoforms (E1A, E1B) were reduced similarly in diamide-exposed cells ( Fig. 5C). Reductions in the measures of E1 and E2 activities remained unabated after 10 min of exposure ( Fig. 5B; compare lane 3 with lane 1), despite the decrease in the GSSG:GSH ratio ( Table 1). These diminished E1 and E2 activities in supernatant from cells exposed to diamide for 5 and 10 min were reflected in >75% reductions in levels of all ¹²⁵I-Ub-protein conjugates ( Fig. 5A; compare lanes 2 and 3 with lane 1). Endogenous Ub-protein conjugates detected on immunoblots were also reduced by more than 75% (data not shown). Coincident with recovery of the GSSG:GSH ratio to 0.16 ( Table 1), we noted partial recovery of E1 and E2 activities and levels of ¹²⁵I-Ub-protein conjugates in cells incubated with diamide for 30 min ( Fig. 5A, B; compare lane 4 with lanes 2 and 3). However, these measures were still <50% of sham values ( Fig. 5A, B; compare lane 4 with lane 1). After a 2 h recovery, E1 and E2 activities returned to preoxidation levels ( Fig. 5B; compare lane 5 with lane 1), and levels of ¹²⁵I-Ub-protein conjugates synthesized de novo were equivalent to or in some cases exceeded levels measured in supernatants from sham-treated cells ( Fig. 5A; compare lane 5 with lane 1). This recovery of Ub-conjugating activity was coincident with full reestablishment of the cellular GSSG:GSH ratio and total glutathione ( Table 1).

View larger version (35K): [in this window] [in a new window]   Figure 5. Diamide-induced loss of Ub-conjugating activity is transient. RPE cells were treated with diamide (250 µM) for the times indicated and either harvested immediately or allowed to recover in complete medium after diamide washout. Cell supernatants were incubated with ATP and 125I-labeled Ub and analyzed by autoradiography after SDS-PAGE. A) autoradiogram after reducing SDS-PAGE (15%). Ub-con, 125I-labeled Ub-protein conjugates; molecular mass markers are at left. B) Nonreducing SDS-PAGE (15%); abbreviations are as in Fig. 4; molecular mass markers are at left. C) Nonreducing SDS-PAGE (6%); thiol esters of E1A (E1A-Ub) and E1B (E1B-Ub) are resolved. One of three experiments is presented.

To determine whether redox-related alterations in Ub-conjugating activity affect Ub-dependent proteolytic activity, we determined the effect of diamide treatment and recovery from diamide on Ub-dependent degradation of an exogenous substrate (¹²⁵I-labeled ß-lactoglobulin). ATP-supplemented and ATP-depleted supernatants from sham-exposed, diamide-exposed, and `recovered' RPE cells were incubated with substrate, after which acid-soluble cpm were counted and used to calculate percent of total degradation (ATP-supplemented) and percent of ATP-independent degradation (ATP-depleted). ATP-dependent degradation (total degradation minus ATP-independent degradation) was entirely Ub dependent, since the proteasome inhibitor MG132 (22, 24, 25) completely blocked ATP-dependent degradation (data not shown).

Ub-dependent proteolysis was blocked almost entirely (90% reduction) in supernatants derived from diamide-treated cells ( Fig. 6). Reduced rates of UPP-dependent proteolysis were coincident with an increased GSSG:GSH ratio ( Table 1) and reduced Ub-conjugating activity ( Fig. 5A, B, lane 3). The capacity for Ub-dependent proteolysis was reestablished during the recovery from diamide treatment. Ub-conjugating activity and the GSSG:GSH ratio were similarly reestablished at this time ( Fig. 5A, B, lane 5; Table 1). These results suggest that redox-associated fluctuations in Ub-conjugating activity can have dramatic effects on processes under UPP control, including Ub-dependent proteolysis. In contrast to the effects of the redox state on UPP-dependent degradation of ß-lactoglobulin, no diamide-associated changes were observed in the magnitude of ATP-independent degradation.

View larger version (34K): [in this window] [in a new window]   Figure 6. Diamide inhibits Ub-dependent proteolysis. RPE cells were treated for 10 min with PBS (sham) or with 250 µM diamide (diamide). Some diamide-treated cells were allowed to recover for 2 h after diamide washout (recovery). ATP-supplemented and ATP-depleted cell supernatants obtained from these cells were incubated (1 h, 37°C) with 125I-labeled ß-lactoglobulin and 100 ng/µl Ub. Percent degradation (mean±SEM, n=4) was determined from acid-soluble cpm. UPP-dependent degradation was defined as the reduction in ATP-dependent proteolysis in the presence of MG132. Statistically significant differences between treatment means were determined by t tests, using Bonferroni adjustments for multiple comparisons. UPP-dependent degradation (hatched bars); UPP-independent degradation (blank bars). ***P < 0.001, diamide vs. sham. Data are from two experiments performed in duplicate with different preparations of supernatant.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Redox regulation through reactive sulfhydryls is emerging as an important mechanism by which activities of thiol-dependent enzymes and transcription factors are modulated in vivo (26–28). We recently proposed that Ub-conjugating activity is also regulated in vivo by cellular redox status (9). Our model predicts that upon exposure to an oxidant (or oxidants), increases in the cellular ratio of oxidized to reduced glutathione (GSSG:GSH ratio) result in rapid S-thiolation of E1 and E2 active-site sulfhydryls by GSSG (glutathiolation), with a resultant reduction in the formation of E1~Ub and E2s~Ub. These thiol esters are requisite intermediates in the protein ubiquitinylation pathway. This model is consistent with the previously demonstrated effects of exogenous alkylating agents and chemical reductants on E1 and E2 activities (29, 30). Our model further predicts that E1 and E2 activities will remain compromised until reestablishment of GSSG:GSH ratios, which more closely approximate preoxidation levels, as would occur upon cellular recovery from oxidant insult. Although our earlier data were consistent with a thiol-dependent mechanism for regulation of E1 and E2 activities (9–11), use of H₂O₂ as the oxidant left open the possibility that nonsulfhydryl amino acid side chains as well as the protein backbones of E1 and E2s were also oxidatively modified in these studies (15). In the present study, use of the thiol-specific oxidant diamide allowed us to test more rigorously the hypothesis that E1 and E2 activities are directly regulated by the thiol redox status (GSSG:GSH) after oxidant challenge to cells.

As previously observed in studies of H₂O₂-treated retina tissue and RPE cells (9), levels of endogenous Ub-protein conjugates and ¹²⁵I-Ub-protein conjugates formed de novo diminished when the the GSSG:GSH ratio increased to 0.5 in diamide-treated cells. This was due to diminished E1 and E2 activities, as indicated by a diminished capacity to generate E1~¹²⁵I-Ub and E2~¹²⁵I-Ub thiol esters in vitro. Activity was partially restored when the GSSG:GSH ratio fell below 0.4 ( Table 1and Fig. 5). The consistency of these results with those of a previous study in which exogenous GSSG and GSH were added to RPE supernatant to mimic the effects of oxidation (9) indicates that Ub-conjugating activity is significantly inhibited when the cellular GSSG:GSH ratio increases to 0.4–0.5. Full reestablishment of E1 and E2 activities was observed during posttreatment recovery, when the GSSG:GSH ratio was fully restored to its preoxidation value (0.02).

Based on loss of PSH and the detection of disulfide-linked aggregates on SDS-gels, diminished E1 activity appears to involve formation of E1-protein mixed disulfides. E1 possesses 16 cysteine residues, and thus contains multiple targets for S-thiolation in addition to the active-site sulfhydryl. Consistent with data from bovine lens epithelial cells (10), E2 thiol ester activity was inhibited in part independently of reduced E1 activity, since enhancing the level of E1~Ub by addition of purified E1 did not restore levels of E2s~Ub formed de novo in supernatants from diamide-treated cells. Inhibition of E2s in these experiments is almost certain to involve S-thiolation of the active site: with the exception of E2–25K (31), the active-site sulfhydryl is the only one found in E2s in RPE cells. Similar data were previously gathered using kinetic analysis (M. Obin, J. Jahngen-Hodge, and A. Taylor, unpublished results).

Although RPE cells treated with 500 µM diamide exhibited the largest increase in GSSG:GSH ratio and the most attenuated de novo Ub-conjugating activity, levels of endogenous conjugates were actually elevated in these cells relative to cells treated with 250 µM diamide. Steady-state levels of Ub-protein conjugates reflect net rates of conjugate assembly by ubiquitinylation as well as rates of conjugate degradation and/or disassembly by the 26 S proteasome and actions of Ub-specific isopeptidases. The highest concentration of diamide may have inhibited conjugate degradation or disassembly, thereby stabilizing extant conjugates. Ub-isopeptidases are likely candidates for inhibition by diamide, since these enzymes are also thiol dependent.

Ubiquitinylation is involved in a variety of critical biological processes. The best understood is Ub-dependent proteolysis, which regulates levels of numerous regulatory proteins and cell effectors (1–4). Data ( Fig. 6) suggest that only the Ub-dependent proteolytic capacity was altered in response to oxidation, consistent with previous studies in this laboratory (10, 11). Inhibition and subsequent reestablishment of the capacity for Ub-dependent proteolysis were coincident with the increase and subsequent recovery of the GSSG:GSH ratio. These data suggest that rates of Ub-dependent proteolysis can be regulated by the thiol redox state of cells. This regulation is likely to be nonselective with respect to substrates, since it apparently involves regulation of all E2s.

Reversible thiolation/dethiolation (and thus transient inhibition) of Ub-conjugating enzymes may serve several potentially adaptive functions in cells. First, as suggested for other regulatory molecules (32, 33), thiolation/dethiolation may protect E1 and E2 from permanent oxidative damage and promote their reactivation when oxidant stress has abated. Second, rapid down-regulation of Ub-dependent proteolysis could protect proteins whose conformations may be only transiently altered by oxidation (e.g., protein disulfides) from Ub-dependent proteolysis, the predominant cellular mechanism for degradation of damaged and aberrant proteins (1–4). If oxidant insult is not too severe or prolonged, oxidatively altered proteins can reattain functional conformations through the actions of a variety of dethiolases, reductases, and chaperonins (34, 35). It is conceivable that down-regulation of Ub-dependent proteolysis protects potentially reparable proteins from untimely and wasteful degradation. Third, since Ub-dependent proteolysis of cyclins and regulators of cyclin-dependent kinases is intimately involved in cell cycle progression (1–4), diminished Ub-dependent proteolysis in response to oxidant stress may facilitate an adaptive cellular `checkpoint' response characterized by cell cycle arrest. Upon cellular recovery from stress, reestabishment of the GSSG:GSH ratio and UPP activity is expected to restore normal proteolytic regulation of cell cycle effectors.

However, reductions in UPP activity could conceivably become pathological if the GSSG:GSH ratio is chronically increased. Prolonged reductions in rates of ubiquitinylation could lead to the accumulation of damaged and/or obsolete proteins. Such proteolytic inactivation may be causally related to diseases and to pathologic states (e.g., cataract, neurodegenerations, `normal' aging) that are characterized by chronic reductions in cellular GSH and by accumulation of nondegraded proteins within cells (36–38).

	ACKNOWLEDGMENTS

 
This research was supported by USDA contract # 53–3K06–0-1 and National Institutes of Health grant EY11735–01 (M.O.)

	FOOTNOTES

 
¹ Correspondence: Laboratory for Nutrition and Vision Research, JMUSDA-HNRCA at Tufts University, 711 Washington Street, Boston, MA 02111, USA. E-mail: Taylor_c1{at}hnrc.tufts.edu

² Abbreviations: UPP, ubiquitin–proteasome pathway; Ub, ubiquitin; E1, ubiquitin-activating enzyme; E1~Ub, thiol ester of E1 and ubiquitin; E2s, ubiquitin carrier proteins; E2~Ub, thiol ester of E2 and ubiquitin; RPE cell, retinal pigment epithelial cell; GSH, reduced glutathione, GSSG, oxidized glutathione; PBS, phosphate-buffered saline; -Glu-Glu, -glutamylglutamate; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; PSH, protein sulfhydryl; HPLC, high-performance liquid chromatography; PVDF, polyvinylidene difluoride.

Received for publication October 15, 1997. Accepted for publication January 5, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Wilkinson, K. D. (1995) Roles of ubiquitinylation in proteolysis and cellular regulation. Annu. Rev. Nutr. 15, 161–189[Medline]
2. Hochstrasser, M. (1996) Ubiquitin-dependent protein degradation. Annu. Rev. Genet. 30, 405–439[Medline]
3. Hershko, A., and Ciechanover, A. (1992) The ubiquitin system for protein degradation. Annu. Rev. Biochem. 61, 761–807[Medline]
4. Ciechanover, A., and Schwartz, A. L. (1994) The ubiquitin-mediated proteolytic pathway: mechanisms of recognition of the proteolytic substrate and involvement in the degradation of native cellular proteins. FASEB J. 8, 182–191[Abstract]
5. Chen, Z. J., Parent, L., and Maniatis, T. (1996) Site-specific phosphorylation of IB by a novel ubiquitination-dependent protein kinase activity. Cell 84, 853–862[Medline]
6. Hicke, L., and Riezman, H. (1996) Ubiquitination of a yeast plasma membrane receptor signals its ligand-stimulated endocytosis. Cell 84, 277–287[Medline]
7. Matunis, M. J., Coutavas, E., and Blobel, G. (1996) A novel ubiquitin-like modification modulates the partitioning of the Ran-GTPase-activating protein RanGAP1 between the cytosol and the nuclear pore complex. J. Cell Biol. 135, 1457–1470[Abstract/Free Full Text]
8. Cheng, L., Watt, R., and Piper, P. W. (1994) Polyubiquitin gene expression contributes to oxidative stress resistance in respiratory yeast (Saccharomyces cerevisiae). Mol. Gen. Genet. 246, 358–362
9. Jahngen-Hodge, J., Obin, M., Gong, X., Shang, F., Nowell, T., Gong, J., Abasi, H., Blumberg, J., and Taylor, A. (1997) Regulation of ubiquitin conjugating enzymes by glutathione following oxidative stress. J. Biol. Chem. 272, 28218–28226[Abstract/Free Full Text]
10. Shang, F., Gong, X., and Taylor, A. (1997) Activity of ubiquitin dependent pathway in response to oxidative stress: ubiquitin activating enzyme (E1) is transiently upregulated. J. Biol. Chem. 272, 23086–23093[Abstract/Free Full Text]
11. Shang, F., and Taylor, A. (1995) Oxidative stress and recovery from oxidative stress are associated with altered ubiquitin conjugating and proteolytic activities in bovine lens epithelial cells. Biochem. J. 307, 297–303
12. Haas, A. L., and Rose, I. A. (1982) The mechanism of ubiquitin activating enzyme. A kinetic and equilibrium analysis. J. Biol. Chem. 257, 10329–10337[Free Full Text]
13. Hershko, A., Heller, H., Elias, S., and Ciechanover, A. (1983) Components of the ubiquitin-protein ligase system. Resolution, affinity purification, and role in protein breakdown. J. Biol. Chem. 258, 8206–8214[Abstract/Free Full Text]
14. Scheffner, M., Nuber, U., and Huibregtse, J. M. (1995) Protein ubiquitination involving a E1-E2-E3 thiolester cascade. Nature (London) 373, 81–83[Medline]
15. Berlett, B. S., and Stadtman, E. R. (1997) Protein oxidation in aging, disease, and oxidative stress. J. Biol. Chem. 272, 20313–20316[Free Full Text]
16. Kosower, N. S., and Kosower, E. M. (1995) Diamide: an oxidant probe for thiols. Methods Enzymol. 251, 123–133[Medline]
17. Obin, M., Nowell, T., and Taylor, A. (1995) A comparison of ubiquitin-dependent proteolysis of rod outer segment proteins in reticulocyte lysate and a retinal pigment epithelial cell line. Curr. Eye Res. 14, 751–760[Medline]
18. Handley-Gearhart, P. M., Stephen, A. G., Trausch-Azar, J. S., Ciechanover, A., and Schwartz, A. L. (1994) Human ubiquitin-activating enzyme, E1. Indication of potential nuclear and cytoplasmic subpopulations using epitope-tagged cDNA constructs. J. Biol. Chem. 269, 33171–33178[Abstract/Free Full Text]
19. Handley, P. M., Mueckler, M., Siegel, N. R., Ciechanover, A., and Schwartz, A. L. (1991) Molecular cloning, sequence and tissue distribution of human ubiquitin-activating enzyme E1. Proc. Natl. Acad. Sci. USA 88, 258–262[Abstract/Free Full Text]
20. Fariss, M. W., and Reed, D. J. (1987) High performance liquid chromatography of thiols and disulfide: dinitrophenol derivatives. Methods Enzymol. 143, 101–109[Medline]
21. Sedlak, J., and Lindsay, R. H. (1968) Estimation of total, protein-bound, and nonprotein sulfhydryl groups in tissue with Ellmans's reagent. Anal. Biochem. 25, 192–205[Medline]
22. Obin, M., Jahngen-Hodge, J., Nowell, T., and Taylor, A. (1996) Ubiquitinylation and ubiquitin-dependent proteolysis in vertebrate photoreceptors (rod outer segments): evidence for ubiquitinylation of Gt and rhodopsin. J. Biol. Chem. 271, 14473–14484[Abstract/Free Full Text]
23. Carelli, S., Ceriotti, A., Cabibbo, A., Fassina, G., Ruvo, M., and Sitia, R. (1997) Cysteine and glutathione secretion in response to protein disulfide bond formation in ER. Science 277, 1681–1684[Abstract/Free Full Text]
24. Rock, K. L., Gramm, C., Rothstein, L., Clark, K., Stein, R., Dick, L., Hwang, D., and Goldberg, A. L. (1994) Inhibitors of the proteasome block the degradation of most cell proteins and the generation of peptides presented on MHC class I molecules. Cell 78, 761–771[Medline]
25. Palombella, V. J., Rando, O. J., Goldberg, A. L., and Maniatis, T. (1994) The ubiquitin-proteasome pathway is required for processing of NF-B procursor protein and the activation of NF-B. Cell 78, 773–785[Medline]
26. Akamatsu, Y., Ohno, T., Hirota, K., Kagoshima, H., Yodoi, J., and Shigesada, K. (1997) Redox regulation of the DNA binding activity in transcription factor PEBP2. The roles of two conserved cysteine residues. J. Biol. Chem. 272, 14497–14500[Abstract/Free Full Text]
27. Shrimpton, C. N., Glucksman, M. J., Lew, R. A., Tullai, J. W., Margulies, E. H., Roberts, J. L., and Smith, A. I. (1997) Thiol activation of endopeptidase EC 3.4..24.15. J. Biol. Chem. 272, 17395–17399[Abstract/Free Full Text]
28. Liu, H., Lightfoot, R., and Stevens, J. L. (1996) Activation of heat shock factor by alkylating agents is triggered by glutathione depletion and oxidation of protein thiols. J. Biol. Chem. 271, 4805–4812[Abstract/Free Full Text]
29. Haas, A. L., Warms, J. V. B., Hershko, A., and Rose, I. A. (1982) Ubiquitin-activating enzyme, mechanism and role in protein-ubiquitin conjugation. J. Biol. Chem. 257, 2543–2548[Abstract/Free Full Text]
30. Pickart, C. M., and Rose, I. A. (1985) Functional heterogeneity of ubiquitin carrier proteins. J. Biol. Chem. 260, 1573–1581[Abstract/Free Full Text]
31. Chen, Z., Niles, E. G., and Pickart, C. M. (1991) Isolation of a cDNA encoding a mammalian multiubiquitinating enzyme (E2_25k) and overexpression of the functional enzyme in Escherichia coli. J. Biol. Chem. 266, 15698–15704[Abstract/Free Full Text]
32. Miller, R. M., Sies, H., Park, E. M., and Thomas, J. A. (1990) Phosphorylase and creatine kinase modification by thiol-disulfide exchange and by xanthine oxidase-initiated S-thiolation. Arch. Biochem. Biophys. 276, 355–363[Medline]
33. Vessey, D. A., Lee, K. H., and Blacker, K. L. (1992) Characterization of the oxidative stress initiated in cultured human keratinocytes by treatment with peroxides. J. Invest. Dermat. 99, 859–863[Medline]
34. Brot, N., Weissbach, L., Werth, J., and Weissbach, H. (1981) Enzymatic reduction of protein-bound methionine sulfoxide. Proc. Natl. Acad. Sci. USA 78, 2155–2158[Abstract/Free Full Text]
35. Holmgren, A. (1985) Thioredoxin. Annu. Rev. Biochem. 54, 237–271[Medline]
36. Taylor, A., Dorey, C. K., and Nowell, T. J. (1997) Oxidative stress and ascorbate in relation to risk for cataract and age-related maculopathy. In Vitamin C in Health and Disease (Packer, L. A. F., Jr., ed) pp. 267–283, Marcel Dekker, New York
37. Di Monte, D. A., Chan, P., and Sandy, M. S. (1992) Glutathione in Parkinson's disease: a link between oxidative stress and mitochondrial damage? Ann. Neurol. 32 (Suppl.), 111–115
38. Jenner, P. (1996) Oxidative stress in Parkinson's disease and other neurodegenerative disorders. Pathol. Biol. 44, 57–64[Medline]