Regulation of NF-κB activity and inducible nitric oxide synthase by regulatory particle non-ATPase subunit 13 (Rpn13)

Abstract

Human Rpn13, also known as adhesion regulating molecule 1 (ADRM1), was recently identified as a novel 19S proteasome cap-associated protein, which recruits the deubiquitinating enzyme UCH37 to the 26S proteasome. Knockdown of Rpn13 by siRNA does not lead to global accumulation of ubiquitinated cellular proteins or changes in proteasome expression, suggesting that Rpn13 must have a specialized role in proteasome function. Thus, Rpn13 participation in protein degradation, by recruiting UCH37, is rather selective to specific proteins whose degradation critically depends on UCH37 deubiquitination activity. The specific substrates for the Rpn13/UCH37 complex have not been determined. Because of a previous discovery of an interaction between Rpn13 and inducible nitric oxide synthase (iNOS), we hypothesized that iNOS is one of the substrates for the Rpn13/UCH37 complex. In this study, we show that Rpn13 is involved in iNOS degradation and is required for iNOS interaction with the deubiquitination protein UCH37. Furthermore, we discovered that IκB-α, a protein whose proteasomal degradation activates the transcription factor NF-κB, is also a substrate for the Rpn13/UCH37 complex. Thus, this study defines two substrates, with important roles in inflammation and host defense for the Rpn13/UCH37 pathway.

The ubiquitin-proteasome system provides the main nonlysososmal degradation pathway for intracellular proteins in eukaryotes. Ubiquitin is covalently linked to proteins targeted for degradation. The ubiquitinated substrate is rapidly hydrolyzed by the 26S proteasome. The latter is an ATP-dependent complex containing the core 20S proteasome, responsible for proteolysis, and enclosed by two 19S regulatory complexes (caps). The 19S regulatory complex is in charge of recognizing, unfolding, deubiquitinating, and threading the substrate into the 20S core for proteolysis (1–3). The presence of proteasome-associated deubiquitination activity is well recognized, but its role in protein degradation is being actively investigated (4–6). Deubiquitination can serve to rescue proteins from degradation. Alternatively, deubiquitination could be essential for protein degradation, either for editing the ubiquitin chain to an appropriate length or to allow the protein to unfold before entering the 20S proteasome core.

Recent studies using affinity purification methods have led to the discovery of additional proteins associated with the proteasome. One such protein is adhesion regulating molecule (ADRM1), which is recently identified as a novel proteasome-associated protein, which recruits the deubiquitinating enzyme UCH37 to the 26S proteasome (7–10). Before this discovery, ADRM1 had been described as a heavily glycosylated membrane protein, which regulates cell adhesion (11). ADRM1 was also identified in a yeast two-hybrid screen in search for inducible nitric oxide synthase (iNOS)-interacting proteins, termed NAP110 (NOS Associated Protein) and suggested to inhibit nitric oxide (NO) synthesis by interfering with iNOS dimerization (12). In the context of inflammation and cell-mediated immunity, iNOS is responsible for high-output NO synthesis and is induced in response to proinflammatory cytokines, lipopolysaccharide (LPS), and other inflammatory mediators (13, 14).

However, in 2006, Jørgensen et al. (8) showed that ADRM1 is associated with the 19S regulatory cap. ADRM1 is expressed constitutively and in a nonglycosylated form in almost all tissues. Further, and in contrast to some earlier reports, ADRM1 is not induced by IFN-γ and is mostly cytosolic. Based on these findings, the function of ADRM1 as an adhesion molecule is now doubted and, hence, the name is rather a misnomer. Deletion of ADRM1 by siRNA does not lead to global accumulation of ubiquitinated cellular proteins or changes in proteasome expression, suggesting that ADRM1 must have a specialized role in proteasome function (8). Shorty after the above report, three independent groups confirmed that ADRM1 is a novel 19S proteasome cap-associated protein, but they further revealed a unique role for ADRM1 in protein degradation (7, 9, 10). These groups found that ADRM1 interacts with the carboxyl-terminal tail of ubiquitin C-terminal hydrolase UCH37 and recruits it to the proteasome. UCH37 is the principal deubiquitinating enzyme associated with mammalian proteasomes. The N-terminal one-third of ADRM1 is 41% similar to Rpn13, which is a subunit of the 19S regulatory complex in yeast and, therefore, ADRM1 has been renamed Rpn13 (7, 9, 10). We will use the term Rpn13 henceforth. The yeast ortholog is much shorter than human Rpn13. However, among mammals, Rpn13 is well conserved with 97% identity between the human and murine proteins. The above studies suggest that Rpn13 C-terminal domain has evolved to recruit UCH37 to the proteasome. Thus, the N-terminal one-third of Rpn13 binds to the proteasome, whereas the C-terminal portion binds to UCH37 (7, 9, 10). In this context, Rpn13 knockdown decreases the deubiquitination activity of the 26S proteasomes, indicating that UCH37 is the dominant deubiquitination enzyme associated with the 26S proteasome and that recruitment of UCH37 by Rpn13 is required for this activity.

All of the four groups that discovered Rpn13 studied the effect of Rpn13 knockdown on global protein degradation. Although Rpn13 is a receptor for ubiquitin (15, 16), three of the four groups (7, 8, 10) found that knockdown of Rpn13 does not affect general short-lived protein degradation, whereas one group observed a slight but consistent reduction (9). These studies, taken together, suggest that Rpn13 is not required for degradation of all proteins but its participation in protein degradation, by recruiting UCH37, is rather selective to specific proteins whose degradation may critically depend on UCH37 deubiquitination activity. The specific substrates for the Rpn13/UCH37 complex have not been determined. Because of previously discovered association between Rpn13 and iNOS (12), we hypothesized that iNOS, an important protein involved in host defense and inflammation, is one of the substrates for the Rpn13/UCH37 complex. In this study, we show that Rpn13 is involved in iNOS degradation and is required for iNOS interaction with the deubiquitination enzyme UCH37. Furthermore, we discovered that IκB-α, a protein whose proteasomal degradation activates the transcription factor NF-κB, is also a substrate for Rpn13/UCH37 complex. Thus, this study defines two substrates in the Rpn13/UCH37 pathway, both with important roles in inflammation and host defense.

Results and Discussion

Knockdown of Rpn13 Decreases iNOS Cellular Level.

We hypothesized that Rpn13 is involved in the degradation of specific proteins that require the UCH37 deubiquitination activity during their degradation by the 26S proteasome. Because of the previously reported interaction between Rpn13 and iNOS (12), we speculated that knockdown of Rpn13 might reduce iNOS degradation and, thus, increase the iNOS steady-state cellular level. To investigate this possibility, we used small interference RNA (siRNA/shRNA) to knockdown Rpn13 in RAW264.7 cells, A549 cells, and primary normal human bronchial epithelial (NHBE) cells grown to full differentiation at an air-liquid interphase. We then induced iNOS expression in these cells with a mixture of cytokines (IL-1β and TNF-α) or with LPS. Contrary to our expectation, knockdown of Rpn13 markedly reduced iNOS cellular levels in all cell models tested (Fig. 1 A and C). Because iNOS is the product of an inducible gene, we investigated whether iNOS mRNA induction was affected by Rpn13 knockdown. Real-time PCR analysis indicated that iNOS mRNA was reduced by Rpn13 knockdown, suggesting that Rpn13 is required for iNOS induction (Fig. 1B).

Fig. 1.

Knockdown of Rpn13 decreases iNOS cellular level. (A and B) RAW264.7 cells were transduced with lentivirus expressing either Rpn13-specific or control shRNA. A549 cells and primary normal human bronchial epithelial (NHBE) cells were transfected for 72 h with either Rpn13-specific or control siRNA. iNOS was induced in RAW264.7 cells by 8 h of incubation with 100 ng/mL LPS and in A549 cells and primary NHBE cells with 24 h of incubation in a cytokine mixture of IL-1β (0.5 ng/mL) and TNF-α (10 ng/mL). Cells were lysed and Western blot analysis was done to evaluate Rpn13, iNOS, or β-actin (A). In parallel experiments, total RNAs were isolated from cell lysates and iNOS mRNA level was measured by using real-time PCR (B). Data represents mean ± SD, n = 3. *P < 0.05, compared with control condition. (C) RAW264.7 cells were transfected for 72 h with Rpn13-specific siRNA or control siRNA. iNOS was induced by 8 h of incubation of cells with 100 ng/mL LPS.

Knockdown of Rpn13 Down-Regulates NF-κB Activity.

NF-κB is a major transcription factor involved in the inducible expression of iNOS in both murine and human cells (17, 18). It is a ubiquitous transcription factor that is activated by myriad proinflammatory stimuli and cytokines. Proinflammatory stimuli activate NF-κB through a tightly regulated cascade of phosphorylation, ubiquitination, and proteasomal proteolysis of a physically associated class of inhibitor molecules. The best characterized of these inhibitors is IκB-α that physically associates with the NF-κB proteins, p65 and p50, in the cytosol and prevents their translocation to the nucleus. The ubiquitin proteasome pathway is required for degradation of IκB-α, thus resulting in NF-κB activation (19). We therefore hypothesized that attenuated iNOS induction, caused by Rpn13 knockdown, is secondary to reduced NF-κB activity caused by retarded degradation of IκB-α. To test this hypothesis, we evaluated the effect of Rpn13 knockdown on cellular levels of IκB-α. Rpn13 knockdown resulted in an increased level of IκB-α in RAW264.7 cells, A549 cells, and primary NHBE cells (Fig. 2 A–C). It is known that two distinct pools of IκB-α exist in cells. The larger IκB-α pool is associated with NF-κB protein, p65, and the minor pool remains as a “free” protein. Both pools of IκB-α are degraded by the proteasomes, and changes in the levels of either can affect NF-κB activity (20). To confirm that the increased amount of IκB-α, observed after Rpn13 knockdown, is associated with an increase in p65-bound IκB-α, we evaluated the effect of Rpn13 knockdown on both pools of IκB-α in RAW264.7 cells. The p65-bound IκB-α was evaluated by immunoprecipitation of cell lysates by using p65-specific antibody followed by Western blotting with IκB-α antibody. Free IκB-α was detected by performing a second immunoprecipitation of the supernatant of the first immunoprecipitation, i.e., of proteins that did not bind to the p65 antibody. Rpn13 knockdown resulted in increased levels of both p65-bound IκB-α and free IκB-α (Fig. 2D). These data suggest that Rpn13 is required for degradation of IκB-α and, thus, in cells deficient in Rpn13, NF-κB activity was reduced. To directly confirm that NF-κB activity was reduced in Rpn13-deficient cells, we used a luciferase reporter vector containing an NF-κB response element and transfected it into RAW264.7 cells. Rpn13 knockdown significantly reduced NF-κB activity at resting state (Fig. 2E) and after activation with LPS (Fig. 2F). Taken together, these results indicate that Rpn13 is required for NF-κB activity through its role in proteasomal degradation of IκB-α.

Fig. 2.

Knockdown of Rpn13 down-regulates NF-κB activity. RAW264.7 cells (A) were transduced with lentivirus vector expressing either Rpn13-specific or control shRNA. A549 cells (B) or primary NHBE cells (C) were transfected for 72 h with either Rpn13-specific or control siRNA. Cells were lysed and Western blot analysis was done to evaluate Rpn13, IκB-α, or β-actin. (D) Cell lysates of RAW264.7 cells, treated as in A, were immunoprecipitated (IP) with p65 antibody and analyzed with Western blot by using antibodies against p65 or IκB-α. The first two lanes represent 10% input of cell lysates, and the middle two lanes represent IκB-α bound to p65. In the last two lanes, flow-through of the IP experiment were subjected to another IP to pull down free IκB-α by using IκB-α antibody. (E and F) RAW264.7 cells, treated as in A, were transfected for 24 h with luciferase reporter vector containing an NF-κB response element. Luciferase activity was measured before and after cells were stimulated for 8 h with LPS (100 ng/mL). *P < 0.05, compared with control condition.

Knockdown of UCH37 Down-Regulates NF-κB Activity.

A proposed function of Rpn13 is to recruit the deubiquitination enzyme UCH37 to the 26S proteasome (7, 9, 10). We therefore hypothesized that Rpn13’s role in regulating NF-κB activity is to recruit UCH37 for the proteasomal degradation of IκB-α. Thus, UCH37 should also be required for NF-κB activity and, in turn, for iNOS induction. We studied the effect of UCH37 knockdown by siRNA on iNOS induction by LPS in RAW264.7 cells and on NF-κB reporter activity in HEK293 cells. Knockdown of UCH37 aborted induction of iNOS by LPS in RAW264.7 cells (Fig. 3A and Fig. S1A). It also significantly reduced NF-κB activity in HEK293 cells both at resting state (Fig. 3B) and after stimulation with IL-1β (Fig. 3C). These data clearly establish that the Rpn13/UCH37 complex is required for IκB-α proteasomal degradation and, thus, for NF-κB activation.

Fig. 3.

Knockdown of UCH37 down-regulates NF-κB activity. (A) RAW264.7 cells were transfected for 72 h with UCH37-specific or control siRNA. Cell lysates were evaluated by RT-PCR using specific primers for UCH37 or GAPDH (Upper) and by Western blot analysis using antibodies against iNOS or β-actin (Lower). (B and C) HEK293 cells were transfected for 72 h with UCH37-specific or control siRNA, and then transfected for 24 h with luciferase reporter vector containing NF-κβ response element. Luciferase activity was measured before (B) and after (C) 4 h of stimulation with IL-1β. *P < 0.05, compared with control condition.

Rpn13 Is Required for iNOS Degradation.

Before the discovery of Rpn13 association with the proteasome, Rpn13 was shown to be an iNOS-interacting protein (12). We have shown that iNOS is degraded primarily by the ubiquitin proteasome pathway (21, 22). Thus, iNOS might also be a substrate for the Rpn13/Uch37 complex. We therefore designed experiments to study the role of Rpn13 in iNOS degradation in iNOS-transfected HEK293 cells. These cells do not express endogenous iNOS from their own genome but only that coded for by the plasmid. Thus, iNOS transcription is independent of NF-κB (21, 22). In these cells, knockdown of Rpn13 resulted in increased iNOS steady-state levels, an effect that was further confirmed by a concomitant increase in NO production (Fig. 4A and Fig. S2). To directly determine whether the increase in iNOS level in Rpn13-deficient cells was due to retarded iNOS degradation, we measured iNOS half-life by using pulse–chase analysis (23, 24). In Rpn13-deficient cells, iNOS half-life was significantly increased compared with control cells (Fig. 4B). These results suggest that Rpn13 is involved in iNOS degradation. We then investigated whether the effect of Rpn13 knockdown on iNOS degradation was part of a global effect on protein degradation. We tested the effect of Rpn13 knockdown on short-lived bulk protein degradation in HEK293 cells (Fig. 4C) (25). There were no significant differences in bulk protein degradation caused by Rpn13 knockdown. These results are consistent with prior studies (7, 8, 10) and suggest that Rpn13 is mostly required for degradation of specific substrates such as iNOS and IκB-α. Some proteins become degraded when bound to proteasomes (26). However, although iNOS binds to proteasome subunit Rpn13, degradation of iNOS still depends on its coupling to ubiquitin (21, 23).

Fig. 4.

Rpn13 is required for iNOS degradation. HEK293 cells, stably expressing Rpn13 shRNA or control shRNA, were transfected for 18 h with iNOS. (A) Cell lysates were analyzed by Western blot using antibodies against Rpn13, iNOS, or GAPDH (Upper). iNOS activity was evaluated by measuring accumulation of nitrite in culture media (Lower). (B) Cells were pulse-labeled with [³⁵S]methionine/cysteine for 1 h and chased with unlabeled media. iNOS was immunoprecipitated from lysates of cells harvested at intervals. Eluted proteins were analyzed by SDS/PAGE (Upper). Bands, representing ³⁵S-labeled iNOS, were quantitated to calculate iNOS half-life (Lower). (C) Rate of global short-lived protein degradation in cells was measured. Data represent mean ± SD, n = 3. **P < 0.01.

UCH37 Is Required for iNOS Degradation.

We then investigated whether Rpn13’s role in degradation of iNOS is linked to Rpn13’s ability to recruit the deubiquitination enzyme UCH37. Consistent with this hypothesis, we found increased levels of ubiquitinated iNOS in HEK293 cells subjected to Rpn13 knockdown (Fig. 5A and Fig. S3). It should be noted, however, that the increase in ubiquitinated iNOS in these cells could merely reflect the retarded degradation of iNOS. To directly confirm the effect of UCH37 on iNOS degradation, we examined the effect of UCH37 knockdown in HEK293 cells, stably expressing iNOS. Knockdown of UCH37 accelerates degradation of bulk cell protein (6), as expected for a deubiquitylating enzyme. In contrast, we found that knockdown of UCH37 led to an increase in iNOS steady-state levels (Fig. 5B and Fig. S1B) and slowed the rate of iNOS degradation (Fig. 5C). These data suggest that UCH37, similar to Rpn13, participates in iNOS degradation. Thus, knockdown of UCH37 causes accelerated degradation of bulk protein, an effect consistent with UCH37 function as a deubiquitination protein. In contrast, knockdown of UCH37 stabilized iNOS, an observation that suggests that iNOS is rather a specific substrate for UCH37, requiring UCH37 deubiquitination activity for its degradation. It is intriguing that the Rpn13–UCH37 complex has opposite effects by increasing iNOS expression, via degradation of IκB-α, and on promoting the degradation of iNOS by the proteasome.

Fig. 5.

UCH37 is involved in iNOS degradation. (A) HEK293 cells, stably expressing Rpn13 shRNA or control shRNA, were transfected for 18 h with iNOS plasmid. Cell lysates were immunoprecipitated (IP) with iNOS antibody and then immunoblotted with antibodies against ubiquitin (UB; Top), iNOS (Middle), or Rpn13 (Bottom). (B) HEK293 cells stably expressing iNOS were transfected with UCH37 siRNA or control siRNA for 72 h. Cell lysates were evaluated by RT-PCR using specific primers for UCH37 or GAPDH (Upper) and by Western blot analysis using antibodies against iNOS or β-actin (Lower). (C) HEK293 cells stably expressing iNOS were transfected with UCH37 siRNA for 72 h and then treated with the translation inhibitor emetine (10 μM). Cells were harvested at intervals and lysates were analyzed by Western blotting using iNOS antibody (Upper). The reduction of iNOS signal by Western blot analysis was quantitated (Lower). Data represent mean ± SD, n = 3. *P < 0.05 and **P < 0.01.

Rpn13 Is Required for iNOS/UCH37 Interaction.

Rpn13 was initially isolated as an iNOS-interacting protein of unknown function (12). Our data above suggested that both Rpn13 and UCH37 participate in iNOS degradation. Because Rpn13 has been shown to recruit UCH37 to the proteasome, we reasoned that the role of Rpn13 in iNOS degradation is to facilitate interaction between iNOS and UCH37 at the proteasome interface. To confirm this hypothesis, we tested for interaction between iNOS and UCH37. In HEK293 cells, cotransfected with iNOS and flag-tagged UCH37, immunoprecipitation of iNOS resulted in coprecipitation of UCH37 (Fig. 6 A and B). These data implied an interaction between iNOS and UCH37. To verify that Rpn13 was required for the interaction between iNOS and UCH37, we conducted similar experiments in HEK293 stably transfected with plasmid expressing Rpn13 shRNA or control shRNA. Immunoprecipitation of iNOS resulted in coprecipitation of UCH37 in cells transduced with control shRNA but not in cells with knockdown of Rpn13 (Fig. 6 C and D). These results indicate that Rpn13 is required for iNOS to interact with UCH37 and, thus, provide a mechanism for Rpn13’s role in iNOS degradation.

Fig. 6.

Rpn13 facilitates iNOS interaction with UCH37. HEK293 cells (A and B) or HEK293 cells stably transfected with plasmid vectors expressing Rpn13 shRNA or control shRNA (C and D) were cotransfected for 18 h with iNOS and with UCH37-flag or control vector. Cell lysates (input) were analyzed by Western blot (A and C) and by IP (B and D). IP was done by using iNOS antibody and immunoblotted with flag antibody or iNOS antibody. To verify the stringency of IP, parallel experiments were done in which IP antibody was replaced with isotype control antibody (B) or cells transfected with empty vector were used (D).

Our study provides unique insight into the recently noted association of Rpn13 with the proteasome. This report demonstrates specific substrates for the Rpn13/UCH37 complex. Thus, the study supports the notion that whereas Rpn13 and UCH37 are proteasome-associated proteins, they participate in degradation of specific substrates. We identify two such substrates as IκB-α and iNOS. Degradation of IκB-α is the trigging event in activation of NF-κB, a pleiotropic transcription factor in inflammation and immunity. These findings place a unique and important function for Rpn13, namely regulation of NF-κB activation by the proteasome. The other substrate for Rpn13/UCH37 is iNOS, which is also an important protein for inflammation and immunity and a downstream target for activation of NF-κB. Rpn13 interacts with iNOS and recruits UCH37 to iNOS and, thus, the Rpn13/UCH37 complex is important for iNOS degradation. Because Rpn13 is also required for NF-κB activation, a major transcription factor for iNOS, the net effect of Rpn13 on iNOS steady state will be determined by the relative contribution of NF-κB to iNOS transcription in a given physiological system. In the studies described above, using IL-1β or LPS to induce iNOS via the NF-κB transcription factor, Rpn13 knockdown almost prevented iNOS induction. It should be noted that these two substrates were discovered because of their pathway association. There are probably other substrates yet to be identified, which require the Rpn13/UCH37 complex for their proteasomal degradation.

Materials and Methods

Cell Culture.

HEK293 cells were cultured in improved Modified Eagle Medium (Mediatech). RAW264.7 cells were maintained in DMEM. A549 cells were cultured in F12-K medium (Mediatech). Each medium was supplemented with 2 mM glutamine, 1× antibiotic-antimycotic mixture, and 10% heat-inactivated FBS (Invitrogen). Primary NHBE were cultured in bronchial epithelial cell growth medium (BEGM; Lonza) containing 130 ng/mL bovine pituitary extract, 5 × 10⁻⁵ mM retinoic acid, 1.5 μg/mL BSA, 20 IU/mL Nystatin, 0.5 mg/mL hydrocortisone, 25 ng/mL hEGF, 0.5 μg/mL epinephrine, 10 μg/mL transferrin, 5 μg/mL insulin, 6.5 ng/mL triiodothyronine, and 50 μg/mL gentamycin. Second passage cells were cultured at the air-liquid interface onto semipermeable membrane inserts (Transwell-clear; Corning) in serum-free, 1:1 mixture of DMEM (Invitrogen) and BEGM supplemented as above except that 0.5 ng/mL hEGF was used. The cultures were grown submerged until cells reached 70% confluence. Thereafter, cell culture media were changed daily by replacing fresh media to the basal compartment only. Cultures were maintained at 37 °C in humidified 5% CO₂ for an additional 2 wk after confluence until they reached fully differentiated mucociliary plateau phase. For iNOS induction, RAW264.7 cells were stimulated with 100 ng of LPS for 8 h. A549 cells and differentiated NHBE cells were stimulated for 24 h by a mixture of IL-1β (0.5 ng/mL), and TNF-α (10 ng/mL). All cells were maintained at 37 °C in 5% CO₂.

Measurement of NO.

Accumulation of NO in cell culture media was evaluated by measuring nitrite accumulation in cell culture media, as described (21–24). Briefly, 100 μL of culture media was mixed with 100 μL of Griess reagent for 10 min at room temperature, and absorbance at 543 nm was recorded in a microplate reader. Serial dilutions of sodium nitrite were used as standards.

Reagents and Antibodies.

Mouse monoclonal iNOS antibody 1E8-B8 (used for Western blots) was from Research and Diagnostic Antibodies. Rabbit polyclonal iNOS antibody 06–573 (used for IP) was from Millipore-Upstate. LPS of Escherichia coli serotype 0111:B4 was from Sigma. Polyclonal IκB-α antibody was purchased from Cell Signaling and polyclonal Rpn13 antibody was from Enzo Life Sciences. Mouse monoclonal Flag M2 antibody was from Sigma. Monoclonal β-actin antibody was from Ambion.

siRNA/shRNA Knockdown of Rpn13 or UCH37.

Rpn13 knockdown in RAW 264.7 cells.

Lentiviral vectors expressing shRNA against mouse Rpn13 and encoding AACGGAAAGGTCTCGTGTA were used (Figs. 1 and 2). Additional experiments were done by using siRNA oligos against mouse Rpn13 and encoding CGGACGACUCCCUUAUUCAtt (Fig. 1).

Rpn13 knockdown in human cells.

Plasmid vectors expressing shRNA against Rpn13 and encoding CCAGTGTGCTGACGCCGGAGATAATGGCT were used (Fig. 4, Fig. 5A, and Fig. 6 C and D). Additional experiments were done by using siRNA oligos against Rpn13 and encoding GGGCUGGUGUACAUUCAGCtt (Fig. 1, Fig. 2 B and C, and Fig. S2).

UCH37 knock down in RAW264.7 cells.

Mouse UCH37 siRNA oligos encoding GGAUUCAAAAGUAUAGUGAtt were used (Fig. 3A). Additional experiments were done by using UCH37 siRNA encoding GCAGGUAAUUAAUAAUGCUtt (Fig. S1A).

UCH37 knockdown in human cells.

Human UCH37 siRNA encoding GCCAGUUCAUGGGUUAAUUtt was used. Additional experiments were done by using human UCH37 siRNA encoding GGCCUGUCAUAGAAAAAAGtt (Fig. S1B).

Plasmids and Cell Transfection.

Vectors encoding full-length iNOS cDNA were described (27). Full-length human UCH37 was cloned by RT-PCR from HEK293 cells and fused with a flag-tag in pcDNA3.1 expression vector (Invitrogen), Vector-encoding full-length Rpn13 cDNA and vector-based shRNA were from Origene or Dharmacon. NF-κB cis reporting system (luciferase reporter plasmid) and luciferase assay kit were from Stratagene. Cationic lipid-mediated transfection of plasmids was done in HEK293 cells by using Lipofectamine 2000 (Invitrogen). RAW264.7 cells were transfected by electroporation using program D-32 of Amaxa Nucleofector.

Cell Lysis.

Cells were rinsed twice with PBS and lysed on ice for 30 min in the presence of a protease inhibitor mixture (BD Biosciences Pharmingen) in high stringency buffer A (40 mM Bis·Tris propane buffer at pH 7.7, 150 mM NaCl, 10% glycerol, 1% Triton X-100). Lysates were centrifuged (16,000 × g, 15 min, 4 °C), and supernatants were used for further analysis. Total protein concentrations were determined by using a bicinchoninic acid reagent, following the manufacturer's instructions (Pierce).

Immunoprecipitation.

Cells were lysed in lysis buffer A and cell lysates containing 0.5 mg of protein were incubated at 4 °C with rabbit polyclonal anti-iNOS antibody in a total volume of 1 mL of lysis buffer for 3 h before protein A-Sepharose beads (40 μl of 10% solution) were added to the samples. After further incubation for 60 min at 4 °C, beads were washed three times in ice-cold lysis buffer. Immunoprecipitated proteins were eluted by heating at 95 °C for 5 min in 2-fold concentrated Laemmli sample buffer (50 mM Tris·HCl at pH 6.8, 2% SDS, 0.001% bromophenol blue, 10% glycerol, 100 mM DTT). To investigate iNOS–UCH37 interaction, cells were lysed in low stringency buffer B (20 mM Tris·HCl at pH 8.0, 100 mM NaCl, 1 mM EDTA, and 0.5% Nonidet P-40) and cell lysates (1 mg of protein) was precleared by incubation at 4 °C with 20 μL of 10% protein A-Sepharose beads in a total volume of 1 mL. Protein A Sepharose beads were discarded after 30 min of incubation, and correspondent antibodies were added. After 2 h of incubation, 50 μL of 10% protein A Sepharose beads were added to the samples. Samples were further incubated for 1 h at 4 °C, and beads were washed three times in ice-cold lysis buffer. Immunoprecipitated proteins were eluted as above.

Western Blot Analysis.

Cell lysate (50 μg of protein) or immunoprecipitated proteins were heated at 95 °C for 5 min in Laemmli sample buffer and resolved on SDS/PAGE (Invitrogen). Proteins were transferred to nitrocellulose membranes by using semidry transfer cell (Bio-Rad). Membranes were blocked with blocking buffer (Li-Cor) or 5% nonfat milk for 1 h, followed by 2 h of incubation of primary antibody and 1 h of incubation of secondary antibody. Immunoreactive bands were detected by infrared imaging system (Odyssey) or visualized with enhanced chemiluminescence system (SuperSignal West Pico; Pierce), and images were acquired with a cooled charge-coupled device camera (Eagle Eye II Still Video System; Stratagene).

iNOS Degradation.

iNOS degradation was measured by pulse–chase analysis. Cells were pulse-labeled with 0.25 mCi/mL ³⁵S-labeled methionine/cysteine mixture for 1 h. Incorporation of radioactive methionine/cysteine was terminated by incubating cells in DMEM with excess unlabeled methionine and cysteine (300 mg/L of each) before harvesting at intervals. Cell lysates were immunoprecipitated by using iNOS antibody and protein A-Sepharose. Eluted proteins were separated by SDS/PAGE. Gels were dried for 3 h. Gel bands, representing³⁵S-labeled iNOS, were quantitated by using a phosphoimager. In some experiments, cells were instead incubated in medium with 10 μM translation inhibitor emetine. The decay of iNOS was then estimated by measuring the rate of decrease of the iNOS signals on immunoblots of cells samples, taken at intervals.

Degradation of Short-Lived Proteins.

Degradation of short-lived proteins was detected by pulse–chase analysis as described (25) with a few modifications. Cells were pulse-labeled with 25 μCi/mL ³⁵S-labeled methionine/cysteine mixture for 1 h as described above. Five hundred microliters of medium were sampled at intervals and mixed with trichloroacetic acid (TCA) to 10%. After centrifugation, the radioactivity in supernatants was determined by scintillation counting. At the end of the chase, the cells were lysed and radioactive proteins were precipitated with 10% TCA for measurement of total radioactivity inside the cells. The percentage of protein degradation was calculated from the amount of radioactivity released in the media and the total radioactivity present inside the cells.

Real-Time PCR Analysis of Gene Expression.

Total RNAs were purified by using a RNEasy Mini Kit (Qiagen), and cDNA synthesis was performed by using cDNA Reverse Transcription Kit (Applied Biosystems). iNOS mRNA expression was measured by using a real-time PCR (Applied Biosystems; StepOnePlus Real-Time PCR Systems) in 96-well optical plates by using SYBR GREEN Universal PCR Master Mix. Real-time PCR primers are listed in Table S1.

Luciferase Reporter Assay.

To evaluate the NF-κB activity, cells were transfected with a luciferase reporter plasmid, controlled by a synthetic promoter that contains the binding sites for NF-κB. After 24 h, cells were stimulated with IL-1β (HEK cells) or LPS (RAW264.7 cells) for 4 h. The luciferase activity was measured by using a luciferase assay kit (Stratagene) using a luminometer.

Statistical Analysis.

Values are means ± SD of n independent experiments. Student's t test was used to evaluate differences for significance, and P values of <0.05 were considered to be statistically significant.