Copper is a potent inhibitor of both the canonical and non-canonical NFκB pathways

Abstract

Copper is an essential trace element that plays key roles in many metabolic processes. Homeostatic regulation of intracellular copper is normally tightly controlled, but deregulated copper levels are found in numerous metabolic and neurodegenerative diseases, as well as in a range of neoplasms. There are conflicting reports regarding the exact role of copper in the regulation of NFκB-responsive genes, specifically whether copper leads to increased activation of the NFκB pathways, or downregulation. Here we show that increased intracellular levels of copper, using the ionophore clioquinol, leads to a potent inhibition of NFκB pathways, induced by multiple distinct stimuli. Addition of copper to cells inhibits ubiquitin-mediated degradation of IκBα by preventing its phoshorylation by the upstream IKK complex. Intriguingly, copper-dependent inhibition of NFκB can be reversed by the addition of the reducing agent, N-acetylcysteine (NAC). These results suggest that the oxidative properties of excess copper prevent NFκB activation by blocking IκBα destruction, and that NFκB activity should be assessed in diseases associated with copper excess.

Introduction

Copper is an essential trace metal that is a required cofactor for a variety of enzymes involved in many critical cellular processes.^1,2 Despite being essential for many metabolic functions, the ability of ionic copper to easily exchange electrons makes free copper highly toxic within the cell. Consequently, free copper in the cell is almost undetectable, and complex mechanisms of transporters, chaperones, and chelators have evolved to regulate intracellular copper distribution and excretion.³

A link between copper metabolism and malignancies has been established with the observation that elevated copper is observed in biopsy samples from various types of cancer, including breast, cervical, ovarian, lung, prostate, stomach, colon, brain, and leukemia.^4,5 As such, efforts have focused on reducing the levels of copper-dependent, malignant cells as a strategy to specifically target cancer cells and spare the surrounding normal tissue. Several studies have focused on decreasing intracellular copper levels to confer a therapeutic benefit, by modulating the activity of the nuclear factor-κB (NFκB) family of transcription factors.^4,6,7

NFκB is the collective term for a family of transcription factors that can regulate the expression of a variety of genes involved in immunity, inflammation, and cancer.^8,9 The mammalian NFκB family includes 5 members: RelA/p65, RelB, c-Rel, NFκB1 (p105/p50), and NFκB2 (p100/p52) that form homo- or heterodimers to transactivate gene expression.¹⁰ In the absence of an activating stimulus, the NFκB subunits are sequestered in the cytoplasm via their association with the inhibitor of κB (IκB) proteins.^10,11 When a cell receives any of a multitude of activating stimuli, the IκB proteins are rapidly phosphorylated, ubiquitinated, and subsequently degraded by the proteasome, allowing NFκB dimers to translocate to the nucleus and modulate target gene expression.^10,11 Due to the association between NFκB and tumorogenesis, many efforts are ongoing to identify how these important signaling pathways can be targeted in malignancies.⁹

In earlier studies NFκB activity has been measured under conditions of elevated or decreased copper, with differing results being reported. The copper chelator, tetrathiomolybdate, has been reported to limit angiogenesis, and therefore tumor growth, by specifically reducing NFκB activity in multiple experimental models.^4-7 However, the effects on increasing copper levels are less clear. Disulfiram, a drug known to increase intracellular copper levels, has been shown to limit NFκB-dependent signaling by interfering with the destruction of the IκB proteins.^12-14 In contrast, other studies have shown that increasing copper levels by direct addition to culture media or using pyrolidine dithiocarbamate exert a positive effect on NFκB signaling.¹⁵

Due to the cell’s inherent ability to regulate and buffer intracellular copper, it has been difficult to functionally examine the consequences of copper overload in the regulation of the NFκB signaling pathway. Here we have utilized the copper ionophore clioquinol to overcome the natural copper buffering, and to efficiently transport copper into cultured cells to assess NFκB activation. Our studies have revealed that increased copper levels severely inhibit cytokine- and genotoxic stress-induced NFκB. Elevated intracellular copper blocks NFκB gene transcription by increasing the production of intracellular reactive oxygen species that results in the inhibition of IKK-dependent phosphorylation of IκBα. These findings provide some clarity to the role of copper in regulating NFκB, and suggests the rationale for the use of copper chelators in a clinical setting may require a reassessment.

Results and Discussion

Copper potently inhibits canonical and non-canonical NFκB activation in response to multiple stimuli

To study the effects of elevating intracellular copper levels on NFκB activity, we used the copper ionophore clioquinol to aid the normally inefficient uptake of copper ions from the culture medium into cells.¹⁶ We pretreated HEK293 cells with copper sulfate (CuSO₄) either alone or in combination with clioquinol. Nuclear extracts were isolated from these cells following a 2 h treatment with the pro-inflammatory cytokine TNF or the chemotherapeutic drug etoposide, both known activators of canonical NFκB.¹⁰ NFκB activation status was then assessed by electrophoretic mobility shift assay (EMSA) using a radiolabeled probe designed to preferentially bind canonical NFκB subunits. As expected, both etoposide- and TNF-induced a robust NFκB binding activity (Fig. 1A), which could be super shifted by an antibody against RelA, indicating that this was indeed canonical NFκB (Fig. 1B). Supplementing the culture media with copper or clioquinol alone prior to NFκB stimulation resulted in no significant alteration in the induced NFκB (Fig. 1A). In contrast, NFκB activation was greatly diminished in cells pretreated with copper in combination with clioquinol (Fig. 1A). To explore the dose of copper required to inhibit NFκB induction, we titrated increasing levels of copper while using a constant dose of clioquinol, followed by stimulation with TNF. NFκB activity, as measured by EMSA, was reduced in a dose-dependent manner in response to increasing copper levels (Fig. 1C). Maximal reduction of NFκB activity was seen when copper levels exceeded 10 μM, a concentration that exceeds physiological copper concentrations.¹⁷

Figure 1. Copper potently inhibits NFκB activation in response to multiple stimuli. (A)HEK293 cells were treated with CuSO₄ (40 μM) and clioquinol (5 μM) for 2 h, then treated with etoposide (10 uM) or TNF (1000 units/mL) for 2 additional hours. Nuclear extracts were prepared and analyzed using EMSA with NFκB probe and octamer. (B)HEK293 cells were treated with etoposide (10 uM) or TNF (1000 units/mL) for 2 h. Nuclear extracts were incubated with the indicated antibodies and subsequently analyzed using EMSA with an NFκB probe. (C)HEK293 cells were treated with increasing concentrations of CuSO₄ (in μM) and clioquinol (5 μM) as indicated, with or without TNF (1000 units/mL). Nuclear extracts were prepared and analyzed using EMSA with NFκB probe.

The impact of elevated copper levels on canonical NFκB-dependent gene induction patterns was then examined. mRNA was isolated from HEK293 cells, pretreated with clioquinol and increasing concentrations of CuSO₄ that were subsequently stimulated with TNF. In agreement with the defects in NFκB DNA binding activity, cells treated with the combination of copper and clioquinol showed decreased expression levels of the classical NFκB target genes c-IAP2 and IκBα in a copper-dependent manner (Fig. 2A). Mouse embryonic fibroblasts (MEF) and the anaplastic large cell lymphoma (ALCL) cell line Karpas 299¹⁸ pretreated with copper and clioquinol also had defective TNF-induced NFκB-dependent gene expression, indicating that this phenomenon is seen in different cell lines from different species (Fig. 2B and C). Collectively, these data indicate that increasing intracellular copper suppresses the canonical NFκB activation pathway.

Figure 2. NFκB-dependent gene expression is impaired by copper in multiple cell lines. (A)HEK293 cells were treated with CuSO₄ as above (in μM) and clioquinol (5 μM) for 2 h, and then treated with TNF for 2 additional hours. RNA was isolated from samples and assessed using qRT-PCR with probes for the NFκB target genes c-IAP2 and IκBα, with fold change graphed as above. (B)MEFs and (C)Karpas 299 cells were treated as in (A). (D)Karpas 299 cells were treated with CuSO4 (40 μM) and clioquinol (5 μM) for 2 h, as indicated, then exposed either to CHO cells or CHO cells expressing CD30L as described in “Materials and Methods”. Whole-cell lysates were prepared and extracts were analyzed by western blot using specific antibodies against p100/p52 or β-actin.

Having observed such a profound copper-dependent inhibition of canonical NFκB signaling, we next sought to examine if copper altered non-canonical NFκB induction. The effect of copper on non-canonical NFκB activation was evaluated by examining the processing of the p100 protein to its transcriptionally active p52 form. Karpas 299 cells were pretreated with copper with or without clioquinol and then exposed to CD30L, a potent inducer of canonical and non-canonical NFκB.¹⁹ Stimulation of the CD30 receptor by CD30L resulted in a significant increase in p100 processing in both untreated cells and cells pre-incubated with copper or clioquinol alone. However, CD30-induced p100 processing was reduced in cells pretreated with the combination of copper and clioquinol (Fig. 2D). Taken together, these results demonstrate that the ionophore clioquinol, when incubated with pathologically relevant levels of copper, suppresses both the canonical and non-canonical NFκB activation pathways.

Copper blocks NFκB nuclear translocation

Data from numerous in vitro and cell culture studies are largely supportive of copper’s capacity to initiate oxidative damage. The DNA binding capabilities of the NFκB subunit p50 can be severely inhibited by oxidative modification of a conserved cysteine residue in its DNA-binding domain.²⁰ Given the dramatic impairment of TNF-induced NFκB activity observed in Figure 1A, we decided to test the possibility that addition of copper and clioquinol might simply be blocking the DNA binding activity of the canonical RelA/p50 heterodimer in the assay. We added clioquinol and increasing amounts of copper directly to nuclear extracts collected from cells that were untreated or treated with TNF. Even at the highest concentration of copper (40 μM) in combination with clioquinol, no significant reduction of NFκB DNA binding activity was observed (Fig. 3A). Therefore, copper and clioquinol are not acting to directly inhibit NFκB heterodimers interacting with DNA.

Figure 3. Copper blocks NFκB nuclear translocation. (A)HEK293 cells were treated with or without TNF (1000 units/mL), then nuclear extracts were prepared and CuSO₄ (in μM) and clioquinol (5 μM) were added to the extracts as indicated, and analyzed using EMSA with NFκB probe. (B and C) HEK293 cells were treated with CuSO₄ (40 μM) and clioquinol (5 μM) for 2 h, then stimulated with TNF for 30 min. Nuclear and cytoplasmic extracts were subjected to immunoblotting analysis for the levels of the proteins as indicated.

In its inactive form, NFκB is sequestered in the cytoplasm, bound by members of the IκB family of proteins.²¹ NFκB-inducing stimuli result in the release of this inhibition and subsequent nuclear translocation of NFκB. Having observed that copper does not alter NFκB DNA binding, we examined if elevated copper levels alter the concentration of nuclear NFκB following TNF treatment. Nuclear and cytoplasmic extracts were prepared from TNF-treated HEK 293 cells pretreated with copper with or without clioqinol. Western blot analysis of these extracts revealed that in both control cells and cells treated with copper or clioquinol alone, a robust translocation of RelA and p50 from the cytoplasm to the nuclear compartment occurred within 30 min of TNF stimulation. However, in cells pretreated with the combination of copper and clioquinol, the level of TNF-induced nuclear RelA and p50 was significantly decreased (Fig. 3B). As expected, immunoblot analysis of cytoplasmic extracts from cells treated with TNF revealed a significant reduction in the levels of RelA, p50, and IκBα (Fig. 3C). Cells with high intracellular copper failed to degrade IκBα in response to TNF, consistent with the lack of NFκB activation, indicating a defect in the upstream signaling pathway (Fig. 3C).

Copper prevents IκBα phosphorylation and degradation

In addition to acting as a metal ionophore, clioquinol is capable of directly inhibiting the proteasome at higher concentrations.^22-29 It has long been established that inhibition of the proteasome prevents activation of NFκB by stabilizing the phosphorylated form of IκBα that is still bound to the NFκB heterodimer, retaining it in the cytoplasm.³⁰ Western blot analysis of lysates prepared from cells treated with the well-described proteasome inhibitor lactacystine³¹ revealed a prolonged IκBα phosphorylation following TNF induction, consistent with impaired IκBα degradation (Fig. 4A). To test the possibility that the combination of copper and clioquinol might be preventing NFκB induction by preventing IκBα degradation by the proteasome inhibition, we prepared whole-cell lysates from HEK293 cells that had been pretreated with copper, clioquinol, or a combination of both, then assessed TNF-dependent IκBα degradation (Fig. 4B–D). The samples treated with clioquinol or copper alone showed a normal pattern of phosphorylation and degradation of IκBα, with phosphorylated protein appearing at 5 and 10 min and then being degraded as expected (Fig. 4B and C). IκBα phosphorylation and degradation was impaired in cells pretreated with both copper and clioquinol, suggesting that increased intracellular copper impairs IκBα degradation upstream of the proteasome. Although our results indicate that the NFκB inhibitory properties are upstream of IκBα phosphorylation and subsequent degradation, we sought to determine if the activity of the proteasome was altered in copper/clioquinol-treated cells. Proteasome activity was detected with the fluorescence substrate Suc-LLVY-AFC. Lactacystin greatly reduced the proteasome-dependent cleavage of the fluorescent substrate, while the addition of copper and clioquinol, either alone or in combination, had no significant effect on proteasome activity (Fig. 4E). Therefore copper and clioquinol do not reduce NFκB activity by simply preventing proteasomal destruction of the ubiquitinated IκBα protein.

Figure 4. Copper prevents IκBα phosphorylation and degradation. (A)HEK293 cells were treated with the selective proteasome inhibitor lactacystin for 30 min, then treated with TNF (1000 units/mL) for the indicated times. Whole-cell lysates were prepared and analyzed by western blot using specific antibodies against phospo-IκBα and β-actin. (B)HEK293 cells were treated with CuSO₄ (40 μM) for 2 h, then treated with TNF (1000 units/mL) for the indicated times. Whole-cell lysates were prepared and analyzed by western blot using specific antibodies against phospo-IκBα, total IκBα, and β-actin. (C)HEK293 cells were treated with clioquinol (5 μM) for 2 h, then treated with TNF (1000 units/mL) for the indicated times. Whole-cell lysates were prepared and analyzed by western blot using specific antibodies against phospo-IκBα, total IκBα, and β-actin. (D)HEK293 cells were treated with CuSO₄ (40 μM) and clioquinol (5 μM) for 2 h, then treated with TNF (1000 units/mL) for the indicated times. Whole-cell lysates were prepared and analyzed by western blot using specific antibodies against phospo-IκBα, total IκBα, and β-actin. (E)HEK293 cells were treated with CuSO₄ (40 μM), clioquinol (5 μM), vehicle control (DMSO), or combination of CuSO₄ and clioquinol for 3 h, then cell lysates were analyzed for proteasome activity using Suc-LLVY-AFC fluorogenic substrate. For control, HEK293 cells were also treated with the specific proteasome inhibitor lactacystin for 30 min and analyzed. (F) HEK293 cells were treated with copper and clioquinol, followed by treatment with TNF (1000 units/ml) for 10 min as indicated. The IKK precipitates were blotted using the indicated antibodies.

The IKK kinase complex is upstream kinase responsible for phosphorylating IκBα in response to NFκB-inducing agents. It is comprised of 2 kinases (IKKα and IKKβ) and a regulatory subunit, NEMO/IKKγ, which is activated by phosphorylation of key serine residues in the activation loops of the catalytic subunits. The IKK complex was precipitated from lysates treated with TNF in the absence or presence of clioquinol. Levels of phosphorylated IKK were markedly increased in cells exposed to TNF compared with untreated controls (Fig. 4F, lane 1 and 2). However, pre-treatment with copper and clioqinol resulted in a marked reduction of activated IKK in response to TNF (Fig. 4F, compare lanes 2 and 4). These data suggest that elevated copper in cells reduces NFκB activity by preventing the activation of the upstream kinase complex.

Inhibition of NFκB by copper is independent of copper’s effect on XIAP

We and others have previously shown that the X-linked inhibitor of apoptosis (XIAP) protein binds copper directly and undergoes a substantial conformational change in the copper-bound state, resulting in decreased protein stability.^32,33 This form of XIAP has reduced ability to interact with caspases and increased rates of cell death in response to apoptotic stimuli.³³ Since XIAP has been suggested to be a modulator of NFκB activity,^34,35 we hypothesized that changes in XIAP may account for the copper-dependent defects in NFκB activation. Addition of copper in combination with the ionophore allowed for a much more rapid change in the mobility of XIAP than with copper alone in both HEK293 cells and mouse embryonic fibroblasts (MEFs) (Fig. 5A and B).

Figure 5. Inhibition of NF κB by copper is independent of copper’s effect on XIAP. (A)HEK293 cells were treated with CuSO₄ (30 μM) and clioquinol (5 μM) as indicated and whole-cell lysates were prepared after the indicated time points. Extracts were analyzed by western blot using specific antibodies against XIAP or β-actin. (B)Primary MEF cells were treated with CuSO4 at indicated concentrations (in μM) and clioquinol (5 μM) for 3 h, then whole-cell lysates were prepared and analyzed by western blot using specific antibodies against XIAP or β-actin. (C and D) Primary wild-type MEFs (C) and littermate XIAP KO MEFs (D) were treated with CuSO₄ as above (in μM) and clioquinol (5 μM) for 2 h, then treated with TNF (1000 units/mL) for 1 additional hour. RNA was isolated from samples and assessed using qRT-PCR with probes for the NFκB target genes c-Iap2, and IκBα, with fold change graphed as above.

To determine definitively if XIAP was required to prevent NFκB induction, MEFs prepared from XIAP-null mice and wild-type controls were treated with CuSO₄ with clioquinol or vehicle, and qRT-PCR was used to assess transcription of NFκB-responsive genes (Fig. 5C and D). NFκB-dependent gene expression was similarly suppressed by treatment with copper and cliquinol in the XIAP-null and wild-type samples, despite the known changes occurring in XIAP caused by excess levels of intracellular copper.

Inhibition of NFκB by copper can be reversed by the ROS inhibitor N-acetyl cysteine

Cell culture models have shown that copper is a potent oxidant that generates ROS within the cells, triggering oxidative damage and disruption of key cellular signaling pathways. Additionally, copper-induced oxidative damage has been implicated in disorders associated with deregulated copper metabolism, including the autosomal recessive genetic disorder Wilson disease, and in various forms of cancers such as lymphoma, breast cancer, and gastrointestinal tract cancer.^36,37 We hypothesized that the effects of excessive copper on NFκB signaling may be due to increased oxidative stress overwhelming the cell’s intrinsic ability to regulate copper levels. To counteract the oxidation induced by copper overload, we utilized the antioxidant N-acetylcysteine (NAC). As expected, addition of copper and clioquinol caused a severe reduction in the NFκB binding activity as compared with the control. Addition of the antioxidant completely reversed the copper-dependent suppression of NFκB activity (Fig. 6A). Addition of NAC reversed the copper-induced suppression of the NFκB target genes, IκBα and c-IAP2, as measured by qRT-PCR (data not shown).

Figure 6. Inhibition of NFκB by copper can be reversed by the ROS inhibitor NAC. (A)HEK293 cells were treated with CuSO₄ (40 μM), clioquinol (5 μM), NAC (10 mM), or combination as indicated, for 2 h, followed by treatment with TNF (1000 units/ml) for one hour as indicated. Nuclear extracts were prepared and analyzed using EMSA with NFκB probe. (B)HEK293 cells were treated as in (A), followed by treatment with TNF (1000 units/ml) for 10 min as indicated. The IKK precipitates were blotted using the indicated antibodies.

To further define the point in the NFκB signaling cascade that NAC counteracts the effect of excess copper, the activity of IKK was assessed. The IKK complex was precipitated from lysates treated with TNF in combination with NAC and excess copper. Pre-treatment with copper and clioqinol resulted in a marked reduction of activated IKK in response to TNF as previously shown (Figs. 4F and 6B). Addition of NAC restored phosphorylation of the IKK complex, suggesting that elevated intracellular copper reduces NFκB activity by preventing the activation of the upstream kinase complex (Fig. 6B). Together these results suggest that the oxidative stress induced by excess copper is a potent suppressor of NFκB-dependent gene expression.

In summary our data suggest that elevated copper is a general inhibitor of NFκB activity induced by multiple stimuli, with both canonical and non-canonical NFκB blocked by excess copper levels. It has become clear that aberrant regulation of NFκB and the signaling pathways that control its activity are implicated in cancer progression, as well as in resistance to chemo- and radiotherapy. This is in part due to the wide range of NFκB target genes that are implicated in the prevention of programmed cell death.⁹ Copper chelators are currently in clinical trials as anti-cancer therapeutics, in part through their described ability to suppress NFκB.³⁸ Our data indicate that a possible alternative therapeutic strategy is to elevate copper to block this pro-survival pathway in tumor cells.

Materials and Methods

Cell lines and culture conditions

Human embryonic kidney (HEK) 293 and mouse embryonic fibroblasts (MEFs) were maintained in DMEM (Mediatech) supplemented with 10% fetal bovine serum and 2 mM L-glutamine at 37° in an atmosphere of 5% CO₂. Generation of Xiap-deficient MEFs, along with wild-type controls, has been previously described.³⁹ The anaplastic large cell lymphoma (ALCL) cell line Karpas 299 has been previously described¹⁸ and was grown in RPMI 1640 medium (Mediatech) supplemented with 10% fetal bovine serum and 2 mM L-glutamine at 37° in an atmosphere of 5% CO₂.

Materials

Reagents were obtained from the following sources: copper sulfate (CuSO₄) and 5-chloro-7-iodo-8-quinolinol (clioquinol; Sigma-Aldrich), N-acetyl-L-(+)cysteine (NAC; Fisher Scientific), lactacystin (Calbiochem), etoposide (Bedford Laboratories), human tumor necrosis factor (TNF; Roche), Suc-LLVY-AFC fluorogenic substrate (Sigma-Aldrich).

Physiologic CD30 stimulation

As described previously,¹⁹ Karpas 299 cells were exposed to either CHO cells (negative control) or CD30L⁺ cells which had been previously seeded in 6-well plates, and incubated for the specified time. Following CD30 stimulation, the Karpas 299 cells were removed from the CHO cells with gentle pipetting and collected and centrifuged at 100 × g for 5 min. Medium was aspirated, and cells were washed with PBS then centrifuged again at 100 × g for 5 min, and then processed as indicated in the figure legend.

Cell lysate preparation and immunoblot analysis

Cell lysates were prepared with RIPA lysis buffer (PBS containing 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS) supplemented with protease inhibitors, for 30 min on ice to ensure complete lysis. Protein quantification was determined by Bradford assay (Bio-Rad) or BCA. RIPA cell lysates of equal protein concentrations were prepared in LDS sample buffer, separated on denaturing NuPAGE 4–12% polyacrylamide gradient gels, and transferred to 0.45 μm nitrocellulose membranes (Invitrogen). Membranes were blocked in 5% milk in Tris-buffered saline (TBS) with 0.2% Tween 20 (Bio-Rad) or with Odyssey^® blocking buffer, depending on the antibody requirements, followed by incubation with the indicated antibodies for 1 h at room temperature or overnight at 4 °C. After washing with TBS containing 0.2% Tween, membranes were incubated with secondary antibodies for 1 h at room temperature. Enhanced chemiluminescence (ECL; Thermo Scientific) was used to visualize the blots on Kodak XAR film or blots were scanned using Li-Cor Odyssey infrared scanner according to the manufacturer’s instructions. The following primary antibodies were used for western blot analysis: anti-XIAP (Enzo Life Sciences), anti-p100/p52 (Millipore), anti-IκBα (Upstate), anti-phospho-IκBα (Cell Signaling), anti-phospho-IKKα/β (Cell Signaling), anti-IKKβ (Cell Signaling), and anti-β-actin (Sigma-Aldrich).

Subcellular fractionation for immunoblotting

HEK293 cells were stimulated as described, then removed from plate with gentle washing and rinsed with ice-cold PBS. The resulting pellet was resuspended in 1 mL of hypertonic buffer [20 mM Hepes (pH 7.4), 10 mM KCl, 1 mM DTT, 0.1% Triton X-100, 20% glycerol, 2 mM PMSF] supplemented with protease inhibitors, and the cell suspension was subjected to 10 slow strokes in a Dounce homogenizer. After centrifugation at 800 × g for 5 min at 4 °C, the supernatant was collected as the cytoplasmic fraction. The pellet was washed once with 1 ml of hypertonic buffer, before being resuspended in 100 μL of cold nuclear extraction buffer [20 mM HEPES, pH 7.9, 0.42 M NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 0.1 mM PMSF, 0.5 mM DTT, and 25% glycerol] supplemented with protease inhibitors. The suspension was incubated on ice for 20 min. After centrifugation at 16 000 × g for 10 min at 4 °C, the supernatant was collected as the nuclear fraction.

IKK immunoprecipitations

Cells were washed in ice-cold PBS and scraped into IP buffer (20 mM Tris pH 8.0, 5 mM EDTA, 10 mM NaF, 150 mM NaCl, 1% NP-40, 0.5 mM PMSF, 1 mM sodium vanadate) supplemented with protease inhibitors. The suspension was incubated on ice for 20 min then clarified by centrifugation at 16 000 × g for 10 min at 4 °C. The IKK complex was precipitated from 500 µg of each lysate using anti-IKKγ IgG conjugated to agarose beads (Santa Cruz Biotechnologies). Supernatants were removed and beads washed 3 times with IP buffer before being resuspended in 1× LDS sample buffer and analyzed by western blot using the indicated antibodies.

Electrophoretic mobility shift assay

HEK293 cells were stimulated as described, then removed from plate with gentle washing and rinsed with PBS. Cells were washed once with buffer A (10 mM HEPES, pH 7.9, 1.5 mM MgCl₂, 10 mM KCl, 0.1 mM PMSF, and 0.5 mM DTT) and pelleted by centrifugation at 1500 × g for 2 min. The supernatant was aspirated, and the pellet was resuspended in 15 μL cold buffer A supplemented with 0.1% NP-40 and incubated on ice for 5 min. The nuclear pellet was isolated by centrifugation at 16 100 × g in a refrigerated microcentrifuge at 4 °C for 15 min. Pellet was resuspended in 10 μL buffer C (20 mM HEPES, pH 7.9, 0.42 M NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 0.1 mM PMSF, 0.6 mM DTT, and 25% glycerol), and incubated at 4 °C for 15 min. The nuclear extract was clarified by centrifugation at 16 100 × g for 15 min, and 10 μL was transferred to a fresh tube and diluted with 60 μL of modified buffer D (20 mM HEPES, pH 7.9, 50 mM KCl, 0.2 mM EDTA, 0.1 mM PMSF, 0.5 mM DTT, and 20% glycerol), flash frozen and stored at −80 °C.

Two complimentary oligonucleotides containing NFκB consensus binding sites, κBEMSAoligo1 (5′-GATCCAGGGA CTTTCCGCTG GGGACTTTCC A-3′) and κBEMSAoligo2 (5′-GATCTGGAAA GTCCCCAGCG GAAAGTCCCT G-3′), were annealed and radiolabeled using T4 polynucleotide kinase (New England BioLabs) in the presence of [α-³²P]dCTP. The radiolabeled probe was then purified using illustra Microspin G-25 Columns (GE Healthcare) according to manufacturer’s instructions. To test for the presence of NFκB in the nuclear extracts, prepared as described above, 2 μL of nuclear extracts were incubated for at room temperature with 1 μg of poly(dI-dC)•poly(dI-dC) in modified buffer D (minus glycerol) in a total volume of 20 μl. 0.1 μl of ³²P-radiolabeled probe was added, and the entire reaction was separated on a non-denaturing 4% polyacrylamide gel. Supershifts were performed by adding 1 μg of the indicated antibodies to the reaction mixture and incubating for 5 min at room temperature before loading onto the gel. Autoradiography was performed overnight at −20 °C.

Real-time reverse transcription-PCR

HEK293 cells or MEF cells were treated as described above. The HEK293 cells were rinsed with PBS and removed with gentle pipetting. The MEF cells were rinsed with PBS and removed with trypsin, which was then inactivated with DMEM containing FBS, which was then removed with centrifugation and rinsed with PBS. RNA isolation was performed using the RNeasy minikit (Qiagen) according to the manufacturer’s instructions. One hundred ng of total RNA was subjected to a reverse transcription reaction using random hexamer primers and MTLV reverse transcriptase (Applied Biosystems). One microliter of the resulting cDNA was analyzed with the indicated target assay using the Applied Biosystems 7500 Real-Time PCR System. Each target assay was normalized to GAPDH levels.

Proteasome activity analysis

HEK293 cells were treated as described above and whole-cell lysates were prepared. Whole-cell lysates were incubated with Suc-LLVY-AFC in 96-well plates at 37 °C for 1 h. AFC release was then measured using a Cytofluor 4000 multiwell plate reader (Applied Biosystems) with an excitation wavelength of 400 nm at an emission wavelength of 505 nm.