Nuclear factor kappa B (NF-κB) transcription factors regulate several important physiological processes, including inflammation and immune responses, cell growth, apoptosis, and the expression of certain viral genes. Therefore, the NF-κB signaling pathway has also provided a focus for pharmacological intervention, primarily in situations of chronic inflammation or in cancer, where the pathway is often constitutively active and plays a key role in the disease. Now that many of the molecular details of the NF-κB pathway are known, it is clear that modulators of this pathway can act at several levels. As described herein, over 750 inhibitors of the NF-κB pathway have been identified, including a variety of natural and synthetic molecules. These compounds include antioxidants, peptides, small RNA/DNA, microbial and viral proteins, small molecules, and engineered dominant-negative or constitutively active polypeptides. Several of these molecules act as general inhibitors of NF-κB induction, whereas others inhibit specific pathways of induction. In addition, some compounds appear to target multiple steps in the NF-κB pathway. Compounds designed as specific NF-κB inhibitors are not yet in clinical use, but they are likely to be developed as treatments for certain cancers and neurodegenerative and inflammatory diseases. Moreover, the therapeutic and preventative effects of many natural products may, at least in part, be due to their ability to inhibit NF-κB.
The nuclear factor kappa B (NF-κB) transcription factors control the expression of genes involved in many critical physiological responses, including immune and acute phase inflammatory responses, cell adhesion, differentiation, oxidative stress responses and apoptosis (reviewed in Pahl, 1999). Vertebrate NF-κB transcription complexes can be any of a variety of homo- and heterodimers formed by the subunits p50, p52, c-Rel, RelA (p65) and RelB (see Gilmore, 2006). These complexes bind to DNA regulatory sites called κB sites, generally to activate specific target gene expression. The target gene specificity of NF-κB dimers is thought to arise from a number of considerations, including the specific NF-κB complexes that are in different cell types, the distinct κB target site binding specificities of different NF-κB complexes, and the different protein–protein interactions and posttranslational modifications that NF-κB complexes undergo in different contexts (Hoffmann et al., 2006; Perkins, 2006).
In most cell types, NF-κB dimers are located in the cytoplasm in an inactive form through association with any of several IκB inhibitor proteins (IκBα, -β, -ɛ, -γ, p105 and p100; see Gilmore, 2006, this issue). In response to a wide array of stimuli (Pahl, 1999; www.nf-kb.org), many of which are involved in intercellular communication such as proinflammatory molecules, IκB is rapidly phosphorylated, ubiquitinated and degraded by the proteasome. The freed NF-κB dimer then translocates to the nucleus where it can modulate specific gene expression.
The phosphorylation and degradation of IκB have received great attention as key steps for the regulation of NF-κB complexes (see Scheidereit, 2006). The IκB kinase (IKK) complex contains two kinase subunits, IKKα and IKKβ, and an associated scaffold-like regulatory protein called NEMO (aka IKKγ). Other proteins, including immunophilins, ELKS and heat-shock proteins, may also be present in the IKK complex under different conditions, but the roles of such accessory molecules are not clearly defined. IKKα and β share much sequence similarity and can form homo- and heterodimers. After stimulation of cells by agents such as tumor necrosis factor α (TNFα), interleukin-1 (IL-1) or various pathogens, the IKK complex is activated in part by phosphorylation of specific serine residues in the activation loop of each IKK subunit (see Scheidereit, 2006). The activated IKK complex can then phosphorylate IκB on two serine residues (Ser 32 and 36 in human IκBα). Phosphorylation of the IκB by IKK signals it for ubiquitination at specific lysine residues by the SCF-β-TrCP E3 ubiquitin ligase complex, which targets the IκB for degradation by the 26S proteasome. NF-κB is now free to enter the nucleus.
However, there are multiple pathways for activation of NF-κB. The two most common pathways are the canonical (or classical) and the non-canonical (or alternative) pathways (Gilmore, 2006; Scheidereit, 2006). In the canonical pathway, a complex such as p50-RelA/IκBα is activated by an IKK complex containing IKKα/IKKβ/NEMO, with IKKβ being the primary kinase for IκBα. In the canoncial pathway proceeding from TNF stimulation to NF-κB, the IKK complex is brought to the TNF-R by binding of NEMO to a K63-ubiquitinated RIP1 adaptor molecule on the TNF-R. The activity of the IKKβ kinase is then enhanced by two phosphorylations in its activation loop (Ser 177, 181), and the downstream events of NF-κB signaling ensue. In the non-canonical pathway, the latent cytoplasmic NF-κB complex is p100/RelB. Upon activation by certain receptor signals, an IKKα homodimer complexbecomes activated and phosphorylates p100 at two C-terminal serine residues. This then promotes the ubiquitination of p100 and the proteasomal processing of the complex to p52/RelB. There are also additional non-IKK-dependent NF-κB pathways: for example, ultraviolet irradiation-induced NF-κB activation does not appear to use the IKK complex, and phosphorylation of IκBα at tyrosine residue 42 can lead to activation of NF-κB (Imbert et al., 1996; Li and Karin, 1998).
Therefore, several steps leading to the activation of an NF-κB target gene are regulated by protein complexes that act as homo- and heterodimers and/or by families of closely related, interacting proteins. This adds additional levels of complexity to the regulation of NF-κB activity. For example, as described above, IKKα and IKKβ are active as dimers. However, they appear to have distinct physiological functions in that disruption of each IKK gene in mice leads to quite different phenotypes (Gerondakis et al., 2006). Furthermore, IKKα and IKKβ can be activated independently by different upstream signals. Moreover, the IKKs can phosphorylate substrates in addition to the IκBs (Scheidereit, 2006). Similarly, the different IκB proteins have different affinities for the diverse NF-κB dimers. Thus, different IκB proteins can indirectly regulate the expression of distinct sets of genes, which are transcriptional targets of specific NF-κB complexes (Hoffmann et al., 2006). Finally, NF-κB is a family of dimeric transcription factors, which have distinct sets of target genes. Therefore, the activities of all interacting proteins in the NF-κB signaling pathway are dependent on the precise nature and intensity of the upstream activating signals. These interacting levels must be considered when one wants to inhibit a given NF-κB response. Thus, there are broad-range inhibitors of NF-κB with a wide range of action, which act at an early step of NF-κB induction. In addition, there are more specific inhibitors, often acting at later stages in the NF-κB molecular signaling pathway (Figure 1).
Because of the multitude of cellular and organismal processes affected by NF-κB signaling, there has been great interest in modulators of this pathway. For example, NF-κB complexes are known to play key roles in the inflammatory response, in the inhibition of apoptosis, and in cell proliferation, and as a consequence, mis-regulated, usually sustained, NF-κB activity also contributes to human disease states, including most chronic inflammatory diseases and many cancers (Pahl, 1999; Basse(c)res and Baldwin, 2006; Dutta et al., 2006; Hayden et al., 2006). Therefore, the identification of specific and potent inhibitors of NF-κB has been the goal of many researchers and pharmaceutical companies. Due to the large number of inducers of NF-κB (Pahl, 1999; www.nf-kb.org) and the various levels of regulation of this pathway (Perkins, 2006; Scheidereit, 2006), inhibitors include a wide variety of molecules that act at any of several levels in the pathway.
The molecular cascade of signaling events provides several steps for specific inhibition of NF-κB activity. Generally speaking, inhibition of NF-κB activation can occur by three mechanisms: (1) blockage of the incoming stimulating signal at an early stage (e.g., binding of ligand to its receptor) resulting in complete abrogation of the signal's effect; (2) interference with a cytoplasmic step in the NF-κB activation pathway by blockage of a specific component of the cascade (e.g., the recruitment of an adaptor to the receptor complex, the activation of the IKK complex, or degradation of IκB); or (3) blockage of NF-κB nuclear activity, that is, inhibiting its translocation to the nucleus, its binding to DNA, a nuclear modification of NF-κB that affects its activity or specificity, or an interaction of NF-κB on DNA with specific or basal transcription machinery. In most cases, each of these three major steps is not susceptible to the same types of inhibitors. This review attempts to classify the multitude of NF-κB inhibitors that have been reported, and describes the mechanism of action of some inhibitors and implications of their use for human health.
A plethora of NF-κB inhibitors
The number of reported inhibitors of NF-κB signaling is staggering. Seven years ago, we categorized 125 NF-κB inhibitors (Epinat and Gilmore, 1999). The current list has grown to 785, which we have now classified according to both the nature of the inhibitor and the step at which NF-κB activation is blocked (see Table 1 for an overview, and Supplementary Tables 1–8 for details). These inhibitors include a variety of natural products, chemicals, metals, metabolites, synthetic compounds, peptides, proteins (cellular, viral, bacterial, fungal) and physical conditions. We have broadly divided these inhibitors into agents that act on NF-κB signaling at the following levels: upstream of IKK (e.g., at a receptor or adaptor level); directly at the IKK complex or IκB phosphorylation; ubiquitination or proteasomal degradation of IκB; nuclear translocation of NF-κB; NF-κB DNA binding; and NF-κB-directed gene transactivation. In most cases, inhibition is based on a given agent's ability to block one or more steps of NF-κB signaling in a tissue culture-based system after stimulation of resting cells with an NF-κB inducer, most commonly either TNF, IL-1 or lipopolysaccharide (LPS). However, it is important to note that the step at which a given compound has been reported to exert its block (e.g., at the level of DNA binding) often is only known within the limit of the assays performed, and thus does not necessarily suggest the molecular target of the inhibitor. That is, many researchers have used electrophoretic mobility shift assays (EMSAs) or κB-site reporter gene assays to measure induced NF-κB activity, without further characterizing upstream NF-κB signaling steps. In the text below, we primarily discuss compounds whose molecular mechanisms of action are known.
Inhibitors that act upstream of the IKK complex
In most situations, the IKK complex is the first common node in the integration of many NF-κB activating pathways. Therefore, one strategy for inhibiting activation of NF-κB is to block a signal before it activates IKK, that is, at a level upstream of IKK (Supplementary Table S1). Many cytokines (e.g. TNF) signal through distinct cell-surface receptors to activate transcription factor NF-κB. Thus, TNF-induced NF-κB activation can be inhibited by using anti-TNF antibodies or agents that block the TNF-R, and this type of upstream anti-NF-κB therapy can have benefits in diseases such as inflammatory bowel disease, arthritis and Crohn's disease (Song et al., 2002).
At the next step in cytokine signaling, members of the tumor necrosis factor receptor-associated factors (TRAF) protein family act as adaptor molecules for the activation of NF-κB by the TNF and IL-1 receptor superfamilies (Bradley and Pober, 2001). TRAF2 is recruited to the TNF-R1 and TNF-R2 receptors following TNF stimulation, and it is required for CD40- and TNFα-mediated activation of NF-κB. As such, a TRAF2 mutant lacking its N-terminal RING finger domain is a dominant-negative inhibitor of TNFα-(but not IL-1) induced NF-κB activation (Hsu et al., 1996). On the other hand, TRAF6 participates primarily in the IL-1 signal transduction pathway by interacting with the IL-1 receptor, and thus, a dominant-negative mutant of TRAF6 inhibits NF-κB activation signaled by IL-1 but not by TNFα (Cao et al., 1996).
Two kinases that can lead to activation of the IKK complex are NIK and MEKK1. Overexpression of NIK potently induces NF-κB (Malinin et al., 1997), primarily through the non-canoncial pathway (Scheidereit, 2006). In addition, some studies have demonstrated that MEKK1 is activated by TNFα and IL-1 and can potentiate the stimulatory effect of TNFα on IKK and NF-κB activation (Hirano et al., 1996; Lee et al., 1998).
NIK has also been reported to interact with several members of the NF-κB signaling pathway, including TRAF2, TRAF6 and IKK. More specifically, NIK can directly phosphorylate IKKα for activation of the non-canonical p52/RelB pathway (see Perkins, 2006; Scheidereit, 2006). As such, overexpression of kinase-deficient NIK mutants can block the ability of IL-1 and TNFα to induce NF-κB (Malinin et al., 1997; Song et al., 1997).
When overexpressed, MEKK1 stimulates the activities of both IKKα and IKKβ in transfected HeLa cells and directly phosphorylates the IKKs in vitro (Lee et al., 1998). Furthermore, a dominant-negative mutant of MEKK1 partially blocks activation of the IKK complex by TNFα (Hirano et al., 1996; Lee et al., 1997). However, the physiological relevance of activation of IKK and NF-κB by overexpressed MEKK1 has been questioned (Karin and Delhase, 1998).
Inhibitors of IKK complex activities
IKK has been a prime target for the development of NF-κB signaling inhibitors, in part due to its central role in funneling upstream signals into the NF-κB activation pathways and in part due to other successes in developing kinase inhibitors for therapeutic applications (e.g., imatinib/Gleevec; O'Hare et al., 2006). Indeed, there are over 150 agents that have been shown to inhibit activation of NF-κB at the IKK step based either on the lack of IκBα phosphorylation or on the lack of stimulus-induced IKK activity in immune complex kinase assays (Supplementary Table S2). Several studies have also used a constitutively active form of IKKβ (SS177,181E) to demonstrate that a given compound is a direct inhibitor of IKKβ. Nevertheless, many studies have not investigated the mechanism by which a given agent can inhibit IKK or its activation.
Where a mechanism of action is known, chemical IKKα/β inhibitors are of three general types: ATP analogs that show some specificity for interaction with these kinases, compounds with allosteric effects on IKK structure, and compounds that interact with a specific Cys residue (Cys-179) in the activation loop of IKKβ. ATP analogs include natural products such as β-carboline, and several synthetic compounds developed by pharmaceutical companies, such as SC-839, which has an approximately 200-fold preference for IKKβ as compared to IKKα (reviewed in Karin et al., 2004; Pande and Ramos, 2005). On the other hand, the synthetic compound BMS-345541 binds to an allosteric site on both kinases, but shows an approximately 10-fold greater inhibitory effect on IKKβ (Burke et al., 2003). Several thiol-reactive compounds, such as parthenolide, certain epoxyquinoids and arsenite, have been shown to block IKKβ activity through Cys-179 (Kapahi et al., 2000; Kwok et al., 2001; Liang et al., 2003, 2006), probably in most cases through a direct conjugation to the thiol group of this cysteine. Although not formally proven, it is likely that interaction of these compounds with Cys-179 interferes with phosphorylation-induced activation of IKKβ, in that Cys-179 is located between Ser 177 and Ser 181 which are part of the kinase activation loop and are required for activation of IKKβ in response to many upstream signals, such as TNF and LPS (see Perkins, 2006; Scheidereit, 2006). Of note, however, several such thiol-reactive compounds (Kwok et al., 2001; Liang et al., 2003) can also block the kinase activity of the constitutively active SS171,181EE IKKβ mutant, which does not require activation; therefore, in some cases, the mechanism of action of these Cys-179-reacting compounds may be more involved than simple steric hindrance for the accessibility of Ser 177, 181 to induced phosphorylation. For example, Cys-179 has recently been shown to be a site of reversible S-glutathionylation, which regulates the kinase activity of IKKβ (Reynaert et al., 2006).
IKK activation can also be blocked by gene-based inhibitors. Specifically, dominant-negative forms of IKKα and β, which are capable of blocking activation of NF-κB, can be created either by mutations in the ATP-binding site or by mutations in the kinase activation loop (DiDonato et al., 1997; Mercurio et al., 1997; Regnier et al., 1997; Woronicz et al., 1997; Zandi et al., 1997; Ling et al., 1998). Because of their distinct roles in the canonical (IKKβ-dependent) and non-canonical (IKKα-dependent) pathways, dominant-negative IKK mutants can show stimulus-dependent inhibition. For example, overexpression of dominant-negative IKKα or NIK (non-canonical pathway inhibitors), but not a dominant-negative IKKβ, blocks caspase-induced NF-κB activation (Shikama et al., 2003). Conversely, overexpression of a dominant-negative IKKβ, but not a dominant-negative IKKα, inhibits LPS induction of κB site-dependent transcription in THP-1 monocytic cells and the proliferative response of anti-CD3-stimulated T cells (O'Connell et al., 1998; Ren et al., 2002). Adenoviral-mediated delivery of an IKKβ dominant-negative kinase may have therapeutic potential for airway inflammatory diseases such as asthma (Broide et al., 2005; Catley et al., 2005).
NEMO can also serve as a target for IKK complex inhibition. In particular, introduction of a cell-permeable 10 amino-acid peptide corresponding to the NEMO-binding domain of IKKβ can block both the binding of NEMO to IKK and induction of NF-κB canonical pathway by TNF (May et al., 2000). Moreover, this peptide has shown efficacy in mouse models of inflammation by both topical and systemic administration (May et al., 2000; di Meglio et al., 2005). Similarly, introduction of peptides corresponding to the region of NEMO including the coiled-coil-2 and leucine zipper, which are required for oligomerization of NEMO, can also block NF-κB activation (Agou et al., 2004).
Agents that stabilize IκB or block its degradation
Several molecules inhibit NF-κB by maintaining a high level of IκB proteins in the cytoplasm and thereby preventing NF-κB nuclear translocation. Among these molecules, some promote synthesis of IκBα, some inhibit IκBα ubiquitination, while others block IκBα degradation. The molecules mentioned here and several others are listed in Supplementary Tables S3–S5.
Upregulation of IκB
A few molecules have been found to inhibit NF-κB by upregulating IκB expression (Supplementary Table S5). One such molecule is the β-amyloid peptide, which is deposited in the neuronal plaques that are a characteristic feature of Alzheimer's disease (AD). β-Amyloid peptide appears to show a cell type-specific effect on NF-κB, acting as an inhibitor in some cells and an activator in others. The constitutive NF-κB activity in fetal rat cortical neurons decreases following exposure to β-amyloid (Bales et al., 1998). IκBα mRNA and protein are increased in these cells following treatment with β-amyloid, and this increase is likely responsible for the decrease in activated NF-κB. This hypothesis is supported by the finding that pretreatment of cortical cultures with an antisense oligonucleotide to IκBα mRNA is neuroprotective towards β-amyloid toxicity. In contrast to cortical neurons, exposure of rat primary astroglial cultures to β-amyloid peptide results in activation of NF-κB with the subsequent increased transcription of the NF-κB target genes encoding IL-1β and IL-6. These data suggest that β-amyloid-induced neurotoxicity as well as astrocyte activation may be mediated by NF-κB, and thus alterations in NF-κB-directed gene expression may contribute to both neurodegeneration and the inflammatory response that occur in AD (Bales et al., 1998).
The cytokines IL-10, IL-11 and IL-13, which have powerful anti-inflammatory activities in vitro and in vivo, all suppress nuclear localization of NF-κB and increase IκBα mRNA expression levels (Lentsch et al., 1997; Ehrlich et al., 1998; Trepicchio and Dorner, 1998).
Blockers of IκB degradation: ubiquitination and proteasome inhibitors
The final common steps before NF-κB leaves the cytoplasm are the ubiquitination of IκB by the SCF-β-TrCP ubiquitin ligase complex followed by the rapid degradation of ubiquitinated IκB by the 26S proteasome (Scheidereit, 2006). Thus, inhibitors of either step in the ubiquitin–proteasome pathway suppress activation of NF-κB by stabilizing IκB. As shown in Supplementary Tables S3 and S4, there are many more blockers of the proteasome than of IκB ubiquitination. There are also a number of inhibitors that have been reported to inhibit degradation of IκB (Supplementary Table S3), but for most of these compounds it is not known that they inhibit IκB degradation per se.
Among blockers of IκB ubiquitination, the YopJ protein of the bacterial pathogen Yersinia stabilizes IκBα (and prevents NF-κB nuclear translocation) by acting as a dequbiquitinase for IκBα (Zhou et al., 2005). In addition, the small molecule R0196-9920 has been reported to be an inhibitor of IκBα ubiquitination, and can act as an oral inflammation inhibitor in two mouse models of induced inflammation (Swinney et al., 2002).
In early studies, Yaron et al. (1997) were able to block TNFα-induced degradation of IκBα by microinjecting phosphopeptides corresponding to the signal-dependent phosphorylation site of IκBα, presumably these acted as competitive inhibitors for binding to the ubiquitin ligase complex required for degradation of IκBα. More recently, inhibition of β-TrCP (the recognition subunit of the SCF E3 ligase complex) has been effected by specific RNAi treatment or by overexpression of dominant-negative β-TrCP mutants, and reduction of β-TrCP by these methods both block NF-κB activity and sensitize breast cancer cells to chemotherapeutic agents (Tang et al., 2005).
Non-genetic proteasome inhibitors that can block NF-κB activity are listed in Supplementary Table S4. These inhibitors can all penetrate cells and inhibit NF-κB activation in a dose-dependent manner by blocking proteasome-mediated degradation of IκB (but do not block its phosphorylation).
One class of proteasome inhibitors includes peptide aldehydes. These molecules inhibit the chymotrypsin-like activity of the proteasome complex (one of the five protease activities of the eukaryotic proteasome), but with distinct efficiencies. ALLnL, also called calpain inhibitor I or MG101, is a cysteine protease inhibitor, but is a less potent inhibitor of the proteasome than MG132 and MG115 (Palombella et al., 1994; Jobin et al., 1998a; Grisham et al., 1999).
Lactacystin and its synthetic precursor, β-lactone, represent a second class of inhibitors of the proteasome. These molecules irreversibly block proteasome activity by acylating a threonine residue in the active site of the mammalian proteasome subunit X/MB1 (Fenteany and Schreiber, 1998; Grisham et al., 1999). For this reason, lactacystin is considered to be a more specific inhibitor of the proteasome than the aldehyde peptides.
A third class of proteasome inhibitors is comprised of peptide boronic acids (or dipeptidyl boronates) named PS-262, PS-273, PS-341, PS-402, etc. These molecules were originally used as inhibitors of serine proteases, but were also found to act as proteasome inhibitors by blocking the chymotrypsin-like site in the 20S subunit core and to be more potent than their aldehyde analogs (Iqbal et al., 1995; Grisham et al., 1999; Adams, 2004). Most effective among these is PS-341 (now called bortezomib or Velcade), which has been shown to have significant efficacy against multiple myeloma, as well as several other hematologic and solid tumors (Adams, 2004). Although bortezomib is clearly an efficient blocker of NF-κB in many in vitro systems including myeloma cells (Cusack et al., 2001; Hideshima et al., 2002), it is not entirely clear whether its antimyeloma effects are mediated entirely, or even in part, through inhibition of NF-κB. For example, bortezomib has non-NF-κB effects on cancer cell growth and additional cancer-related protein targets are affected by its proteasome inhibitory activity (Hideshima et al., 2002; Adams, 2004; Zheng et al., 2004; Takigawa et al., 2006); moreover, in some cases bortezomib can activate, rather than inhibit, NF-κB (Németh et al., 2004). Thus, whether the therapeutic effect of bortezomib in the treatment of multiple myeloma (and other cancers) is entirely or in part due to its effect on NF-κB is not clear.
Several serine proteases inhibitors with chymotrypsin-like specificity (DCIC, TPCK, TLCK, BTEE, APNE) are also able to block proteasome function. However, unlike other protease inhibitors, those serine protease inhibitors can also block IκB phosphorylation as well as degradation, suggesting that a proteolytic step could (in some cases) be necessary for IKK activation. Finally, it is important to note that not all serine protease inhibitors can inhibit NF-κB activation (Higuchi et al., 1995; D'Acquisto et al., 1998; Rossi et al., 1998).
Downregulation of NF-κB nuclear functions: nuclear translocation, DNA binding and transcriptional activation
Direct inhibition of NF-κB-specific transactivation could involve blocking either nuclear translocation, DNA binding or transactivation by NF-κB dimers. In many studies, these last steps of NF-κB activation have been used as indicators of NF-κB inhibition (either as reduced binding in an EMSA or reduced κB-site reporter gene activity) (see Supplementary Tables S5–S7). However, inhibition of nuclear activities can also occur by maintaining a pool of active IκB proteins in the cytoplasm to thus reduce nuclear NF-κB activity.
Inhibitors of NF-κB nuclear translocation
Certain NF-κB inhibitors block its nuclear translocation. One approach to block this step has used cell-permeable peptides that contain the nuclear localizing sequence of p50. These peptides are thought to inhibit nuclear translocation of p50-containing dimers by saturating the nuclear import machinery responsible for the uptake of NF-κB dimers containing p50 (Lin et al., 1995; Torgerson et al., 1998; Letoha et al., 2005). However, one of the more commonly used peptides of this type, SN-50, has been reported to block nuclear translocation of a number of non-NF-κB transcription factors as well (Torgerson et al., 1998). An allosteric drug, o,o′-bismyristoyl thiamine disulfide, suppresses HIV-1 replication through prevention of nuclear translocation of both HIV-1 Tat and NF-κB (Shoji et al., 1998), and a fungal epoxyquinoid, DHMEQ, which has anti-inflammatory and antitumor activity in several mouse models, has been reported to be a specific inhibitor of NF-κB nuclear translocation (Umezawa, 2006).
Effective blocking of NF-κB DNA binding (and generally inhibition of its nuclear transport) can be accomplished by using mutant forms of IκBs, called super-repressors (SRs), which cannot be phosphorylated or degraded; thus, these mutant IκBs stably bind to NF-κB complexes. The most common SRs of NF-κB are ones derived as mutants of IκBα: such IκBα-SRs carry mutations of the signal-induced phosphorylation sites Ser 32 and 36 (wherein alanine residues replace the two serines), of the lysine ubiquitination sites (Lys → Arg mutations) or are deleted for their first 40 amino acids and thus, can be neither phosphorylated nor ubiquitinated (Wang et al., 1996; van Antwerp et al., 1996; Jobin et al., 1998b; Bentires-Alj et al., 1999). In addition, specific C-terminal Ser to Ala mutations are sometimes included to reduce the constitutive turnover of IκBα (van Antwerp et al., 1996). Such SR molecules have been used successfully to inhibit NF-κB activity in a variety of in vitro cell culture studies (e.g., van Antwerp et al., 1996; Wang et al., 1996; Bushdid et al., 1998; Kanegae et al., 1998) and in several in vivo transgenic mouse studies (see Gerondakis et al., 2006). In general, the use of the IκBα-SR has been taken as one of the best evidences for the involvement of NF-κB in a given physiological process; however, because IκBα does not bind well to RelB-containing complexes, the IκBα-SR is probably primarily a blocker of the canonical NF-κB pathway. Moreover, even the use of this ‘specific’ pathway inhibitor must be taken with some caution, in that the IκBα-SR can interact with and affect the activity of non-NF-κB pathway proteins including p53 (Chang, 2002), cyclin-dependent kinase 4 (Li et al., 2003) and HDACs (Aguilera et al., 2004). Furthermore, in terms of clinical use, overexpression of the IκBα-SR has been associated with the spontaneous development of squamous cell carcinoma in a murine model (van Hogerlinden et al., 1999).
A p105-based SR has also been created, which, like the IκBα-SR, is also a broad inhibitor of NF-κB; however, the p105-SR has the added ability to inhibit p50 homodimers and RelB-containing complexes, both of which are generally not inhibited by the IκBα-SR (Fu et al., 2004). Interestingly, several truncated p100 proteins, which would lack key proteasomal processing signals and could act as p100-SRs, are created by chromosomal alterations in leukemias/lymphomas (see Courtois and Gilmore, 2006).
Inhibitors of NF-κB DNA binding
The largest class of non-specified NF-κB inhibitors block DNA binding and κB site-dependent gene expression. However, for several of the molecules listed in Supplementary Tables S6 and S7, inhibition of DNA binding does not necessarily indicate that the DNA-binding step is specifically impaired, rather it is usually that the effect of the inhibitor on NF-κB has been measured by assaying the amount of NF-κB bound to a κB-site probe in an EMSA, as compared to control cells which have not been treated with the inhibitor.
Several sesquiterpene lactones (SLs) have anti-inflammatory activity and act as inhibitors of NF-κB DNA binding (Zhang et al., 2005). Some SLs can directly inhibit NF-κB DNA binding, primarily through interaction with Cys-38 in the DNA-binding loop of RelA (García-Piñeres et al., 2001, 2004). Most SLs can also inhibit DNA binding by p50 and c-Rel through an analogous Cys residue in the DNA-binding loop, and mutations of this Cys residue to Ser generally make p50, RelA or c-Rel refractory to inhibition by such thiol-reactive compounds. Recently, a computer-based structural comparison of 103 SLs predicted that a methylene-carbonyl substructure is important for SL-based inhibition of RelA at Cys-38 (Wagner et al., 2006). Interestingly, some SLs, including the natural product parthenolide, have been shown to also inhibit IKKβ through a reactive Cys residue (Cys-179), which is in the kinase activation loop (Kwok et al., 2001; García-Piñeres et al., 2004). Thus, the SLs (and some epoxyquinoids; Liang et al., 2006) have multistep inhibitory activity within the NF-κB signaling pathway, targeting both IKK activity and NF-κB subunit DNA binding.
A molecular method to block specific NF-κB DNA binding is through the use of decoy oligonucleotides that have κB sites, that is , to complete out NF-κB dimer binding to specific genomic promoters (e.g., Morishita et al., 1997; Khaled et al., 1998; Kupatt et al., 2002). Generally, these oligonucleotides have modifications to increase their stability and their affinity for NF-κB in vivo (Tomita et al., 2003; Crinelli et al., 2004; Isomura and Morita, 2006). These κB-site decoy oligonucleotides have had therapeutic efficacy in a number of animal models of inflammation and cancer.
In principle, agents that can interfere with NF-κB dimerization could also be effective blockers of NF-κB activity, and because the various NF-κB subunits show different dimerization affinities (see Hoffmann et al., 2006), it should be possible to develop agents that selectively disrupt different NF-κB dimers. Although trans-dominant p50 mutants (that can dimerize but not bind DNA) have been used in tissue culture models (Logeat et al., 1991), no chemical or natural product inhibitors of NF-κB dimerization have been described.
Inhibitors of NF-κB transactivation
The final step in NF-κB signaling that can be blocked is transactivation of specific target genes. In theory, this should be a step that could provide extreme specificity and efficacy, given that different NF-κB dimers target different promoters/enhancers in tissue-, stimulus-, and promoter-specific manners (Hoffmann et al., 2006). In addition, NF-κB subunit modification can influence their target gene transactivation ability (Perkins, 2006). Based on these considerations, one may be able to interfere with the specific subset of activated (or repressed) NF-κB target genes that effects a given biological response.
Although many compounds have been reported as NF-κB transactivation inhibitors (Supplementary Table S7), often this is simply because the readout for inhibition is transactivation of an NF-κB-responsive reporter gene. Nevertheless, some compounds do appear to be specific inhibitors of the NF-κB-dependent transactivation step. For example, D609, RO31-8220 and SB203580 are compounds that selectively inhibit phosphatidylcholine-phospholipase C inhibitor, protein kinase C and p38 MAPK, respectively. In the human A549 lung cancer cell line, they are also able to block NF-κB-dependent transcription after stimulation by IL-1 and TNFα. However, none of these molecules inhibits IκBα degradation, NF-κB nuclear translocation or DNA binding (Bergmann et al., 1998). In another study, LY294,002 and wortmannin (both PI-3 kinase inhibitors) have been used to block IL-1-induced NF-κB activation of a reporter gene (Reddy et al., 1997; Sizemore et al., 1999). LY294,002 did not inhibit IκBα degradation or NF-κB DNA binding, but it did block IL-1-stimulated phosphorylation of NF-κB, especially the RelA subunit.
Similar to those chemical inhibitors, the natural products mesalamine, mesuol and pertussis toxin appear to specifically block RelA phosphorylation, which is required for optimal RelA-mediated transactivation (Egan et al., 1999; Iordanskiy et al., 2002: Marquez et al., 2005). Moreover, the antiapoptosis protein Bcl-2 was shown to specifically inhibit transactivation by RelA in one study (Grimm et al., 1996), and antithrombin was reported to inhibit NF-κB transactivation by interfering with RelA-CREB co-activator interaction (Uchiba et al., 2004). Experiments using NF-κB target gene cDNA expression microarrays to assess the gene-specific effects of transactivation inhibitors may provide insights into methods to interfere with NF-κB target gene subpathways that control specific biological responses.
In many, but not all cell types, NF-κB can be activated by direct treatment with oxidants, such as treatment with hydrogen peroxide; in addition, the induction of NF-κB activity in response to a variety of stimuli (e.g., IL-1β, LPS, TNFα) in some cell types can be inhibited by antioxidants (reviewed in Bubici et al., 2006; Gloire et al., 2006). However, neither the target for oxidant-induced activation nor the exact pathway used by such molecules to activate NF-κB is known. Similarly, it is not known precisely how antioxidants block activation of NF-κB, but it is likely that they act at different steps in the NF-κB pathway in different cell types.
As such, many antioxidant compounds such as thiol antioxidants (e.g., NAC, PDTC), calcium chelators (e.g., EGTA, lacidipine), vitamin C and E derivatives, and α-lipoic acid have been used to inhibit hydrogen peroxide- or stimulus-induced NF-κB activation. Presumably, many of these agents act by scavenging reactive oxygen intermediates (ROIs) (Sen et al., 1996b). In addition, inhibitors of mitochondrial electron transport that suppress ROI production (like rotenone) or overexpression of antioxidizing enzymes, such as MnSOD and catalase, can block TNFα-induced activation of NF-κB (Schulze-Osthoff et al., 1993; Manna et al., 1998, 1999). Caffeic acid phenethyl ester, a phenolic antioxidant and a structural relative of flavonoids, may directly interfere with DNA binding by NF-κB (Natarajan et al., 1996), and this effect on DNA binding can be reversed by reducing agents like dithiothreitol (Singh and Aggarwal, 1995). Therefore, all antioxidants probably do not act at the same level to inhibit NF-κB. Moreover, at least one report has suggested that the NF-κB inhibitory activity of PDTC can be separated from its antioxidant activity (Hayakawa et al., 2003).
Supplementary Table 8 lists the many antioxidant molecules that have been shown to inhibit NF-κB. Additionally, the anti-NF-κB activity of the many natural plant and food extracts listed in other Supplementary Tables S1–S7 may well be due to antioxidants in these extracts. In general, antioxidants may function in two ways to block activation of NF-κB. On one hand, antioxidants could act as scavangers for ROIs that act as signaling molecules to activate the NF-κB pathway. Alternatively, under certain circumstances, antioxidants may directly inhibit IKK kinase activity or NF-κB DNA-binding by affecting the redox state of critical Cys residues in the IKK kinase activation loop and NF-κB DNA-binding loop (Gloire et al., 2006; Bubici et al., 2006). For example, the oxidation state of NF-κB seems important for its interaction with DNA (Yang et al., 1995), and its DNA binding can be blocked by interaction with thiol-reactive metals (Shumilla et al., 1998).
A large number of studies have indicated that compounds with antioxidant activity can have anti-inflammatory and/or antitumor activity in animal models, and there is much correlative data for similar effects in humans (Na and Surh, 2006). Because of its key role in inflammation and cancer (Aggarwal et al., 2006), the NF-κB pathway certainly seems a reasonable target for such preventative and therapeutic effects of antioxidants, even though the mechanism(s) by which antioxidants block NF-κB signaling is still a mystery and, to some, their effects are controversial.
NF-κB inhibitors encoded or produced by viruses, bacteria and fungi
The innate immune response is initiated in large part because most microorganisms and viruses activate the NF-κB pathway through pattern-associated molecular pattern receptors called Toll-like receptors (see Hayden et al., 2006; Hiscott et al., 2006). Nevertheless, several microorganisms and viruses also encode proteins that can inhibit NF-κB, and this can often either enhance their replication or contribute to their pathogenicity.
Viruses have developed a number of mechanisms to inhibit NF-κB signaling (see Hiscott et al., 2006). At least three virsuses – African swine fever virus (ASFV) (Powell et al., 1996), rabbit myxoma virus (Camus-Bouclainville et al., 2004), insect Microplitis demolitor bracovirus (Thoetkiattikul et al., 2005) – encode IκB-like inhibitors of NF-κB. In the case of ASFV, the viral A238L IκB-like protein lacks the analogous serine residues that direct signal-induced phosphorylation and degradation of cellular IκBs, and thus, A238L can stably interact with RelA to inhibit NF-κB DNA binding as induced by TNFα, IFN-γ and phorbol ester (Revilla et al., 1998). The poliovirus 3C protease cleaves RelA to reduce NF-κB signaling (Neznanov et al., 2005). In addition, several viruses have adaptor-like or small proteins that inhibit IKK activity (see Hiscott et al., 2006, this issue). Interestingly, the MC160 protein of molluscum contagiosum virus (Nichols and Shisler, 2006) and the NS5B protein of hepatitis C virus (Choi et al., 2006) appear to be specific for IKKα, and thus, may provide prototypes for inhibitors of the non-canonical NF-κB pathway.
Among bacteria, as described above, the YopJ protein encoded by the enteropathogen Yersinia pseudotuberculosis inhibits NF-κB activation by deubiquitinating IκBα, which prevents its degradation. Consequently, eukaryotic cells infected with YopJ-expressing Yersinia become impaired in NF-κB-dependent cytokine expression, which coincides with YopJ-dependent induction of apoptosis (Schesser et al., 1998). The Salmonella typhimurium AvrA protein, a homolog YopJ, also inhibits NF-κB activation, although the mechanism may be different than that used by YopJ (Collier-Hyams et al., 2002).
In addition, several small molecule metabolites synthesized by microorganisms (or designed as derivatives of such compounds) can inhibit NF-κB. These include panepoxydone (from Lentinus crinitus), 5,6 epoxycyclohexenone compounds (from Amycolatopsis), gliotoxin (Aspergillus fumigatus), and cycloepoxydon (reviewed in Umezawa et al., 2000). In spite of their sometimes similar structures, such compounds may affect distinct parts of the NF-κB pathway, including DNA binding, nuclear translocation and phosphorylation/degradation of IκBα.
Commonly used human drugs that may act, at least in part, by inhibiting NF-κB activity
NF-κB is one of the key transcription factors in the inflammatory response, and many NF-κB target genes are involved in the acute phase or inflammatory response (Pahl, 1999, see also www.nf-kb.org, under Target Genes; Hayden et al., 2006). Therefore, inhibition of NF-κB can reduce the pathologic effects induced by chronic stimulation of the inflammatory response, as seen in many chronic inflammatory diseases or animal models of inflammation. Consistent with this, several commonly-used anti-inflammatory agents have been shown to inhibit NF-κB activity, although whether this is their primary mode of action in vivo is not clear. Similarly, some anticancer agents have been shown to have anti-NF-κB activity, which may be involved with their antitumor effects, likely either by lessening a proliferative, antiapoptotic or inflammatory activity of NF-κB.
Several nonsteroidal anti-inflammatory drugs (NSAIDs), such as aspirin (sodium salicylate), ibuprofen, sulindac and indomethacin, can inhibit activation of NF-κB in cell culture (Kopp and Ghosh, 1994; Grilli et al., 1996; Palayoor et al., 1998; Takada et al., 2004). Aspirin is commonly thought to act pharmacologically primarily via inhibition of prostaglandin synthesis (Weissmann, 1991). However, at higher concentrations, aspirin has also been shown to block NF-κB activity by directly binding to and inhibiting the kinase activity of IKKβ by reducing its ability to bind ATP (Yin et al., 1998); more recently, aspirin has also been reported to inhibit proteasome activity and consequently to interfere with degradation of IκB (Dikshit et al., 2006). As such, high-dose aspirin therapy may have applications to diseases where NF-κB activity is involved, including cancer (McCarty and Block, 2006), diabetes (Yuan et al., 2001) and heart disease (Li and Fang, 2004).
Glucocorticoids, such as dexametasone, prednisone and methylprednisolone, are used for their anti-inflammatory properties and to prevent allograft rejection. Their physiological effects appear to be, at least partially, mediated through inhibition of NF-κB. As described by De Bosscher et al. (2006), glucocorticoids, acting via the glucocorticoid receptor, can inhibit NF-κB by a variety of methods, including inhibition of DNA binding, IKK activity and transactivation, in different cell contexts. Similarly, estrogen and certain selective estrogen receptor modulators (SERMs), such as raloxifene, can act through the estrogen receptor to inhibit NF-κB by a variety of mechanisms (Kalaitzidis and Gilmore, 2005; Olivier et al., 2006).
Several well-known immunosuppressants target NF-κB. Cyclosporin A (CsA), inhibits B- and T-cell proliferation by blocking the activity of calcineurin, a calcium and calmodulin-dependent serine/threonine phosphatase (Frantz et al., 1994). Several reports have shown that CsA can also inhibit NF-κB induction, although the mechanism by which CsA exerts this inhibitory effect may differ in different situations. In vitro, CsA has been reported to act as a non-competitive inhibitor of the chymotrypsin-like activity of the proteasome, enabling it to block LPS-induced IκB degradation and p105 processing in vivo (Meyer et al., 1997). Similarly, CsA prevents NF-κB nuclear translocation in stimulated T cells (McCaffrey et al., 1994) by preventing the inducible degradation of IκBα and IκBβ; however, in these cells, CsA does not inhibit the processing of p105 to p50 (Marienfeld et al., 1997). CsA is able to block the activation of IL-2 and IL-8 gene expression by NF-κB in T cells (Wechsler et al., 1994; Nishiyama et al., 2005). Like aspirin, CsA has also been shown to be neuroprotective (Meyer et al., 1997).
FK506 (aka tacrolimus) is an immunosuppressant that acts as a potent blocker of B- and T-cell proliferation. At least in part, FK506, like CsA, acts by blocking the activity of calcineurin. However, unlike CsA, the inhibitory effect of tacrolimus on NF-κB appears, in some cases, to be specific for c-Rel, among the NF-κB family members. That is, FK506 can specifically block c-Rel nuclear translocation (but not p50/RelA) after treatment of cells with phorbol esters and ionomycin (Sen et al., 1995; Venkataraman et al., 1995). Therefore, the antiproliferative effects of FK506 in T cells results from inhibition of NF-κB, which consequently blocks transcription of the IL-2 and IL-2 receptor genes by interfering with the induction of c-Rel-dependent transcription of their promoters (Serfling et al., 1995; Venkataraman et al., 1995). Interestingly, the effectiveness of topical administration of FK506 ointment in keratinocytes correlates with its anti-NF-κB activity (Lan et al., 2005), suggesting that inhibition of NF-κB may be responsible for FK506's effectiveness in psoriasis.
Certain other immunosuppressants act at different levels than CsA and FK506 to block immune cell proliferation. For example, PG490 (pure triptolide, a diterpene triepoxide) is an immunosuppressant molecule that can synergize with CsA to inhibit transcriptional activation by NF-κB. However, PG490 does not appear to interfere with the induction of NF-κB DNA binding, suggesting that it acts to inhibit transcriptional activation by NF-κB (Qiu et al., 1999). Moreover, the immunosuppressant deoxyspergualin inhibits NF-κB nuclear translocation by a mechanism involving the heat-shock protein Hsp70 (Tepper et al., 1995).
Thus, different immunosuppressants appear to inhibit NF-κB by distinct mechanisms. Some inhibit NF-κB nuclear translocation by stabilizing IκBα (Meyer et al., 1997), some may act on NF-κB through inhibiting calcineurin (Frantz et al., 1994), some by binding heat-shock proteins (Tepper et al., 1995) and some by modulating the DNA binding or transactivation potential of NF-κB (McCaffrey et al., 1994; Wechsler et al., 1994; Kunz et al., 1995; Qiu et al., 1999).
Many human drugs have been reported to have off-target anti-NF-κB activity
Many human drugs that have been primarily characterized for activities other than anti-inflammatory or antitumor activity can also inhibit NF-κB. A partial list of such drugs is presented in Table 2. These drugs have a variety of molecular protein targets, and how and why they can also affect NF-κB is often not clear. In cases such as these, a component of the NF-κB pathway may be an off-target for these drugs, these compounds may simply show anti-NF-κB activity at non-physiological doses in vitro and in vivo, or NF-κB may be affected due to cross-talk between the drug's primary target and the NF-κB pathway.
Common problems with identifying and characterizing NF-κB inhibitors
It is now clear that there are several experimental and therapeutic problems that arise when characterizing NF-κB inhibitors. First, many NF-κB pathway inhibitors may be cell type- or stimulus-dependent (see Epinat and Gilmore, 1999). For example, in the vast majority of studies, NF-κB inhibitors have been characterized based solely on their ability to block TNFα or LPS-induced NF-κB activity. Second, in many cases, these compounds are likely to affect a variety of other targets and pathways. This is especially likely to be the case with highly reactive compounds such as antioxidants or thiol-reactive chemicals, but may even be the case with molecular inhibitors such as the IκB SR, which has been shown to interact with and affect the activity of non-NF-κB pathway proteins including p53 (Chang, 2002), cyclin-dependent kinase 4 (Li et al., 2003) and HDACs (Aguilera et al., 2004). In many cases, the intense interest in NF-κB and the multistep activation pathway may have led to the abundance of reports on NF-κB inhibitors. Third, the concentrations of compounds used to inhibit NF-κB in in vitro studies may be substantially different (i.e., often much higher) than could ever be used or achieved in vivo.
As detailed in this review, numerous inhibitors of the NF-κB activity have been described. In a limited number of cases, the molecular basis for this inhibitory activity is known, however, more often it is simply known that such molecules inhibit some step in induced NF-κB activation, for example, induced nuclear NF-κB DNA binding. Some NF-κB inhibitors, such as the IKK or proteasome inhibitors, are broad-range inhibitors, which can block most NF-κB activating signals since they target common steps in NF-κB activation. However, the broad-range inhibitors may affect a variety of pathways, given that IKK and the proteasome are now known to be involved in both the canonical and non-canonical NF-κB signaling pathways, as well as in non-NF-κB signaling pathways and cellular events. Other inhibitors of NF-κB block activation when induced by only certain stimuli, for example, by acting on a particular protein in the signaling cascade or by blocking specific NF-κB complexes. For example, inhibitors of NIK are likely specific for the non-canonical pathway, while NEMO-directed inhibitors are expected to be specific for the canonical NF-κB pathway. Nevertheless, even these specific inhibitors may block other pathways, either because their target participates in several intracellular pathways or because they block a certain class of protein (i.e., have off-target activity). Indeed, for most inhibitors listed in this review, neither the molecular target in the NF-κB signaling pathway nor the cell-type specificity of the inhibitor is known.
Nature's use of single molecular components in overlapping signaling pathways makes it a challenge to find molecules that block specific pathways leading to NF-κB activation without interfering with other signaling cascades. A future goal will be to discover molecules that can inhibit distinct NF-κB complexes induced by select stimuli in specific cell types. To cite one example, antisense oligonucleotides to RelA have been shown to have therapeutic effects in various animal models (e.g., Choi et al., 2006; Isomura and Morita, 2006), but given that RelA is probably the most general NF-κB subunit, the effectiveness of this approach for specific diseases is not clear. Similarly, the use of compounds that block target gene-specific events for NF-κB complexes may provide exquisite specificity (e.g., the phosphorylation of RelA seems to modulate its transactivating activity on specific genes). Alternatively, the combined use of low doses of inhibitors that target multiple steps in the NF-κB pathway (e.g., thiol-reactive drugs that target both IKK and NF-κB DNA binding) may be an effective strategy with reduced side effects.
To us, it seems likely that the most promising near-term uses for NF-κB inhibitors will be in cases where such inhibitors can be applied topically (e.g., cylindromatosis or other skin inflammatory diseases), locally (e.g., airway inflammation), or in a highly directed fashion (e.g., to a specific cell or tumor type). Systemic application of potent NF-κB inhibitors will likely have unwarranted side effects. Nevertheless, it is certainly possible that the long-term systemic ingestion of low-dose NF-κB inhibitors may have general beneficial effects in reducing inflammation and cancer. At least in part, chronic dampening of NF-κB activity may explain the promoted anti-inflammatory, antiaging, and anticancer effects of routine, long-term ingestion of natural compounds, for example, green tea, curcumin, others, which have been used for centuries.
Acarin L, Gonzalez B, Castellano B . (2000). Neurosci Lett 288: 41–44.
Adams J . (2004). Cancer Cell 5: 417–421.
Aggarwal BB, Shishodia S, Sandur SK, Pandey MK, Sethi G . (2006). Biochem Pharmacol, in press.
Agou F, Courtois G, Chiaravalli J, Baleux F, Coïc YM, Traincard F et al. (2004). J Biol Chem 279: 54248–54257.
Aguilera C, Hoya-Arias R, Haegeman G, Espinosa L, Bigas A . (2004). Proc Natl Acad Sci USA 101: 16537–16542.
Aravindan N, Natarajan M, Shaw AD . (2006). J Cardiothorac Vasc Anesth 20: 179–186.
Azuma RW, Suzuki J, Ogaa M, Futamatsu H, Koga N, Onai Y et al. (2004). Cardiovasc Res 64: 412–420.
Bales KR, Du Y, Dodel RC, Yan GM, Hamilton-Byrd E, Paul SM . (1998). Brain Res Mol Brain Res 57: 63–72.
Basse(c)res D, Baldwin Jr AS . (2006). Oncogene 25: 6817–6830.
Bentires-Alj M, Hellin AC, Ameyar M, Chouaib S, Merville MP, Bours V . (1999). Cancer Res 59: 811–815.
Bergmann M, Hart L, Lindsay M, Barnes PJ, Newton R . (1998). J Biol Chem 273: 6607–6610.
Bradley JR, Pober JS . (2001). Oncogene 20: 6482–6491.
Broide DH, Lawrence T, Doherty T, Cho JY, Miller M, McElwain K et al. (2005). Proc Natl Acad Sci USA 102: 17723–17728.
Bubici C, Papa S, Dean K, Franzoso G . (2006). Oncogene 25: 6731–6748.
Burke JR, Pattoli MA, Gregor KR, Brassil PJ, MacMaster JF, McIntyre KW et al. (2003). J Biol Chem 278: 1450–1456.
Bushdid PB, Brantley DM, Yull FE, Blaeuer GL, Hoffman LH, Niswander L et al. (1998). Nature 392: 615–618.
Camus-Bouclainville C, Fiette L, Bouchiha S, Pignolet B, Counor D, Filipe C et al. (2004). J Virol 78: 2510–2516.
Cao Z, Xiong J, Takeuchi M, Kurama T, Goeddel DV . (1996). Nature 383: 443–446.
Catley MC, Chivers JE, Holden NS, Barnes PJ, Newton R . (2005). Br J Pharmacol 145: 114–122.
Chang NS . (2002). J Biol Chem 277: 10323–10331.
Choi SH, Park KJ, Ahn BY, Jung G, Lai MM, Hwang SB . (2006). Mol Cell Biol 26: 3048–3059.
Collier-Hyams LS, Zeng H, Sun J, Tomlinson AD, Bao ZQ, Chen H et al. (2002). J Immunol 169: 2846–2850.
Courtois G, Gilmore TD . (2006). Oncogene 25: 6831–6843.
Crinelli R, Bianchi M, Gentilini L, Palma L, Magnani M . (2004). Curr Drug Targets 5: 745–752.
Cusack Jr JC, Liu R, Houston M, Abendroth K, Elliott PJ, Adams J et al. (2001). Cancer Res 61: 3535–3540.
D'Acquisto F, Sautebin L, Iuvone T, Di Rosa M, Carnuccio R . (1998). FEBS Lett 440: 76–80.
De Bosscher K, Vanden Berge W, Haegeman G . (2006). Oncogene 25: 6868–6886.
di Meglio P, Ianaro A, Ghosh S . (2005). Arthritis Rheum 52: 951–958.
DiDonato JA, Hayakawa M, Rothwarf DM, Zandi E, Karin M . (1997). Nature 388: 548–554.
Dikshit P, Chatterjee M, Goswami A, Mishra A, Jana NR . (2006). J Biol Chem 25: 6868–6886.
Dutta J, Fan Y, Gupta N, Gan G, Gélinas C . (2006). Oncogene 25: 6800–6816.
Egan LJ, Mays DC, Huntoon CJ, Bell MP, Pike MG, Sandborn WJ et al. (1999). J Biol Chem 274: 26448–26453.
Ehrlich LC, Hu S, Peterson PK, Chao CC . (1998). Neuroreport 9: 1723–1726.
Epinat J-C, Gilmore TD . (1999). Oncogene 18: 6896–6909.
Fenteany G, Schreiber SL . (1998). J Biol Chem 273: 8545–8548.
Frantz B, Nordby EC, Bren G, Steffan N, Paya CV, Kincaid RL et al. (1994). EMBO J 13: 861–870.
Fu D, Kobayashi M, Lin L . (2004). J Biol Chem 279: 12819–12826.
García-Piñeres AJ, Castro V, Mora G, Schmidt TJ, Strunck E, Pahl HL et al. (2001). J Biol Chem 276: 39713–39720.
García-Piñeres AJ, Lindenmeyer MT, Merfort I . (2004). Life Sci 75: 841–856.
Gerondakis S, Grumont R, Gugasyan R, Wong L, Isomura I, Ho W et al. (2006). Oncogene 25: 6781–6799.
Gilmore TD . (2006). Oncogene 25: 6680–6684.
Gloire G, Legrand-Poels S, Piette J . (2006). Biochem Pharmacol, in press.
Grilli M, Pizzi M, Memo M, Spano P . (1996). Science 274: 1383–1385.
Grimm S, Bauer MKA, Baeuerle PA, Schulze-Osthoff K . (1996). J Cell Biol 134: 13–23.
Grisham MB, Palombella VJ, Elliott PJ, Conner EM, Brand S, Wong HL et al. (1999). Methods Enzymol 300: 345–363.
Hayakawa M, Miyashita H, Sakamoto I, Kitagawa M, Tanaka H, Yasuda H et al. (2003). EMBO J 22: 3356–3366.
Hayden MS, West AP, Ghosh S . (2006). Oncogene 25: 6758–6780.
Hideshima T, Chauhan D, Richardson P, Mitsiades C, Mitsiades N, Hayashi T et al. (2002). J Biol Chem 277: 16639–16647.
Higuchi M, Singh S, Chan H, Aggarwal BB . (1995). Blood 86: 2248–2256.
Hirano F, Kobayashi A, Makino I . (2003). Int Immuopharmacol 3: 225–232.
Hirano M, Osada S-i, Aoki T, Hirai S-i, Hosaka M, Inoue J-i et al. (1996). J Biol Chem 271: 13234–13238.
Hiscott J, Nguyen T-LA, Arguello M, Nakhaei P, Paz S . (2006). Oncogene 25: 6844–6867.
Hoffmann A, Natoli G, Ghosh G . (2006). Oncogene 25: 6758–6780.
Hsu H, Shu HB, Pan MG, Goeddel DV . (1996). Cell 84: 299–308.
Ikezoe T, Hisatake Y, Takeuchi T, Ohtsuki Y, Yang Y, Said JW et al. (2004). Cancer Res 64: 7426–7431.
Imbert V, Rupec RA, Livolsi A, Pahl HL, Traenckner EB, Mueller-Dieckmann C et al. (1996). Cell 86: 787–798.
Iordanskiy S, Iordanskaya T, Quivy V, Van Lint C, Bukrinsky M . (2002). Virology 302: 195–206.
Iqbal M, Chatterjee S, Kauer JC, Das M, Messina P, Freed B et al. (1995). J Med Chem 38: 2276–2277.
Isomura I, Morita A . (2006). Microbiol Immunol 50: 559–563.
Jobin C, Hellerbrand C, Licato LL, Brenner DA, Sartor RB . (1998a). Gut 42: 779–787.
Jobin C, Panja A, Hellerbrand C, Iimuro Y, Didonato J, Brenner DA et al. (1998b). J Immunol 160: 410–418.
Johanson V, Arvidsson Y, Kolby L, Bernhardt P, Sward C, Nilsson O et al. (2005). Neuroendocrinology 82: 171–176.
Kalaitzidis D, Gilmore TD . (2005). Trends Endocrinol Metab 16: 46–52.
Kanegae Y, Tavares AT, Izpisua Belmonte JC, Verma IM . (1998). Nature 392: 611–614.
Kapahi P, Takahashi T, Natoli G, Adams SR, Chen Y, Tsien RY et al. (2000). J Biol Chem 275: 36062–36066.
Karin M, Delhase M . (1998). Proc Natl Acad Sci USA 95: 9067–9069.
Karin M, Yamamoto Y, Wang QM . (2004). Nat Rev Drug Discov 3: 17–26.
Khaled AR, Butfiloski EJ, Sobel ES, Schiffenbauer J . (1998). Clin Immunol Immunopathol 86: 170–179.
Kopp E, Ghosh S . (1994). Science 265: 956–959.
Kunz D, Walker G, Eberhardt W, Nitsch D, Pfeilschifter J . (1995). Biochem Biophys Res Comm 216: 438–446.
Kupatt C, Wichels R, Deiss M, Molnar A, Lebherz C, Raake P et al. (2002). Gene Therapy 9: 518–526.
Kwok BH, Koh B, Ndubuisi MI, Elofsson M, Crews CM . (2001). Chem Biol 8: 759–766.
Lan CC, Yu HS, Wu CS, Kuo HY, Chai CY, Chen GS . (2005). Br J Dermatol 153: 725–732.
Lawrence DM, Singh RS, Franklin DP, Carey DJ, Elmore JR . (2004). J Vasc Surg 40: 334–338.
Lee FS, Hagler J, Chen ZJ, Maniatis T . (1997). Cell 88: 213–222.
Lee FS, Peters RT, Dang LC, Maniatis T . (1998). Proc Natl Acad Sci USA 95: 9319–9324.
Lentsch AB, Shanley TP, Sarma V, Ward PA . (1997). J Clin Invest 100: 2443–2448.
Letoha T, Somlai C, Takacs T, Szabolcs A, Jarmay K, Rakonczay Jr Z et al. (2005). World J Gasteroenterol 11: 990–999.
Li J, Joo SH, Tsai MD . (2003). Biochemistry 42: 13476–13483.
Li JJ, Fang CH . (2004). Med Hypotheses 62: 499–506.
Li N, Karin M . (1998). Proc Natl Acad Sci USA 95: 13012–13017.
Li X, Meng Y, Yang XS, Mi LF, Cai SX . (2005). World J Gastroenterol 11: 4807–4811.
Liang M-C, Bardhan S, Li C, Pace EA, Porco Jr JA, Gilmore TD . (2003). Mol Pharmacol 64: 123–131.
Liang M-C, Bardhan S, Pace EA, Rosman D, Beutler JA, Porco Jr JA et al. (2006). Biochem Pharmacol 71: 634–645.
Lin YZ, Yao SY, Veach RA, Torgerson TR, Hawiger J . (1995). J Biol Chem 270: 14255–14258.
Ling L, Cao Z, Goeddel DV . (1998). Proc Natl Acad Sci USA 95: 3792–3797.
Logeat F, Israël N, Ten R, Blank V, Le Bail O, Kourilsky P et al. (1991). EMBO J 10: 1827–1832.
Malinin NL, Boldin MP, Kovalenko AV, Wallach D . (1997). Nature 385: 540–544.
Manna SK, Kuo MT, Aggarwal BB . (1999). Oncogene 18: 4371–4382.
Manna SK, Zhang HJ, Yan T, Oberley LW, Aggarwal BB . (1998). J Biol Chem 273: 13245–13254.
Marienfeld R, Neumann M, Chuvpilo S, Escher C, Kneitz B, Avots A et al. (1997). Eur J Immunol 27: 1601–1609.
Marquez N, Sancho R, Bedoya LM, Alcami J, Lopez-Perez JL, Feliciano AS et al. (2005). Antiviral Res 66: 137–145.
May MJ, D'Acquisto F, Madge LA, Glockner J, Pober JS, Ghosh S . (2000). Science 289: 1550–1554.
McCaffrey PG, Kim PK, Valge-Archer VE, Sen R, Rao A . (1994). Nucleic Acids Res 22: 2134–2142.
McCarty MF, Block KI . (2006). Integr Cancer Ther 5: 252–268.
Mercurio F, Zhu H, Murray BW, Shevchenko A, Bennett BL, Li J et al. (1997). Science 278: 860–866.
Meyer S, Kohler NG, Joly A . (1997). FEBS Lett 413: 354–358.
Miyanohara T, Ushikai M, Matsune S, Ueno K, Katahira S, Kurono Y . (2000). Laryngoscope 110: 126–131.
Morishita T, Sugimoto T, Aoki M, Kida I, Tomita N, Moriguchi A et al. (1997). Nat Med 3: 894–899.
Na HK, Surh YJ . (2006). Mol Nutr Food Res 50: 152–159.
Natarajan K, Singh S, Burke Jr TR, Grunberger D, Aggarwal BB . (1996). Proc Natl Acad Sci USA 93: 9090–9095.
Németh ZH, Wong HR, Odoms K, Deitch EA, Szabo C, Vizi ES et al. (2004). Mol Pharmacol 65: 342–349.
Neznanov N, Chumakov KM, Nennanova L, Almasan A, Banerjee AK, Gudkov AV . (2005). J Biol Chem 280: 24153–24158.
Nichols DB, Shisler JL . (2006). J Virol 80: 578–586.
Nishiyama S, Manabe N, Kubota Y, Ohnishi H, Kitanaka A, Tokuda M et al. (2005). Int Immunopharmacol 5: 699–710.
O'Hare T, Corbin AS, Druker BJ . (2006). Curr Opin Genet Dev 16: 92–99.
O'Connell MA, Bennett BL, Mercurio F, Manning AM, Mackman N . (1998). J Biol Chem 273: 30410–30414.
Olivier S, Close P, Castermans E, de Leval L, Tabruyn S, Chariot A et al. (2006). Mol Pharmacol 69: 1615–1623.
Pahl HL . (1999). Oncogene 18: 6853–6866.
Palayoor ST, Bump EA, Calderwood SK, Bartol S, Coleman CN . (1998). Clin Cancer Res 4: 763–771.
Palombella VJ, Rando AL, Goldberg AL, Maniatis T . (1994). Cell 78: 773–786.
Pan Q, Kleer CG, van Golen KL, Irani J, Bottema KM, Bias C et al. (2002). Cancer Res 62: 4854–4859.
Pande V, Ramos MJ . (2005). Curr Med Chem 12: 357–374.
Perkins ND . (2006). Oncogene 25: 6717–6730.
Powell PP, Dixon LK, Parkhouse RME . (1996). J Virol 70: 8527–8533.
Pu Q, Amiri F, Gannon P, Shiffrin EL . (2005). J Hypertension 23: 401–409.
Qiu D, Zhao G, Aoki Y, Shi L, Uyei A, Nazarian S et al. (1999). J Biol Chem 274: 13443–13450.
Reddy SA, Huang JH, Liao WS . (1997). J Biol Chem 272: 29167–29173.
Regnier CH, Song HY, Gao X, Goeddel DV, Cao Z, Rothe M . (1997). Cell 90: 373–383.
Ren H, Schmalstieg A, van Oers NS, Gaynor RB . (2002). J Immunol 168: 3721–3731.
Reynaert NL, van der Vliet A, Guala AS, McGovern T, Hristova M, Pantano C et al. (2006). Proc Natl Acad Sci USA, in press.
Revilla Y, Callejo M, Rodriguez JM, Culebras E, Nogal ML, Salas ML et al. (1998). J Biol Chem 273: 5405–5411.
Rossi A, Elia G, Santoro MG . (1998). J Biol Chem 273: 16446–16452.
Ruan H, Pownall HJ, Lodish HE . (2003). J Biol Chem 278: 28181–28192.
Scheidereit C . (2006). Oncogene 25: 6685–6705.
Schesser K, Spiik AK, Dukuzumuremyi JM, Neurath MF, Pettersson S, Wolf-Watz H . (1998). Mol Microbiol 28: 1067–1079.
Schrek R, Albermann K, Baeuerle PA . (1992). Free Radical Res Commun 17: 221–237.
Schulze-Osthoff K, Beyaert R, Vandervoorde V, Haegeman G, Fiers W . (1993). EMBO J 12: 3095–3104.
Sen CK, Roy S, Packer L . (1996b). FEBS Lett 85: 58–62.
Sen J, Venkataraman L, Shinkai Y, Pierce JW, Alt FW, Burakoff SJ et al. (1995). J Immunol 154: 3213–3221.
Serfling E, Avots A, Neumann M . (1995). Biochim Biophys Acta 1263: 181–200.
Severa M, D'Ambrosio A, Biordani L, Qintieri F, Coccia E . (2005). Biochem Pharmacol 69: 425–432.
Shikama Y, Yamada M, Miyashita T . (2003). Eur J Immunol 33: 1998–2006.
Shoji S, Furuishi K, Ogata A, Yamataka K, Tachibana K, Mukai R et al. (1998). Biochem Biophys Res Comm 249: 745–753.
Shumilla JA, Wetterhahn KE, Barchowski A . (1998). Arch Biochem Biophys 349: 356–362.
Singh S, Aggarwal BB . (1995). J Biol Chem 270: 10631–10639.
Sizemore N, Leung S, Stark GR . (1999). Mol Cell Biol 19: 4798–4805.
Song HY, Regnier CH, Kirschning CJ, Goeddel DV, Rothe M . (1997). Proc Natl Acad Sci USA 94: 9792–9796.
Song XR, Torphy TJ, Grisowold DE, Shealy D . (2002). Mol Interv 2: 36–46.
Swinney DC, Xu YZ, Scarafia LE, Lee I, Mak AY, Gan QF et al. (2002). J Biol Chem 277: 23573–23581.
Takada Y, Bhardwaj A, Potdar P, Aggarwal BB . (2004). Oncogene 23: 9247–9258.
Takigawa N, Vaziri SA, Grabowski DR, Chikamori K, Rhbicki LR, Bukowski RM et al. (2006). Anticancer Res 26: 1869–1876.
Tang W, Li Y, Yu D, Thomas-Tikhonenko A, Spiegelman VS, Fuchs SY . (2005). Cancer Res 65: 1904–1908.
Tepper MA, Nadler SG, Esselstyn JM, Sterbenz KG . (1995). J Immunol 155: 2427–2436.
Thoetkiattikul H, Beck MH, Strand MR . (2005). Proc Natl Acad Sci USA 102: 11426–11431.
Tomita N, Ogihara T, Morishita R . (2003). Curr Drug Targets 4: 603–608.
Torgerson TR, Colosia AD, Donahue JP, Lin YZ, Hawiger J . (1998). J Immunol 161: 6084–6092.
Trepicchio WL, Dorner AJ . (1998). Ann NY Acad Sci 856: 12–21.
Uchiba M, Okajima K, Kaun C, Wojta J, Binder BR . (2004). Thromb Haemost 92: 1420–1427.
Umezawa K, Ariga A, Matsumoto N . (2000). Anti-Cancer Drug Design 15: 239–244.
Umezawa K. (2006). Cancer Sci, in press.
Van Antwerp DJ, Martin SJ, Kafri T, Green DR, Verma IM . (1996). Science 274: 787–789.
van Hogerlinden M, Rozell BL, Åhrlund-Richter L, Toftgård R . (1999). Cancer Res 59: 3299–3303.
Venkataraman L, Burakoff SJ, Sen R . (1995). J Exp Med 181: 1091–1099.
Wagner S, Hofmann A, Siedle B, Terfloth L, Merfort I, Gasteiger J . (2006). J Med Chem 49: 2241–2252.
Wang C-Y, Mayo MW, Baldwin Jr AS . (1996). Science 274: 784–787.
Wechsler AS, Gordon MC, Dendorfer U, LeClair KP . (1994). J Immunol 153: 2515–2523.
Weiss T, Shalit I, Blau H, Werber S, Halperin D, Levitov A et al. (2004). Antimicrob Agents Chemother 48: 1974–1982.
Weissmann G . (1991). Hosp Pract 26: 60–76.
Weyrich AS, Denis MM, Kuhlmann-Eyre JR, Spencer ED, Dixon DA, Marathe GK et al. (2005). Circulation 111: 633–642.
Wolf AM, Wolf D, Rumpold H, Ludwiczek S, Enrich B, Gastl G et al. (2005). Proc Natl Acad Sci USA 102: 13622–13627.
Woronicz JD, Gao X, Cao Z, Rothe M, Goeddel DV . (1997). Science 278: 866–869.
Yang J, Merin JP, Nakano T, Kato T, Kitade Y, Okamoto T . (1995). FEBS Lett 361: 89–96.
Yao HW, Li J, Chen JQ, Xu SY . (2004). Acta Phamacol Sin 25: 915–920.
Yaron A, Gonen H, Alkalay I, Hatzubai A, Jung S, Beyth S et al. (1997). EMBO J 16: 6486–6494.
Yasuda H, Yamaya M, Sasaki T, Inoue D, Nakayama K, Yamada M et al. (2006). Eur Respir J 28: 51–58.
Yin M-J, Yamamoto Y, Gaynor RB . (1998). Nature 396: 77–80.
Yuan M, Konstantopoulos N, Lee J, Hansen L, Li ZW, Karin M et al. (2001). Science 293: 1673–1677.
Zandi E, Rothwarf DM, Delhase M, Hayakawa M, Karin M . (1997). Cell 91: 243–252.
Zhang S, Won YK, Ong CN, Shen HM . (2005). Curr Med Chem Anticancer Agents 5: 239–249.
Zheng B, Georgakis GV, Li Y, Bharti A, McConkey D, Aggrawal BB et al. (2004). Clin Cancer Res 10: 3207–3215.
Zhou H, Monack DM, Kayagaki N, Wertz I, Yin J, Wolf B et al. (2005). J Exp Med 202: 1327–1332.
We thank Jean-Charles Epinat for his contributions to the 1999 version of this article. We especially thank John Porco for help with classifying the inhibitors and Melissa Chin for help with Figure 1. MH was partially supported by a Pre-doctoral Fellowship from the Natural Sciences & Engineering Research Council of Canada. Research in our laboratory is supported by a grant from the National Institutes of Health (to TDG). For a continued updating of the lists of NF-κB inhibitors, the reader is referred to our lab website at www.nfkb.org (click on INHIBITORS).
About this article
Cite this article
Gilmore, T., Herscovitch, M. Inhibitors of NF-κB signaling: 785 and counting. Oncogene 25, 6887–6899 (2006). https://doi.org/10.1038/sj.onc.1209982
- NF-kappa B
- signal transduction
- cancer therapy
Phytochemistry Reviews (2020)
Thiazolidine-2,4-dione-based irreversible allosteric IKK-β kinase inhibitors: Optimization into in vivo active anti-inflammatory agents
European Journal of Medicinal Chemistry (2020)
Bright and Early: Inhibiting Human Cytomegalovirus by Targeting Major Immediate-Early Gene Expression or Protein Function
Oleanolic acid oxime derivatives and their conjugates with aspirin modulate the NF-κB-mediated transcription in HepG2 hepatoma cells
Bioorganic Chemistry (2019)
3D-QSAR and docking studies on ursolic acid derivatives for anticancer activity based on bladder cell line T24 targeting NF-kB pathway inhibition
Journal of Biomolecular Structure and Dynamics (2019)