beyond ikbs alternative regulation of nfkb activity 22

7/27/2019 Beyond IkBs Alternative Regulation of NFkB Activity 22

http://slidepdf.com/reader/full/beyond-ikbs-alternative-regulation-of-nfkb-activity-22 1/13

The FASEB Journal • Review

Beyond IBs: alternative regulation of NF-B activity

Manfred Neumann and Michael Naumann

Institute of Experimental Internal Medicine, Otto-von-Guericke University, Magdeburg, Germany

ABSTRACT The transcription factor nuclear factor-kappa B (NF-B) is a crucial regulator of many physi-ological and patho-physiological processes, including control of the adaptive and innate immune responses,inflammation, proliferation, tumorigenesis, and apoptosis. Thus, the tight regulation of NF-B activity within a cell is extremely important. The central mech-anism of NF-⟩ regulation is the signal-induced proteo-lytic degradation of a family of cytoplasmic inhibitorsof NF-B, the IBs. However, with the discovery of anIB-independent noncanonical or “alternative” pathway of NF-B activation, the importance of otherregulatory mechanisms responsible for the fine-tuning of NF-B became clear. Post-translational modifica-tion, especially phosphorylation, of the Rel proteins, of which dimeric NF-B is composed, are such alternativeregulatory mechanisms. The best analyzed example isRelA phosphorylation, which takes place at specificamino acids resulting in distinct functional changes of this gene regulatory protein. The interaction of NF-B with other proteins such as glucocorticoid receptors is very important for the regulation of NF-B activity.Recently, exciting new concepts of IB-independent NF-B control like dimer exchange and nucleolar se-questration of RelA have been described, indicating

that many aspects of NF-

B control are waiting to bediscovered.—Neumann, M., Naumann, M. BeyondIBs: alternative regulation of NF-B activity. FASEB J.21, 2642–2654 (2007)

Key Words: RelA phosphorylation ⅐ RelA acetylation ⅐ GR/ NF- B cross-talk ⅐ nucleolar sequestration ⅐ dimer exchange

nf-b is a pleiotropic transcription factor activated by a huge array of stimuli (1, 2), includinginflammatory cytokines recognized by specific mem-brane receptors such as tumor necrosis factor receptor(TNF-R) and microbial pathogens, which are recog-

nized by the Toll-like receptors (TLRs) as well asreceptors of the NACHT (domain present in NAIP,CIITA, HET-E, and TP1) -LRR (leucine-rich repeat)family (NLRs). The NLR family includes nucleotidebinding oligomerization domain (NOD) proteins andNACHT-, LRR-, and pyrin-domain-containing proteins(NALPs). These receptor proteins function in a similarmanner in that they recognize microbial componentsat the cell surface, in vesicles, and in the cytosol (3).Other important NF-B inducers are phorbol esterssuch as TPA, cytotoxic agents like chemotherapeuticdrugs, oxidative stress, UV light, and ionizing radiation.

NF-B is a dimer composed of the five mammalian Relproteins p65/RelA, c-Rel, RelB, NF-B1/p50, and NF-B2/p52 in almost any combination. The last two aresynthesized as precursor molecules, p105 and p100,respectively, and processing of these precursors isneeded to generate the mature transcription factors.Only RelA, c-Rel, and RelB, but not p50 or p52, possessC-terminal transactivation domains. All Rel proteinsshare the N-terminal Rel homology domain (RHD)mediating dimerization, DNA binding, and nuclearlocalization and interaction with the inhibitors of NF-B, the IBs. This family of mainly cytoplasmic inhibi-tors currently has eight members: IB␣, IB␤, IB␥,IBε, IB, Bcl-3, as well as p105 and p100. Theseprecursors possess ankyrin repeats in their C terminus,the characteristic structural feature of all IBs (1, 2).The IB-dependent NF-B activation pathways outlinedbelow certainly play the dominant role in controllingNF-B activity. However, the question remains: how isthe specificity of NF-B function achieved under cer-tain physiological conditions or in a specific cell type? It is the purpose of this review to demonstrate the com-plex mechanisms contributing to the fine-tuning of NF-B activity.

IB-DEPENDENT NF-B ACTIVATION

The IBs (inhibitors of NF-B) are the centerpiece of the canonical (classical) pathway (1, 4) (Fig. 1 A ). Inunstimulated cells, NF-B is complexed with the IBsand thereby locked into the cytoplasm. Upon stimula-tion of the cell, multiple intracellular signaling pathways are activated that converge at the IB kinase (IKK)complex. The most common form of this complexconsists of two functionally nonredundant kinases,IKK ␣ (IKK1) and IKK ␤ (IKK2), as well as a regulatory subunit, IKK ␥, also called NEMO (“NF-B essential

modulator”). Upon activation, the IKK complex phos-phorylates the IBs at specific amino acid residues. Thephosphorylation of IB␣ at Ser-32 and Ser-36 is pre-dominantly mediated by IKK ␤. This site-specific phos-phorylation is a prerequisite for a subsequent post-translational modification, the ubiquitinylation of IB␣, which tags the NF-B inhibitor for degradation in

1Correspondence: Institute of Experimental Internal Medi-cine, Otto-von-Guericke University, Medical Faculty, LeipzigerStrasse 44, 39120 Magdeburg, Germany. E-mail: [email protected]

doi: 10.1096/fj.06-7615rev

2642 0892-6638/07/0021-2642 © FASEB



the 26S proteasome. NF-B is now free to translocateinto the nucleus, where it regulates the expression of genes involved in fundamental physiological andpatho-physiological cellular processes such as control of the immune system, especially of the innate immuneresponse, as well as the regulation of inflammation andapoptosis. Various MAP3kinases (mitogen-activated

protein kinase kinase kinases) located immediately upstream of the IKK complex have been thought to bekey components of IKK activation. However, recent studies have demonstrated that the binding of NEMOto Lys-63-linked polyubiquitinylated RIP1 is the essen-tial step in IKK activation mediated by TNF-R1 (5–7).This finding provided a key step to understanding thelink between cytokine receptor proximal signaling andthe activation of IKK/NF-B.

The canonical NF-B activation pathway outlinedabove is generally regarded to be the main mechanismby which NF-B activity is regulated. However, this

fundamental pathway is not the only one controllingNF-B. Recently, a noncanonical (alternative) signalingpathway of NF-B activation was discovered that involves processing of the p52 precursor p100 (1, 4) (Fig.1B ). This pathway is completely independent of IKK ␤or NEMO, whereas IKK ␣ and the NF-B-inducing ki-nase (NIK) are essential. NIK is positioned immediately

upstream of the IKK ␣ homodimer. Cytokines such aslymphotoxin ␤ (LT-␤) and B cell-activating factor,CD40 ligand, as well as viruses such as the human T cellleukemia virus 1 (HTLV-1) or the Epstein-Barr virus(EBV), are among the few select stimuli able to activatethe noncanonical pathway. The phosphorylation of p100 at two specific C-terminal serine residues by IKK ␣homodimers is a key event in this pathway. Again, thissite-specific phosphorylation is essential for polyubiq-uitinylation and proteasomal degradation. However,the entire molecule is not degraded; only the C termi-nus of p100 is destroyed. RelB/p52 heterodimers are

Figure 1. Schematic representation of four IB-dependent NF-B pathways. A) The canonical (“classical”) pathway depends onthe site-specific phosphorylation of IBs by activated IKK complexes. The TNF-R1-mediated NF-B activation is shown as anexample. Activation of NF-B takes place near the membrane. One recently discovered critical step—the interaction of NEMO

with Lys-63 polyubiquitinylated RIP1—is shown. It should be noted that the TNF-RI complex contains more components. Someof them, such as the transforming growth factor beta-activated kinase 1 (TAK1) and the TAK1 binding protein 2 (TAB2), whichalso bind polyubiquitinylated RIP1, are able to interact with and activate the IKK complex. B) The noncanonical (alternative)pathway based on the processing of the p100 precursor. NIK and IKK ␣, but not IKK ␤, are critical kinases. Mainly RelB/p52NF-B dimers are generated. C) The unique NF-B activation pathway induced by UV light and a few select chemotherapeutic

agents. The MAP kinase p38 and CKII are needed to phosphorylate IB␣ at sites different from the amino acid residues usedin the canonical pathway. These phosphorylation events are a prerequisite for signal-induced IB␣ degradation. D) A stress-activated pathway based on the sequential, nuclear modification of NEMO. Nuclear NEMO first becomes sumoylated,mediated by PIDD and RIP1, then phosphorylated by the stress sensor ATM kinase. This phosphorylation is a prerequisite formono-ubiquitinylation of NEMO, which serves as nuclear export signal. ATM is also translocated to the cytoplasm along withNEMO interacting with and activating IKK.

2643IB-INDEPENDENT NF-B REGULATION



the main NF-B factors generated by p100 processingso that not only the signals inducing this pathway, but also its products, are specific.

The mechanism of UV light-induced NF-B induc-tion is unique (Fig. 1C ). Exposure of cells to UV light leads to the degradation of IB␣, but this phosphory-lation-dependent process is not mediated by the IKK complex (8). It could be demonstrated that activationof a p38-MAPkinase-dependent pathway is critical forUV-induced IB␣-degradation. One of the main targetsof this pathway is casein kinase II (CKII), a serine/threonine kinase that can phosphorylate IB␣ at acluster of C-terminal sites. Like the IKK-mediated N-terminal phosphorylation, this post-translational IB␣modification induces degradation of this NF-B inhib-itor. Thus, CKII is a stress-activated rather than aconstitutive IB kinase playing a crucial role in UV protection by inducing NF-B-dependent antiapoptoticgene expression (8).

UV light can induce NF-B activation in a DNA damage-independent manner. In contrast, other DNA-damaging stimuli such as ionizing radiation or chemo-

therapeutic drugs also induce an NF-B response.However, this type of NF-B activation is not only induced by DNA damage, particularly the formation of double strand breaks (genotoxic stress), but is alsofound to be associated with other stress conditions,such as oxidative stress as well as heat or electric shock(9) (Fig. 1 D ). Recent studies have shed new light ontothese “nuclear-to-cytoplasmic” pathways of NF-B activation (4, 10, 11). The regulatory IKK component NEMO and its tripartite post-translational modificationstand in the center of this NF-B activation pathway.Upon cellular stress, NEMO is sumoylated within thenucleus. This means it is modified by covalent attach-

ment of the small ubiquitin-like modifier (SUMO).PIASy (protein inhibitor of activated STATy) was re-cently identified to be the NEMO-specific SUMO ligase(12). Sumoylation is increased in the presence of PIDD(p53-inducible death domain-containing protein), aprotein implicated in cell cycle control. PIDD forms acomplex with NEMO and RIP1 (11, 13). Then a secondataxia telangiectasia-mutated (ATM) -dependent pathway results in the phosphorylation of NEMO andsubsequently its mono-ubiquitinylation. The last modification does not mark NEMO for proteasomaldegradation, but rather is a signal for cytoplasmictranslocation, where modified NEMO activates the

IKK complex.The NF-B activation pathway originating from DNA

damage described above is an excellent example of thefact that protein ubiquitinylation has emerged as animportant intracellular mechanism regulating variousbiological processes (14, 15). Ubiquitin is a highly conserved 76 amino acid polypeptide that can becovalently attached to a lysine residue of a target protein by an E3-ubiquitin ligase, which is primarily responsible for providing substrate specificity. Subse-quent to the attachment of the first ubiquitin, polyu-biquitin chains are often generated by the ligation of

multiple ubiquitin moieties. Ubiquitin has seven lysineresidues, of which Lys-48 and Lys-63 are used forgenerating most polyubiquitinylation chains. WhereasLys-48-linked polyubiquitinylation is a marker for pro-teasomal degradation, the attachment of Lys-63-linkedpolyubiquitin chains regulates protein kinase activitiesand protein trafficking (7).

The zinc finger protein A20 functions as a ubiq-uitin editor and a potent NF-B inhibitor; both

functions are connected (16). The central roleplayed by A20 in regulating NF-B is demonstratedby the fact that A20-deficient mice develop severeinflammation and cachexia, are hypersensitive toLPS as well as TNF-␣, and die prematurely (17). A20interferes with TNF-␣-induced NF-B activation up-stream of the IKK complex (18). A20 interacts withTRAF2 (TNF receptor-associated factor 2) andNEMO, both critical signaling components of thepathway originating from the TNF receptor. Other A20 binding proteins such as ABIN-1 (A20 bindinginhibitor of NF-B 1) also bind to NEMO andcooperate with A20 in inhibiting NF-B (19). How-

ever, the critical A20 target for inhibition of TNF-␣-induced NF-B activation seems to be the serine/threonine protein kinase RIP1 (receptor-interactingprotein 1), which also interacts with A20. RIP1 isactivated by TRAF2-driven Lys-63-linked polyubiquiti-nylation. A deubiquitinylation activity located in theso-called OTU (ovarian tumor) domain at the Nterminus of A20 can remove these polyubiquitinchains. The subsequent attachment of Lys-48-linkedpolyubiquitin chains to RIP1 is mediated by the Cterminus of A20, which encompasses seven novel zincfinger structures. This domain functions as an E3-ubiquitin ligase. RIP1 is marked for proteasomal

degradation by the latter post-translational modifica-tion interrupting the NF-B activation pathway (18).Thus, A20 is an astonishing molecule combining twodistinct, but connected, functions of ubiquitin edit-ing.

IB-INDEPENDENT REGULATION OF NF-B:POST-TRANSLATIONAL MODIFICATIONS OF REL PROTEINS

Not only the IB inhibitors, but also the Rel proteinsthemselves, are subject to signal-induced post-transla-tional modifications (20, 21) (Fig. 2). These modificationsalter various physiological functions of the NF-B factors.The most important post-translational modification isphosphorylation, and the basal and signal-induced phos-phorylation of RelA is by far the best-characterized mod-ification. Site-specific phosphorylation is often a prereq-uisite for other modifications regulating RelA activity.Such subsequent post-translational modifications includeacetylation, ubiquitinylation, and isomerization of specificamino acid residues.

2644 Vol. 21 September 2007 NEUMANN ET AL.The FASEB Journal



REGULATION OF NF-B ACTIVITY BY MODIFICATIONS OF RelA

The first description of an inducible phosphorylationof RelA in vivo already suggested an important mecha-nism in the control of NF-B activity (22, 23). Today,nine putative sites of RelA phosphorylation are known:six serine and three threonine residues (Fig. 2). Three

of these sites (Ser-276, Ser-311, and Thr-254) are lo-cated within the N-terminal RHD, whereas six acceptorsites (Ser-468, Ser-529, Ser-535, Ser-536, Thr-435, andThr-505) are found within the C-terminal transactiva-tion domain. The effects of RelA phosphorylation at these specific sites vary dramatically, ranging fromtranscriptional activation to the complete repression of certain genes. The pattern of RelA phosphorylation ismost likely stimulus- and cell type-specific. It becomesmore and more apparent that RelA phosphorylation at different sites serves as an integrator for multipleincoming signals, which could control both the kinetics

and strength of the transactivating potency of RelA. We will now look at each RelA modification, its functionalconsequences, and biological significance. Since many reviews comprehensively describe RelA phosphoryla-tion (20, 21), the focus here is laid on the most recently discovered phosphorylation events. We start with serinephosphorylation, specifically with the most C-terminalphosphorylation, followed by threonine phosphoryla-

tion, ubiquitinylation, and finally acetylation of RelA.The evolutionary conserved Ser-536 of RelA is the

target of five known protein kinases representing fivedifferent signaling pathways in vitro and/or in vivo . Inaddition to IKK ␣/␤ (24–26), these protein kinasesinclude IKK ε and TBK1 (TANK binding kinase 1) (27,28). Ser-536 can be phosphorylated by IKK ε after T cellcostimulation (29). Both TBK1 and IKK ε show se-quence homology to IKK ␣/␤, but are not componentsof the IKK complex. DNA-damaging agents, such as theanticancer drugs doxorubicin or etoposide, activateNF-B through Ser-536 phosphorylation of RelA medi-

Figure 2. Compilation of post-translational modifications of Rel proteins and their functional consequences. Œ ϭ increase,up-regulation; ϭ decrease, down-regulation.




ated by ribosomal S6 kinase 1 (RSK1) involving ap53-dependent pathway (30). The physiological conse-quence is a weaker interaction, with nuclear IB␣resulting in a drastic decrease of nuclear RelA export and thus enhanced RelA-dependent transcription. Inline with this observation, a recent study claims that Ser-536-phosphorylated RelA does not interact at all with cytosolic IB␣ and that this RelA species regulatesa distinct set of target genes emphasizing the impor-tance of the Ser-536 phosphorylation (31).

Ser-535 of RelA is specifically targeted by calmodulin-dependent kinase IV (CaMKIV) (32, 33). This phos-phorylation results in the activation of gene expression,especially of antiapoptotic genes. This is in line with theobservation that CaMKIV activity, like NF-B, has anti-apoptotic and proproliferative effects, suggesting that NF-B is a potential effector of CaMKIV (33).

CKII phosphorylates Ser-529 of RelA upon stimula-tion with TNF-␣ or IL-1␤ (34, 35). Ser-529 phosphory-lation is dependent on IB degradation, and thus takesplace exclusively within the cytoplasm. This RelA mod-ification appears to play a minor role in regulating

RelA-dependent transactivation and appears to be rel-evant only for an optimal cellular response induced by HTLV-1 infection (25).

Phosphorylation of the serine residue at position 468located in the TAD2 of RelA appears to have opposingeffects. RelA phosphorylated at Ser-468 is located pre-dominantly within the nucleus, whereas most other versions of phosphorylated RelA are found mainly inthe cytoplasm, indicating a function for this specificphosphorylation in the control of RelA-dependent transactivation. This site is the target of three proteinkinases: IKK ␤, IKK ε, and GSK-3␤ (36–38). GSK-3␤mediates basal Ser-486 phosphorylation inhibiting the

basal activity of RelA (38). In line with this, expressionof a constitutive-active form of GSK-3␤ suppressesNF-B activity (39, 40). However, a contradictory find-ing appears to be of greater physiological relevance:NF-B cannot be activated in cells derived from GSK-3␤-deficient embryos despite an unaltered profile of IB␣ degradation, which suggests a positive regulatory role of GSK-3␤ in NF-B activation (41). A positiveimpact on RelA activity was also observed for theIKK ε-mediated Ser-468 phosphorylation on T cell co-stimulation (37), whereas the IKK ␤-dependent phos-phorylation induced in HeLa cells stimulated withTNF-␣ or IL-1␤ had a negative effect on RelA-mediated

transactivation (36). This contradiction might be ex-plained by the influence of other phosphorylationevents, which differ significantly in both experimentalsystems.

Phosphorylation of both serine residues 311 and 276located within the N-terminal RHD of RelA results in anincrease in transactivation potency. Ser-311 is phos-phorylated by PKC upon TNF-␣ stimulation, leadingto an enhanced interaction of Ser-311-phosphorylatedRelA with the cAMP response element binding (CREB)binding protein (CBP) and its recruitment to theinterleukin 6 (IL-6) promoter together with RNA poly-

merase II (42). Ser-276 is phosphorylated exclusively within the cytoplasm by the catalytic subunit of proteinkinase A (PKAc) (43). Since PKAc is complexed to andinactivated by IB␣, the stimulus-induced degradationof IB␣ is a prerequisite for the activation of PKAc.Ser-276 is also targeted by the mitogen- and stress-activated protein kinase-1 (MSK1) (44). In this case, thephosphorylation takes place exclusively within the nu-cleus. Thus, a single serine residue is phosphorylated by different protein kinases in distinct cellular compart-ments. In both cases, RelA phosphorylation results inan enhanced recruitment of histone acetyltransferases(HATs) such as CBP and p300 and to increased dis-placement of inhibitory p50/histone deacetylase-1(HDAC1) complexes from the DNA (45). In addition,Ser-276-phosphorylated RelA can interact with RelB inthe nucleus (46). These heterodimers cannot bindDNA, and thus represent inactive NF-B. The physio-logical consequence is the inhibition of RelB-mediatedNF-B activity.

RelA can also be phosphorylated at three specificthreonine residues. Two are located within the C-

terminal TADs (Thr-435 and Thr-505) and one (Thr-254) is found at the N-terminal RHD of RelA (47–49).Phosphorylation of each of the two C-terminal aminoacids has a suppressive effect on RelA activity. Phos-phorylation of Thr-505 is induced by the ARF tumorsuppressor (human p14 ARF). ARF induces RelA phos-phorylation via activation of the ATM/Rad3-related(ATR) checkpoint kinase and checkpoint kinase 1(Chk1) in an p53-independent fashion. Phosphoryla-tion of Thr-505 results in an increased association of RelA with HDAC1, and this complex formation isresponsible for the ARF-induced inhibition of RelA activity. In osteosarcoma cells, an induction of the

inhibitory Thr-505 phosphorylation of RelA is observedupon treatment with the chemotherapeutic drug cispla-tin. Since it uses the same signaling pathways as ARF,cisplatin can be regarded as an ARF mimic (49).

Phosphorylation of Thr-254 leads to the activation of RelA by a novel mechanism. The nuclear peptidyl-prolyl isomerase Pin1 [protein-interacting NIMA (never in mitosis arrest) 1] binds and isomerizes spe-cific phosphorylated serine or threonine residues that precede a proline residue (50). These Pin1-inducedconformational changes have profound functional ef-fects on the target proteins. One of these targets isRelA, which is bound by Pin1 at phosphorylated Thr-

254-Pro-255. Binding precedes the isomerization of theproline residue. This alteration of RelA structure re-sults in a strong decrease in its affinity to IB␣, whereasthe binding to p50 remains unchanged. Further, phos-phorylation-dependent Pin1/RelA interaction in-creases the nuclear translocation of RelA and, most important, results in stabilization of RelA, as demon-strated by the extreme instability of the T254A mutant of RelA. The counterpart of Pin1 in the regulation of RelA stability is the E3-ubiquitin ligase suppressor of cytokine signaling 1 (SOCS-1), which has been shownto ubiquitinylate RelA, thereby inducing its proteolysis.




Pin1 and SOCS-1 bind RelA at sites extremely close toone another, making it likely that both proteins com-pete for the same binding site (50). The role of SOCS-1in RelA ubiquitinylation was confirmed in a recent study demonstrating that COMMD1 (copper metabo-lism gene MURR1 domain containing 1) acceleratesubiquitinylation and degradation of RelA through itsinteraction with a multimeric ubiquitin ligase contain-ing Elongins B and C, cullin 2 (Cul2), and SOCS-1(ECS(SOCS-1)) (51). COMMD1 binds to Cul2 in astimulus-dependent manner, thereby strengtheningthe interaction between SOCS-1 and RelA. The impor-tant role of COMMD1 for RelA ubiquitinylation isstressed by the finding that RelA is stabilized inCOMMD1-deficient cells (51). However, it is still amatter of discussion whether SOCS-1 is the only ubiq-uitin ligase mediating ubiquitinylation of RelA (7).Functionally, RelA ubiquitinylation is a potential mech-anism regulating the termination of NF-B-mediatedinflammatory responses (52) and may also contributeto the oscillation of nuclear NF-B activity (53).

The finding that the phosphorylation status of Relproteins, especially of RelA, is the decisive parameterfor their association with either activating HATs (e.g.,CBP/p300) or suppressing HDACs was a hallmarkdiscovery in the field of NF-B regulation (45) (Fig.3 A ). RelA is also reversibly modified by these enzymes.Upon TNF-␣ stimulation, RelA is acetylated at fivespecific lysine residues: Lys-122, -123, -218, -221, and-310 (54, 55). Modification of Lys-218 and Lys-221 weakens the interaction of RelA with IB␣ and in-creases RelA DNA binding (56). Acetylation of Lys-310strongly enhances the transactivation potency of RelA (56), whereas deacetylation by histone deacetylase 3(HDAC3) or SIRT1 (sirtuin 1) inhibits the transcrip-tional activity of NF-B and augments TNF-␣-mediatedapoptosis (56, 57). Stimulation with TPA specifically acetylates RelA at Lys-122 and -123. PCAF (p300/CBP-associated factor) and p300 are involved in these post-translational modifications, which decreases RelA-spe-cific DNA binding and facilitates its nuclear export (58). The most important aspect of these studies of

Figure 3. Changes in NF-B-dependent transcription induced by the interaction with GR. A) All cytoplasmic and nuclear kinasesthat can phosphorylate RelA upon TNF-RI stimulation are shown. Only phosphorylated RelA is able to interact with coactivatorssuch as CBP/p300. The recruitment of such coactivators changes the chromatin structure by histone acetylation and facilitatesassembly of the basic transcription apparatus. DNA-bound RelA can induce an activating phosphorylation of RNA polymeraseII. B) The interaction of activated, monomeric GR with DNA-bound NF-B results in the active recruitment of corepressors suchas HDAC2, whose expression is also up-regulated. The activity of CBP/p300 is suppressed by the GR/NF-B complex. Further,histone 3 methylation adds to the inactive chromatin state. Finally, the activating RNA polymerase II phosphorylation isinhibited. hsp, heat shock protein.




RelA acetylation is that modification of specific lysineresidues regulates distinct functions of NF-B. Thus,modification of RelA by site-specific acetylation is an-other mechanism whereby different signals controllingNF-B activity are integrated.

It was recently demonstrated that decreased phos-phorylation of Ser-276 or Ser-536 by ectopic expressionof enzymatically inactive forms of PKAc/MSK1 (target:Ser-276) or IKK ␣/IKK ␤ (target: Ser-536), respectively,drastically reduced RelA acetylation at Lys-310. In line with this, reconstitution of RelA-deficient murine em-bryonic fibroblasts with mutated forms of RelA (RelA S276A or S536A) resulted in a sharp decrease of RelA acetylation. Reconstitution with wild-type RelA led to anincreased assembly of phospho-RelA with p300 and, asa consequence, to an up-regulation of Lys-310 acetyla-tion compared with cells expressing mutant RelA (59).Phosphorylation of DNA-bound RelA at Ser-536 occursprior to RelA acetylation at lysine 310. In parallel, therepressor SMRT (silencing mediator for retinoic acidand thyroid hormone receptor), which is bound toRelA/p50 heterodimers, is also phosphorylated and

thereby derepressed (55). Both phosphorylation eventsare mediated by nuclear IKK ␣ recruited to the chroma-tin. The derepression of SMRT potentiates RelA acety-lation by p300, leading to a further increase in NF-Bactivity. In line with these findings, histone deacetylaseinhibitors induce phosphorylation and thus the activa-tion of p300 (54).

REGULATION OF NF-B ACTIVITY BY MODIFICATIONS OF RelB

RelB is in many respects an unusual member of the

Rel/NF-B family. For example, RelB preferentially binds to p50 and p52, although an inhibitory interac-tion with RelA and homodimerization have been dem-onstrated. Further, it could be shown that the RelBmonomer is an extremely unstable protein that isalmost instantly degraded (60).

RelB is phosphorylated at Thr-84 and Ser-552 (61)(Fig. 2). These phosphorylation events have functionalconsequences that are contrary to the effects of RelA phosphorylation. RelB phosphorylation is induced by specific signals and is a prerequisite for the subsequent proteasomal degradation of RelB. The inducing signalsare T cell receptor activation by CD3/CD28-specific

antibodies or TPA/ionomycin stimulation, which issupposed to mimic T cell receptor activation. At first sight, induced proteolysis of RelB resembles the signal-induced degradation of the IBs. However, there arealso differences between the two processes. The phos-phorylation-dependent RelB degradation takes place inat least two distinct steps: an N-terminal processing isfollowed by the complete proteasomal degradation of the protein. Further, RelB degradation is stimulusspecific and cannot be induced by cytokines like TNF-␣or IL-1 (61).

Another prominent post-translational modification

of RelB is Ser-368 phosphorylation. A murine plasma-cytoma cell line lacking endogenous RelB expression was substituted with wild-type and Ser-368-mutatedRelB (62). The authors observed that Ser-368 phos-phorylation is not required for the nuclear transloca-tion of RelB, but rather appears to be critical for RelBdimerization with the p52 precursor p100. RelB phos-phorylated at Ser-368 stabilizes p100, suggesting a rolein regulating the noncanonical pathway of NF-B activation. As in the case of the two other RelB phosphorylation sites, the kinase (or kinases) mediating Ser-368phosphorylation has not been identified (62).

REGULATION OF NF-B ACTIVITY BY MODIFICATIONS OF C-REL

Regulation of c-Rel activity is achieved to a large extent by its signal-induced phosphorylation (Fig. 2). An ini-tial hint stressing the importance of c-Rel phosphoryla-tion came from a study reporting that tyrosine phos-phorylation can be rapidly induced by G-CSF

(granulocyte colony-stimulating factor) in human neu-trophils (63). However, neither the protein tyrosinekinase mediating the G-CSF effect nor the target ty-rosine residue (or residues) has been identified. Six years later, the complex formation of a 65 kDa serine/threonine protein kinase with murine c-Rel was re-ported (64). This kinase apparently has two bindingsites and mediated C-terminal phosphorylation of c-Rel. Again, the specific phospho-acceptor sites within c-Reland the protein kinase responsible for c-Rel phosphorylation have not been identified.

Two prominent serine/threonine signaling kinases,PKC and the C␤ isoform of PKA, have been shown to

phosphorylate c-Rel (65, 66). Both phosphorylationevents have the same result: they enhance the transac-tivation potential of c-Rel. The target serine/threonineresidue of PKA is still unknown, whereas PKC phos-phorylates Ser-471 of c-Rel in Jurkat T cells stimulatedby TNF-␣. The signaling pathway leading to this c-Relmodification also involves the PI3K/AKT kinase lo-cated upstream of PKC (67). The mutation of eitherSer-471 or Ser-460 to a nonphosphorylatable residueresulted in a version of c-Rel with higher transformingpotency, whereas the transactivation ability was not influenced (68). However, the expression of selectedgenes was differentially affected by mutated vs. wild-

type c-Rel, indicating that phosphorylation inhibits thetransforming potency of this proto-oncoprotein by en-abling it to correctly express target genes.

A dysregulated phosphorylation of c-Rel might alsocontribute to the development of certain human B celllymphomas. In two patients with different types of Bcell lymphoma, point mutations affecting Ser-525 weredetected (69). The mutant c-Rel proteins display al-tered transactivation properties depending on the celltype analyzed, and the potency of mutant c-Rel totransform chicken spleen cells was drastically en-hanced.




Phosphorylation of specific sites within c-Rel as wellas of RelA appears to be crucial for the resolution of inflammation. A clear division of tasks between IKK ␤and IKK ␣ was noted in a study analyzing NF-B activa-tion in macrophages during inflammatory processes(52). IKK ␤ mediates NF-B activation in response toproinflammatory cytokines and microbial products, whereas IKK ␣ contributes to the termination of inflam-mation by accelerating the turnover of c-Rel as well asRelA in a phosphorylation-dependent manner. BothRel proteins have been shown to be ubiquitinylated andsubsequently proteolytically degraded. This study dem-onstrated that IKK ␣ is able to phosphorylate c-Rel andRelA, and thereby induce their degradation. The serineresidue 536 of RelA appears to be one IKK ␣ target, whereas the phospho-acceptor site of c-Rel was not identified in detail. It is known only that phosphoryla-tion takes place at the C terminus of c-Rel.

Recent studies identified NIK, TBK1, and IKK ε asc-Rel-specific kinases (70, 71). TBK1 and IKK ε phos-phorylate c-Rel at its C terminus with the functionalconsequence of an increased nuclear c-Rel accumula-

tion. The degradation of I

B␣

is not affected, but rather IKK ε-mediated phosphorylation leads to dissoci-ation of the IB␣-c-Rel complex.

REGULATION OF NF-B ACTIVITY BY MODIFICATIONS OF p50/NF-B1

Similar to the IBs, p105, the precursor of the Relprotein p50, is site-specifically phosphorylated, thenpolyubiquitinylated and proteolytically degraded in the26S proteasome. Although it has been known for many years that p50 as well as its p105 precursor are inducibly

phosphorylated upon stimulation of various cell types(23), reports about site-specific phosphorylation of themature p50 and the functional consequences of such apost-translational modification are rare. So far, thefinding that the catalytic subunit of PKA (PKAc) caninducibly phosphorylate Ser-337, which is located within the RHD of p50, stands alone. Since p50 has notransactivation domains, the only functional conse-quence appears to be an enhanced p50 DNA binding(72).

IB-INDEPENDENT REGULATION OF NF-B:

CROSS-TALK BETWEEN NF-B AND OTHER PROTEINS

Complex formation of NF-B with transcriptional reg-ulators such as AP-1 (activator protein 1), SP1 (speci-ficity protein 1), and C/EBP␤, termed cross-couplingor cross-talk, is important for the control of NF-Bactivity. Cross-talk is defined as the interaction betweendifferent transcription factors resulting in either thesynergistic activation or antagonistic inhibition of geneexpression. The interaction of NF-B with AP-1 leads toactivation of both transcription factors, whereas NF-

B/C/EBP␤ complex formation leads to the enhance-ment of C/EBP␤ activity but to the repression of NF-Bvia inhibition of RelA phosphorylation (73). A recent study showed that interaction of RelA with breast cancer-associated protein 3 (BCA3) results in the tran-scriptional repression of NF-B (74). This complexformation is dependent on neddylation of BCA3. Ned-dylated BCA3 can recruit SIRT1, a histone deacetylasethat might be responsible for the suppression of NF-Bactivity.

The binding of NF-B to glucocorticoid receptors(GR), ligand-induced transcription factors belongingto the nuclear receptor superfamily, is an excellent example of the regulation of NF-B activity by cross-coupling to unrelated proteins (75, 76) (Fig. 3B ).Glucocorticoids are potent immunosuppressants that work mainly by inhibiting cytokine gene expression.Surprisingly, it could be shown that the DNA bindingand transactivating ability of GR are not needed tomediate most of these anti-inflammatory effects. It ismainly the interaction with NF-B that finally leads todown-regulation of the transcription of cytokine genes

due to GR-mediated NF-B repression. This inhibitionresults in decreased transactivation potency, DNA bind-ing, and nuclear localization of NF-B.

The precise molecular mechanisms mediating theinhibiting effects of glucocorticoids on NF-B vary with the cell type and depend on the availability of specific cofactors, the promoter context of a giventarget gene, the status of the chromatin, the so-calledhistone code, and other parameters. The general im-portance of cytoplasmic mechanisms of GR-mediatedNF-B transrepression, such as an glucocorticoid-in-duced increase in IB␣ levels, is still a matter of controversy (77). In most cases, gene repression by

activated GR takes place through interaction with DNA-bound NF-B. GR DNA binding and homodimerizationare not required for this process (78). Evidence of acompletely nuclear mechanism of glucocorticoid re-pression of specific promoters, such as the IL-8 or theICAM promoter, came from in vivo footprinting andchromatin immunoprecipitation experiments showingthat the pattern of RelA/NF-B binding remains un-changed under repression conditions (79, 80).

One crucial nuclear mechanism is based on changesin the chromatin environment of the regulatory ele-ments of the respective genes called chromatin remod-eling (75). In the resting cell, DNA is tightly wrapped

around a protein core. This chromatin structure iscomposed of nucleosomes, which are particles consist-ing of ϳ146 bp DNA associated with an octamer of twomolecules each of core proteins (histones H2A, H2B,H3, and H4). This “closed” chromatin structure can beloosened by modification of the N-terminal tails of thecore histones enabling access of RNA polymerase II andthe rest of the transcription machinery. These modifi-cations include methylation, ubiquitinylation, phos-phorylation, and (of special importance) acetylation. It is known that NF-B activated by TNF-␣ or IL-1␤ caninduce histone acetylation preferentially on histone H4




at lysine residues 8 and 12. This takes place at NF-B-responsive regulatory elements. Upon DNA binding,NF-B recruits large coactivator complexes containingthe histone acetylase (HAT) proteins CBP and p300. Inoverexpression experiments, PCAF was also shown tobe critical for the RelA-associated increase in HATactivity. Further, several other HATs are known tointeract with NF-B such as transcriptional intermedi-ary factor 2 (TIF-2), aka GR interacting protein 1, andsteroid receptor coactivator 1 (SRC-1). In addition tothe site-specific acetylation of histone H4, the phos-phorylation of histone H3 at Ser-10 is associated withgene induction. Surprisingly, this phosphorylation wasfound to be mediated by nuclear IKK ␣, which is re-cruited to the promoters of NF-B-regulated genes inassociation with CBP and RelA upon stimulation withTNF-␣ (81, 82).

The interaction between transcription factors, suchas NF-B and GRs, may result in differential effects onhistone acetylation/deacetylation (75). Several mecha-nisms, which are probably not exclusive, are currently discussed. Activated GRs can directly modify the chro-

matin structure. It is known that low glucocorticoidconcentrations are able to repress TNF-␣-induced his-tone H4 acetylation via direct inhibition of CBP-associ-ated HAT activity. In addition, high concentrations of glucocorticoids lead to an increase in HDAC2 expres-sion and the activated GR can be recruited to activatedinflammatory gene complexes, switching off NF-B-driven gene expression (83). Using a bioluminiscenceresonance energy transfer assay, it was shown that treatment with the GR antagonist RU486 efficiently recruited the corepressor NcoR (nuclear corepressor)to GR (84). These and other findings led to theconclusion that suppression of NF-B-driven gene ex-

pression may involve recruitment of a fully functionalcorepressor complex that minimally contains HDAC2and NcoR (76).

The GR/NF-B cross-talk can also change the chro-matin structure by methylating histone H3. Thischange of the “histone code,” the sequence of histonemodifications within a distinct chromatin area, might be critical for switching off NF-B-mediated inflamma-tory gene transcription. Another level of GR actionseems to be the reduction of an activating C-terminalserine phosphorylation of RNA polymerase II inducedby DNA-bound NF-B (80). This phosphorylation ismediated by the kinase pTEFb and promotes glucocor-

ticoid-mediated repression (85). This indicates that theGR/NF-B cross-talk has effects downstream of NF-BDNA binding and even downstream of coactivatorfunction (Fig. 3B ).

NUCLEOLAR SEQUESTRATION AND DIMER EXCHANGE: NEW CONCEPTS IN IB-INDEPENDENT REGULATION OF NF-B

The compartmentalization of transcription-associatedproteins within nuclear bodies is increasingly recog-

nized as an important mechanism for the control of gene expression, proliferation, and apoptosis. Twoprominent examples are regulation of the p53 tumorsuppressor gene by the nucleolar sequestration of itsinhibitor MDM2 (86, 87) and the inhibition of c-myc -induced cell cycle progression by sequestration of thisproto-oncoprotein to the nucleolus (88). With regardto the NF-B system, NIK (89) and the transactingNF-B-repressing factor (NRF) (90) have been shownto shuttle between the nucleoplasm and nuclear bod-ies. Furthermore, two prominent nucleolar proteins,NF-B binding protein (NFBP) (91) and nucleophos-min/B32 (92), strongly interact with NF-B, indicatinga possible role for nucleolar sequestration of NF-B inregulating NF-B activity.

The final proof of this hypothesis came from a study demonstrating that RelA translocates into the nucleo-lus of colon cancer cells in response to proapoptoticstimuli like aspirin, serum withdrawal, and UV-C radia-tion (93). In contrast, RelA is excluded from thenucleolus in response to the cytokines TNF-␣ or TRAIL(TNF-related apoptosis-inducing ligand). Nucleolar

RelA sequestration is functionally accompanied by asharp decrease in NF-B transcriptional activity and astrong increase in apoptosis. An N-terminal motif seemsto be responsible for nucleolar translocation of RelA,but the exact molecular mechanism has yet to beidentified. The current model suggests the existence of an unknown nuclear cofactor (corepressor) that selec-tively associates with the cytoplasmic NF-B pool activated by proapoptotic stimuli such as aspirin. A secondNF-B population, the basal pool, constitutively drivesthe transcription of antiapoptotic genes. The inducedRelA/corepressor complex does not activate transcrip-tion but interacts with basal NF-B; subsequently, all

nuclear RelA translocates into the nucleolus. As aconsequence, NF-B is in a cellular location different from its target promoters, resulting in a decreasedtranscription of antiapoptotic genes and finally in apo-ptosis. In contrast to TNF-␣ and TRAIL, proapototicstimuli induce NF-B with a delay, which might beresponsible for the selective association of the hypo-thetical cofactor with this NF-B pool (93).

Dimer exchange at specific NF-B-dependent pro-moters represents a new and intriguing concept for themodulation of NF-B activity. There are two types of NF-B-regulated promoters. One type is selectively con-trolled by only one specific dimer, whereas other

promoters are regulated by more than one NF-Bcomplex. Using dendritic cell maturation as a modelsystem, it could be demonstrated that rapidly activateddimers like p50/RelA are replaced by more slowly activated dimers such as p52/RelB at a subset of promoters (94). p52/RelB dimers are less stringently controlled by IBs and are less sensitive to resynthe-sized IBs, enabling them to mediate a sustained activation. Thus, dimer exchange contributes to both thefine-tuning of the NF-B response as well as a pro-longed activation at specific NF-B target promoters.Further, it was shown that each NF-B complex has a




specific transcriptional activity at a specific promoter, which is not simply due to affinity variations of different NF-B dimers. At some promoters, a complete replace-ment of specific dimers in an ordered sequence wasnotable, whereas multiple dimers were recruited toothers without an obvious temporal order. The occur-rence of dimer exchange was also dependent on theapplied stimulus. These findings indicate that the reg-ulated disassembly of existing promoter complexes andthe controlled reassembly of new complexes containingother components would make the expression of spe-cific genes responsive to changes in the concentrationof Rel proteins as well as other cooperating transcrip-tion factors (94).

The same research group discovered a mechanismthat might promote the exchange of NF-B dimers at specific target genes (95). They found that RelA polyu-biquitinylation and subsequent proteasomal degrada-tion is a dominant control mechanism for the post-transcriptional repression of NF-B activity in theabsence of IB␣. This polyubiquitinylation-dependent degradation specifically affects DNA-bound RelA. Thus,

it also provides an elegant mechanism for how thecomposition of NF-B dimers at specific promoters canbe altered in an ordered, sequential manner.

PROSPECTS AND PREDICTIONS

A few years ago most of the regulatory aspects in theNF-B system appeared to be solved. An elegant activa-tion mechanism centering around the signal-induceddegradation of the IBs had been elucidated in detail.However, research within the last few years has unrav-eled a far more sophisticated picture of NF-B activa-

tion. For example, IB-independent mechanisms of NF-B regulation exist and their importance for thecontrol of nuclear NF-B activity is increasingly recog-nized. What can we expect in the future from thisexciting research field? First, new types of post-transla-tional modification of Rel proteins important for regu-lating NF-B activity will be discovered; one strongcandidate is sumoylation. The precise molecular mech-anisms underlying the controlled nucleolar sequestra-tion of Rel proteins should be discovered in the nearfuture. The same is true for another exciting concept of NF-B regulation, the dimer exchange at specific pro-moters. Last, but not least, the research area of micro-

RNAs (miRNAs) promises to become increasingly im-portant for unraveling novel mechanisms regulatingNF-B activity independent of IBs. miR-146 is such aninducible miRNA inhibiting NF-B-activating pathwaysinvolved in the innate immune response (96). Potentialtargets of miR-146 are TRAF6 and IRAK1 (interleukin-1receptor-associated kinase 1). The expression of thismiRNA is NF-B-dependent, suggesting a feedbackregulatory loop.

Analysis of the physiological significance of alterna-tive mechanisms of NF-B regulation will be anotherarea of intense future research. There are already clear

indications that the regulation of phosphorylation of Rel proteins plays an important role in the pathogene-sis of inflammatory and malignant diseases (95, 97).Physiological and biological processes controlled by alternative regulatory mechanisms of NF-B are thespecific NF-B-dependent gene expression by control-ling the transactivating potency of NF-B, proliferationincluding tumorigenesis, apoptosis, immune responses,inflammatory processes, and intracellular viral re-sponses. The ultimate goal of research focusing onphysiological effects of NF-B modifications has to bethe identification of novel targets to interfere withdisease processes. Such targets could be enzymes (i.e.,protein kinases) known to modify Rel proteins inpathological conditions. For example, the activity of these enzymes could be blocked by specific inhibitory drugs. The prerequisite for such an interference withdisease processes is a profound understanding of thephysiological role played by each NF-B control mech-anism. Progress in this field of research will becomeincreasingly faster in the near future.

The expected discoveries will certainly add to the

fascination surrounding the field of I

B-independent regulation of NF-B and will provide new insights intothe regulation of NF-B-dependent physiologicalprocesses.

The authors thank Dr. Jonathan Lindquist for criticalreading of the manuscript. The work described from theauthors’ laboratory was funded by grants of the DeutscheForschungsgemeinschaft NE608/3–2 to Ma.N. and NA292/7–3 to Mi.N., respectively. The authors apologize to thosecolleagues whose work could not be cited due to spacelimitations.

REFERENCES

1. Hayden, M. S., and Ghosh, S. (2004) Signaling to NF-B. Genes Dev. 18, 2195–2224

2. Gilmore, T. D. (2006) Introduction to NF-B: players, pathways,perspectives. Oncogene 25, 6680–6684

3. Strober, W., Murray, P. J., Kitani, A., and Watanabe, T. (2006)Signalling pathways and molecular interactions of NOD1 andNOD2. Nat. Rev. Immunol. 6, 9–20

4. Janssens, S., and Tschopp, J. (2006) Signals from within: theDNA-damage-induced NF-B response. Cell Death Differ. 13,773–784 (review)

5. Wu, C.-J., Conze, D. B., Li, T., Srinivasula, S. M., and Ashwell, J. D. (2006) Sensing of Lys 63-linked polyubiquitinylation by NEMO is a key event in NF-B activation. Nat. Cell Biol. 8,398–406

6. Ea, C.-K., Deng, L., Xia, Z.-P., Pineda, G., and Chen, Z. J. (2006) Activation of IKK by TNF␣ requires site-specific ubiquitinationof RIP1 and polyubiquitin binding by NEMO. Mol. Cell 22,245–257

7. Chen, F., Bhatia, D., and Castranova, V. (2006) Finding NEMOby K63-linked polyubiquitin chain. Cell Death Differ. 13, 1835–1838

8. Kato, T., Jr., Delhase, M., Hoffmann, A., and Karin, M. (2003)CK2 is a C-terminal IB kinase responsible for NF-B activationduring UV response. Mol. Cell 12, 829–839

9. Wuerzberger-Davis, S. M., Nakamura, Y., Seufzer, B. J., andMiyamoto, S. (2007) NF-B activation by combinations of NEMO sumoylation and ATM activation stresses in the absenceof DNA damage. Oncogene 26, 641–651




10. Huang, T. T., Wuerzberger-Davis, S. M., Wu, Z.-H., and Miya-moto, S. (2003) Sequential modification of NEMO/IKK ␥ by SUMO-1 and ubiquitin mediates NF-B activation by genotoxicstress. Cell 115, 565–576

11. Janssens, S., Tinel, A., Lippens, S., and Tschopp, J. (2005) PIDDmediates NF-B activation in response to DNA damage. Cell 123,1–14

12. Mabb, A. M., Wuerzberger-Davis, S. M., and Miyamoto, S. (2006)PIASy mediates NEMO sumoylation and NF-B activation inresponse to genotoxic stress. Nat. Cell Biol. 8, 986–993

13. Wu, Z.-H., Shi, Y., Tibbetts, R. S., and Miyamoto, S. (2006)Molecular linkage between the kinase ATM and NF-B signaling

in response to genotoxic stimuli. Science 311, 1141–114614. Karin, M., and Ben-Neriah, Y. (2000) Phosphorylation meets

ubiquitinylation: the control of NF-B activity. Annu. Rev. Immu- nol. 18, 621–663

15. Gao, M., and Karin, M. (2005) Regulating the regulators:control of protein ubiquitinylation and ubiquitin-like modifica-tions by extracellular stimuli. Mol. Cell 19, 581–593

16. Heynick, K., and Beyaert, R. (2005) A20 inhibits NF-B activa-tion by dual ubiquitin-editing functions. Trends Biochem. Sci. 30,1– 4

17. Lee, E. G., Boone, D. L., Chai, S., Libby, S. L., Chien, M.,Lodolce, J. P., and Ma, A. (2000) Failure to regulate TNF-induced NF-B and cell death responses in A20-deficient mice.Science 289, 2350–2354

18. Wertz, I. E., O’Rourke, K. M., Zhou, H., Eby, M., Aravind, L.,Seshagiri, S., Wu, P., Wiesmann, C., Baker, R., Boone, D. L., Ma,

A., et al. (2004) De-ubiquitinylation and ubiquitin ligase do-mains of A20 downregulate NF-B signalling. Nature 430, 694–699

19. Mauro, C., Pacifico, F., Lavorgna, A., Mellone, S., Iannetti, A., Acquaviva, R., Formisano, S., Vito, P., and Leonardi, A. (2006) ABIN-1 binds to NEMO/IKK ␥ and co-operates with A20 ininhibiting NF-B. J. Biol. Chem. 281, 18482–18488

20. Viatour, P., Merville, M.-P., Bours, V., and Chariot, A. (2005)Phosphorylation of NF-B and IB proteins: implications incancer and inflammation. Trends Biochem. Sci. 30, 43–52

21. Perkins, N. D. (2006) Post-translational modifications regulat-ing the activity and function of the NF-B pathway. Oncogene 25,6717–6730

22. Mellits, K. H., Hay, R. T., and Goodbourn, S. (1993) Proteolyticdegradation of MAD3 (IB␣) and enhanced processing of theNF-B precursor p105 are obligatory steps in the activation of NF-B. Nucleic Acids Res. 21, 5059–5066

23. Naumann, M., and Scheidereit, C. (1994) Activation of NF-B in vivo is regulated by multiple phosphorylations. EMBO J. 13,4597–4607

24. Sakurai, H., Chiba, H., Miyoshi, H., Sugita, T., and Toriumi, W.(1999) IB kinases phosphorylate NF-B p65 subunit on serine536 in the transactivation domain. J. Biol. Chem. 274, 30353–30356

25. O’Mahony, A. M., Montano, M., vanBeneden, K., Chen, L. F.,and Greene, W. C. (2004) Human T-cell lymphotropic virustype 1 tax induction of biologically active NF-B requires IBkinase-1-mediated phosphorylation of RelA/p65. J. Biol. Chem.279, 18137–18145

26. Jiang, X., Takahashi, N., Matsui, N., Tetsuka, T., and Okamoto,T. (2003) The NF-B activation in lymphotoxin ␤ receptorsignaling depends on the phosphorylation of p65 at serine 536.

J. Biol. Chem. 278, 919–926

27. Buss, H., Dorrie, A., Schmitz, M. L., Hoffmann, E., Resch, K.,and Kracht, M. (2004) Constitutive and interleukin-1-induciblephosphorylation of p65 NF-B at serine 536 is mediated by multiple protein kinases including IB kinases IKK ␣, IKK ␤,IKK ε, TRAF family member-associated (TANK)-binding kinase 1(TBK1), and an unknown kinase and couples p65 to TATA-binding protein-associated factor II31-mediated interleukin-8transcription. J. Biol. Chem. 279, 55633–55643

28. Fujita, F., Taniguchi, Y., Kato, T., Narita, Y., Furuya, A., Ogawa,T., Sakurai, H., Joh, T., Itoh, M., Delhase, M., Karin, M., andNakanishi, M. (2003) Identification of NAP1, a regulatory subunit of IB kinase-related kinases that potentiates NF-Bsignaling. Mol. Cell Biol. 23, 7780–7793

29. Adli, M., and Baldwin, A. S. (2006) IKK-i/IKK ε controls consti-tutive, cancer cell-associated NF-B activity via regulation of

Ser-536 p65/RelA phosphorylation. J. Biol. Chem. 281, 26976–26984

30. Bohuslav, J., Chen, L. F., Kwon, H., Mu, Y., and Greene, W. C.(2004) p53 induces NF-B activation by an IB kinase-indepen-dent mechanism involving phosphorylation of p65 by ribosomalS6 kinase 1. J. Biol. Chem. 279, 26115–26125

31. Sasaki, C. Y., Barberi, T. J., Ghosh, P., and Longo, D. L. (2005)Phosphorylation of RelA/p65 on serine 536 defines an IB␣-independent NF-B pathway. J. Biol. Chem. 280, 34538–34547

32. Qin, C., Nguyen, T., Stewart, J., Samudio, I., Burghardt, R., andSafe, S. (2002) Estrogen up-regulation of p53 gene expressionin MCF-7 breast cancer cells is mediated by calmodulin kinase

IV-dependent activation of a NF-B/CCAAT-binding transcrip-tion factor-1 complex. Mol. Endocrinol. 16, 1793–180933. Bae, J. S., Jang, M. K., Hong, S., An, W. G., Choi, Y. H., Kim,

H. D., and Cheong, J. (2003) Phosphorylation of NF-B by calmodulin-dependent kinase IV activates anti-apoptotic geneexpression. Biochem. Biophys. Res. Commun. 305, 1094–1098

34. Bird, T. A., Schooley, K., Dower, S. K., Hagen, H., and Virca,G. D. (1997) Activation of nuclear transcription factor NF-Bby interleukin-1 is accompanied by casein kinase II-mediatedphosphorylation of the p65 subunit. J. Biol. Chem. 272,32606–32612

35. Wang, D., Westerheide, S. D., Hanson, J. L., and Baldwin, A. S., Jr. (2000) Tumor necrosis factor alpha-induced phosphoryla-tion of RelA/p65 on Ser529 is controlled by casein kinase II.

J. Biol. Chem. 275, 32592–3259636. Schwabe, R. F., and Sakurai, H. (2006) IKK ␤ phosphorylates p65

at S468 in transactivation domain 2. FASEB J. 19, 1758–176037. Mattioli, I., Geng, H., Sebald, A., Hodel, M., Bucher, C., Kracht,M., and Schmitz, M. L. (2006) Inducible phosphorylation of NF-B p65 at serine 468 by T cell costimulation is mediated by IKK ε. J. Biol. Chem. 281, 6175–6183

38. Buss, H., Dorrie, A., Schmitz, M. L., Frank, R., Livingstone, M.,Resch, K., and Kracht, M. (2004) Phosphorylation of serine 468by GSK-3␤ negatively regulates basal p65 NF-B activity. J. Biol.Chem. 279, 49571–49574

39. Bournat, J. C., Brown, A. M., and Soler, A. P. (2000) Wnt-1dependent activation of the survival factor NF-B in PC12 cells.

J. Neurosci. Res. 61, 21–3240. Sanchez, J. F., Sniderhan, L. F., Williamson, L. F., Fan, S.,

Chakraborty-Sett, S., and Maggirwar, S. B. (2003) Glycogensynthase kinase 3␤-mediated apoptosis of primary cortical astro-cytes involves inhibition of NF-B signaling. Mol. Cell. Biol. 23,4649–4662

41. Hoeflich, K. P., Luo, J., Rubie, E. A., Tsao, M. S., Jin, O., and Woodgett, J. R. (2000) Requirement for glycogen synthasekinase-3␤ in cell survival and NF-B activation. Nature 206,86–90

42. Duran, A., Diaz-Meco, M. T., and Moscat, J. (2003) Essential roleof RelA Ser311 phosphorylation by zetaPKC in NF-B transcrip-tional activation. EMBO J. 22, 3910–3918

43. Zhong, H., SuYang, H., Erdjument-Bromage, H., Tempst, P.,and Ghosh, S. (1997) The transcriptional activity of NF-B isregulated by the IB-associated PKAc subunit through a cyclic AMP-independent mechanism. Cell 89, 413–424

44. Vermeulen, L., De Wilde, G., van Damme, P., van den Berghe, W., and Haegeman, G. (2003) Transcriptional activation of theNF-B p65 subunit by mitogen- and stress-activated proteinkinase-1 (MSK1). EMBO J. 22, 1313–1324

45. Zhong, H., May, M. J., Jimi, E., and Ghosh, S. (2002) The

phosphorylation status of nuclear NF-B determines its associa-tion with CBP/p300 or HDAC-1. Mol. Cell 9, 625–636

46. Jacque, E., Tchenio, T., Piton, G., Romeo, P. H., and Baud, V.(2005) RelA repression of RelB activity induces selective geneactivation downstream of TNF receptors. Proc. Natl. Acad. Sci.U. S. A. 102, 14635–14640

47. Yeh, P. Y., Yeh, K. H., Chuang, S. E., Song, Y. C., and Cheng, A. L. (2004) Suppression of MEK/ERK signaling pathway en-hances cisplatin-induced NF-B activation by protein phospha-tase 4-mediated NF-B p65 Thr dephosphorylation. J. Biol. Chem.279, 26143–26148

48. Rocha, S., Garrett, M. D., Campbell, K. J., Schumm, K., andPerkins, N. D. (2005) Regulation of NF-B and p53 throughactivation of ATR and Chk1 by the ARF tumour suppressor.

EMBO J. 24, 1157–1169




49. Campbell, K. J., Witty, J. M., Rocha, S., and Perkins, N. D. (2006)Cisplatin mimics ARF tumor suppressor regulation of RelA (p65) NF-B transactivation. Cancer Res. 66, 929–935

50. Ryo, A., Suizu, F., Yoshida, Y., Perrem, K., Liou, Y. C., Wulf, G.,Rottapel, R., Yamaoka, S., and Lu, K. P. (2003) Regulation of NF-B signaling by Pin1-dependent prolyl isomerization andubiquitin-mediated proteolysis of p65/RelA. Mol. Cell 12, 1413–1426

51. Maine, G. N., Mao, X., Komarck, C. M., and Burstein, E. (2007)COMMD1 promotes the ubiquitination of NF-B subunitsthrough a cullin-containing ubiquitin ligase. EMBO J. 26, 436–447

52. Lawrence, T., Bebien, M., Liu, G. Y., Nizet, V., and Karin, M.(2005) IKK ␣ limits macrophage NF-B activation and con-tributes to the resolution of inflammation. Nature 434, 1138–1143

53. Nelson, D. E., Ihekwaba, A. E., Elliot, M., Johnson, J. R., Gibney,C. A., Foreman, B. E., Nelson, G., See, V., Horton, C. A., Spiller,D. G., et al. (2004) Oscillations in NF-B signaling control thedynamics of gene expression. Science 306, 704–708

54. Liu, Y., Denlinger, C. E., Rundall, B. K., Smith, P. W., and Jones,D. R. (2006) Suberoylanilide hydroxamic acid induces Akt-mediated phosphorylation of p300, which promotes acetylationand transcriptional activation of RelA/p65. J. Biol. Chem. 281,31359–31368

55. Hoberg, J. E., Popko, A. E., Ramsey, C. S., and Mayo, M. W.(2006) IKK ␣-mediated derepression of SMRT potentiates acetylation of RelA/p65 by p300. Mol. Cell Biol. 26, 457–471

56. Chen, L. F., Mu, Y., and Greene, W. C. (2002) Acetylation of

RelA at discrete sites regulates distinct nuclear functions of NF-B. EMBO J. 21, 6539–6548

57. Yeung, F., Hoberg, J. E., Ramsey, C. S., Keller, M. D., Jones,D. R., Frye, R. A., and Mayo, M. W. (2004) Modulation of NF-B-dependent transcription and cell survival by the SIRT1deacetylase. EMBO J. 23, 2369–2380

58. Kiernan, R., Bres, V., Nig, R. W., Coudart, M. P., El Mes-saoudi, S., Sardet, C., Jin, D. Y., Emiliani, S., and Benkirane,M. (2003) Post-activation turn-off of NF-B-dependent tran-scription is regulated by acetylation of p65. J. Biol. Chem. 278,2758–2766

59. Chen, L. F., Williams, S. A., Mu, Y., Nakano, H., Duerr, J. M.,Buckbinder, L., and Greene, W. C. (2005) NF-B RelA phos-phorylation regulates RelA acetylation. Mol. Cell. Biol. 25, 7966–7975

60. Huang, D. B., Vu, D., and Gosh, G. (2005) NF-B RelB forms an

intertwined homodimer. Structure 13, 1365–137361. Marienfeld, R., Berberich-Siebelt, F., Berberich, I., Denk, A.,Serfling, E., and Neumann, M. (2001) Signal-specific and phos-phorylation-dependent RelB degradation: a potential mecha-nism of NF-B control. Oncogene 20, 8142–8147

62. Maier, H. J., Marienfeld, R., Wirth, T., and Baumann, B. (2003)Critical role of RelB serine 368 for dimerization and p100stabilization. J. Biol. Chem. 278, 39242–39250

63. Druker, B. J., Neumann, M., Okuda, K., Franza, B. R., Jr., andGriffin, J. D. (1994) rel Is rapidly tyrosine-phosphorylatedfollowing granulocyte-colony stimulating factor treatment of human neutrophils. J. Biol. Chem. 269, 5387–5390

64. Fognani, C., Rondi, R., Romano, A., and Blasi, F. (2000)cRel-TD kinase: a serine/threonine kinase binding in vivo andin vitro c-Rel and phosphorylating its transactivation domain.Oncogene 19, 2224–2232

65. Martin, A. G., and Fresno, M. (2000) Tumor necrosis factor-alpha activation of NF-B requires the phosphorylation of Ser-471 in the transactivation domain of c-Rel. J. Biol. Chem. 275,24383–24391

66. Yu, S. H., Chiang, W. C., Shih, H. M., and Wu, K. J. (2004)Stimulation of c-Rel transcriptional activity by PKA catalyticsubunit beta. J. Mol. Med. 82, 621–628

67. Martin, A. G., San-Antonio, B., and Fresno, M. (2001) Regula-tion of NF-B transactivation. Implication of phosphatidylinosi-tol 3-kinase and PKC in c-Rel activation by TNF␣. J. Biol. Chem.276, 15840–15849

68. Starczynowski, D. T., Reynolds, J. G., and Gilmore, T. D. (2005)Mutations of TNF␣-responsive serine residues within the C-terminal transactivation domain of human transcription factorREL enhance its in vitro transforming ability. Oncogene 24,7355–7368

69. Starczynowski, D. T., Trautmann, H., Pott, C., Harder, L., Arnold, N., Africa, J. A., Leeman, J. R., Siebert, R., and Gilmore,T. D. (2006) Mutation of an IKK phosphorylation site within thetransactivation domain of REL in two patients with B-celllymphoma enhances REL’s in vitro transforming activity. Onco- gene In press

70. Harris, J., Oliere, S., Sharma, S., Sun, Q., Lin, R., Hiscott, J., andGrandvaux, N. (2006) Nuclear accumulation of c-Rel followingC-terminal phosphorylation by TBK1/IKK ε. J. Immunol. 177,2527–2535

71. Sanchez-Valdepenas, C., Martin, A. G., Ramakrishnan, P., Wallach, D., and Fresno, M. (2006) NF-B-inducing kinase is

involved in the activation of the CD28 responsive element through phosphorylation of c-Rel and regulation of its transac-tivating activity. J. Immunol. 176, 4666–4674

72. Guan, H., Hou, S., and Ricciardi, R. P. (2005) DNA binding of repressor NF-B p50/p50 depends on phosphorylation of Ser337 by the protein kinase A catalytic subunit. J. Biol. Chem.280, 9957–9962

73. Zwergal, A., Quirling, M., Saugel, B., Huth, K. C., Sydlik, C., Poli, V., Neumeier, D., Ziegler-Heitbrock, H. W., and Brand, K.(2006) C/EBP beta blocks p65 phosphorylation and thereby NF-B-mediated transcription in TNF-tolerant cells. J. Immunol.177, 665–672

74. Gao, F., Cheng, J., Shi, T., and Yeh, E. T. (2006) Neddylation of a breast cancer-associated protein recruits a class III histonedeacetylase that represses NF-B-dependent transcription. Nat.Cell Biol. 8, 1171–1178

75. Adcock, I. M., Ito, K., and Barnes, P. J. (2004) Glucocorticoids:effects on gene transcription. Proc. Am. Thorac. Soc. 1, 247–25476. De Bosscher, K., Vanden Berghe, W., and Haegeman, G. (2006)

Cross-talk between nuclear receptors and NF-B. Oncogene 25,6868–6886

77. De Bosscher, K., Vanden Berghe, W., and Haegeman, G. (2003)The interplay between the glucocorticoid receptor and NF-Bor AP-1: molecular mechanisms for gene repression. Endocr. Rev.24, 488–522

78. Martens, C., Bilodeau, S., Maira, M., Gauthier, Y., and Drouin, J.(2005) Protein-protein interactions and transcriptional antago-nism between the subfamily of NGFI-B/Nur77 orphan nuclearreceptors and glucocorticoid receptor. Mol. Endocrinol. 19, 885–897

79. Liden, J., Rafter, I., Truss, M., Gustafsson, J. A., and Okret, S.(2000) Glucocorticoid effects on NF-B binding in the tran-scription of the ICAM-1 gene. Biochem. Biophys. Res. Commun.

273, 1008–101480. Nissen, R. M., and Yamamoto, K. R. (2000) The glucocorticoid

receptor inhibits NF-B by interfering with serine-2 phosphorylation of the RNA polymerase II carboxy-terminal domain.Genes Dev. 14, 2314–2329

81. Park, K. J., Krishnan, V., O’Malley, B. W., Yamamoto, Y., andGaynor, R. B. (2005) Formation of an IKK ␣-dependent tran-scription complex is required for estrogen receptor-mediatedgene activation. Mol. Cell 18, 71–82

82. Park, G. Y., Wang, X., Hu, N., Pedchenko, T. V., Blackwell, T. S.,and Christman, J. W. (2006) NIK is involved in nucleosomalregulation by enhancing histone H3 phosphorylation by IKK ␣.

J. Biol. Chem. 281, 18684–1869083. Ito, K., Barnes, P. J., and Adcock, I. M. (2000) Glucocorticoid

receptor recruitment of HDAC 2 inhibits interleukin-1␤-in-duced histone H4 acetylation on lysines 8 and 12. Mol. Cell. Biol.

20, 6891–690384. Garside, H., Stevens, A., Farrow, S., Normand, C., Houle, B.,

Berry, A., Maschera, B., and Ray, D. (2004) Glucocorticoidligands specify different interactions with NF-B by allostericeffects on the glucocorticoid receptor DNA binding domain.

J. Biol. Chem. 279, 50050–5005985. Luecke, H. F., and Yamamoto, K. R. (2005) The glucocorti-

coid receptor blocks P-TEFb recruitment by NF-B to effect promoter-specific transcriptional repression. Genes Dev. 19,1116–1127

86. Rubbi, C. P., and Milner, J. (2000) Non-activated p53 co-localizes with sites of transcription within both the nucleoplasmand the nucleolus. Oncogene 19, 85–96

87. Weber, J. D., Kuo, M. L., Bothner, B., DiGiammarino, E. L.,Kriwacki, R. W., Russel, M. F., and Sherr, C. J. (2000) Cooper-




ative signals governing ARF-Mdm2 interaction and nucleolarlocalization of the complex. Mol. Cell. Biol. 20, 2517–2528

88. Datta, A., Nag, A., Pan, W., Hay, N., Gartel, A. L., Colamonici,O., Mori, Y., and Raychaudhuri, P. (2004) Myc-ARF interactioninhibits the functions of Myc. J. Biol. Chem. 35, 36698–36707

89. Birbach, A., Bailey, S. T., Ghosh, S., and Schmid, J. A. (2004)Cytosolic, nuclear and nucleolar localization signals determinesubcellular distribution and activity of the NF-B inducingkinase NIK. J. Cell Sci. 117, 3615–3624

90. Niedick, I., Froese, N., Oumard, A., Muller, P. P., Nourbakhsh,M., Hauser, H., and Koster, M. (2004) Nucleolar localizationand mobility analysis of the NF-B repressing factor NRF. J. Cell

Sci. 117, 3447–345891. Sweet, T., Khalili, K., Sawaya, B. E., and Amini, S. (2003)Identification of a novel protein from glia cells based on itsability to interact with NF-B subunits. J. Cell Biochem. 90,884–891

92. Dhar, S. K., Lynn, B. C., Daosukho, C., and St. Clair, D. K.(2004) Identification of nucleophosmin as an NF-B co-activa-tor for the induction of the human SOD2 gene. J. Biol. Chem.279, 28209–28219

93. Stark, L., and Dunlop, M. G. (2005) Nucleolar sequestration of RelA (p65) regulates NF-B-driven transcription and apoptosis.Mol. Cell. Biol. 25, 5985–6004

94. Saccani, S., Pantano, S., and Natoli, G. (2003) Modulation of NF-B activity by exchange of dimers. Mol. Cell 11, 1563–1574

95. Saccani, S., Marazzi, I., Beg, A. A., and Natoli, G. (2004)Degradation of promoter-bound p65/RelA is essential for theprompt termination of the nuclear factor kappaB response. J.

Exp. Med. 200, 107–11396. Taganov, K. D., Boldin, M. P., Chang, K.-J., and Baltimore, D.

(2006) NF-B-dependent induction of microRNA miR-146, aninhibitor targeted to signaling proteins of innate immune

responses. Proc. Natl. Acad. Sci. U. S. A. 103, 12481–1248697. Rayet, B., Fan, Y., and Gelinas, C. (2003) Mutations in the v-Rel transact ivation domain indicat e altered phosphoryla-tion and identify a subset of NF-B-regulated cell deathinhibitors important for v-Rel transforming activity. Mol. Cell.Biol. 23, 1520–1533

Received for publication January 8, 2007.Accepted for publication March 15, 2007.

2654 Vol 21 September 2007 NEUMANN ET ALThe FASEB Journal

beyond ikbs alternative regulation of nfkb activity 22

Documents