Oncogene (2006) 25, 66806684

& 2006 Nature Publishing Group All rights reserved 0950-9232/06 $30.00
www.nature.com/onc

REVIEW

Introduction to NF-jB: players, pathways, perspectives

TD Gilmore
Biology Department, Boston University, Boston, MA, USA

This article serves as an introduction to the collection of reviews on nuclear factor-kappaB (NF-jB). It provides an overview of the discovery and current status of NF-jB as a research topic. Described are the structures, activities and regulation of the proteins in the NF-jB family of transcription factors. NF-jB signaling is primarily regulated by inhibitor jB (IjB) proteins and the IjB kinase complex through two major pathways: the canonical and non-canonical NF-jB pathways. The organization and focus of articles included in the following reviews are described, as well as likely future areas of research interest on NF-jB. Oncogene (2006) 25, 66806684. doi:10.1038/sj.onc.1209954 Keywords: NF-kappaB; IkappaB; IKK; Rel; signal transduction; transcription factor

Introduction With great pleasure, I have edited this second compilation of reviews on nuclear factor-kappa B (NF-kB) transcription factors and NF-kB signaling. It is approximately 20 years since the coincident discovery of three proteins classical NF-kB, v-Rel and Dorsal that show variable nucleo-cytoplasmic subcellular localization (Gilmore and Temin, 1986; Sen and Baltimore, 1986; Baeuerle and Baltimore, 1988; Steward et al., 1988), which were soon demonstrated to be members of the same family of proteins (Stephens et al., 1983; Wilhelmsen et al., 1984; Steward, 1987; Ghosh et al., 1990; reviewed by Gilmore, 1990; Kieran et al., 1990). Notably, the biological processes immunity (NF-kB), oncogenesis (v-Rel) and development (Dorsal) investigated in those early studies continue to be areas that provoke much of the interest in NF-kB. Today, the study of NF-kB signaling is essentially an industry, complete with website (www.nf-kb.org), patent (Baltimore et al., 2002) and approximately 25 000 publications. For those few unfamiliar with the NF-kB transcription factor family, it includes a collection of proteins conserved from (at least) the phylum Cnidaria to humans. Among model organisms, these
Correspondence: Dr TD Gilmore, Biology Department, Boston University, 5 Cummington Street, Boston, MA 02215, USA. E-mail: gilmore@bu.edu

transcription factors are notably absent in yeast and Caenorhabditis elegans; in the latter, it is likely that NFkB-like genes/proteins have been lost (given that they are present in the more primitive organism, the sea anenome Nematostella vectensis (Sullivan et al., 2006)). As described below, the term NF-kB is somewhat confusing, as it can be used to refer to the superfamily (of Rel and NF-kB proteins across species), the subfamily (e.g., p100, p105 and Relish) or the specic p50-RelA heterodimer, which is the major NF-kB dimer in many cells. The larger NF-kB family of proteins is composed of two subfamilies: the NF-kB proteins and the Rel proteins. All of these proteins share a highly conserved DNA-binding/dimerization domain called the Rel homology domain (RHD) (Gilmore, 1990) (Figure 1). The Rel subfamily includes c-Rel, RelB, RelA (aka p65) and Drosophila Dorsal and Dif. The Rel proteins contain C-terminal transactivation domains, which are often not conserved at the sequence level across species, even though they can activate transcription in a variety of species including yeast. Members of the NF-kB subfamily (p105, p100 and Drosophila Relish) are distinguished by their long C-terminal domains that contain multiple copies of ankyrin repeats, which act to inhibit these proteins. The NF-kB proteins become shorter, active DNA-binding proteins (p105 to p50 and p100 to p52) by either limited proteolysis or, possibly sometimes, by arrested translation. As such, members of the NF-kB subfamily are generally not activators of transcription, except when they form dimers with members of the Rel subfamily. Of note, the nuclear factor of activated T-cell (NFAT) transcription factors also contain the RHD and bind to similar DNA sequences as the Rel/NF-kB dimers, but NFAT proteins generally have not been found to form dimers with Rel and NF-kB proteins. Collectively, NF-kB transcription factor dimers bind to 910 base pair DNA sites (kB sites), which have a great amount of variability (50 -GGGRNWYYCC-30 ; R, A or G; N, any nucleotide; W, A or T; Y, C or T). All vertebrate NF-kB family proteins can form homodimers or heterodimers in vivo, except for RelB, which only forms heterodimers in vivo. This combinatorial diversity contributes to the regulation of distinct, but overlapping, sets of genes for at least three reasons: because the individual dimers have distinct DNA-binding site specicities for a collection of related kB sites, because of the different proteinprotein interactions the

Introduction to NF-jB TD Gilmore

6681
RelA RelB c-Rel Dorsal Dif p50/p105 p52/p100 Relish , , , Bcl-3, IB Cactus RHD TAD

Rel

NF-B

RHD
SS SS

IB

IKK

Kinase

HLH

LZ

NBD

NEMO

()

upstream regulatory step in both of these pathways is activation of an IkB kinase (IKK) complex, which consists of catalytic kinase subunits (IKKa and/or IKKb) and a scaffold, sensing protein called NF-kB essential modulator (NEMO). As such, activation of NF-kB dimers is the result of IKK-mediated, phosphorylation-induced degradation of the IkB inhibitor, which enables the NF-kB dimers to enter the nucleus and activate specic target gene expression. In most cases, the activation of NF-kB is transient and cyclical in the presence of continual inducer. For example, in mouse broblasts maintained in the presence of tumor necrosis factor, nuclear NF-kB DNA-binding activity appears and disappears approximately every 3060 min; these cycles are due to repeated degradation and re-synthesis of IkB and the consequent activation and inactivation of NF-kB, respectively (Hoffmann et al., 2006).

CC1

CC2

LZ

ZF

Figure 1 Structures of core NF-kB signaling proteins. The generalized structures of the two subfamilies (Rel and NF-kB) of NF-kB transcription factors are shown at the top. All have a conserved DNA-binding/dimerization domain called the Rel homology domain (RHD), which also has sequences important for nuclear localization and IkB inhibitor binding. The C-terminal halves of the Rel proteins have transcriptional activation domains (TAD). The C-terminal halves of the NF-kB subfamily proteins have ankyrin repeat-containing inhibitory domains (red bars), which can be removed by proteasome-mediated proteolysis. As with the C-terminal domains of the NF-kB proteins, the independent IkB proteins consist mainly of ankyrin repeats, and several (IkBa, IkBb, IkBe, IkBg) have two N-terminal serine residues (S) that serve as IKK phosphorylation sites, which signal the protein for ubiquitination and degradation. The generalized structures of IKKa and b (kinase domain; HLH, helix-loop-helix; LZ, leucine zipper; NBD, NEMO binding domain) and of NEMO (CC, coiled coil; LZ, leucine zipper; ZF, zinc nger) are also shown.

Organization of this collection of reviews As with the 1999 issue of Oncogene Reviews on NF-kB (Gilmore, 1999), the choice of subjects to review was difcult. Because Oncogene is a journal dedicated primarily to the control of cell growth and oncogenesis, I decided to make the role of NF-kB in these processes the focus of this issue. Nevertheless, to set the stage, it was necessary to include several papers on the regulation of NF-kB: the regulation of NF-kB by upstream IKK pathways (Scheidereit, 2006); the dynamics and direct mechanisms of gene regulation by NF-kB (Hoffmann et al., 2006); post-translational modications that regulate components of the NF-kB pathway (Perkins, 2006); and the controversial role of reactive oxygen species in the regulation of NF-kB activity (Bubici et al., 2006). These papers are followed by three papers in which the normal physiological roles of NF-kB are discussed. Minakhina and Steward describe the role of NF-kB in Drosophila development and immunity; Hayden et al. describe what is known about the extensive role that NF-kB plays in the mammalian immune system; and Gerondakis et al. discuss what has been learned about the physiological roles of NF-kB signaling components through the use of mouse knockouts and transgenics. The next four papers focus on subjects related to NF-kB and pathogenesis/oncogenesis, describing the role of NF-kB in apoptosis (Dutta et al., 2006) and cancer (Basseres and Baldwin, 2006), how mutations in NF-kB pathway genes are related to human disease (Courtois and Gilmore, 2006) and how viruses affect NF-kB signaling for their pathogenesis, their replication, or as part of detection by the host organism (Hiscott et al., 2006). The nal two papers describe ways that the NF-kB pathway can be modulated, in some cases for possible therapeutic purposes: De Bosscher et al. describe the extensive literature on inhibitory crosstalk between the important steroid receptor pathways and NF-kB, and Gilmore and Herscovitch have unearthed an extensive collection of inhibitors of NF-kB signaling and discuss their
Oncogene

individual dimers make at target promoters, and because of the gene activation prole of different dimers under specic physiological conditions. The activity of NF-kB is tightly regulated by interaction with inhibitory IkB proteins. As with the NF-kB transcription factors, there are several IkB proteins (e.g., IkBa, IkBb, IkBg, IkBe and Drosophila Cactus) that have different afnities for individual NF-kB dimers. Individual IkBs are also regulated slightly differently by phosphorylation and proteolysis, and show differences in their tissue-specic expression patterns. From biochemical studies and direct structural determinations (reviewed by Chen and Ghosh, 1999), it is clear that IkBa makes multiple contacts with NF-kB. Generally these interactions cover the nuclear localization signal of the given NF-kB dimer and interfere with sequences involved in DNA binding. Thus, in most cells, NF-kB is present as a latent, inactive, IkB-bound complex in the cytoplasm. There are two, and possibly three, pathways leading to the activation of NF-kB (see Figure 2 for details). The two best-described pathways are called either the canonical and non-canonical pathways or the classical and alternative pathways, respectively. The common

Introduction to NF-jB TD Gilmore

6682
Canonical Pathway Non-canonical Pathway Pathway 3

NEM

NEM NEMO IKK P l B p50 RelA P Proteasomal degradation P

NIK ?

p10

5
p50

p10

P P p100 processing
p5 0

p105 processing

RelB RelA

p50

p50

p5

RelB

Nucleus

Nucleus Bcl-3

Nucleus

Figure 2 NF-kB signal transduction pathways. In the canonical (or classical) NF-kB pathway, NF-kB dimers such as p50/RelA are maintained in the cytoplasm by interaction with an independent IkB molecule (often IkBa). In many cases, the binding of a ligand to a cell surface receptor (e.g., tumor necrosis factor-receptor (TNF-R) or a Toll-like receptor) recruits adaptors (e.g., TRAFs and RIP) to the cytoplasmic domain of the receptor. In turn, these adaptors often recruit an IKK complex (containing the a and b catalytic subunits and two molecules of the regulatory scaffold NEMO) directly onto the cytoplasmic adaptors (e.g., by virtue of the K63-ubiquitinbinding activity of NEMO). This clustering of molecules at the receptor activates the IKK complex. IKK then phosphorylates IkB at two serine residues, which leads to its K48 ubiquitination and degradation by the proteasome. NF-kB then enters the nucleus to turn on target genes. The auto-regulatory aspect of the canonical pathway, wherein NF-kB activates expression of the IkBa gene that leads to resequestration of the complex in the cytoplasm by the newly synthesized IkB protein is not shown. The non-canonical (or alternative) pathway is largely for activation of p100/RelB complexes during B- and T-cell organ development. This pathway differs from the canonical pathway in that only certain receptor signals (e.g., Lymphotoxin B (LTb), B-cell activating factor (BAFF), CD40) activate this pathway and because it proceeds through an IKK complex that contains two IKKa subunits (but not NEMO). In the noncanonical pathway, receptor binding leads to activation of the NF-kB-inducing kinase NIK, which phosphorylates and activates an IKKa complex, which in turn phosphorylates two serine residues adjacent to the ankyrin repeat C-terminal IkB domain of p100, leading to its partial proteolysis and liberation of the p52/RelB complex. Other distinct NF-kB pathways no doubt exist. For example, in Pathway 3, p50 (or p52) homodimers enter the nucleus, where they become transcriptional activators by virtue of interaction with the IkB-like co-activator Bcl-3 (or IkBz). How Pathway 3 is regulated is not known. In all three pathways, various post-translational modications (e.g., phosphorylation, acetylation and prolyl isomerization) of the NF-kB subunits can modulate their transcriptional activity (see Perkins, 2006, this issue).

mechanisms of action and how some may prove important for the treatment of human diseases.

signaling pathway, physiology and disease, therapy and simple model systems. Basic biochemistry Although we have learned a great deal about the intracellular NF-kB signaling pathway, we still have very rudimentary knowledge about the dynamics of the NF-kB pathway in cells in tissue culture and know essentially nothing about these dynamics in whole organisms. In most cell types and signaling conditions, it is still not known what is the contribution of specic NF-kB complexes (e.g., p50-RelA vs p52-c-Rel vs c-Rel-c-Rel) to given physiological responses or how

Future perspectives Even with the many advances since 1999, there continue to be several unsolved mysteries in the saga of NF-kB. If I use my crystal ball to predict what the next 7 years have in store, I suspect that progress will be made in the following four general areas: basic biochemistry of the
Oncogene

Introduction to NF-jB TD Gilmore

6683

different NF-kB dimers are targeted to specic promoters. I suspect that we will have tables of genes, listing the NF-kB dimers that control them under specic conditions. Undoubtedly, there is also much more to be learned about how post-translational modications and proteinprotein interactions modulate and/or specify activated NF-kB responses. Despite the numerous papers on the IKK complex, there is still much to be uncovered about its composition and regulation. What is the precise stoichiometry of core components in the IKK complex, especially, how many molecules of NEMO are in the complex in uninduced vs induced states? What are the three-dimensional structures of components of the IKK complex? How does the structure (and composition) of the IKK complex change upon activation? I suspect that there are dynamic changes in the complex that have not yet been addressed. For example, what are the roles of the various non-core components that have been reported to associate with the IKK complex (e.g., ELKS, heatshock proteins, etc)? What is the nature of the IKKadependent IKK complex used in the non-canonical pathway? What are the biological roles of pathways other than the canonical and non-canonical NF-kB pathways (e.g., Pathway 3 in Figure 2, or constitutive turnover or tyrosine-induced release of IkBa)? What other cellular pathways are regulated by IKK? Physiology and disease There is no question that the use of additional and more nely tuned mouse genetic model systems (e.g., conditional knockouts) will reveal novel and sometimes unexpected roles for NF-kB in normal physiology. For example, by the time of the next review series, there is likely to be a full paper on the role of NF-kB in learning, memory and behavior. Furthermore, it is likely that polymorphisms that play a role in inter-individual susceptibilities to diseases, especially ones involving pathogens, will be identied in components or target genes of NF-kB signaling. Anti-NF-kB therapy Whether or not its patent stands the test of time, NF-kB will no doubt continue to occupy a central focus of therapeutic intervention. By the time of the next review issue, I expect that clear and focused modulators of NF-kB signaling will be in use in humans. I speculate that such inhibitors will rst show efcacy in unusual and accessible diseases, which have NF-kB dependency (e.g., cylindromatosis). Although there has been great focus on potent single-step NF-kB signaling inhibitors, I suspect that low-dose, multi-step inhibition of NF-kB (e.g., compounds or combinations of compounds that act at more than one step) will be more effective. Perhaps, we will also have an appreciation of how chronic use of natural product-derived NF-kB inhibitors (e.g., curcumin, green tea and antioxidants) contribute to human health.

Table 1 Nomenclature for NF-kB core signaling proteins Gene (human gene) Dorsal cactus dif Relish v-rel c-rel (REL) relb (RELB) rela (RELA) nfkb1 (NFKB1) nfkb2 (NFKB2) ikba (IKBA,NFKBA1) ikbb (IKBB) ikbe (IKBE) nfkb1 (NFKB1) ikbz (IKBZ) bcl-3 (BCL3) ikka (IKKA,IKBKA,CHUK) ikkb (IKKB,IKBKB) ikke (IKKE,IKBKE) nemo (NEMO) Protein Dorsal Cactus Dif Relish v-Rel c-Rel RelB RelA (p65) p50, p105 or p50/p105 p52, p100 or p52/p100 IkBa IkBb IkBe IkBg IkBz Bcl-3 IKKa IKKb IKKe NEMO (IKKg)

Abbreviations: IKK, IkB kinase; NEMO, NF-kB essential modulator; NF-kB, nuclear factor-kappaB.

NF-kB in simple organisms Among simpler organisms, we will begin to have an appreciation for the role that NF-kB plays in insect vectors for human and animal diseases (e.g., in mosquitoes carrying disease-causing microbes), and how we might genetically modify components of NFkB signaling in these organisms to decrease their vector effectiveness. Similarly, I suspect that our recent discovery of a primitive NF-kB system in the phylum Cnidaria will lead to insights into the role that NF-kB plays in the response of simple marine organisms and ecosystems to environmental stresses, be they manmade, physical or biotic.

An updated nomenclature for components of the NF-jB signal-transduction pathway Among the many publications on this topic, there are inconsistencies in the naming of genes and proteins in the NF-kB pathway. In this collection of reviews, we have used what I believe is a simple nomenclature (Table 1). The revised nomenclature reects the new members of the pathway, common usage over the past several years, and at times my own judgment.
Acknowledgements I thank Melissa Chin for help with artwork in the gures. I also thank D Kalaitzidis, D Starczynowski and members of my lab for comments on the manuscript. I especially thank the authors for their contributions to this review issue, and collectively, we apologize to any researchers who feel that their work was overlooked. Research in my laboratory on NF-kB is currently supported by the National Institutes of Health. For additional, detailed information on NF-kB, the reader is directed to our lab website at www.nf-kb.org.

Oncogene

Introduction to NF-jB TD Gilmore

6684 References Baeuerle PA, Baltimore D. (1988). Cell 53: 211217. Baltimore D, Sen R, Sharp PA, Singh H, Staudt L, Lebowitz JH et al. (2002) US Patent 6,410,516. Basseres D, Baldwin Jr AS. (2006). Oncogene 25: 68176830. Bubici C, Papa S, Dean S, Franzoso G. (2006). Oncogene 25: 67316748. Chen FE, Ghosh G. (1999). Oncogene 18: 68456852. Courtois G, Gilmore TD. (2006). Oncogene 25: 68316843. De Bosscher K, Vanden Berge W, Haegeman G. (2006). Oncogene 25: 68686886. linas C. (2006). Oncogene Dutta J, Fan Y, Gupta N, Gan G, Ge 25: 68006816. Gerondakis S, Grumont R, Gugasayn R, Wong L, Isomura I, Ho W et al. (2006). Oncogene 25: 67816799. Ghosh S, Gifford AM, Riviere LR, Tempst P, Nolan GP, Baltimore D. (1990). Cell 62: 10191029. Gilmore TD. (1990). Cell 62: 841843. Gilmore TD ed. (1999). NF-kB. Oncogene 18: 68416964. Gilmore TD, Herscovitch M. (2006). Oncogene 25: 68876899. Gilmore TD, Temin HM. (1986). Cell 44: 791800. Hayden MS, West AP, Ghosh S. (2006). Oncogene 25: 67586780. Hiscott J, Nguyen T-LA, Arguello M, Nakhaei P, Paz S. (2006). Oncogene 25: 68446867. Hoffmann A, Natoli G, Ghosh G. (2006). Oncogene 25: xxxxxx. Kieran M, Blank V, Logeat F, Vandekerckhove J, Lottspeich F, Le Bail O et al. (1990). Cell 62: 10071018. Minakhina S, Steward R. (2006). Oncogene 25: 67496757. Perkins ND. (2006). Oncogene 25: 67176730. Scheidereit C. (2006). Oncogene 25: 66856705. Sen R, Baltimore D. (1986). Cell 47: 921928. Stephens RM, Rice NR, Hiebsch RR, Bose Jr HR, Gilden RV. (1983). Proc Natl Acad Sci USA 80: 62296233. Steward R. (1987). Science 238: 692694. Steward R, Zusman SB, Huang LH, Schedl P. (1988). Cell 55: 487495. Sullivan J, Kalaitzidis D, Gilmore TD, Finnerty JR. (2006). Devel Genes Evol (in press). Wilhelmsen KC, Eggleton K, Temin HM. (1984). J Virol 52: 172182.

Oncogene