Вы находитесь на странице: 1из 22

Oncogene (2003) 22, 8961–8982

& 2003 Nature Publishing Group

All rights reserved 0950-9232/03 $25.00

Publishing Group All rights reserved 0950-9232/03 $25.00 www.nature.com/onc To be, or not to be: NF- j

www.nature.com/onc

To be, or not to be: NF-jB is the answer – role of Rel/NF-jB in the regulation of apoptosis

Je´roˆme Kucharczak 1,4 , Matthew J Simmons 1,2,4 , Yongjun Fan 1,4 and Ce´line Ge´linas* ,1,3

1 Center for Advanced Biotechnology and Medicine, 679 Hoes Lane, Piscataway, NJ, USA; 2 Graduate Program in Biochemistry and Molecular Biology, Piscataway, NJ, USA; 3 Department of Biochemistry and Cancer Institute of New Jersey, Robert Wood Johnson Medical School, University of Medicine and Dentistry of New Jersey, Piscataway, NJ, USA

During their lifetime, cells encounter many life or death situations that challenge their very own existence. Their survival depends on the interplay within a complex yet precisely orchestrated network of proteins. The Rel/NF- jB signaling pathway and the transcription factors that it activates have emerged as critical regulators of the apoptotic response. These proteins are best known for the key roles that they play in normal immune and inflammatory responses, but they are also implicated in the control of cell proliferation, differentiation, apoptosis and oncogenesis. In recent years, there has been remark- able progress in understanding the pathways that activate the Rel/NF-jB factors and their role in the cell’s decision to either fight or surrender to apoptotic challenge. Whereas NF-jB is most commonly involved in suppres- sing apoptosis by transactivating the expression of antiapoptotic genes, it can promote programmed cell death in response to certain death-inducing signals and in certain cell types. This review surveys our current understanding of the role of NF-jB in the apoptotic response and focuses on many developments since this topic was last reviewed in Oncogene 4 years ago. These recent findings shed new light on the activity of NF-jB as a critical regulator of apoptosis in the immune, hepatic, epidermal and nervous systems, on the mechanisms through which it operates and on its role in tissue development, homoeostasis and cancer. Oncogene (2003) 22, 8961–8982. doi:10.1038/sj.onc.1207230

Keywords: Rel; NF- k B; apoptosis; transcription factor; tumor necrosis factor; lymphocyte; neurodegenerative disease; hepatocyte; cancer therapy

Introduction

The signaling pathways that control the life and death of a cell are of prime interest in modern biology. Mutations affecting components of these pathways are often

*Correspondence: C Ge´linas, Center for Advanced Biotechnology and Medicine, 679 Hoes Lane, Piscataway, NJ, USA; E-mail: gelinas@cabm.rutgers.edu 4 These authors contributed equally

implicated in diseases arising from too much or too little apoptosis, such as neurodegeneration and cancer. The vertebrate Rel/NF- kB transcription factors are criticalplayers in the controlof the apoptotic response to a variety of stimuli. They are best known for their pivotalroles in immune, inflammatory and acute phase responses, but they also play important roles in cell growth, apoptosis and oncogenesis (reviewed in Silver- man and Maniatis, 2001; Karin and Lin, 2002; Li and Verma, 2002). The Rel/NF- kB family is comprised of the vertebrate c-Rel, RelA, RelB, p105/NF- k B1 and p100/NF- k B2 proteins, the viraloncoprotein v-Rel, Xenopus X-Rel1, and the Drosophila Dorsal, Dif and Relish factors. These structurally related proteins share extensive sequence similarity in their N-terminal Rel homology domain (RHD) that enables them to translocate to the nucleus, form dimers with one another and bind to k B DNA sites (GGGRNNYYCC). The distinct C-terminal ends of vertebrate Rel/NF- k B factors participate in transcriptional activation (v-Rel, c-Rel, RelA, RelB) or in the controlof Relprotein activity (p105/NF- k B1, p100/NF- k B2). The latter group shares homology with the I kB family of Rel protein regulators that control their subcellular localization and inhibit their DNA-

binding activity (IkB a , I kBb , I kBe , Bcl-3, p105/NF- k B1,

p105/NF- k B2).

Studies in recent years have provided important insights into the pathways that lead to NF- kB activation (reviewed in Karin and Ben-Neriah, 2000). In most cells, Rel/NF- k B subunits are sequestered in the cytoplasm as inactive homo- or heterodimers, bound to one of severalI k B factors. The rapid and transient activation of NF- kB complexes in response to a wide variety of stimuli generally involves phosphorylation of I k B by the I kB kinase complex (IKK), which contains the catalytic subunits IKK a and IKK b and the regulatory IKK g /NEMO protein. Phosphorylation targets Ik B for degradation via the ubiquitin– proteasome pathway and culminates in the nuclear translocation of active Rel/NF- k B dimers, their binding to DNA and the transcriptional activation of cellular genes involved in immune responses, inflammation, viral infection, cell proliferation and survival. In turn, nuclear NF- kB factors trigger the resynthesis of I k Ba , giving rise to an autoregulatory loop that terminates the

Rel/NF- j B and apoptosis J Kucharczak et al 8962 activation process. Phosphorylate I k

Rel/NF-jB and apoptosis

J Kucharczak et al

8962

activation process. Phosphorylate I k Ba to promote its degradation (Schwarz et al., 1996; Miyamoto et al., 1998; Romieu-Mourez et al., 2001; Shen et al., 2001; Kato et al., 2003). In addition to the classical NF- k B signaling pathway, a noncanonical NF- k B cascade has been identified to involve signaling via the NF-k B- inducing kinase (NIK) and IKK a , and results in the processing of p100/NF- kB2 to mature p52/NF- k B2

(Senftleben et al., 2001; Xiao et al., 2001). Hundreds of NF- k B-regulated genes have been identified (reviewed

in Pahl, 1999). While it is clear that the particular

combination of NF- k B subunits within distinct NF- k B homo- or heterodimers is in a large part responsible for determining the subsets of genes that they activate, a recent study has added a new dimension to this concept. Experiments with cell lines deficient for single or multiple NF- kB subunits revealed that the in vivo specificity of cellular gene activation does not only lie within the sequence of the kB DNA site, but is also likely to be greatly influenced by combinatorial protein– protein interactions with other promoter-bound factors (Hoffmann et al., 2003). This additionallevelof regulation is likely to help determine the specificity of gene activation to ensure a differentialresponse to different stimuli. The last few years have seen a virtual explosion in the number of studies investigating the role of NF-k B in apoptosis in different systems, and important progress was made in understanding the mechanisms involved. Here, we review the pro- and antiapoptotic effects of NF- k B, its important involvement in different physio- logical contexts, the mechanisms by which it operates and those that regulate its activity.

Lessons in survival: studies of NF- k B’s pro- and antiapoptotic functions in knockout mice and model cell systems

All five members of the mammalian Rel/NF- k B family

and severalcomponents of the NF-k B signaling path- way have been targeted by homozygous inactivation in mice to characterize their function. The phenotypes of

these animals have been reviewed in detail (reviewed in

Li and Verma, 2002). These studies revealed distinct

roles for individual NF- k B subunits in innate and adaptive immune responses, and also unveiled a pivotal role for NF- k B in regulating apoptosis in the immune

and nervous systems, in the liver, and in hair follicles and epidermalappendages. This section focuses on those that revealed a link to apoptosis, on studies in in vitro cell systems that support these findings and emphasizes recent discoveries that shed further light

on the subject.

NF- k B: life and death decisions in the immune system

A protective role for NF-k B in B cells The survivalof peripheral B cells in response to antigen depends on B- cell receptor (BCR)-mediated activation of NF- k B and the induction of antiapoptotic target genes. Early studies in B-lymphocyte cell lines demonstrated a vital

Oncogene

role for NF- kB in this process, as suppression of endogenous NF- k B activity invariably led to apoptosis in cells treated with anti-IgM antibodies, whereas ectopic expression of c-rel restored viability (Wu et al., 1996a, b). Consistent with these findings, primary B cells from c-rel-deficient mice undergo apoptosis upon mitogenic stimulation, and this phenotype is rescued by transgenes encoding c-Relitself, or its antiapoptotic targets Bcl-2, Bcl-xL or Bfl-1/A1 (Kontgen et al., 1995; Grumont et al., 1998, 1999; Tumang et al., 1998; Owyang et al., 2001). Whereas p50/NF- kB1 is important for the survivalof quiescent B cells, combined inactiva- tion of p50 and c-rel in double knockout mice established a critical role for these factors in mitogen- activated B cells (Grumont et al., 1998). The pathway leading to NF-k B activation in response to BCR engagement has been elucidated (Figure 1A). As anticipated, inactivation of individualcomponents of this cascade, including B-lineage-specific depletion of IKK b or IKK g /NEMO, results in dramatic inhibition of NF- k B stimulation following BCR engagement and

increased apoptosis (Petro et al., 2000; Leitges et al., 2001; Petro and Khan, 2001; Tan et al., 2001; Martin et al., 2002; Pasparakis et al., 2002b; Li et al., 2003). In some of these studies, this was correlated with a failure to induce expression of antiapoptotic proteins Bcl-xL and Bcl-2. Contrary to the single knockouts of c-rel or relA, B cells from compound c-rel/relA double knockout embryos failto mature to the IgM(lo)IgD(hi) stage due to premature death by apoptosis, coincident with impaired expression of Bcl-2 and Bfl-1/A1 (Grossmann et al., 2000). A bcl-2 transgene could rescue this phenotype. In contrast, expression of a bcl-xl transgene into bone marrow cells transduced with a transdominant form of I k Ba could not surmount the impairment for maturation to follicular B cells in recipient mice, although it restored viability to pre-B/immature B-cell subsets (HC Liou, personalcommunication). Together, these results emphasize an important role for individual NF- k B subunits in B-cell survival, maturation and function. NF- k B has also been implicated in attenuating apoptosis in response to the cytokine BAFF (also known as BlyS), which is necessary for peripheral B-cell development, as illustrated by the phenotype of mice expressing nonfunctionalBAFF receptor (BAFF-R) or lacking BAFF (Schiemann et al., 2001). Recent studies showed that in quiescent B cells, BAFF activates NF- kB through both the classical and the noncanonicalNF- kB pathways, with predominance for the latter (Figure 1A; Kayagaki et al., 2002; Hatada et al., 2003). Whereas binding of BAFF to BAFF receptor 3 (BR3) leads to processing of p100/NF-

k B2 to p52, one study revealed a requirement for the

NF- k B subunit of p50 for the survivalfunction of BAFF (Hatada et al., 2003). As a consequence of BAFF-mediated NF- k B activation, antiapoptotic genes encoding Bcl-2, Bcl-xL and Bfl-1/A1 are activated, but only in mature B cells (Do et al., 2000; Hsu et al.,

2002).

Rel/NF-jB and apoptosis

J Kucharczak et al

Rel/NF- j B and apoptosis J Kucharczak et al 8963 Figure 1 NF- k B activation

8963

Rel/NF- j B and apoptosis J Kucharczak et al 8963 Figure 1 NF- k B activation

Figure 1 NF-k B activation in adaptative and innate immune responses (A) B-cell activation triggers the NF- kB signaling pathway in the adaptive immune response. (a) BCR activation by antigen leads to the recruitment of a functional complex containing Syk, Btk, BLNK and PLC-g 2. Activation of PLC-g2 provokes the release of calcium and diacylglycerol to activate PKC-b and induce both the classical and noncanonical NF- kB pathways via phosphorylation of IKKa and IKKb. Both pathways lead to the activation of NF-k B- responsive genes. Contrary to PKC-b, PKC-z does not activate NF- kB via IKK, but rather directly phosphorylates RelA to enhance its transcriptional activity. (b) BAFF signaling in B cells induces the noncanonical NF-k B pathway. BAFF-R engagement leads to the activation of NIK that phosphorylates and activates IKKa. In turn, IKKa phosphorylates p100/NF-k B2 in an IKKg /NEMO- independent manner. This initiates ubiquitin-mediated proteolytic processing of p100 to p52. Formation of p52 heterodimers with RelB results in NF-k B-dependent gene activation. (B) Engagement of cell surface TLRs and cytosolic NOD receptors activates NF- kB signaling in the innate immune response. (a) The binding of bacterial LPS to TLR4 triggers recruitment of MyD88 and the serine/ threonine kinase IRAK, leading to its association with adaptor protein TRAF6 and MAP3K. This intracellular signaling cascade converges on activation of the IKK kinase complex and NF- kB to induce gene expression. (b) Ligands such as MDP, released following phagocytosis of bacteria, are recognized by cytosolic NOD receptors. Activation of some NODs can induce two different signaling pathways. One pathway activates the classical NF-k B signaling cascade in a RICK- and RIP-dependent manner, resulting in NF-kB-dependent expression of proinflammatory cytokines, and in the case of NOD2 of prosurvivalfactor Bfl-1/A1. The other pathway leads to caspase activation and cell death

Anti- and proapoptotic roles for NF-kB in T cells The full activation of naı¨ve T cells requires engagement of the T-cell receptor (TCR) and costimulatory signaling by CD28. This leads to NF-kB activation and con- sequent cell survival via induction of antiapoptotic genes, and to proliferation via synthesis of IL-2 and GM-CSF. Studies with wild-type and p50 / c-Rel / CD4 þ T cells demonstrated that activation of NF-k B following TCR engagement in conjunction with CD28 costimulation is critically required to induce expression of antiapoptotic genes bcl-xl and bcl-2 and promote T- cell viability (Zheng et al., 2003). Similar results were observed in CD4 þ T cells expressing a super-repressor form of I k B a (Khoshnan et al., 2000). Whereas cessation of TCR stimulation in wild-type T cells caused a rapid drop in endogenous NF- k B levels along with apoptosis, enforced expression of IKK b enabled cells to survive in the absence of sustained TCR stimulation (Zheng et al., 2003). A recent analysis of activated CD4 þ or CD8 þ T cells indicated that while induction of the NF- k B targets Bcl-xL and its homologue Bfl-1/A1 is transient, that of Bcl-2 is delayed and IL-2 dependent (Verschelde et al., 2003).

In the absence of CD28-derived costimulatory signals,

apoptosis of antigenically stimulated naı¨ve T cells can be inhibited in vivo by adjuvant-induced inflammation.

Microarray analyses revealed that sustained T cell response in a modelof adjuvant-induced survivalwas correlated with increased expression of the I kB-related Bcl-3 protein (Mitchell et al., 2002a). Retroviral- mediated transfer of Bcl-3 itself or a subdomain containing its first and second ankyrin repeats could confer survival to activated T cells in vitro and in vivo

(Mitchell et al., 2001, 2002b). Unlike other I k B molecules, Bcl-3 is generally found in the nucleus where it modulates transcription via interaction with homo- dimers of p50/NF- k B1 or p52/NF- k B2. It was thus postulated that Bcl-3 might enhance cell viability in this context by inducing the expression of NF- kB-regulated genes. These results reveal a mechanism underlying the protective effects of inflammation and innate immunity toward T-cell apoptosis in the adaptive immune response (reviewed in Gerondakis and Strasser, 2001).

A recently published report has unveiled a new way

for NF-k B to antagonize cell death in antigenically stimulated naı¨ve T cells in the absence of costimulatory

Oncogene

Rel/NF- j B and apoptosis J Kucharczak et al 8964 Table 1 Products of NF-

Rel/NF-jB and apoptosis

J Kucharczak et al

8964

Table 1

Products of NF-k B-regulated target genes with anti- or proapoptotic activities

Antiapoptotic proteins

Function in apoptosis

References

IAPs

c-IAP1

Inhibits caspase-3 and -7, prevents activation of

You et al. (1997), Deveraux et al. (1998), Wang et al.

procaspase-9

(1998)

c-IAP2

Inhibits caspase-3 and -7, prevents activation of

Chu et al. (1997), Deveraux et al. (1998), Wang et al.

procaspase-9

(1998)

XIAP

Inhibits caspase-3 and -7, prevents activation of

Liston et al. (1996), Stchlik et al. (1998), Takahashi et al. (1998) Tang et al. (2001)

procaspase-9

Inhibits TNF-induced JNK activation

Bcl-2 family

Bcl-2

Antagonizes

proapoptotic Bcl-2 family members

Tamatani et al. (1999), Grossmann et al. (2000), Catz and Johnson (2001) Lee et al. (1999a), Chen et al. (2000) Zong et al. (1999), Grumont et al. (1999), Wang et al. (1999), Lee et al. (1999a) Lee et al. (1999b)

Bcl-xL

Antagonizes proapoptotic Bcl-2 family members

Bfl-1/A1

Antagonizes

proapoptotic Bcl-2 family members

NR13

Antagonizes proapoptotic Bcl-2 family members

Adaptor molecules

TRAF-1

Adaptor molecule, augments NF- kB activation Adaptor molecule, augments NF- kB activation

Wang et al. (1998), Schwenzer et al. (1999) Wang et al. (1998)

TRAF-2

Others

A20

Blocks the TNFa death pathway

Krikos et al. (1992), Malewicz et al. (2003)

MnSOD

Clears reactive oxygen species Bernard et al. (2002, 2001a), Delhalle et al. (2002), Tanaka et al. (2002)

c-FLIP

Interferes with activation of procaspase-8 and -10

Kreuz et al. (2001), Micheau et al. (2001)

DcR1 (decoy receptor 1)

Decoy TrailDR Inhibits TNFa-induced JNK activation Prevents Fas-induced apoptosis in B cells

Bernard et al. (2001) De Smaele et al. (2001), Jin et al. (2002) Zazzeroni et al. (2003)

GADD45b

Proapoptotic proteins

Ligands

FasL (CD95L)

DR ligand

Matsui et al. (1998), Kasibhatla et al. 1999 Siegmund et al. (2001), Baetu et al. (2001), Rivera- Walsh et al. (2001)

TRAIL

DR ligand

Receptors

Fas (CD95)

DR

Chan et al. (1999), Kasibhatla et al. 1998 Ravi et al. (2001) Ravi et al. (2001) Kasof et al. (2001)

DR4 (TRAIL-R1)

DR

DR5 (TRAIL-R2)

DR

DR6

DR

Transcription factors

p53

Tumor suppressor; transcription factor

Wu and Lozano (1994)

c-Myc

Transcription factor

Duyao et al. (1990), La Rosa et al. (1994)

Bcl-2 family

Bcl-xS

Antagonizes prosurvival Bcl-2 family members

Chen et al. (2003)

signal, that is, by suppressing expression of the p53- related p73 protein (Wan and DeGregori, 2003). In this study, inhibition of NF- k B in T cells by a super- repressor form of I k Ba triggered E2F- and p73- dependent apoptosis upon cell activation. In contrast,

cells lacking p73 survived despite inhibition of NF-k B. The mechanism by which NF- k B downregulates p73 remains to be clarified, but the presence of putative NF-

k B binding sites in the vicinity of p73’s first exon gave

rise to speculation that NF- k B might interfere with E2F- mediated transactivation of p73. This study is consistent with prior work implicating E2F1 and p73 in inducing

apoptosis in activated mature T cells (reviewed in Green, 2003). This newly uncovered interplay between NF- k B

Oncogene

and the Cdk-Rb-E2F pathway to controlthe outcome of T cells following receptor engagement suggests that in the absence of costimulatory signal to activate NF-k B, E2F-mediated induction of p73 could perhaps serve to eliminate autoreactive T cells from the peripheral repertoire (reviewed in Green, 2003). These results provide new insights into the role of NF- kB in the survivalof antigen-stimulated naı¨ve T cells. In contrast to the antiapoptotic activity of NF- k B described above, others showed that NF-k B contributes to apoptosis during activation-induced cell death (AICD) in mature T cells. In this context, NF- k B triggered the Fas-death cascade by directly activating the expression of Fas ligand (FasL) (Table 1; Kasibhatla

et al., 1999; Lin et al., 1999; Zheng et al., 2001). The expression of a dominant I kB molecule (Ik Ba M) inhibited both FasL expression and apoptosis. However,

it remains to be determined how this pathway of AICD

relates to that involving Cdk-Rb-E2F and p73 described above. The protective role of NF- kB during T-cell develop-

ment is also a subject of debate. Studies indicate that

NF- k B is required for the survivalof developing

thymocytes, as pre-T cells in which NF-k B expression

was abolished undergo spontaneous apoptosis, whereas

a constitutively active form of IKK allowed cells to

differentiate to the CD4 þ CD8 þ double-positive stage (Voll et al., 2000). This agrees with the impaired thymocyte development observed in transgenic mice expressing I k B in the T-cell lineage (Boothby et al., 1997; Esslinger et al., 1997). In contrast, others reported a proapoptotic role for NF- k B in double-positive CD4 þ CD8 þ thymocytes (Hettmann et al., 1999; Kim et al., 2002a). T-cell-specific expression of a CD2-I k B a M transgene conferred resistance to anti-CD3-mediated apoptosis (Hettmann et al., 1999). Further studies will help to determine how NF- k B elicits different apoptotic responses in different contexts.

A possible role for NF- k B in suppressing apoptosis in the innate immune response A link between NF- k B and apoptosis has also surfaced in the innate immune response, which plays a fundamental role in the detection and elimination of pathogens. The nucleotide- binding oligomerization domain (NOD) proteins act as cytosolic regulators of inflammation and apoptosis (Figure 1B; reviewed in Inohara and Nunez, 2003). NOD1 and NOD2 activate the canonicalNF- kB path- way in a RICK-dependent manner (Bertin et al., 1999; Geddes et al., 2001; Ogura et al., 2001; Kobayashi et al., 2002). NOD1 and IPAF activate caspase-1 to form an ‘inflammasome’ required for cleavage of IL-1b during the innate immune response (Poyet et al., 2001; Grenier et al., 2002; Martinon et al., 2002; Wang et al., 2002a; Yoo et al., 2002). In addition to promoting expression of proinflammatory cytokines, NOD-mediated activation of NF- k B was shown in one instance to also promote transcription of antiapoptotic genes, as demonstrated by activation of the prosurvival Bcl-2 family member Bfl-1/ A1 following activation of NOD2 by bacterial muramyl dipeptide (MDP) (Inohara et al., 2003). Interestingly NOD1, NOD2, IPAF and DEFCAP can also induce cell death in overexpression studies. In light of the dual capacity of some NOD proteins to both promote apoptosis and activate NF-k B, it is speculated that in

a physiological context NF- k B-dependent activation of

antiapoptotic genes by NOD2 might serve to modulate its proapoptotic effect (Inohara and Nunez, 2003). Further experimentation will test this hypothesis. Toll-like receptors (TLRs) recognize extracellular microbial components in myelomonocytic cells and activate the NF- k B signaling pathway via IL-1-recep- tor-associated kinase (IRAK) and tumor necrosis factor (TNF) receptor-associated factor TRAF6 to induce

Rel/NF-jB and apoptosis

J Kucharczak et al

to induce Rel/NF- j B and apoptosis J Kucharczak et al 8965 production of proinflammatory cytokines

8965

production of proinflammatory cytokines and costimu- latory molecules and trigger a defensive reaction (Figure 1B; reviewed in Takeda and Akira, 2001). TLR activation has also been reported to trigger apoptosis in some cases e.g. (Aliprantis et al., 2000). A recent example comes from the 19-kDa Mycobacterium tuberculosis protein (p19) that increased the rate of apoptosis of infected macrophages in a caspase-8- dependent manner (Lopez et al., 2003). However, it is unclear whether NF- kB induces expression of antia- poptotic genes in this context. A recent report indicated that TLR4 activation increases neutrophilsurvivalin an NF-k B- and mitogen-activated protein kinase (MAPK)- dependent manner via production of survivalcytokines (Sabroe et al., 2003). Thus it appears that similar to the NOD pathway, a picture is emerging in which both NF- k B and apoptosis can be triggered upon TLR activation in some instances. Prolonged cell survival in these cases depends on the proper balance between proapoptotic and prosurvivalsignals induced in NF- kB-dependent or -independent manners.

A critical role for NF- k B in suppressing hepatocyte apoptosis

NF-k B was ascribed both beneficialand damaging effects in the liver, owing to its role in liver development, regeneration and carcinogenesis (reviewed in Taub, 1998). Severalyears ago, the dramatic phenotype of RelA-deficient mice provided compelling evidence to support a protective role for NF-k B in suppressing liver apoptosis. RelA þ animals succumb by days 15.5–16.5 of embryogenesis due to massive apoptosis of hepatocytes (Beg and Baltimore, 1996; reviewed in Li and Verma, 2002). Mouse embryos lacking both RelA and NF-k B1 (i.e. the classical p50/p65 NF- k B dimer) died earlier, at about E13 of gestation (Horwitz et al., 1997). The combined inactivation of RelA and c-Rel in compound knockouts also led to advanced death compared to RelA þ embryos, suggesting that the lack of RelA can be alleviated to some extent by c-Rel (Grossmann et al., 1999; Grossmann et al., 2000). Not surprisingly, the recent analysis of animals deficient for upstream components of the NF- kB signaling cascade revealed a similar phenotype. Mice lacking IKK b , IKK g /NEMO or both IKK a /IKK b showed embryonic lethality between E12.5–E14.5 (Li et al., 1999a, c, 2000; Tanaka et al., 1999; Rudolph et al., 2000). Homozygous deletion of the IKK kinase homologue T2K (TBK1, NAK), which associates with TRAF2 but is not a component of the IKK complex, also led to severe liver degeneration (Bonnard et al., 2000). The massive hepatocyte apopto- sis in RelA and IKK b knockout animals was attributed to increased sensitivity to circulating TNF a , as embryo- nic liver apoptosis was not observed in compound knockouts for RelA and TNF a , or IKK b and TNF-R1 (Doi et al., 1999; Alcamo et al., 2001). This agrees with the increased sensitivity of mouse embryonic fibroblasts (MEFs) and macrophages from relA þ mice to TNFa - induced apoptosis (Beg and Baltimore, 1996). It is interesting however that MEFs from T2K / mice are

Oncogene

Rel/NF- j B and apoptosis J Kucharczak et al 8966 not sensitized to TNF a

Rel/NF-jB and apoptosis

J Kucharczak et al

8966

not sensitized to TNF a nor is NF- k B DNA-binding activity affected, although NF- k B’s transcriptional activity is impaired in these cells (Bonnard et al., 2000). These recent findings suggest a role for T2K in suppressing embryonic liver apoptosis that apparently does not depend on degradation of I k B or NF- k B DNA binding. Studies in hepatocytes treated with transforming growth factor- b (TGF-b ) further support a protective role for NF- k B. Prior studies indicated that TGF- b induces apoptosis by stabilizing I k Ba and inactivating NF- k B (Arsura et al., 1997). Recent work has shed further light on the subject by showing that TGF-b 1 inhibits the ability of CK2 to stabilize I k B a in immortalized hepatocytes (Cavin et al., 2003). Transient activation of NF- kB via a TAK1/IKK pathway was also described to play a role in increasing I k B a synthesis to repress NF- k B in liver epithelial cells, allowing TGF- b 1- induced cell death to proceed through AP-1/SMAD (Arsura et al., 2003).

An essential role for NF- kB in the development and survival of hair follicles and epidermal appendages

An important role for NF- k B in epidermalhomeostasis and the development of skin appendages has only recently come to light. Five independent lines of investigation support this conclusion: (1) Mice that ubiquitously express a super-repressor of NF- k B (I k BD N) display impaired development of hair follicles and exocrine glands due to increased apoptosis (Schmidt-Ullrich et al., 2001). This phenotype is reminiscent of the human epidermaldisorder hypohi- drotic (anhydrotic) ectodermaldysplasia (HED), and is indistinguishable from that of mice with mutations in the genes encoding the TNF-related ligand ectodermal dysplasia (EDA) (tabby), its TNFR-related receptor ectodysplasin receptor (EDAR) (Downless) and crinkled (Cr or Edaradd for EDAR-associated death domain (DD)), a novelDD-containing adaptor that binds EDAR and mediates engagement of the NF- k B pathway (Headon et al., 2001; Yan et al., 2002). (2) Heterozygous IKK g /NEMO females survive to birth but show traits reminiscent of those associated with human incontinentia pigmenti (IP), the first genetic disorder attributed to NF-kB dysfunction (Schmidt- Supprian et al., 2000; reviewed in Smahi et al., 2002). Mice heterozygous for IKK g have skin lesions with granulocyte infiltration, keratinocyte apoptosis and abnormaldevelopment of teeth, eyes and hair. TNF a plays an essential role in the development of IP lesions as human IP cells are highly sensitive to TNF-induced apoptosis and failto elicit an NF- k B response. Patients with EDA also show abnormal development of ecto- derm-derived structures (teeth, hair, nails, sweat glands) and display hypomorphic NEMO mutations (reviewed in Smahi et al., 2002). (3) Conditionaldeletion of IKK b in the epidermis and hair follicles recapitulates many of the traits seen in IKK g /NEMO þ / embryos, with the notable exception that the extensive keratinocyte apop- tosis seen in NEMO knockouts was not observed in the

Oncogene

skin of IKK b -deficient embryos (Pasparakis et al.,

2002a). The mutant phenotype caused by conditional inactivation of IKK b in the epidermis was suppressed in

a TNFR-deficient background. (4) TRAF6-deficient

mice suffer from HED due to impaired development

of epidermalappendages (Naito et al., 2002). Although

TRAF6 does not associate with EDAR, it associates with X-linked ectodysplasin-A2 receptor (XEDAR) expressed in epidermalappendages and is essentialfor XEDAR-mediated NF- k B activation. (5) Loss of tumor suppressor CYLD causes the autosomaldomi-

nant syndrome cylindromatosis (turban tumor syn- drome), that predisposes patients to benign tumors of hair follicles and cells of the sweat and scent glands (reviewed in Wilkinson, 2003). Interestingly, CYLD was recently shown to interact directly with TRAF2 and IKK g /NEMO and to function as a TRAF2 deubiquitinating enzyme to negatively regulate IKK- mediated activation of NF- k B downstream of CD40, XEDAR and EDAR (Brummelkamp et al., 2003; Kovalenko et al., 2003; Trompouki et al., 2003). Loss

of CYLD activates TNF a signaling through NF- k B and

decreases apoptosis, whereas agents that antagonize NF- k B signaling interfere with cell death inhibition caused by loss of CYLD (Brummelkamp et al., 2003). Altogether, the studies in this section highlight a critical role for NF- kB in suppressing apoptosis to allow differentiation and survival of hair follicles and epider- malappendages.

A double-edged sword: anti- and proapoptotic effects of NF- kB in the nervous system

FunctionalNF-k B complexes (mostly p50 and p65) are

present in essentially all cell types of the nervous system, including neurons, astrocytes, microglia and oligoden- drocytes. Among the many stimuli that can activate NF-

k B in this system are membrane depolarization, nerve

growth factor (NGF), opioids, the secreted form of b - amyloid protein, TNF and glutamate (reviewed in Mattson and Camandola, 2001). However, the role of

NF- k B within the developing and mature central nervous system has been controversial. While some studies report a proapoptotic role, most indicate a

cytoprotective effect. Accumulating evidence points to NF-k B as a survival determinant for neurons. Consistent with its expression pattern during development of the nervous system, NF-

k B was implicated in NGF-mediated protection of

developing sensory neurons to cytokines (Hamanoue et al., 1999; Middleton et al., 2000). Moreover, inhibition of NF- k B in primary rat corticalneurons with a super-repressor form of I kBa , a dominant- negative form of NIK, or NF- k B inhibitors partheno- lide, SN50 or BAY 11-7082 triggers rapid activation of an intrinsic apoptotic program and is accompanied by reduced levels of NF- kB-regulated transcripts for antiapoptotic genes Bcl-2, Bcl-xL and Bfl-1/A1, in particular (Bhakar et al., 2002; Chiarugi, 2002). Similarly, cultured relA / neurons show reduced survivalcompared to wild type (Hamanoue et al.,

1999). Incidentally, despite the fact that homozygous inactivation of bcl-2 or bcl-x is embryonic lethal, embryos show severe loss of neurons in the central and peripheralnervous systems, suggesting a protective

role for these NF- k B targets in developing neurons (Motoyama et al., 1995; Michaelidis et al., 1996). In the PC12 model cell system, NGF stimulation was shown to activate NF- k B via the p75(NTR) and TrkA receptors and to result in different physiological effects (Foehr et al., 2000b). Whereas NF-k B activation via the p75(NTR) receptor significantly blocked apoptosis, its activation via TrkA induced neurite process formation. Similarly, overexpression of NIK induced PC12 cell differentiation and antagonized cell death via the IKK and MAPK signaling cascades (Foehr et al., 2000a). Consistent with these findings in cultured cells, embryos from IKK a /IKK b double knockout mice show defects in neurulation due to increased apoptosis. These were not apparent in the single knockouts, and it should be noted that embryos from IKK g /NEMO / mice lack neuronaldefects (Li et al., 2000). It was proposed that IKK a and IKK b have redundant effects during neuronaldevelopment and that NF- kB-independent mechanisms might be involved in the neuronal pheno- type of the IKK a /IKK b double knockout, akin to the

NF- k B-independent effects observed on the skin phe-

notype of IKK a / mice (Hu et al., 1999; Li et al., 1999b;

Takeda et al., 1999). Severalexperiments indicate that NF- k B is also neuroprotective in response to injury and that NF-k B decoys abolish protection (reviewed in Mattson and Camandola, 2001). NF-k B was shown to antagonize neurodegeneration in this context by triggering expres- sion of various antiapoptotic genes including inhibitor of apoptosis proteins (IAPs) that protect neurons in experimentalmodels of stroke or seizure, manganese superoxide dismutase (MnSOD) a mitochondrialanti- oxidant enzyme that is neuroprotective, and cell death inhibitors Bcl-2 and Bcl-xL (Mattson and Camandola, 2001). NF-k B was also ascribed a cytoprotective effect when induced by amyloid deposits in vivo and in cell culture studies of neurons (reviewed in Mattson and Camandola, 2001). Indeed, exposure of cortical neurons to the amyloid beta-peptide associated with Alzheimer’s disease leads to neurodegeneration and is accompanied by increased levels of I k Ba and decreased NF-k B activity (Bales et al., 1998). In contrast, the neuropro- tective alpha-secretase-derived form of the beta-amyloid precursor protein activates NF-kB and protects against apoptosis induced by the neurotoxic amyloid beta- peptide (Guo et al., 1998). Increased NF- k B activity was also observed in other neurodegenerative disorders, such as Parkinson’s disease and amyotrophic lateral sclerosis, whereby it may correspond to an early defense mechan- ism against oxidative stress and mitochondrialdysfunc- tion (reviewed in Mattson and Camandola, 2001). Consistent with this notion, p50-deficient animals treated with a mitochondrialtoxin in an experimental modelof Huntington’s disease showed increased striatal neuronal damage and failure to upregulate MnSOD, in contrast to wild-type mice (Yu et al., 2000).

Rel/NF-jB and apoptosis

J Kucharczak et al

al ., 2000). Rel/NF- j B and apoptosis J Kucharczak et al 8967 Contrary to the

8967

Contrary to the work described above, other reports described a proapoptotic effect for NF- k B in in vitro and in vivo models of neuronal cell injury (reviewed in Barkett and Gilmore, 1999; Denk et al., 2000). Many of them converge on tumor suppressor p53, which was implicated as a detrimental factor for neuronal viability in response to ischemia/reperfusion and excitotoxic insult (Crumrine et al., 1994; Morrison et al., 1996; Xiang et al., 1996). For instance, activation of the NMDA receptor by the excitotoxin quinolic acid triggers NF- k B activation and a concomitant increase of p53 and c-Myc expression in rat striatum, whereas its suppression prevented cell death (Qin et al., 1999). A more recent study involving stimulation of rat striatal neurons with the glutamate receptor agonist kainic acid produced similar results (Nakai et al., 2000). Since NF- k B was described as a positive regulator of both p53 and c-Myc expression (Duyao et al., 1990; La Rosa et al., 1994; Wu and Lozano, 1994), activation of NF- k B in these excitotoxic models of Huntington’s disease was proposed to induce expression of c-Myc and p53 to trigger cell death. Recently, individual NF-k B subunits were suggested to have opposing roles in modulating neuronal survival following exposure to the neurotoxin glutamate. Whereas glutamate activated p50/NF- kB1 and p65/ RelA in cultured rat cerebellar granule cells and mouse hippocampalslices, treatment with IL-1 b that promotes neuron survivalactivated expression of c-Relalong with p50/NF- kB1 and p65/RelA (Pizzi et al., 2002). NF-k B inhibition with antisense oligonucleotides implicated RelA in glutamate-mediated cell death, while c-Rel was essentialfor IL-1 b -mediated survival in cerebellar granule cells. These results imply that the particular composition of NF- k B complexes may be a key factor in establishing neuronal cell survival or cell death. In this regard, a neuron-specific factor distinct from NF-k B was previously described to bind to k B DNA sites (neuronal k B binding factor, N kBF) (Moerman et al., 1999). A recent follow-up on this work showed that N k BF consists of Sp1-related proteins and that its binding to kB DNA motifs can be competed by oligonucleotides containing an Sp1-binding motif (Mao et al., 2002). Activation of ionotropic glutamate receptors reduced the activity of N kBF, consistent with the proteolytic degradation of Sp1-related factors in this context. It therefore appears that non-NF- kB-related factors may also play a role in determining neuronal cell fate. Future work will undoubtedly help to establish the contribution of this factor relative to that of bona fide NF-k B in the response of neurons to apoptotic signals. Another important factor in the balance between the pro- and antiapoptotic effects of NF- kB in the nervous system appears to be the cell type. While many studies have proposed that NF- k B guards neurons from degeneration, its activation in microglia appears to foster neuronalcelldeath (reviewed in Mattson and Camandola, 2001). Indeed, whereas homozygous in- activation of p50/NF- k B1 in mice significantly reduced ischemic neuronaldeath (Schneider et al., 1999), NF-k B activation in microglia promoted ischemic neuronal

Oncogene

Rel/NF- j B and apoptosis J Kucharczak et al 8968 degeneration most likely due to

Rel/NF-jB and apoptosis

J Kucharczak et al

8968

degeneration most likely due to production of neuro- toxic reactive oxygen species (ROS) and excitotoxins by activated microglia. Thus, while NF- k B may suppress the death of cells in which it is activated, its ability to induce expression of cytotoxic agents such as nitric oxide may promote the demise of other cells (reviewed in Mattson and Camandola, 2001). Although the con- sequences of NF- kB activation in the nervous system remain somewhat controversial, it seems that some of the discrepancies might derive from the analysis of different cell types, experimental conditions, inducing signals and the particular NF- kB subunits involved (reviewed in Denk et al., 2000). These issues deserve further investigation.

Implication of NF- k B in the cell response to viral infection

For many viruses, activation of NF- k B has been linked to their transforming activity. The first identified member of the Rel/NF- k B family, v-Rel, is in fact a viral protein derived from the cellular gene c-Rel. As the oncogene of the avian Rev-T retrovirus that causes fatal lymphoma in infected young birds, v-Rel suppresses apoptosis from a variety of death signals, and this appears to be important for its oncogenic effect (reviewed in Gilmore, 1999). Indeed, cells that con- ditionally express v-Rel readily undergo apoptosis upon inactivation of the protein (White et al., 1995; Zong et al., 1997). This is consistent with the resilience of tumor-derived cells transformed by v-Rel to proapop- totic stimuli (Neiman et al., 1991). Recent work suggests that the oncogenic potentialof Relproteins arises at least in part from their ability to upregulate antiapop- totic proteins. Both v-Relmutants with impaired transforming activity due to transactivation domain mutations or weakly transforming Rel proteins could be rescued by coexpression of death suppressors Bcl-xL or Bcl-2 (White and Gilmore, 1996; Gilmore et al., 2001; Rayet et al., 2003). In support of this idea, v-Rel upregulates expression of c-IAP1 in transformed cells and does so at higher levels than its cellular homologue c-Rel(Kralova et al., 2002). The Tax oncoprotein of human T-cell leukemia virus type-1 (HTLV-1) immortalizes T cells in an NF-k B- dependent manner and is implicated in adult T-cell leukemia (ATL). Suppression of NF- k B in Tax-induced mouse tumor cells sensitizes them to proapoptotic stimuli (Portis et al., 2001). As seen in HTLV-1- immortalized T cells and those of ATL patients, Tax- mediated activation of NF- k B induces death-suppres- sive genes, such as Bcl-xL, Bcl-2 and Bfl-1/A1 (Harhaj et al., 1999; Tsukahara et al., 1999; Nicot et al., 2000). NF- k B is also linked to transformation by the oncogenic human herpesvirus Epstein–Barr Virus (EBV) that is implicated in Burkitt’s lymphoma. Its latent membrane protein 1 (LMP1) suppresses apoptosis and is required for human B-cell transformation by EBV (Cahir- McFarland et al., 2000; Feuillard et al., 2000; He et al., 2000; reviewed in Hiscott et al., 2001). LMP1 was recently shown to upregulate expression of Bfl-1/A1 via a novelNF- k B-like binding site and this increase

Oncogene

conferred protection against apoptosis induced by growth factor deprivation in an EBV-positive cell line exhibiting a latency type I infection (D’Souza et al., 2000, 2003). Contrary to these findings and others, LMP1 was recently reported to induce cell death via NF- k B-dependent activation of caspase-3 (Nitta et al., 2003). In this study, LMP1 failed to induce death in absence of RelA or upon inhibition of NF-kB. Although the reason for this discrepancy remains to be deter- mined, the use of different cell types should be noted. The human herpesvirus 8 is linked to Kaposi’s sarcoma and primary effusion lymphoma. Its v-FLIP protein suppresses apoptosis triggered by growth factor depri- vation by persistently activating the IKK kinase complex thereby inducing expression of Bcl-xL (Liu et al., 2002; Sun et al., 2003). Lastly, gammaherpes- viruses establish latent infections in lymphocytes and are implicated in nasopharyngeal carcinoma, Kaposi’s sarcoma and B-cell lymphomas. NF- kB was recently implicated in promoting latency in gammaherpesvirus- infected cells, as its overexpression in epithelial and fibroblast cells suppressed replication of murine herpes- virus 68 (MHV68) and blocked activation of lytic promoters from MHV68 and Kaposi’s sarcoma-asso- ciated herpesvirus (KSHV) (Brown et al., 2003). In contrast, NF- k B inhibition in KSHV-infected cells led to lytic protein synthesis and virus reactivation. There are a number of other examples in which viruses utilize NF- kB to block apoptosis as a means to enhance their replication or their pathogenicity (re- viewed in Hiscott et al., 2001; Santoro et al., 2003). The effects of HIV-encoded proteins Tat, Vpr, Nef and envelope glycoprotein gp120 on the NF- k B pathway, and in turn those that NF-k B exerts on HIV replication and apoptosis are well documented and were reviewed extensively (reviewed in Barkett and Gilmore, 1999; Hiscott et al., 2001; Santoro et al., 2003). The herpes simplex virus 1 envelope glycoprotein D and hepatitis C virus core protein induce NF- kB and confer resistance to apoptotic stimuli-like TNF a (Tai et al., 2000; Goodkin et al., 2003; Medici et al., 2003). Encephalo- myocarditis virus (EMCV) also abrogates apoptosis via NF- k B as a means to enhance its pathogenic effects. In fact, p50 / mice survive EMCV infections that would otherwise kill wild-type animals (Schwarz et al., 1998). Interestingly, African swine fever virus (ASFV) encodes two proteins that have opposite effects on NF- k B activation (reviewed in Gilmore et al., 2003). The early viral protein A238L acts like a degradation- resistant mutant of I k B to blunt the initial burst of NF- k B activity that follows infection and suppress host immune and inflammatory responses. On the other hand, the IAP-like ASFV protein A224L is expressed later during the course of infection and activates NF- kB via the IKK kinase complex to seemingly suppress apoptotic caspases and lengthen the infection (Rodriguez et al., 2002). On the flip side, certain viruses exploit the NF- k B cascade to induce cell death. Precedents include Dengue, Sindbis and Reovirus that promote apoptosis in infected cells, whereas death was inhibited by NF-k B decoys or

in cells lacking p50 or p65 (Lin et al., 1995; Marianneau et al., 1997; Connolly et al., 2000; Jan et al., 2000). However, the generality of NF- k B’s proapoptotic effects in the pathogenicity of flaviviruses was recently brought into question, as contrary to Dengue, Taiwanese Japanese Encephalitis virus could replicate and trigger apoptosis in cells in which NF- k B activity was stably suppressed (Liao et al., 2001). Lastly, other viral proteins such as human adenovirus E1A suppress NF- kB to sensitize cells to proapoptotic stimuli (Shao et al., 1997, 1999). E1A blocks the activity of the IKK complex, and was recently shown to antagonize expression of the NF-k B-regulated antia- poptotic protein c-FLIP(s) that inhibits caspase-8 activation during TNF a signaling (Shao et al., 1999; Perez and White, 2003). Overall, although distinct viruses interface differently with the NF- k B signaling cascade, it is clear that many of them have evolved means to influence this important pathway to advantage the host or the virus during infection.

NF- k B weighs in the balance to regulate life and death decisions in death receptor (DR) pathways: new insights

TNF a is a key cytokine that regulates immune responses, cell differentiation and apoptosis. Activation of TNF receptors TNFR1 or TNFR2 by TNF a represents the quintessentialpathway balancing the induction of cell death with its inhibition through activation of NF-k B. Although many players involved

Rel/NF-jB and apoptosis

J Kucharczak et al

involved Rel/NF- j B and apoptosis J Kucharczak et al 8969 in the TNFR signaling cascade

8969

in the TNFR signaling cascade have been identified, the mechanism(s) that determine whether the death or survivalbranch of the pathway willprevailhas not been entirely clarified. A recent study by Micheau and Tschopp (2003) sheds new light on the issue. Their study demonstrates that two separate complexes can form following trimerization of TNFR1 in response to TNF a and recruitment of TRADD and RIP1 (Figure 2). Complex I binds TRAF2, leading to NF-kB activation and consequent transcriptionalactivation of caspase-8 inhibitor FLIP (CASPER). Complex II arises when TNFR1, TRADD and RIP1 are modified, possibly via mono- and/or diubiquitination, and contrary to expec- tation, the complex dissociates from TNFR1. When released freely into the cytoplasm, complex II recruits FADD, likely through an interaction between the TRADD and FADD DDs. Caspase-8 and either the proapoptotic caspase-10 (for death) or the caspase regulator c-FLIP(L) (for survival) are then recruited in a mutually exclusive interaction. The onset of apoptosis depends upon how effectively complex I induces c- FLIP(L) synthesis to suppress death signaling by complex II. These new findings imply that the death- regulating complex forms in the cytosol, consistent with prior work that failed to demonstrate a direct physical association of FADD and caspase-8 with TNFR1 (Harper et al., 2003). The results also agree with earlier studies with c-FLIP (CASPER) knockout mice, whereby elimination of FLIP highly sensitized cells to TNF a - and FasL-induced apoptosis (Yeh et al., 2000).

TNF a - and FasL-induced apoptosis (Yeh et al ., 2000). Figure 2 TNF-R1 activation upon

Figure 2 TNF-R1 activation upon binding of cytokine TNFa leads to the formation of two sequential complexes. Complex I is comprised of TRADD, RIP1 and TRAF2, and is responsible for activating NF- kB signaling, which upregulates antiapoptotic genes. Following modification (possibly mono- or diubiquitination) and endocytosis, the protein complex recruits FADD, procaspase-8 and - 10, producing complex II, which initiates an apoptotic signal pathway. The fate of the cell is determined by the balance between NF- kB-dependent antiapoptotic genes and proapoptotic factors that act in the death cascade, through interactions that can occur at different steps along the death pathway. Binding of FLIP competes with procaspase-10 in complex II, while the antiapoptotic Bcl-2 family members Bfl-1/A1, Bcl-xL, NR13 and Bcl-2 block the release of cytochrome c from mitochondria. IAPs, including XIAP, c- IAP1 and c-IAP2, can bind and silence effector caspases. A complete list of antiapoptotic NF- kB target genes is found in Table 1

Oncogene

Rel/NF- j B and apoptosis J Kucharczak et al 8970 Other factors were found to

Rel/NF-jB and apoptosis

J Kucharczak et al

8970

Other factors were found to also weigh in the balance to favor the life- or death-promoting branches of this pathway. These include caspase-3-mediated proteolysis of IKK b to suppress NF- kB activation during TNFa - induced apoptosis (Tang et al., 2001b). Another mechanism by which NF- k B can protect cells against TNF a -induced demise involves inhibition of lysosome- mediated apoptosis (Liu et al., 2003). Cell treatment with TNF a triggers the lysosomal release of cathepsin B, shown in recent years to play an important role in mitochondria-independent cell death via caspase-depen- dent and -independent pathways. The inhibitory effect of NF- k B in this context is mediated by upregulation of the serine protease inhibitor Spi2A, a potent inhibitor of cathepsin B (Liu et al., 2003). NF-k B has also been found to help scavenge oxygen radicals induced by TNFa by upregulating new enzymes involved in ROS metabo- lism (G Franzoso, personal communication).

Distinct effects for distinct subunits Parenthetically, it seems that not all NF- kB subunits have the same capacity to suppress apoptosis induced by DRs. Whereas RelA is required to protect cells from TNFa , some groups reported that c-Relis dispensable for viability in response to TNFR- and Fas-mediated apoptosis. Indeed, c-Rel / fibroblasts are refractory to TNF-mediated cell death, as are c-Rel / B cells from Fas-mediated apoptosis (Owyang et al., 2001; Chen et al., 2003). In contrast, c-Relis wellrecognized to be necessary for protecting B cells from antigen receptor- mediated cell death (Wu et al., 1996b; Grumont et al., 1998; Owyang et al., 2001). Subunit-specific regulation was also suggested in the Apo2L/TRAIL signaling pathway, although the protective activity of NF- k B in this context has been the subject of controversy. One report showed that T-cell death induced by HTLV-1 Tax is dependent on NF-k B signaling and involves activated expression of DR ligand TRAIL (Rivera- Walsh et al., 2001). In contrast, inhibition of NF-k B in breast cancer cells induced expression of DR5 (TRAIL- R2) and adaptor protein TRADD, sensitizing cells to TRAIL-induced apoptosis (Chen et al., 2003). Over- expression of RelA could protect MEFs from TRAIL- induced apoptosis by inhibiting expression of caspase-8, DR4 (TRAIL-R1) and DR5 (TRAIL-R2), and enhan- cing expression of inhibitors of apoptosis c-IAP1 and c- IAP2 to decrease simultaneously formation of DR complexes and inhibit active pathways. Conversely when c-Relwas expressed under the same conditions, DR4, DR5 and the proapoptotic Bcl-xS protein were upregulated, concomitant with inhibition of c-IAP1, c- IAP2 and survivin. However, these effects may be cell specific as c-Rel expression in HeLa cells increased resistance to TRAIL by upregulating the decoy receptor protein DcR1 (Bernard et al., 2001b). In another study, RelA-mediated induction of Bcl-xL was implicated in suppressing TRAIL-mediated apoptosis (Ravi et al., 2001), although others do not support an important role for Bcl-xL in this pathway (Walczak et al., 2000). Since many of these studies evaluated the effects of NF- k B subunits using protein overexpression and in different

Oncogene

cell systems, future analyses looking at their activity under more physiological conditions will certainly help to clarify their effects.

Crosstalk between NF- k B and JNK: something to live for

In addition to turning on NF-k B signaling, TNF induces the stress-activated Jun kinase (JNK), a member of the MAPK family. Although JNK can cause cell death by activating the mitochondrialapoptotic path- way, it can also promote cell survival depending on the cell context. The existence of negative crosstalk between the NF- k B and JNK pathways has recently come to light and was implicated to be responsible for the transient activation of JNK in response to TNF a (De Smaele et al., 2001; Javelaud and Besancon, 2001; Tang et al., 2001a; Reuther-Madrid et al., 2002; Papa et al., 2003; reviewed in Franzoso et al., 2003). While some reports indicate that this negative interplay is specific to TNF a signaling and does not affect JNK activation by IL-1 or UV rays (Tang et al., 2001a; Lin, 2003; A Lin, personalcommunication), others also observed NF-k B- mediated suppression of JNK activation in cells stimulated with IL-1 b (Reuther-Madrid et al., 2002). Different NF- k B-regulated genes were proposed to be responsible for inhibition of the JNK signaling cascade. Tang et al. (2001a) studies with IKK b / and RelA / MEFs showed that NF- k B suppresses JNK activity by upregulating expression of the caspase inhibitor X- chromosome-linked inhibitor of apoptosis (XIAP), although others noted that the protective activity of XIAP depends upon JNK1 activation (Sanna et al., 2002). Other studies found that GADD45 b /MyD118, an NF- k B target gene that belongs to the GADD45 family of factors involved in cell cycle control and DNA repair, is involved in suppressing JNK signaling at the level of JNKK2/MKK7 (De Smaele et al., 2001; Papa et al., 2003). The NF- k B-regulated zinc-finger protein A20, which antagonizes TNF induction of JNK, is another possible contender (Lademann et al., 2001). Never- theless, elimination of XIAP in MEFs was not sufficient to abolish inactivation of JNK by NF- kB (A Lin, personal communication). Similarly, Amanullah et al. (2003) commented that mice deficient for gadd45 b apparently show no defect in TNF a signaling or cell survival. Some differences in experimental conditions have been noted between this study and that of De Smaele et al., but it remains to be determined whether these might explain these discrepancies (Zazzer- oni et al., 2003b). Recent work by Sakon et al. (2003) proposed that NF- k B prevents sustained JNK activity in TNF-treated cells by inhibiting accumulation of reactive oxygen species. Based on these analyses, it appears that a combination of these or other NF-k B- regulated genes might be necessary to suppress JNK activity in response to TNF. There is a consensus that further analyses will be needed to clarify the contribu- tion of these and other NF- k B-regulated factors to the controlof JNK signaling. It is undisputed that suppression of NF- kB activity leads to prolonged activation of JNK, yet the outcome

of this activation on the fate of the cell remains somewhat controversial. Several analyses found that in the absence of NF- k B, persistent JNK activity is proapoptotic (De Smaele et al., 2001; Javelaud and Besancon, 2001; Tang et al., 2001a; Lamb et al., 2003). In contrast Reuther-Madrid et al. (2002) showed that persistent activation of JNK is antiapoptotic in TNF a - treated NF- k B-null cells, and that inhibitors of JNK enhanced TNF-induced killing in this context. Thus, opposite results were obtained in studies that both used the same RelA / MEFs activated by TNF and treated with JNK inhibitor SP600125, although there were some differences in the dose and duration of the TNF a treatment (Tang et al., 2001a; Reuther-Madrid et al., 2002). It should perhaps be noted that not 100% of the cells died upon suppression of persistently activated JNK in TNF-treated RelA / MEFs in the former study, and that cell death protection was not 100% effective following JNK inhibition in the latter. Recent work in JNK / fibroblasts indicates that JNK induces expression of transcription factor JunD and that NF- k B collaborates with this JNK/JunD pathway in wild-type cells to enhance expression of c-IAP2, a caspase inhibitor that also interacts with TRAF2 and inhibits TNF-induced apoptosis (Lamb et al., 2003). These findings imply that NF- k B-mediated cell survival in cells treated with TNF requires the transient activation of JNK, and that when NF- k B is absent JNK activation is persistent and apoptosis ensues (Lamb et al., 2003). Clearly, the interplay between the JNK and NF- k B signaling pathways is important but nonetheless complex. It is a topic of intense investigation that will undoubtedly continue to shed new light on the regulation of cell fate.

Answering the suicide hotline: how NF- k B regulates programmed cell death

As illustrated above, the relationship between NF-k B and apoptosis is intricate. To complicate things further, while NF- k B generally exploits its transcriptional activity to regulate the expression of genes that function in death signaling, there are an increasing number of reports that indicate other modes of action. This section reviews recent developments in this arena.

Mechanisms for NF- kB-mediated protection from apoptosis

Transcriptional activation of antiapoptotic genes As summarized in Table 1, NF- kB activates the transcrip- tion of many genes capable of suppressing cell death (reviewed in Barkett and Gilmore, 1999; Karin and Lin, 2002; Burstein and Duckett, 2003). By and large, this is the most common way for NF- k B to antagonize apoptosis. Genes in this category contain functional

NF- k B-binding sites in their regulatory region, undergo

NF- k B-mediated activation in response to death-indu-

cing stimuli, and can suppress apoptosis under condi- tions where NF- k B is inactivated, although several show partial activity in this context. This most likely reflects

Rel/NF-jB and apoptosis

J Kucharczak et al

reflects Rel/NF- j B and apoptosis J Kucharczak et al 8971 the need for cooperative action

8971

the need for cooperative action to antagonize different apoptotic challenges in different cells efficiently. Among them are prosurvival members of the mammalian Bcl-2 gene family Bcl-xL, NR13 and Bfl-1/A1 (Grumont et al., 1999; Lee et al., 1999a, b; Wang et al., 1999b; Zong et al., 1999; Chen et al., 2000). Bcl-2 itself recently joined this group, owing to its NF- kB-dependent induction that conferred protection against hypoxia- and nitric oxide- induced injury in primary hippocampalneurons and its ability to promote the survival of peripheral B cells in response to c-Reland RelA (Tamatani et al., 1999; Grossmann et al., 2000). Factors in this group suppress the release of proapoptotic cytochrome c and Smac/ Diablo from mitochondria and can block programmed cell death in response to varied stimuli including TNF a , antigen receptor ligation and chemotherapeutic agents like etoposide (Grumont et al., 1999; Lee et al., 1999a; Wang et al., 1999b; Zong et al., 1999; Reed and Green, 2002; Sun et al., 2002). Other antiapoptotic NF-k B target genes include XIAP that inhibits the processing of procaspase-9 and the activities of caspase-7 and -3, and was implicated in NF-k B-mediated suppression of JNK signaling (Stehlik et al., 1998; Tang et al., 2001a). The cellular inhibitors of apoptosis c-IAP1, c-IAP2 and the TNF receptor- associated factors TRAF1 and TRAF2, are also activated in response to NF- kB and were implicated to act jointly with one another to suppress TNF-induced cell demise in RelA / MEFs, although coexpression of c-IAP1 and c-IAP2 alone was adequate to confer resistance to etoposide-induced death (Chu et al., 1997; You et al., 1997; Deveraux et al., 1998; Stehlik et al., 1998; Wang et al., 1998; Schwenzer et al., 1999; Chen et al., 2003). The zinc-finger protein A20 was described severalyears ago as a dualinhibitor of NF-k B activation and TNF-provoked death (reviewed in Beyaert et al., 2000). Recent work uncovered a role for A20 in controlling the balance between the Fas and TNF death pathways during AICD in T cells (Malewicz et al., 2003). c-FLIP is a potent negative regulator of DR-induced apoptosis (Kreuz et al., 2001; Micheau et al., 2001). Gadd45 b is a new recruit to the antiapoptotic NF-k B-regulated targets. It was impli- cated in NF- k B-mediated suppression of JNK in response to TNF, and in mediating the cytoprotective activity of CD40 against Fas-induced death in B cells (De Smaele et al., 2001; Zazzeroni et al., 2003a). MnSOD also belongs to this category, as an NF- k B- induced enzyme that scavenges cytotoxic ROS generated by various apoptotic pathways (Bernard et al., 2001a, 2002; Delhalle et al., 2002; Tanaka et al., 2002). Lastly, there have been new developments regarding the controversial role of the immediate-early response gene iex-1 in NF- k B-mediated survival. A variant of IEX-1 (IEX-1L) was previously reported to mediate cytopro- tection by NF-kB (Wu et al., 1998), but its role in suppressing apoptosis has been the subject of contro- versy (Schafer et al., 1999; Arlt et al., 2001; Schilling et al., 2001; B Rayet and C Ge´linas, unpublished data). Of late, Wu (2003) published that whereas IEX-1 sensitizes some cells to apoptosis when overexpressed

Oncogene

Rel/NF- j B and apoptosis J Kucharczak et al 8972 in vitro , its constitutive

Rel/NF-jB and apoptosis

J Kucharczak et al

8972

in vitro, its constitutive expression in the lymphocytes of IEX-1 transgenic mice prevents activated T cells from undergoing stimuli-induced apoptosis. In this in vivo context, IEX-1 gave rise to a lupus-like condition and the mice developed T-cell lymphomas. These provoca- tive new findings deserve further investigation.

Alternative means to enhance cell survival A few reports have suggested alternative means for NF- k B to enhance cell viability. Three recent examples come to mind. The first comes from studies of p53-mediated cell death in IKK a / IKK b / double knockout MEFs, in which IKK b -mediated activation of NF- k B led to increased levels of Mdm2, destabilization of p53 and resistance to doxorubicin-induced apoptosis (Tergaonkar et al., 2002). The second is the suppression of caspase-8, DR4 (TRAIL-R1) and DR5 (TRAIL-R2) gene expres- sion by RelA in cells stimulated with the death-inducing ligand TRAIL (Chen et al., 2003). This came along with enhanced expression of c-IAP1 and c-IAP2. The third example is one in which expression of the proapoptotic Bcl-2 family member Bax was increased following expression of a dominant inhibitor of NF- k B in cancer cell lines (Bentires-Alj et al., 2001). Although a functionalNF- k B-binding site was identified in the Bax promoter region, this site was dispensable for NF- k B-mediated suppression of Bax promoter activation by p53. Nevertheless, it is conceivable that suppression of Bax activation by NF- k B might contribute to the survival of some tumor cells, although the mechanism

for this repression needs to be clarified. Finally, NF- k B

and I k B subunits have been observed to localize to mitochondria, and to suppress mitochondrialgene expression (Bottero et al., 2001; Cogswell et al., 2003).

Bottero et al. (2001) showed an interaction between mitochondrialI k B and the ANT adenine nucleotide transporter, which regulates the mitochondrial perme- ability transition. In light of the critical role of this organelle in the execution and amplification of many apoptotic programs, these intriguing new findings give rise to speculation that NF- kB-dependent mechanisms operating at the level of mitochondria may perhaps contribute to its protective activity. However, further investigation is needed to ascertain this.

Mechanisms through which NF- kB promotes cell death

Despite a large body of evidence supporting the antiapoptotic role of NF- k B, there are a growing number of scenarios in which NF- k B behaved in a

proapoptotic fashion. One such study published a few years ago concluded that tumor suppressor p53 activates NF- k B via MEK1 and pp90 rsk , and that NF- k B plays a vitalrole in p53-mediated apoptosis (Ryan et al., 2000). Inhibition of MEK1 or absence of RelA abrogated p53- induced cell death. However, overexpression of RelA in a p53-deficient background failed to induce apoptosis, suggesting that RelA might be necessary but not sufficient for the proapoptotic phenotype of p53. A recently published pair of papers from the group of

N Perkins offers an interesting counterpoint to this

Oncogene

theory. In the first, Rocha et al. (2003a) demonstrate that the tumor suppressor ARF, which normally activates p53 by inhibiting Mdm2, can also inhibit the transcriptionalactivity of NF- k B in cells stimulated with TNF a or the chemotherapeutic agent etoposide (N Perkins, personalcommunication). This inhibitory effect is independent of p53 and Mdm2, and serves to downregulate expression of antiapoptotic proteins con- trolled by NF- k B, such as Bcl-xL. The study shows that ARF induces association of the RelA transactivation domain with histone deacetylase 1 (HDAC1) to repress NF- kB’s transcriptionalactivity in a promoter-specific manner. The second paper addresses the effect of p53 on NF- k B-mediated transcription and cell cycle regulation (Rocha et al., 2003b). Cyclin D1, involved in the G1– S phase transition, is transcriptionally regulated by NF-

k B primarily through a Bcl-3/p52 complex (Westerheide

et al., 2001). Rocha et al. (2003b) show that p53 suppresses expression of Bcl-3, an I kB-related protein that acts as a transcriptionalcoactivator with p52. At the same time, p53 promotes association of p52 with HDAC1 to repress cyclin D1 transcription. Since ARF is able to regulate p53, these findings create a scenario in which ARF negatively regulates the antiapoptotic and proproliferative activities of NF- k B, thereby creating a situation where NF-k B is no longer able to inhibit cell death without necessarily promoting it either. When extrapolated to the findings of Ryan et al., these exciting new results lead one to speculate that perhaps p53 does not activate a proapoptotic activity of NF- k B, but rather neutralizes its capacity to stimulate expression of antiapoptotic genes that would otherwise counteract p53-dependent apoptosis. In this scenario, ARF and p53 might be thought of as guardians of the antiapoptotic and proproliferative activities of NF- kB. It will be intriguing to see if future experiments confirm these predictions. There exist severalother studies in which NF- k B sensitized cells to death-inducing signals (for example, Jung et al., 1995; Grimm et al., 1996; Lin et al., 1998; Hettmann et al., 1999). In some of these cases, NF- k B induced expression of death-promoting genes (Table 1). These include tumor suppressor p53, the DR Fas and its ligand FasL and TNF a (Collart et al., 1990; Shakhov

et al., 1990; Wu and Lozano, 1994; Kasibhatla et al., 1998, 1999; Matsui et al., 1998; Chan et al., 1999; Zheng et al., 2001). For instance, NF- kB induces FasL expression in T cells in response to AICD (Kasibhatla et al., 1999). Recently, nitric oxide-induced activation of p38 was implicated in triggering NF-k B-mediated expression of p53, leading to upregulation of the proapoptotic Bcl-2 family member Bax, just as cya- nide-induced activation of NF-k B resulted in activation of the proapoptotic Bcl-xS and Bax proteins (Kim et al., 2002b; Shou et al., 2002). Other examples include NF-

k B-dependent activation of DR ligand TRAIL and

TNF receptor family member DR6 (Baetu et al., 2001; Kasof et al., 2001; Rivera-Walsh et al., 2001; Siegmund et al., 2001). In one study, c-Relinduced expression of TRAIL receptors DR4 (TRAIL-R1), DR5 (TRAIL-R2)

and the proapoptotic Bcl-2 family member Bcl-xS to

sensitize cells to TRAIL-induced apoptosis (Chen et al., 2003). Although the antiapoptotic bcl-xl transcript was previously shown to be transcriptionally regulated by

NF- k B, this the first report of the proapoptotic product

Bcl-xS being induced by NF-k B. Since Bcl-xL and Bcl- xS are derived by alternative splicing, this raises the possibility that in some instances c-Rel is perhaps capable of influencing the splicing reaction to benefit production of one form over the other. Activation of death-causing genes is not the only way in which NF- k B is involved in promoting apoptosis. While c-Rel upregulates expression of MnSOD to

suppress cell death by converting toxic O

over time the buildup of hydrogen peroxide itself triggers apoptosis (Bernard et al., 2002). This under- scores the idea that timing and environment can convert an antiapoptotic signalinto a death signal. NF- k B can also work indirectly to upregulate the expression of other transcription factors that in turn activate genes that regulate cell death and/or proliferation. Examples include p53, IRF-1 that works along with NF-k B to stimulate production of the death-promoting inducible nitric oxide synthase after ischemic injury, and E2F3a that induces expression of cyclin E to promote cell cycle progression (Wu and Lozano, 1994; Cheng et al., 2003; reviewed in Denk et al., 2000). Finally, one study last year claimed that the presence of a DD in p100/NF- k B2

allows it to promote apoptosis via recruitment to DRs and activation of apicalcaspase-8 (Wang et al., 2002b). It was suggested that the loss of this proapoptotic effect might be necessary for the oncogenic activity of rearranged nf- k B2 genes observed in human lympho- mas. However, others have expressed concerns regard- ing the physiological relevance of these findings in light of experimentalconditions and of the phenotype of mice in which expression p100/NF- k B2 was inactivated (Hacker and Karin, 2002). The involvement of p100/

NF- k B2 as an activator of cell death therefore remains

contentious.

2 to H 2 O 2 ,

Factors that influence NF- k B’s decision between life and death

There are severalfactors that influence the transcrip- tionalactivity and biologicalfunction of NF-k B and this in turn might affect its capacity to weigh in the balance between life and death. This section highlights those that were recently implicated in modulating its transcrip- tional activity, its relationship with factors that regulate death signaling, and its effect on cell cycle control.

Modulation of NF- k B’s transcriptional activity

Post-translational modification has surfaced as an important means to modulate NF- k B’s transcriptional activity. The IKK complex, casein kinase II and AKT are among the kinases that can phosphorylate the transactivation domains of RelA or c-Rel in response to NF-k B-inducing stimuli (Wang and Baldwin, 1998; Sakurai et al., 1999; Martin and Fresno, 2000; Wang et al., 2000; Mayo et al., 2002; Sizemore et al., 2002;

Rel/NF-jB and apoptosis

J Kucharczak et al

et al ., 2002; Rel/NF- j B and apoptosis J Kucharczak et al 8973 Yang et

8973

Yang et al., 2003). On the other hand, the catalytic

subunit of protein kinase A (PKAc), MSK1 and PKC z phosphorylate the RHD of RelA (Zhong et al., 1997; Duran et al., 2003; Vermeulen et al., 2003). In the case of PKAc, this modification is necessary for RelA interac- tion with transcriptionalcoactivator p300 (Zhong et al., 1998, 2002). Another RelA-interacting factor, protein phosphatase 2A, was implicated in its dephosphoryla- tion (Yang et al., 2001). Acetylation of NF- k B subunits adds another level of regulation, as acetylation of RelA prevented its association with newly synthesized I k B a and interaction with corepressor HDACs (Ashburner

et al., 2001; Chen et al., 2001). The effects of these different modifications on the pro- and antiapoptotic activities of NF- k B remain to be explored. However, a RelA mutant that could no longer be phosphorylated by PKAc (S276A) failed to rescue TNF-induced apoptosis of RelA / MEFs, and a knockin of this mutation is lethal in mice (Okazaki et al., 2003; S Ghosh, personal communication). Moreover, Akt was shown to suppress cell death by stimulating the transactivation potential of RelA (Madrid et al., 2000). c-Rel and RelA functionally interact with diverse transcriptionalcoactivators. The TAFII105 component of TFIID was shown to mediate NF- kB-dependent transactivation of antiapoptotic genes like A20, whereas

a

dominant-negative TAFII105 mutant sensitized cells

to

TNF-induced killing (Yamit-Hezi and Dikstein, 1998;

Yamit-Hezi et al., 2000). Novelinteractions between NF-k B and tumor suppressors ARF and BRCA1 have recently surfaced. ARF inhibited NF- kB-mediated transcription in a promoter-specific manner and down- regulated expression of the antiapoptotic Bcl-xL protein (Rocha et al., 2003a). Conversely, interaction of the RHD of RelA with BRCA1 enhanced NF- kB-mediated transcription of the proapoptotic gene Fas (Benezra et al., 2003). Lastly, c-Rel was recently shown to act in concert with AP-1 and C/EBP b to recruit coactivators TAFII250 and p300 and upregulate Bfl-1 gene expres- sion in activated T cells via an enhanceosome-like complex (Edelstein et al., 2003). In studies with null fibroblasts, c-Myc was revealed to sensitize cells to TNF-induced death by inhibiting p65-mediated trans- activation of Bfl-1/A1 (You et al., 2002).

Interplay between NF- k B and factors that regulate death signaling

NF-k B participates in a number of other interactions with proteins involved in the apoptotic cascade, which affect its protective activity to either break or amplify the survivalresponse.

Caspase-mediated cleavage of NF- k B and I k B: alter- native means to promote cell death Cysteine proteases known as caspases are major effectors of apoptosis that are activated upstream or downstream of mitochondria and cleave substrates in a sequence-specific manner. Among their targets are structuralproteins, nuclear lamins and a number of apoptosis regulators, such as Bcl-2 (Johnson and Boise, 1999). It is perhaps not

Oncogene

Rel/NF- j B and apoptosis J Kucharczak et al 8974 surprising that in order to

Rel/NF-jB and apoptosis

J Kucharczak et al

8974

surprising that in order to blunt the cell’s survival response, NF- k B is a substrate for caspase cleavage. Caspase-3-mediated cleavage of RelA, p50 and c-Rel upon treatment with death-inducing ligands FasL or TNF a has come to light (Ravi et al., 1998; Barkett et al., 2001). Caspase-mediated processing of I k B was por- trayed as another means to suppress the protective activity of NF- k B and promote an apoptotic response. Independent studies showed that the N-terminus of I k B a undergoes caspase-3-mediated cleavage, coincident with apoptosis (White et al., 1995; White and Gilmore, 1996; Barkett et al., 1997; Jung et al., 1998; Reuther and Baldwin, 1999). The product of this cleavage is an I kB a molecule that is resistant to signal-induced degradation by the proteasome, and that acts as a constitutive inhibitor of NF- k B to promote cell death (Reuther and Baldwin, 1999). Lastly, the Drosophila caspase homo- logue Dredd interacts with the Drosophila NF- k B factor Relish to promote its endoproteolytic cleavage (Stoven et al., 2003). This processing of Relish generates a RHD fragment capable of DNA binding and a stable fragment resembling I k B. In this particular case, this caspase action contributes a gain-of-function toward Relish and is independent of apoptosis.

A new role for FADD as a regulator of NF- k B signaling FADD is best known for its role as an adaptor that recruits caspase-8 to activated Fas recep- tors to signaldownstream activation of celldeath. However, there have been reports of late implicating FADD in the regulation of NF- k B signaling, although the outcome of this regulation is contentious. Initial reports showed that FADD and caspase-8 activate NF- kB via IKK, independent of the proteolytic activity of caspase-8 but dependent on its death effector domain- containing prodomain (Chaudhary et al., 2000; Hu et al., 2000; Schaub et al., 2000). Contrary to this claim, recent work showed that FADD activates caspase-8 in response to TNF or TLR-4 signaling, but that it simultaneously suppresses NF-k B activation from TLR-4 possibly via interaction with adaptor protein MyD88 (Bannerman et al., 2002). Consistent with this model, FADD / MEFs showed enhanced activation of NF- k B upon lipopolysaccharide (LPS) treatment. The discrepancies between these studies might be explained by differences in experimentalconditions (transient vs stable expression), in cell type and/or stimulation (reviewed in Duckett, 2002). Undoubtedly, further studies into this novel function of FADD will help to understand its effects on NF-k B and its implication in the controlof the celldeath response.

Bcl-2 feedsback to downregulate NF- k B activity Wher- eas NF- k B transactivates prosurvivalmembers of the Bcl-2 family (Table 1), there have been some reports in which Bcl-2 suppressed NF- k B activity by stabilizing I kB a , by blunting the transactivation potential of RelA or by influencing the nuclear accumulation of NF- kB subunits (Ivanov et al., 1995; Lin et al., 1995; Grimm et al., 1996; de Moissac et al., 1998; Badrichani et al.,

1999).

Oncogene

Positive feedback of IAPs to amplify NF-k B’s protective activity IAPs were reported to increase NF-k B activ- ity, and by the same token, enhance its antiapoptotic function. H-IAP1, a member of the IAP family, could promote I k B a degradation, thereby potentiating the ability of NF- kB to suppress TNF a -induced apoptosis (Chu et al., 1997). These results imply that the IAPs, which are transcriptionally activated by NF-k B, may provide positive feedback to amplify its cytoprotective activity.

Effects of NF- kB on cell cycle control

Since aberrant proliferation signals are often conducive to cell suicide, it is relevant to review recent findings that pertain to the effects that NF-k B exerts on cell cycle control. Consistent with the implication of NF- k B in promoting cell proliferation, NF-k B was shown to regulate the expression of cyclins D1, D2, D3 and more recently cyclin E (Guttridge et al., 1999; Hinz et al., 1999, 2001; Hsia et al., 2002). For example, B cells from c-rel / or NF- k B1 / c-rel / double-knockout mice show defective proliferative response to mitogenic stimulation that results in G1 phase arrest (Kontgen et al., 1995; Hsia et al., 2002; Pohl et al., 2002). Hsia et al. (2002) linked the proliferative defect of c-rel / B cells to a failure to express cyclins D3 and E. Recently, the same group reported that c-Relactivates expression of transcription factor E2F3a that, in turn, is likely to transactivate cyclin E (Cheng et al., 2003). Others showed that NF-k B-mediated transactivation of cyclin D1 is essentialfor the transforming activity of the oncoprotein v-Abl, and is increased in mammary carcinomas arising in MMTV-c-Rel transgenic mice (Nakamura et al., 2002; Romieu-Mourez et al., 2003). In agreement with these findings tumor suppressor p53, which plays a central role in cell cycle control, was revealed to inhibit cyclin D1 expression by favoring formation of nonfunctionalp52/HDAC1 complexes over transcriptionally active p52/Bcl-3 complexes (Ro- cha et al., 2003b). Furthermore, RelA participates in protein–protein interactions with the cell cycle regula- tory complex cyclin E-Cdk2 via its association with CBP/p300, and with the p16Ink4 cyclin-dependent kinase inhibitor (Perkins et al., 1997; Wolff and Naumann, 1999). An additionalinterplay between NF- k B and the Cdk/E2F pathway is highlighted by the direct interaction of E2F1 with RelA that antagonized the antiapoptotic activity of NF- k B, by preventing activation of MnSOD (Tanaka et al., 2002). While NF-k B is often associated with increased cell proliferation, it can suppress cell proliferation in certain cell types (Abbadie et al., 1993; Seitz et al., 1998; Sheehy and Schlissel, 1999; van Hogerlinden et al., 1999; Bernard et al., 2001c). For example, the epidermis of mice expressing p50/p65 is hypoplastic, contrary to that of mice expressing a super-repressor I k B in the skin that is hyperplastic (Seitz et al., 1998). NF-k B appears to inhibit the proliferation of keratinocytes by inducing the cyclin-dependent kinase inhibitor p21/Waf1 (van Ho- gerlinden et al., 1999; Seitz et al., 2000; Hinata et al.,

2003). This agrees with earlier studies showing that overexpression of c-Rel in the human epithelial cell line HeLa arrested proliferation, coincident with stabiliza- tion of p53 and induction of p21/WAF1 (Bash et al., 1997). In light of these findings, it will be interesting to unravel the molecular mechanisms that govern the pro- and antiproliferative actions of NF- k B in different cell types.

Antiapoptotic activity of NF- kB: its implication in cancer and therapy

A link between aberrant NF- k B activity and cancer was initially suggested by the acute oncogenicity of its viral oncoprotein v-Relthat causes aggressive lymphomas in animal models (reviewed in Gilmore, 1999). Since then, constitutive nuclear NF-k B activity has emerged as a hallmark for many human leukemias, lymphomas and solid tumors, most commonly due to persistent activa- tion of the IKK complex (reviewed in Rayet and Gelinas, 1999; Karin et al., 2002). These include Hodgkin’s lymphoma, diffuse large B-cell lymphoma (DLBCL), acute lymphoblastic leukemia, chronic mye- logenous leukemia, ATL and breast cancer among many

others (Bargou et al., 1997; Nakshatri et al., 1997; Sovak et al., 1997; Cogswell et al., 2000; Kordes et al., 2000; Davis et al., 2001; Hinz et al., 2001; Kalaitzidis et al., 2002). Amplification and overexpression of rel/nf- k B genes is another means by which NF- k B is deregulated in cancer (reviewed in Rayet and Gelinas, 1999). Moreover, severaloncoproteins including Ha-Ras and Bcr-Ablrely on NF- k B to mediate their transforming activity, although recent work showed that NF-k B activation by Ras and Bcr-Abldoes not involve IKK (Finco et al., 1997; Mayo et al., 1997; Reuther et al., 1998; Arsura et al., 2000; Hanson et al., 2003). Since human tumors frequently contain mutations in ras genes, and Bcr-Abl is involved in acute lymphoblastic leukemia and chronic myelogenous leukemia, it is reasonable to expect that activation of these oncogenes in human tumors may be yet another way to activate

NF- k B in cancer constitutively.

Lately, studies in primary cell cultures and animal models provided evidence that mammalian NF-k B proteins can have a proactive role in carcinogenesis. MMTV-driven expression of the mouse c-rel gene gave rise to late-onset mammary carcinomas in transgenic mice, whereas retroviral-mediated overexpression of the human or mouse c-rel genes malignantly transformed primary chicken spleen cells that could induce tumors in vivo (Gilmore et al., 2001; Fan et al., 2003; Romieu- Mourez et al., 2003). Although the exact role of NF- k B in the pathogenesis of human tumors remains to be determined, it is clear that in many instances suppres- sion of apoptosis is an important part of its contribu- tion. Indeed, inhibition of NF-k B sensitizes many tumor cells to death-inducing stimuli, including chemothera- peutic agents (reviewed in Baldwin, 2001; Lin and Karin, 2003). For example, delivery of a recombinant super-repressor Ik B to chemoresistant tumors in mouse xenograft models provoked tumor regression by sensi-

Rel/NF-jB and apoptosis

J Kucharczak et al

by sensi- Rel/NF- j B and apoptosis J Kucharczak et al 8975 tizing them to chemotherapeutic

8975

tizing them to chemotherapeutic treatment with the topoisomerase I inhibitor CPT-11 (Wang et al., 1999a; Cusack et al., 2000). Consistent with this work, NF-k B inhibition in many human tumor-derived cell lines, including malignant Reed–Sternberg (H-RS) cells of Hodgkin’s disease and DLBCLs, induces spontaneous apoptosis or/and sensitizes them to killing by TNF a or anticancer drugs (Table 2; for example, Bargou et al., 1997; Davis et al., 2001). These results agree with the elevated levels of NF- kB-dependent antiapoptotic pro- teins observed in many human tumors, including Bfl-1/ A1, c-IAP2, Bcl-xL and Bcl-2. These findings suggest that suppression of NF-k B might enhance the efficacy of anticancer treatments. In this regard, it is tempting to speculate that part of the therapeutic effects of anti-Her- 2/neu monoclonal antibodies (Herceptin) used in con- junction with chemotherapy to treat Her-2/neu-positive breast cancers might perhaps derive from suppression of NF-k B’s anti-apoptotic effects, as decreased NF- k B- binding activity was observed following treatment of a breast cancer cell line with anti-neu antibodies (Pianetti et al., 2001). However, the exact role of NF- kB in Bcr- Abl-induced leukemias remains to be clarified, since its antiapoptotic activity is not necessary for Bcr-Abl- mediated cell survival (Reuther et al., 1998). Since many chemotherapeutic agents trigger NF-k B activation, hence decreasing their effectiveness, many investigators have sought to characterize the mechan- isms involved and to identify compounds that could circumvent this problem (reviewed in Baldwin, 2001; Garg and Aggarwal, 2002). A study by Panta et al. (2003) identified that anthracyclines induce a novel pathway of ATM and DNA-PK signaling, which leads to NF- kB activation and chemoresistance via a MEK/ ERK/p90 rsk cascade. Recently, the same group devel- oped a new class of cytoplasmic-targeted anthracyclines that can provoke apoptosis with improved cytotoxicity and are unaffected by NF- kB-induced antiapoptotic genes (Bilyeu et al., 2003). The discovery that IKK- mediated NF- k B activation decreases p53 stability brings up another way for NF-k B to counter the efficiency of chemotherapeutic treatment with agents like doxorubicin (Tergaonkar et al., 2002). Since I k B undergoes ubiquitin-mediated degradation via the pro- teasome, proteasome inhibitors have been used as an alternative approach to suppress NF-k B in tumor cells. Among them, bortezomib (PS-341; Velcade) is a potent and selective proteasome inhibitor that was shown to sensitize tumor cells to chemotherapy and radiation in preclinical models, for example, in human colorectal cancer cells in a xenograft model (Cusack et al., 2001; reviewed in Lenz, 2003). PS-341 has since been approved for treatment of multiple myeloma and has shown promising results in early phase clinical trials for lymphoma, prostate cancer and lung cancer (reviewed in Lenz, 2003; Richardson, 2003). However, since the proteasome is involved in the degradation of many other cellular factors, including cyclins, cyclin-dependent kinase inhibitors p21Waf1 and p27Kip1 and tumor suppressor p53, there is a great dealof interest in the quest for more specific NF-k B inhibitors.

Oncogene

Rel/NF- j B and apoptosis J Kucharczak et al 8976 Table 2 A sample of

Rel/NF-jB and apoptosis

J Kucharczak et al

8976

Table 2

A sample of human tumor cell lines that respond to suppression of NF- kB

Tumor cell lines

Response to super-repressor IkB a

References

Hematopoietic tumors EBV-transformed lymphoblastoid cells (IB4) DLBCL (activated B-cell type) Multiple myeloma (U266) Hodgkin’s disease (H-RS cells – L428) Acute T-cell leukemia (Jurkat) Hodgkin’s disease (H-RS) (L428, L540, HD-MyZ)

Spontaneous apoptosis Spontaneous apoptosis Spontaneous apoptosis Spontaneous apoptosis Sensitized to TNFa -induced apoptosis Sensitized to CD30-induced apoptosis

Sensitized to death by growth factor withdrawalBargou et al. (1997)

Cahir-McFarland et al. (2000) Davis et al. (2001) Ni et al. (2001) Hinz et al. (2001) Van Antwerp (1996) Mir et al. (2000)

Solid tumors Pancreatic ductaladenocarcinoma (MiaPaCa2, Panc1)

Spontaneous apoptosis

Liptay et al. (2003),

Fibrosarcoma (HT1080)

Colorectal cancer (LOVO, SW1463, SW837, SW620, SW480, KM12-L4, KM12-SM, HT-29, WiDR, CL188, CCD841, NCI H508)

Sensitized to TNFa -induced apoptosis Sensitized to CPT-11-induced apoptosis Sensitized to CPT-11-induced apoptosis

Fujioka et al. (2003) Wang et al. (1996) Wang et al. (1999a) Cusack et al. (2000)

Hepatocellular carcinoma Lung adenocarcinoma (A549) Prostate carcinoma (PC-3 and LNCaP) Ewing sarcoma (EW7) Pancreatic insulinoma (MIN6N8) Epidermoid carcinoma (A431)

Sensitized to TNFa -induced apoptosis Sensitized to TNFa -induced apoptosis Sensitized to TNFa -induced apoptosis Sensitized to TNFa -induced apoptosis Sensitized to TNFa -induced apoptosis Sensitized to EGF-induced apoptosis

Tietze et al. (2000) Cheng et al. (2000) Muenchen et al. (2000) Javelaud et al. (2002) Chang et al. (2003) Anto et al. (2003)

It should be noted that while NF- k B contributes to oncogenesis in a majority of cell types, its I k B a - mediated suppression in keratinocytes was recently demonstrated to be necessary for Ras-mediated induc- tion of invasive epidermaltumors resembling squamous cell carcinoma (Dajee et al., 2003). Since NF- k B mediates Ras-induced growth arrest in keratinocytes, antagonizing NF- kB in these cells is required to bypass this roadblock to malignant transformation. It therefore appears that depending on the cell type, NF-

k B can be viewed as a positive or negative factor for oncogenesis.

Conclusions

It is now clear that in addition to its widely recognized role as a key regulator of immune and inflammatory responses, NF- k B has emerged as a decisive factor in the cell’s response to apoptotic challenge. Indeed, over 1800 papers addressing the role of NF-k B in apoptosis were published since the end of 1999 when this topic was last reviewed in Oncogene. As illustrated in this review, the effects of NF-k B on apoptosis has far-reaching con- sequences for normaldevelopment and/or homeostasis in many cells and tissues, including the immune system, hair follicles and epidermal appendages, the liver and the nervous system. NF-kB’s implication in apoptosis is also vital to the outcome of viral infections, the onset and progression of many cancers and their response to radiation and chemotherapy. While NF- kB is most commonly found to be cytoprotective, there are a number of instances where it is proapoptotic depending on the inducing stimulus and the cell context. A lot of progress has been made in understanding its mode of

Oncogene

action and its interplay with other key factors, including JNK, Arf, p53, p73 and FADD to name a few. These studies identified many anti- and proapoptotic NF-k B- regulated genes that mediate its activity, uncovered alternative means by which it functions and unveiled specific protein interactions and pathways that modulate its activity. These important new insights fuelhope that novel approaches will be developed to control the effects of NF- k B on apoptosis. Since deregulated NF- k B activity is a hallmark of many different cancers, efforts to better understand the mechanisms that drive con- stitutive NF- k B signaling will undoubtedly help the development of new strategies to restrict tumor cell growth and suppress both intrinsic and therapy-induced chemoresistance. These might also be applicable to other disease conditions in which NF- kB is implicated, although serious consideration should be given to circumstances where NF- kB is proapoptotic. The coming years promise to bring significant advances in this quest. In light of its pivotal role in apoptosis, it is clear that for many cells encountering death-inducing stimuli, the answer to the question: ‘To be, or not to be’ truly is NF- k B.

Acknowledgements We are very gratefulto M Arsura, PG Ashton-Rickardt, AS Baldwin Jr, D Baltimore, C Duckett, G Franzoso, S Ghosh, T Gilmore, M Karin, A Lin, HC Liou, N Perkins, GE Sonenshein and D Walls for communicating results and sharing manuscripts before publication. We apologize to many investigators whose work could not be cited due to space limitations. We thank J Dutta, N Gupta and L Lagos for criticalcomments on the review. Research performed in this laboratory on the roles of Rel/NF- k B in apoptosis and oncogenesis and on its antiapoptotic target Bfl-1/A1 is supported by grants from the NationalInstitutes of

Health – National Cancer Institute CA54999 and CA83937 to CG. JK is recipient of a postdoctoral fellowship from the New Jersey Commission on Cancer Research and is partially

References

Abbadie C, Kabrun N, Bouali F, Smardova J, Ste´helin D, Vandenbunder B and Enrietto PJ. (1993). Cell, 75, 899–912. Alcamo E, Mizgerd JP, Horwitz BH, Bronson R, Beg AA, Scott M, Doerschuk CM, Hynes RO and Baltimore D. (2001). J. Immunol., 167, 1592–1600. Aliprantis AO, Yang RB, Weiss DS, Godowski P and Zychlinsky A. (2000). EMBO J., 19, 3325–3336. Amanullah A, Azam N, Balliet A, Hollander C, Hoffman B, Fornace A and Liebermann D. (2003). Nature, 424, 741. Anto RJ, Venkatraman M and Karunagaran D. (2003).

J. Biol. Chem., 278, 25490–25498.

Arlt A, Grobe O, Sieke A, Kruse ML, Folsch UR, Schmidt WE and Schafer H. (2001). Oncogene, 20, 69–76. Arsura M, FitzGerald MJ, Fausto N and Sonenshein GE. (1997). Cell Growth Differ., 8, 1049–1059. Arsura M, Mercurio F, Oliver AL, Thorgeirsson SS and Sonenshein GE. (2000). Mol. Cell. Biol., 20, 5381–5391. Arsura M, Panta GR, Bilyeu JD, Cavin LG, Sovak MA, Oliver AA, Factor V, Heuchel R, Mercurio F, Thorgeirsson SS and Sonenshein GE. (2003). Oncogene, 22, 412–425. Ashburner BP, Westerheide SD and Baldwin Jr AS. (2001). Mol. Cell. Biol., 21, 7065–7077. Badrichani AZ, Stroka DM, Bilbao G, Curiel DT, Bach FH and Ferran C. (1999). J. Clin. Invest., 103, 543–553. Baetu TM, Kwon H, Sharma S, Grandvaux N and Hiscott J. (2001). J. Immunol., 167, 3164–3173. Baldwin Jr AS. (2001). J. Clin. Invest., 107, 241–246. Bales KR, Du Y, Dodel RC, Yan GM, Hamilton-Byrd E and PaulSM. (1998). Brain Res. Mol. Brain Res., 57, 63–72. Bannerman DD, Tupper JC, Kelly JD, Winn RK and Harlan JM. (2002). J. Clin. Invest., 109, 419–425. Bargou R, Emmerich F, Krappmann D, Bommert K, Mapara M, Arnold W, Royer H, Grinstein E, Greiner A, Scheidereit C and Dorken B. (1997). J. Clin. Invest., 100,

2961–2969.

Barkett M, Dooher JE, Lemonnier L, Simmons L, Scarpati

JN, Wang Y and Gilmore TD. (2001). Biochim. Biophys. Acta, 1526, 25–36. Barkett M and Gilmore TD. (1999). Oncogene, 18, 6910–6924. Barkett M, Xue D, Horvitz HR and Gilmore TD. (1997).

J. Biol. Chem., 272, 29419–29422.

Bash J, Zong WX and Gelinas C. (1997). Mol. Cell. Biol., 17,

6526–6536.

Beg AA and Baltimore D. (1996). Science, 274, 782–784. Benezra M, Chevallier N, Morrison D, MacLachlan T, El-Deiry WS and Licht JD. (2003). J. Biol. Chem., 278,

26333–26341.

Bentires-Alj M, Dejardin E, Viatour P, Van Lint C, Froesch B, Reed JC, Merville MP and Bours V. (2001). Oncogene, 20,

2805–2813.

Bernard D, Monte D, Vandenbunder B and Abbadie C. (2002). Oncogene, 21, 4392–4402. Bernard D, Quatannens B, Begue A, Vandenbunder B and Abbadie C. (2001a). Cancer Res., 61, 2656–2664. Bernard D, Quatannens B, Vandenbunder B and Abbadie C. (2001b). J. Biol. Chem., 276, 27322–27328. Bernard D, Slomianny C, Vandenbunder B and Abbadie C. (2001c). Free Radic. Biol. Med., 31, 943–953. Bertin J, Nir WJ, Fischer CM, Tayber OV, Errada PR, Grant JR, Keilty JJ, Gosselin ML, Robison KE, Wong GH,

Rel/NF-jB and apoptosis

J Kucharczak et al

KE, Wong GH, Rel/NF- j B and apoptosis J Kucharczak et al 8977 supported by the

8977

supported by the Foundation of the UMDNJ. MS was partially supported by an NIH predoctoral training grant in Biochemistry and Molecular Biology (GM08360).

Glucksmann MA and DiStefano PS. (1999). J. Biol. Chem., 274, 12955–12958. Beyaert R, Heyninck K and Van HuffelS. (2000). Biochem. Pharmacol., 60, 1143–1151. Bhakar AL, Tannis LL, Zeindler C, Russo MP, Jobin C, Park DS, MacPherson S and Barker PA. (2002). J. Neurosci., 22,

8466–8475.

Bilyeu JD, Panta GR, Cavin LG, Barrett CM, Turner EJ, Sweatman TW, IsraelM, Lothstein L and Arsura M. (2003). (submitted). Bonnard M, Mirtsos C, Suzuki S, Graham K, Huang J, Ng M, Itie A, Wakeham A, Shahinian A, HenzelWJ, Elia AJ, Shillinglaw W, Mak TW, Cao Z and Yeh WC. (2000). EMBO J., 19, 4976–4985. Boothby MR, Mora AL, Scherer DC, Brockman JA and Ballard DW. (1997). J. Exp. Med., 185, 1897–1907. Bottero V, Rossi F, Samson M, Mari M, Hofman P and Peyron JF. (2001). J. Biol. Chem., 276, 21317–21324. Brown HJ, Song MJ, Deng H, Wu TT, Cheng G and Sun R. (2003). J. Virol., 77, 8532–8540. Brummelkamp TR, Nijman SM, Dirac AM and Bernards R. (2003). Nature, 424, 797–801. Burstein E and Duckett CS. (2003). Curr. Opin. Cell Biol., 15, (in press). Cahir-McFarland ED, Davidson DM, Schauer SL, Duong J and Kieff E. (2000). Proc. Natl. Acad. Sci. USA, 97,

6055–6060.

Cavin LG, Romieu-Mourez R, Panta GR, Yuan S, Factor M, Thorgeirsson SS, Sonenshein GE and Arsura M. (2003). Hepatology, (in press). Chan H, Bartos DP and Owen-Schaub LB. (1999). Mol. Cell. Biol., 19, 2098–2108. Chang I, Kim S, Kim JY, Cho N, Kim YH, Kim HS, Lee MK, Kim KW and Lee MS. (2003). Diabetes, 52,

1169–1175.

Chaudhary PM, Eby MT, Jasmin A, Kumar A, Liu L and Hood L. (2000). Oncogene, 19, 4451–4460. Chen C, Edelstein LC and Gelinas C. (2000). Mol. Cell. Biol., 20, 2687–2695. Chen L, Fischle W, Verdin E and Greene WC. (2001). Science, 293, 1653–1657. Chen X, Kandasamy K and Srivastava RK. (2003). Cancer Res., 63, 1059–1066. Cheng Q, Lee HH, Li Y, Parks TP and Cheng G. (2000). Oncogene, 19, 4936–4940. Cheng S, Hsia CY, Leone G and Liou H-C. (2003). Oncogene, (in press). Chiarugi A. (2002). Brain Res. Mol. Brain Res., 109, 179–188. Chu ZL, McKinsey TA, Liu L, Gentry JJ, Malim MH and Ballard DW. (1997). Proc. Natl. Acad. Sci. USA, 94,

10057–10062.

Cogswell PC, Guttridge DC, Funkhouser WK and Baldwin Jr AS. (2000). Oncogene, 19, 1123–1131. Cogswell PC, Kashatus DF, Keifer JA, Guttridge DC, Reuther JY, Bristow C, Roy S, Nicholson DW and Baldwin Jr AS. (2003). J. Biol. Chem., 278, 2963–2968. Collart MA, Baeuerle PA and Vassalli P. (1990). Mol. Cell Biol., 10, 1498–1506. Connolly JL, Rodgers SE, Clarke P, Ballard DW, Kerr LD, Tyler KL and Dermody TS. (2000). J. Virol., 74, 2981–2989.

Oncogene

Rel/NF- j B and apoptosis J Kucharczak et al 8978 Crumrine RC, Thomas AL and

Rel/NF-jB and apoptosis

J Kucharczak et al

8978

Crumrine RC, Thomas AL and Morgan PF. (1994). J. Cereb.

Blood Flow Metab., 14, 887–891. Cusack Jr JC, Liu R and Baldwin Jr AS. (2000). Cancer Res., 60, 2323–2330. Cusack Jr JC, Liu R, Houston M, Abendroth K, Elliott PJ, Adams J and Baldwin Jr AS. (2001). Cancer Res., 61,

3535–3540.

D’Souza BN, Edelstein LC, Pegman P, Smith S, Loughran

AC, Mehl A, Floettmann E, Rowe M, Gelinas C and Walls

D. (2003). (submitted).

D’Souza B, Rowe M and Walls D. (2000). J. Virol., 74,

6652–6658.

Dajee M, Lazarov M, Zhang JY, Cai T, Green CL, Russell AJ, Marinkovich MP, Tao S, Lin Q, Kubo Y and Khavari PA. (2003). Nature, 421, 639–643. Davis RE, Brown KD, Siebenlist U and Staudt LM. (2001). J. Exp. Med., 194, 1861–1874. de Moissac D, Mustapha S, Greenberg AH and Kirshenbaum

LA. (1998). J. Biol. Chem., 273, 23946–23951.

De Smaele E, Zazzeroni F, Papa S, Nguyen DU, Jin R, Jones J, Cong R and Franzoso G. (2001). Nature, 414, 308–313. Delhalle S, Deregowski V, Benoit V, Merville MP and Bours

V. (2002). Oncogene, 21, 3917–3924.

Denk A, Wirth T and Baumann B. (2000). Cytokine Growth Factor Rev., 11, 303–320. Deveraux QL, Roy N, Stennicke HR, Van Arsdale T, Zhou Q, Srinivasula SM, Alnemri ES, Salvesen GS and Reed JC.

(1998). EMBO J., 17, 2215–2223. Do RK, Hatada E, Lee H, Tourigny MR, Hilbert D and Chen-Kiang S. (2000). J. Exp. Med., 192, 953–964.

Doi TS, Marino MW, Takahashi T, Yoshida T, Sakakura T, Old LJ and Obata Y. (1999). Proc. Natl. Acad. Sci. USA, 96,

2994–2999.

Duckett CS. (2002). J. Clin. Invest., 109, 579–580.

Duran A, Diaz-Meco MT and Moscat J. (2003). EMBO J., 22,

3910–3918.

Duyao MP, Buckler AJ and Sonenshein GE. (1990). Proc.

Natl. Acad. Sci. USA, 87, 4727–4731.

Edelstein LC, Lagos L, Simmons M, Tirumalai H and Gelinas

C. (2003). Mol. Cell. Biol., 23, 2749–2761.

Esslinger CW, Wilson A, Sordat B, Beermann F and Jongeneel

CV. (1997). J. Immunol., 158, 5075–5078.

Fan Y, Rayet B and Ge´linas C. (2003). Oncogene, (in press).

Feuillard J, Schuhmacher M, Kohanna S, Asso-Bonnet M, Ledeur F, Joubert-Caron R, Bissieres P, Polack A, Bornkamm GW and RaphaelM. (2000). Blood, 95,

2068–2075.

Finco TS, Westwick JK, Norris JL, Beg AA, Der CJ and Baldwin AS. (1997). J. Biol. Chem., 272, 24113–24116. Foehr ED, Bohuslav J, Chen LF, DeNoronha C, Geleziunas R, Lin X, O’Mahony A and Greene WC. (2000a). J. Biol. Chem., 275, 34021–34024. Foehr ED, Lin X, O’Mahony A, Geleziunas R, Bradshaw RA and Greene WC. (2000b). J. Neurosci., 20, 7556–7563. Franzoso G, Zazzeroni F and Papa S. (2003). Cell Death Differ., 10, 13–15.

Fujioka S, Sclabas GM, Schmidt C, Niu J, Frederick WA, Dong QG, Abbruzzese JL, Evans DB, Baker C and Chiao

PJ. (2003). Oncogene, 22, 1365–1370.

Garg A and AggarwalBB. (2002). Leukemia, 16, 1053–1068. Geddes BJ, Wang L, Huang WJ, Lavellee M, Manji GA, Brown M, Jurman M, Cao J, Morgenstern J, Merriam S, Glucksmann MA, DiStefano PS and Bertin J. (2001). Biochem. Biophys. Res. Commun., 284, 77–82. Gerondakis S and Strasser A. (2001). Nat. Immunol., 2,

377–379.

Oncogene

Gilmore TD. (1999). Oncogene, 18, 6925–6937. Gilmore TD, Cormier C, Jean-Jacques J and Gapuzan ME. (2001). Oncogene, 20, 7098–7103. Gilmore TD and Mosialos G. (2003). Nucl. Factor kB:

Regulation and Role in Disease, Beyaert R (ed). Kluwer Academic Publishers: The Netherlands, pp. 91–115. Goodkin ML, Ting AT and Blaho JA. (2003). J. Virol., 77,

7261–7280.

Green DR. (2003). Mol. Cell, 11, 551–552. Grenier JM, Wang L, Manji GA, Huang WJ, Al-Garawi A, Kelly R, Carlson A, Merriam S, Lora JM, Briskin M, DiStefano PS and Bertin J. (2002). FEBS Lett., 530,

73–78.

Grimm S, Bauer MKA, Baeuerle PA and Schulze-Osthoff K. (1996). J. Cell Biol., 134, 13–23. Grossmann M, Metcalf D, Merryfull J, Beg A, Baltimore D and Gerondakis S. (1999). Proc. Natl. Acad. Sci. USA, 96,

11848–11853.

Grossmann M, O’Reilly LA, Gugasyan R, Strasser A, Adams JM and Gerondakis S. (2000). EMBO J., 19, 6351–6360. Grumont RJ, Rourke IJ and Gerondakis S. (1999). Genes Dev., 13, 400–411. Grumont RJ, Rourke IJ, O’Reilly LA, Strasser A, Miyake K, Sha W and Gerondakis S. (1998). J. Exp. Med., 187,

663–674.

Guo Q, Robinson N and Mattson MP. (1998). J. Biol. Chem., 273, 12341–12351. Guttridge DC, Albanese C, Reuther JY, Pestell RG and Baldwin Jr AS. (1999). Mol. Cell. Biol., 19, 5785–5799. Hacker H and Karin M. (2002). Cancer Cell, 2, 431–433. Hamanoue M, Middleton G, Wyatt S, Jaffray E, Hay RT and Davies AM. (1999). Mol. Cell Neurosci., 14, 28–40. Hanson JL, Anest V, Reuther-Madrid J and Baldwin AS. (2003). J. Biol. Chem., 278, 34910–34917. Harhaj EW, Good L, Xiao G and Sun SC. (1999). Oncogene, 18, 1341–1349. Harper N, Hughes M, MacFarlane M and Cohen GM. (2003).

J. Biol. Chem., 278, 25534–25541.

Hatada EN, Do RK, Orlofsky A, Liou HC, Prystowsky M, MacLennan IC, Caamano J and Chen-Kiang S. (2003).

J. Immunol., 171, 761–768.

He Z, Xin B, Yang X, Chan C and Cao L. (2000). Cancer Res., 60, 1845–1848.

Headon DJ, EmmalSA, Ferguson BM, Tucker AS, Justice MJ, Sharpe PT, Zonana J and Overbeek PA. (2001). Nature, 414, 913–916. Hettmann T, DiDonato J, Karin M and Leiden JM. (1999).

J. Exp. Med., 189, 145–158.

Hinata K, Gervin AM, Jennifer Zhang Y and Khavari PA. (2003). Oncogene, 22, 1955–1964. Hinz M, Krappmann D, Eichten A, Heder A, Scheidereit C and Strauss M. (1999). Mol. Cell. Biol., 19, 2690–2698. Hinz M, Loser P, Mathas S, Krappmann D, Dorken B and Scheidereit C. (2001). Blood, 97, 2798–2807. Hiscott J, Kwon H and Genin P. (2001). J. Clin. Invest., 107,

143–151.

Hoffmann A, Leung TH and Baltimore D. (2003). EMBO J., 22, 5530–5539. Horwitz BH, Scott ML, Cherry SR, Bronson RT and Baltimore D. (1997). Immunity, 6, 765–772. Hsia CY, Cheng S, Owyang AM, Dowdy SF and Liou HC. (2002). Int. Immunol., 14, 905–916. Hsu BL, Harless SM, Lindsley RC, Hilbert DM and Cancro MP. (2002). J. Immunol., 168, 5993–5996. Hu WH, Johnson H and Shu HB. (2000). J. Biol. Chem., 275,

10838–10844.

Rel/NF-jB and apoptosis

J Kucharczak et al

Rel/NF- j B and apoptosis J Kucharczak et al 8979 Hu Y, Baud V, Delhase M,

8979

Hu Y, Baud V, Delhase M, Zhang P, Deerinck T, Ellisman M, Johnson R and Karin M. (1999). Science, 284, 316–320. Inohara N and Nunez G. (2003). Nat. Rev. Immunol., 3,

Lademann U, Kallunki T and Jaattela M. (2001). Cell Death Differ., 8, 265–272. Lamb JA, Ventura JJ, Hess P, Flavell RA and Davis RJ.

Lenz HJ. (2003). Cancer Treat. Rev., 29 (Suppl. 1), 41–48.

371–382.

(2003). Mol. Cell, 11, 1479–1489.

Inohara N, Ogura Y, Fontalba A, Gutierrez O, Pons F, Crespo J, Fukase K, Inamura S, Kusumoto S, Hashimoto M, Foster SJ, Moran AP, Fernandez-Luna JL and Nunez G. (2003). J. Biol. Chem., 278, 5509–5512. Ivanov VN, Deng G, Podack ER and Malek TR. (1995). Int. Immunol., 7, 1709–1720. Jan JT, Chen BH, Ma SH, Liu CI, Tsai HP, Wu HC, Jiang SY, Yang KD and Shaio MF. (2000). J. Virol., 74, 8680–8691.

Lee HH, Dadgostar H, Cheng Q, Shu J and Cheng G. (1999a). Proc. Natl. Acad. Sci. USA, 96, 9136–9141. Lee RM, Gillet G, Burnside J, Thomas SJ and Neiman P. (1999b). Genes Dev., 13, 718–728. Leitges M, Sanz L, Martin P, Duran A, Braun U, Garcia JF, Camacho F, Diaz-Meco MT, Rennert PD and Moscat J. (2001). Mol. Cell, 8, 771–780.

Javelaud D and Besancon F. (2001). Oncogene, 20, 4365–4372. Javelaud D, Poupon MF, Wietzerbin J and Besancon F.

Li

ZW, Chu W, Hu Y, Delhase M, Deerinck T, Ellisman M, Johnson R and Karin M. (1999a). J. Exp. Med., 189,

(2002). Int. J. Cancer, 98, 193–198.

1839–1845.

Jin R, De Smaele E, Zazzeroni F, Nguyen DU, Papa S, Jones J, Cox C, Gelinas C and Franzoso G. (2002). DNA

Li

Q, Estepa G, Memet S, IsraelA and Verma IM. (2000). Genes Dev., 14, 1729–1733.

Cell Biol., 21, 491–503. Johnson BW and Boise LH. (1999). J. Biol. Chem., 274,

Li

Q, Lu Q, Hwang JY, Buscher D, Lee KF, Izpisua-Belmonte JC and Verma IM. (1999b). Genes Dev., 13, 1322–1328.

18552–18558.

Jung M, Zhang Y, Dimtchev A and Dritschilo A. (1998).

621–663.

Li

ZW, Omori SA, Labuda T, Karin M and Rickert RC. (2003). J. Immunol., 170, 4630–4637.

Radiat. Res., 149, 596–601. Jung M, Zhang Y, Lee S and Dritschilo A. (1995). Science,

Li

Q, Van Antwerp D, Mercurio F, Lee KF and Verma IM. (1999c). Science, 284, 321–325.

268, 1619–1621.

Li

Q and Verma IM. (2002). Nat. Rev. Immunol., 2, 725–734.

Kalaitzidis D, Davis RE, Rosenwald A, Staudt LM and Gilmore TD. (2002). Oncogene, 21, 8759–8768. Karin M and Ben-Neriah Y. (2000). Annu. Rev. Immunol., 18,

Liao CL, Lin YL, Wu BC, Tsao CH, Wang MC, Liu CI, Huang YL, Chen JH, Wang JP and Chen LK. (2001). J. Virol., 75, 7828–7839. Lin A. (2003). BioEssays, 25, 17–24.

107–114.

Karin M, Cao Y, Greten FR and Li ZW. (2002). Nat. Rev. Cancer, 2, 301–310. Karin M and Lin A. (2002). Nat. Immunol., 3, 221–227. Kasibhatla S, Brunner T, Genestier L, Echeverri F, Mahboubi

515–524.

Lin KI, DiDonato JA, Hoffmann A, Hardwick JM and Ratan RR. (1998). J. Cell Biol., 141, 1479–1487. Lin A and Karin M. (2003). Semin. Cancer Biol., 13,

A and Green DR. (1998). Mol. Cell, 1, 543–551. Kasibhatla S, Genestier L and Green DR. (1999). J. Biol. Chem., 274, 987–992. Kasof GM, Lu JJ, Liu D, Speer B, Mongan KN, Gomes BC

Lin KI, Lee SH, Narayanan R, Baraban JM, Hardwick JM and Ratan RR. (1995). J. Cell Biol., 131, 1149–1161. Lin B, Williams-Skipp C, Tao Y, Schleicher MS, Cano LL, Duke RC and Scheinman RI. (1999). Cell Death Differ., 6,

and Lorenzi MV. (2001). Oncogene, 20, 7965–7975.

570–582.

Kato T, Delhase M, Hoffmann A and Karin M. (2003). Mol. Cell, (in press). Kayagaki N, Yan M, Seshasayee D, Wang H, Lee W, French DM, GrewalIS, Cochran AG, Gordon NC, Yin J, Starovasnik MA and Dixit VM. (2002). Immunity, 17,

Liptay S, Weber CK, Ludwig L, Wagner M, Adler G and Schmid RM. (2003). Int. J. Cancer, 105, 735–746. Liston P, Roy N, Tamai K, Lefebvre C, Baird S, Cherton- Horvat G, Farahani R, McLean M, Ikeda JE, MacKenzie AE and Korneluk RG. (1996). Nature, 379, 349–353. Liu L, Eby MT, Rathore N, Sinha SK, Kumar A and

24383–24391.

Khoshnan A, Tindell CA, Laux I, Bae D, Bennett BL and Nel AE. (2000). J. Immunol., 165, 1743–1754. Kim D, Xu M, Nie L, Peng XC, Jimi E, Voll RE, Nguyen T, Ghosh S and Sun XH. (2002a). Immunity, 16, 9–21. Kim SJ, Hwang SG, Shin DY, Kang SS and Chun JS. (2002b). J. Biol. Chem., 277, 33501–33508. Kobayashi K, Inohara N, Hernandez LD, Galan JE, Nunez G, Janeway CA, Medzhitov R and Flavell RA. (2002). Nature, 416, 194–199. Kontgen F, Grumont RJ, Strasser A, Metcalf D, Li R, Tarlinton D and Gerondakis S. (1995). Genes Dev., 9,

Chaudhary PM. (2002). J. Biol. Chem., 277, 13745–13751. Liu N, Raja SM, Zazzeroni F, Metkar SS, Shah R, Zhang M, Wang Y, Bromme D, Russin WA, Lee JC, Peter ME, Froelich CJ, Franzoso G and Ashton-Rickardt PG. (2003). EMBO J., 22, 5313–5322. Lopez M, Sly LM, Luu Y, Young D, Cooper H and Reiner NE. (2003). J. Immunol., 170, 2409–2416. Madrid LV, Wang CY, Guttridge DC, Schottelius AJ, Baldwin Jr AS and Mayo MW. (2000). Mol. Cell. Biol., 20, 1626–1638. Malewicz M, Zeller N, Yilmaz ZB and Weih F. (2003). J. Biol.

1965–1977.

Chem., 278, 32825–32833.

Kordes U, Krappmann D, Heissmeyer V, Ludwig WD and Scheidereit C. (2000). Leukemia, 14, 399–402. Kovalenko A, Chable-Bessia C, Cantarella G, Israel A, Wallach D and Courtois G. (2003). Nature, 424, 801–805. Kralova J, Liss AS, Bargmann W, Pendleton C, Varadarajan J, Ulug E and Bose Jr HR. (2002). J. Virol., 76,

Mao X, Moerman AM and Barger SW. (2002). J. Biol. Chem., 277, 44911–44919. Marianneau P, Cardona A, Edelman L, Deubel V and Despres P. (1997). J. Virol., 71, 3244–3249. Martin AG and Fresno M. (2000). J. Biol. Chem., 275,

11960–11970.

Kreuz S, Siegmund D, Scheurich P and Wajant H. (2001). Mol. Cell. Biol., 21, 3964–3973. La Rosa FA, Pierce JW and Sonenshein GE. (1994). Mol. Cell. Biol., 14, 1039–1044.

Martin P, Duran A, Minguet S, Gaspar ML, Diaz-Meco MT, Rennert P, Leitges M and Moscat J. (2002). EMBO J., 21,

4049–4057.

Martinon F, Burns K and Tschopp J. (2002). Mol. Cell, 10,

417–426.

Oncogene

Rel/NF- j B and apoptosis J Kucharczak et al 8980 Matsui K, Fine A, Zhu

Rel/NF-jB and apoptosis

J Kucharczak et al

8980

Matsui K, Fine A, Zhu B, Marshak-Rothstein A and Ju ST. (1998). J. Immunol., 161, 3469–3473. Mattson MP and Camandola S. (2001). J. Clin. Invest., 107,

247–254.

Mayo MW, Madrid LV, Westerheide SD, Jones DR, Yuan XJ, Baldwin Jr AS and Whang YE. (2002). J. Biol. Chem., 277, 11116–11125. Mayo MW, Wang CY, Cogswell PC, Rogers-Graham KS, Lowe SW, Der CJ and Baldwin AS. (1997). Science, 278,

1812–1815.

Medici MA, Sciortino MT, Perri D, Amici C, Avitabile E, Ciotti M, Balestrieri E, De Smaele E, Franzoso G and Mastino A. (2003). J. Biol. Chem., 278, 36059–36067. Michaelidis TM, Sendtner M, Cooper JD, Airaksinen MS, Holtmann B, Meyer M and Thoenen H. (1996). Neuron, 17,

75–89.

Micheau O, Lens S, Gaide O, Alevizopoulos K and Tschopp J. (2001). Mol. Cell. Biol., 21, 5299–5305.

Micheau O and Tschopp J. (2003). Cell, 114, 181–190. Middleton G, Hamanoue M, Enokido Y, Wyatt S, Pennica D, Jaffray E, Hay RT and Davies AM. (2000). J. Cell Biol., 148,

325–332.

Mir SS, Richter BW and Duckett CS. (2000). Blood, 96,

4307–4312.

Mitchell TC, Hildeman D, Kedl RM, Teague TK, Schaefer BC, White J, Zhu Y, Kappler J and Marrack P. (2001). Nat. Immunol., 2, 397–402. Mitchell TC, Teague TK, Hildeman DA, Bender J, Rees WA, KedlRM, Swanson B, Kappler JW and Marrack P. (2002a). Ann. NY Acad. Sci., 975, 114–131. Mitchell TC, Thompson BS, Trent JO and Casella CR. (2002b). Ann. NY Acad. Sci., 975, 132–147. Miyamoto S, Seufzer BJ and Shumway SD. (1998). Mol. Cell Biol., 18, 19–29. Moerman AM, Mao X, Lucas MM and Barger SW. (1999). Brain Res. Mol. Brain Res., 67, 303–315. Morrison RS, WenzelHJ, Kinoshita Y, Robbins CA, Donehower LA and Schwartzkroin PA. (1996). J. Neurosci., 16, 1337–1345. Motoyama N, Wang F, Roth KA, Sawa H, Nakayama K, Negishi I, Senju S, Zhang Q, Fujii S and Loh DY. (1995). Science, 267, 1506–1510. Muenchen HJ, Lin DL, Walsh MA, Keller ET and Pienta KJ. (2000). Clin. Cancer Res., 6, 1969–1977. Naito A, Yoshida H, Nishioka E, Satoh M, Azuma S, Yamamoto T, Nishikawa S and Inoue J. (2002). Proc. Natl. Acad. Sci. USA, 99, 8766–8771. Nakai M, Qin ZH, Chen JF, Wang Y and Chase TN. (2000). J. Neurochem., 74, 647–658. Nakamura Y, Grumont RJ and Gerondakis S. (2002). Mol. Cell. Biol., 22, 5563–5574. Nakshatri H, Bhat-Nakshatri P, Martin DA, Goulet Jr RJ and Sledge Jr GW. (1997). Mol. Cell. Biol., 17, 3629–

3639.

Neiman PE, Thomas SJ and Loring G. (1991). Proc. Natl. Acad. Sci. USA, 88, 5857–5861. Ni H, Ergin M, Huang Q, Qin JZ, Amin HM, Martinez RL, Saeed S, Barton K and Alkan S. (2001). Br. J. Haematol., 115, 279–286. Nicot C, Mahieux R, Takemoto S and Franchini G. (2000). Blood, 96, 275–281. Nitta T, Chiba A, Yamashita A, Rowe M, IsraelA, Reth M, Yamamoto N and Yamaoka S. (2003). Cell Signal., 15,

423–433.

Ogura Y, Inohara N, Benito A, Chen FF, Yamaoka S and Nunez G. (2001). J. Biol. Chem., 276, 4812–4818.

Oncogene

Okazaki T, Sakon S, Sasazuki T, Sakurai H, Doi T, Yagita H, Okumura K and Nakano H. (2003). Biochem. Biophys. Res. Commun., 300, 807–812. Owyang AM, Tumang JR, Schram BR, Hsia CY, Behrens TW, Rothstein TL and Liou HC. (2001). J. Immunol., 167,

4948–4956.

PahlHL. (1999). Oncogene, 18, 6853–6866. Panta GR, Cavin LG, Cortes P, Mercurio F, Lothstein L, Sweatman TW, IsraelM and Arsura M. (2003). (submitted). Papa S, Zazzeroni F, Bubici C, Jayawardena S, Alvarez K, Matsuda S, Nguyen D, Nelsbach A, Melis T, De Smaele E, Tang W, D’Adamio L and Franzoso G. (2003). (submitted). Pasparakis M, Courtois G, Hafner M, Schmidt-Supprian M, Nenci A, Toksoy A, Krampert M, Goebeler M, Gillitzer R, IsraelA, Krieg T, Rajewsky K and Haase I. (2002a). Nature, 417, 861–866. Pasparakis M, Schmidt-Supprian M and Rajewsky K. (2002b).

J. Exp. Med., 196, 743–752.

Perez D and White E. (2003). J. Virol., 77, 2651–2662. Perkins ND, Felzien LK, Betts JC, Leung K, Beach DH and

NabelGJ. (1997). Science, 275, 523–527. Petro JB and Khan WN. (2001). J. Biol. Chem., 276,

1715–1719.

Petro JB, Rahman SM, Ballard DW and Khan WN. (2000).

J. Exp. Med., 191, 1745–1754.

Pianetti S, Arsura M, Romieu-Mourez R, Coffey RJ and Sonenshein GE. (2001). Oncogene, 20, 1287–1299. Pizzi M, Goffi F, Boroni F, Benarese M, Perkins SE, Liou HC and Spano P. (2002). J. Biol. Chem., 277, 20717–20723. PohlT, Gugasyan R, Grumont RJ, Strasser A, Metcalf D, Tarlinton D, Sha W, Baltimore D and Gerondakis S. (2002). Proc. Natl. Acad. Sci. USA, 99, 4514–4519. Portis T, Harding JC and Ratner L. (2001). Blood, 98,

1200–1208.

Poyet JL, Srinivasula SM, Tnani M, Razmara M, Fernandes- Alnemri T and Alnemri ES. (2001). J. Biol. Chem., 276,

28309–28313.

Qin ZH, Chen RW, Wang Y, Nakai M, Chuang DM and Chase TN. (1999). J. Neurosci., 19, 4023–4033. Ravi R, Bedi A and Fuchs EJ. (1998). Cancer Res., 58,

882–886.

Ravi R, Bedi GC, Engstrom LW, Zeng Q, Mookerjee B, Gelinas C, Fuchs EJ and Bedi A. (2001). Nat. Cell. Biol., 3,

409–416.

Rayet B, Fan Y and Gelinas C. (2003). Mol. Cell. Biol., 23,

1520–1533.

Rayet B and Gelinas C. (1999). Oncogene, 18, 6938–6947. Reed JC and Green DR. (2002). Mol. Cell, 9, 1–3. Reuther JY and Baldwin Jr AS. (1999). J. Biol. Chem., 274,

20664–20670.

Reuther JY, Reuther GW, Cortez G, Pendergast AM and Baldwin Jr AS. (1998). Genes Dev., 12, 968–981. Reuther-Madrid JY, Kashatus D, Chen S, Li X, Westwick J, Davis RJ, Earp HS, Wang CY and Baldwin Jr AS. (2002). Mol. Cell. Biol., 22, 8175–8183. Richardson P. (2003). Cancer Treat. Rev., 29 (Suppl. 1),

33–39.

Rivera-Walsh I, Waterfield M, Xiao G, Fong A and Sun SC. (2001). J. Biol. Chem., 276, 40385–40388. Rocha S, Campbell KJ and Perkins ND. (2003a). Mol. Cell, 12, 15–25. Rocha S, Martin AM, Meek DW and Perkins ND. (2003b). Mol. Cell. Biol., 23, 4713–4727. Rodriguez CI, NogalML, Carrascosa AL, Salas ML, Fresno M and Revilla Y. (2002). J. Virol., 76, 3936–3942.

Romieu-Mourez R, Kim DW, Shin SM, Demicco EG, Landesman-Bollag E, Seldin DC, Cardiff RD and Sonenshein GE. (2003). Mol. Cell. Biol., 23, 5738–5754. Romieu-Mourez R, Landesman-Bollag E, Seldin DC, Traish AM, Mercurio AM and Sonenshein GE. (2001). Cancer Res., 61, 3810–3818. Rudolph D, Yeh WC, Wakeham A, Rudolph B, Nallainathan D, Potter J, Elia AJ and Mak TW. (2000). Genes Dev., 14,

854–862.

Ryan KM, Ernst MK, Rice NR and Vousden KH. (2000). Nature, 404, 892–897. Sabroe I, Prince LR, Jones EC, Horsburgh MJ, Foster SJ, VogelSN, Dower SK and Whyte MK. (2003). J. Immunol., 170, 5268–5275. Sakon S, Xue X, Takekawa M, Sasazuki T, Okazaki T, Kojima Y, Piao JH, Yagita H, Okumura K, Doi T and Nakano H. (2003). Embo J., 22, 3898–3909. Sakurai H, Chiba H, Miyoshi H, Sugita T and Toriumi W. (1999). J. Biol. Chem., 274, 30353–30356. Sanna MG, da Silva Correia J, Ducrey O, Lee J, Nomoto K, Schrantz N, Deveraux QL and Ulevitch RJ. (2002). Mol. Cell. Biol., 22, 1754–1766. Santoro MG, Rossi A and Amici C. (2003). EMBO J., 22,

2552–2560.

Schafer H, Arlt A, Trauzold A, Hunermann-Jansen A and Schmidt WE. (1999). Biochem. Biophys. Res. Commun., 262,

139–145.

Schaub FJ, Han DK, Liles WC, Adams LD, Coats SA, Ramachandran RK, Seifert RA, Schwartz SM and Bowen- Pope DF. (2000). Nat. Med., 6, 790–796. Schiemann B, Gommerman JL, Vora K, Cachero TG, Shulga- Morskaya S, Dobles M, Frew E and Scott ML. (2001). Science, 293, 2111–2114. Schilling D, Pittelkow MR and Kumar R. (2001). Oncogene, 20, 7992–7997. Schmidt-Supprian M, Bloch W, Courtois G, Addicks K, IsraelA, Rajewsky K and Pasparakis M. (2000). Mol. Cell, 5, 981–992. Schmidt-Ullrich R, Aebischer T, Hulsken J, Birchmeier W, Klemm U and Scheidereit C. (2001). Development, 128,

3843–3853.

Schneider A, Martin-Villalba A, Weih F, Vogel J, Wirth T and Schwaninger M. (1999). Nat. Med., 5, 554–559. Schwarz EM, Badorff C, Hiura TS, Wessely R, Badorff A, Verma IM and Knowlton KU. (1998). J. Virol., 72,

5654–5660.

Schwarz EM, Van Antwerp D and Verma IM. (1996). Mol. Cell Biol., 16, 3554–3559. Schwenzer R, Siemienski K, Liptay S, Schubert G, Peters N, Scheurich P, Schmid RM and Wajant H. (1999). J. Biol. Chem., 274, 19368–19374. Seitz CS, Freiberg RA, Hinata K and Khavari PA. (2000). J. Clin. Invest.