Oncogene (2004) 23, 2275–2286 & 2004 Nature Publishing Group All rights reserved 0950-9232/04 $25.00 www.nature.com/onc REVIEW The c-Rel and B-cell proliferation: a deal with the devil

Thomas D Gilmore*,1, Demetrios Kalaitzidis1, Mei-Chih Liang1 and Daniel T Starczynowski1

1Department of Biology, Boston University, 5 Cummington Street, Boston, MA 02215, USA

Activation of the Rel/NF-jB signal transduction pathway called the Rel homology domain, which has sequences has been associated with a varietyof animal and human required for DNA binding, dimerization, nuclear malignancies. However, among the Rel/NF-jB family localization, and inhibitor (IkB) binding. Sequences members, onlyc-Rel has been consistentlyshown to be C-terminal to the Rel homology domain in c-Rel, RelA, able to malignantlytransform cells in culture. In addition, and RelB contain transcriptional activation domains, c- has been activated bya retroviral promoter insertion and therefore dimers that contain these Rel/NF-kB in an avian B-cell lymphoma, and amplifications of REL members usually increase transcription of target genes. (human c-rel) are frequentlyseen in Hodgkin’s lympho- Although the Rel homology domain sequences of c-Rel mas and diffuse large B-cell lymphomas, and in some across species are highly conserved, their follicular and mediastinal B-cell lymphomas. Phenotypic C-terminal transactivation domains are not. For exam- analysis of c-rel knockout mice demonstrates that c-Rel ple, the Rel homology domains of chicken and human has a normal role in B-cell proliferation and survival; c-Rel are approximately 85% identical, whereas their moreover, c-Rel nuclear activityis required for B-cell C-terminal transactivation domains are only about 10% development. Few mammalian model systems are avail- identical. Therefore, the DNA-binding specificities of able to studythe role of c-Rel in oncogenesis, and it is still c-Rel proteins across species are likely to be quite not clear what features of c-Rel endow it with its unique similar, but the properties and regulation of their oncogenic activityamong the Rel/NF- jB family. In any transactivation domains may vary. event, REL mayprovide an appropriate therapeutic target The activity of Rel/NF-kB dimers is also highly for certain human lymphoid cell malignancies. regulated. The primary method of Rel/NF-kB regula- Oncogene (2004) 23, 2275–2286. doi:10.1038/sj.onc.1207410 tion is through interaction with the family of IkB Published online 2February 2004 inhibitor proteins (Ghosh and Karin, 2002). Interaction of the Rel/NF-kB dimer with the IkB generally Keywords: c-Rel; Rel; NF-kB; B cell; lymphoma; gene sequesters the complex in the cytoplasm and blocks its amplification ability to bind DNA. Activation of the Rel/NF-kB pathway by any number of inducers activates an IkB kinase (IKK) complex, which then phosphorylates IkB. Phosphorylated IkB is then a preferred substrate for ubiquitination and degradation by the proteasome, Introduction which frees the Rel/NF-kB complex to enter the nucleus and bind to target gene promoters/enhancers. However, The c-Rel transcription factor is a member of the Rel/ in some cases, Rel/NF-kB proteins may also have effects NF-kB family, which also includes RelA (p65), RelB, on gene expression that are not mediated by direct DNA p50, and p52(Gilmore, 2003).As Rel/NF- kB transcrip- binding, such as by interacting with heterologous DNA- tion factors have been the subject of numerous reviews, bound transcription factors. For example, NF-kB this article will focus on the unique role of c-Rel in can enhance bcl-2 gene expression by associating on controlling B-cell proliferation, survival, and oncogen- the bcl-2 enhancer through CREB and Sp1 (Heckman esis. In a variety of cell types, c-Rel exists as et al., 2002). homodimers or heterodimers with p50, or infrequently Like other Rel/NF-kB proteins, c-Rel is also subject with RelA. Dimers containing c-Rel bind to a set of to regulation beyond that imposed by IkB proteins. related 9–10 bp DNA sequences (kB sites), and regulate Phosphorylation has been reported to affect c-Rel DNA the expression of numerous cellular genes, including binding (Neumann et al., 1992; Glineur et al., 2000; Liu many genes involved in lymphoid cell development, and Beller, 2002) and its C-terminal transactivation proliferation, and survival. activity (Martin et al., 2001). Moreover, even though the Rel/NF-kB proteins are related through an approxi- C-terminal transactivation domains of human and mately 300-amino-acid N-terminal domain usually chicken c-Rel are weakly, if at all, conserved (Gilmore et al., 2001), they both contain numerous sites of serine *Correspondence: TD Gilmore; E-mail: [email protected] phosphorylation (Mosialos et al., 1991; Martin et al., Received 6 November 2003; revised 30 November 2003; accepted 1 2001). In the case of REL, phosphorylation of different December 2003 sets of serine residues affects its transactivation activity c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2276 in response to different stimuli, such as phorbol ester/ background of increased IkBa turnover in mature B ionomycin versus tumor necrosis factor (Martin and cells. Fresno, 2000; Martin et al., 2001). In addition, the C- c-rel knockout mice develop normally and have no terminal domain of REL has numerous sites of gross irregularities in hemopoiesis, but show immuno- ubiquitination (Chen et al., 1998), which has recently deficiencies (Ko¨ ntgen et al., 1995; Tumang et al., 1998). been shown to play a role in transactivation in other Although c-rel knockout mice show some alterations in transcription factors (Jin and Harper, 2003). T-cell function (Liou and Hsia, 2003), the primary defects in these mice occur in B cells. In particular, c-rel c-Rel plays a role in B-Cell proliferation and survival knockout mice have greatly reduced B-cell proliferation and reduced survival in response to mitogenic activa- The distinct phenotypes of knockout mice lacking tion, such as by treatment with LPS, anti-IgM, antigen, individual Rel/NF-kB family members and the different or CD40, and show reduced production in expression patterns of these proteins indicate that each response to antigen (Ko¨ ntgen et al., 1995; Harling- family member has unique physiological functions. As McNabb et al., 1999; Tumang et al., 2002). The reduced such, a role for c-Rel in B-cell development, growth, and mitogen-induced proliferation is due to arrested cell- survival has been established by the analysis of c-Rel cycle progression in c-relÀ/À B cells (Hsia et al., 2002). expression and activity and by the characterization of However, c-Rel is not required for antigen-independent c-rel knockout mice. B-cell proliferation that occurs during B-cell depletion The first indication that c-Rel might have a unique (Cabatingan et al., 2002). Consistent with the B-cell role in lymphoid cells came from in situ hybridization proliferation and survival defects, c-rel null mice have studies of mouse embryos and adult tissues, wherein it small/irregular germinal centers and reduced marginal was found that c-rel mRNA was expressed primarily in zone B cells (Cariappa et al., 2000). Recently, Cheng developing hematopoietic tissues (Carrasco et al., 1994). et al. (2003) have shown that overexpression of both The c-Rel protein is expressed at all stages of B-cell cyclin E and Bcl-XL can restore BCR-induced cell-cycle development, but is expressed at the highest levels and is progression and survival in c-relÀ/À B cells; Bcl-XL is constitutively nuclear in mature B cells (Grumont and likely to be a direct target for c-Rel activation, whereas Gerondakis, 1994; Liou et al., 1994; Weih et al., 1994). cyclin E gene expression is likely induced by c-Rel- The nuclear kB site DNA-binding activity in mature B mediated activation of cell-cycle transcription factor lymphocytes consists mainly of p50–c-Rel heterodimers, E2F3a. which is different from pre-B cells that have largely The c-rel null phenotypes are exacerbated in / inducible p50–RelA complexes (Grumont and Geron- c-rel double knockouts, which have reduced numbers of dakis, 1994; Liou et al., 1994; Miyamoto et al., 1994b; follicular, marginal, and peritoneal B cells and have Kistler et al., 1998). Moreover, c-rel mRNA has recently mature B cells that fail to proliferate in response to been shown to be elevated in mature B cells as compared antigen due to a cell-cycle defect (Pohl et al., 2002). to plasma B cells (Tarte et al., 2003; Zhan et al., 2003), Adoptive transfer studies have shown that embryonic paralleling the protein expression data. liver cells lacking both RelA and c-Rel can still The preferential nuclear accumulation of c-Rel in repopulate all hematopoietic cell lineages, although at mature B cells is probably due, at least in part, to a somewhat reduced level compared to wild-type cells increased basal turnover of IkBa in these cells (Miya- (Grossmann et al., 1999), emphasizing that both moto et al., 1994a, 1998; Doerre and Corley, 1999; activating subunits of NF-kB are not required for Shumway et al., 1999; Fields et al., 2000). In addition, hematopoietic cell development. However, mature IgM- Tam et al. (2001) have shown that the nuclear and IgD-positive B cells were absent in these mice, accumulation of p50–c-Rel heterodimers in lieu of because of a survival defect as a result of reduced p50–RelA heterodimers in mature B cells can also be expression of antiapoptotic genes bcl-2 and A1 (Gross- attributed to the types of IkB interactions that occur mann et al., 2000). Indeed, expression of a bcl-2 with the different Rel/NF-kB heterodimers and to the transgene in the c-rel/ double knockout cells can nuclear export rates of different heterodimers. Namely, restore their ability to generate peripheral B cells in RelA contains a nuclear export sequence whereas c-Rel adoptive hosts (Grossmann et al., 2000). Thus, c-Rel does not, and therefore p50–RelA heterodimers are and RelA appear to serve essential but redundant exported more efficiently from the nucleus than p50–c- functions in the development of mature B cells, and Rel heterodimers. Moreover, in pre-B cells and non-B c-rel knockout mice have increased levels of p50–RelA cells, c-Rel complexes are primarily associated with in their splenic B cells, no doubt as a compensation for IkBb, which does not shuttle through the nucleus, the lack of c-Rel (Gerondakis et al., 1996; Liou et al., whereas in mature B cells, a significant portion of c-Rel 1999). is also associated with IkBa, which does shuttle The homozygous disruption of several non-Rel/NF- continuously through the nucleus. Thus, although kB genes has been shown to affect both c-Rel and B-cell IkBa–p50–RelA and IkBa–p50–c-Rel complexes both maturation or proliferation, suggesting that these genes shuttle through the nucleus in mature B cells, the are involved in c-Rel expression, activity, or regulation. reduced nuclear export of c-Rel as compared to RelA Surprisingly, little work has been performed on regula- results in increased nuclear accumulation of p50–c-Rel tion of c-rel gene transcription; however, the mouse c-rel versus p50–RelA. All of these may be accelerated in the promoter has binding sites for the Ets family proteins

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2277 PU.1 and SpiB, and has several Rel/NF-kB binding sites. 2000; Grumont et al., 2002; Okkenhaug and Vanhae- Thus, c-Rel can enhance its own expression (Grumont sebroeck, 2003; Suzuki et al., 2003), the CARD domain et al., 1993). Moreover, PU.1 þ /À,SpiBÀ/À mice have proteins Bcl-10 and Carma-1 (Jun et al., 2003; Newton reduced c-rel mRNA expression and reduced numbers and Dixit, 2003; Xue et al., 2003), and isoforms of of immature and mature B cells (Hu et al., 2001). protein kinase C (Moscat et al., 2003). On the latter, it is Expression of an exogenous c-rel gene in PU.1 þ /À,SpiBÀ/À interesting to note that diffuse B-cell lymphomas with B cells restores their ability to form normal numbers of B poor prognosis specifically overexpress protein kinase cells in bone marrow reconstitution experiments (Hu C-b (Shipp et al., 2002). Activation of c-Rel and B-cell et al., 2001). Similarly, mice deficient in B-cell adaptor for proliferation/survival in response to B-cell PI3 kinase (BCAP) have reduced numbers of mature B stimulation can be abolished by disruption of genes cells due to reduced survival and proliferation, and their encoding the Syk (Harnett, 1996) and Bruton’s tyrosine B cells have reduced levels of c-Rel (Yamazaki and kinases (Kerner et al., 1995; Khan et al., 1995; Petro Kurosaki, 2003). Re-expression of c-Rel in BCAP- et al., 2000), the BLNK adaptor protein (Jumaa et al., deficient B cells restores both the number of mature B 1999; Pappu et al., 1999; Hayashi et al., 2000; Tan et al., cells and their responsiveness to B-cell receptor stimula- 2001), and phospholipase C-g2(Wang et al., 2000; Petro tion (Yamazaki and Kurosaki, 2003). These results and Khan, 2001), which are all involved in the PI3 suggest that there is a PI3 kinase pathway that promotes kinase pathway. On the other hand, Death Receptor 6 mature B cell-specific expression of c-rel,eitherthrough may be a negative regulator of c-Rel function, in that c-Rel itself, PU.1, or SpiB. mice with this gene knocked out have increased c-Rel The upstream pathways required for c-Rel-mediated activity in B cells and also have increased B-cell growth, proliferation, and survival in response to proliferation, survival, and germinal center formation mitogen stimulation have only recently started to be (Schmidt et al., 2003). Not surprisingly, several c-Rel worked out. Based on inhibitor and knockout studies, target genes in B cells are involved in growth, survival, several signaling molecules have been implicated in proliferation, and antibody production (see Figure 1 c-Rel function, including PI3 kinase/Akt (Andjelic et al., and Table 1).

Figure 1 Signaling pathways leading to c-Rel activation and c-Rel target gene expression in B cells. c-Rel dimeric complexes are predominantly held inactive in the cytoplasm of resting B cells until extracellular signals such as antigen (through B-cell Receptor, BCR), CD40 ligand (CD40), bacterial cell wall products (Toll-like Receptors, TLRs), etc. engage and activate their respective receptors. The activated receptors then initiate signaling pathways that lead to nuclear translocation of c-Rel complexes (through the classic IKK-IkB phosphorylation-IkB degradation pathway). For BCR activation, several intracellular pathways have been indicated to be involved in signaling to c-Rel. c-Rel complexes can be either c-Rel homodimers or heterodimers with p50 or RelA, and each of these complexes may regulate unique target genes important for B-cell function as shown. In contrast, Death Receptor 6 (DR6) appears to be a negative regulator of c-Rel complexes. Target genes whose expression is decreased in c-relÀ/À B lymphocytes (Hsia et al., 2002), but have not yet been shown to be direct target genes, are in italics

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2278 Table 1 Some c-Rel target genes in B cells et al., 2003). Taken together, these results suggest that Gene Cellular effect References the strength of the c-Rel transactivation domain, but not type of transactivation domain, is important for c-Rel’s c- Growth Grumont et al. (2002) and Lee et al. (1995) transforming activity. Moreover, they suggest that there IRF-4 Proliferation Grumont and Gerondakis (2000) is a moderate level of transactivation that is optimal for CD21 Proliferation Tolnay et al. (2002) E2F3a Proliferation Cheng et al. (2003) transformation of chicken lymphoid cells by c-Rel bcl-2 Survival Bureau et al. (2002) and Grossmann et al. proteins. Consistent with reduced transactivation by c- (2000) Rel proteins being most compatible with transforma- bfl-1/A1 Survival Grumont et al. (1999) and Zong et al. tion, deletions in sequences encoding C-terminal trans- (1999) activation sequences frequently arise in chicken bcl-Xl Survival Chen et al. (2000) gamma1 Ig heavy chain Kaku et al. (2002) lymphoid cells transformed by viral vectors containing gamma4 Ig heavy chain Agresti and Vercelli (2002) wild-type chicken or human c-rel (Hrdlickova´ et al., 1994; Gilmore et al., 1995; Fan et al., 2003; Starczy- nowski et al., 2003). The analysis of many mutants has shown that in order Transformation of chicken lymphoid cells by v-Rel for v-Rel to transform and immortalize chicken and c-Rel proteins lymphoid cells in culture, v-Rel must be expressed at a high level, form homodimers, enter the nucleus (in part The first indication that c-Rel could be oncogenic came by escaping regulation by IkBa), bind to DNA, and from the study of the avian Rev-T retrovirus, which activate transcription (Gilmore, 1999). REL is also contains v-rel as its sole gene (Gilmore, 1999). An likely to transform chicken cells by activating target understanding of Rel-induced oncogenesis in avian gene expression (Starczynowski et al., 2003). Never- lymphoid cells is likely to continue to provide insights theless, if one looks at the subcellular localization of into the role that REL plays in human B-cell cancers. v- v-Rel or REL in transformed chicken lymphoid cells by Rel causes a rapidly fatal B-cell lymphoma/leukemia in immunofluorescence, one finds that these proteins are young chickens, and in vitro v-Rel can transform and primarily located in the cytoplasm (Gilmore, 1999; immortalize a variety of chicken hematopoietic cell Starczynowski et al., 2003). types from primary spleen, bone marrow, or bursal How then does one resolve the discrepancy between cultures, including B- and T-lymphoid cells, myeloid the mutagenesis studies that indicate that v-Rel and cells, erythroid cells, and dendritic cells. However, the REL must activate transcription to transform cells with most common targets for transformation by v-Rel are the immunofluorescence data that indicate that these cells of the B-cell lineage, including immature and proteins are largely cytoplasmic in transformed cells? mature B cells with normal or abnormal immunoglo- First, some Rel in these cells is clearly in the nucleus: by bulin gene rearrangements. Retroviral vectors for the biochemical fractionation, one can readily detect some overexpression of chicken, mouse, and human c-Rel can v-Rel and REL homodimers in the nuclear fractions of also transform chicken lymphoid cells in vitro, although transformed cells, and these dimers can bind to DNA these normal c-Rel proteins transform less efficiently (Hrdlickova´ et al., 1995a; Starczynowski et al., 2003). than v-Rel (Gilmore, 1999; Gilmore et al., 2001; Fan Second, treatment of v-Rel- or REL-transformed spleen et al., 2003). Of note, overexpression of other human cells with the nuclear export inhibitor leptomycin B family members (p50, p52, RELA, RELB) or active causes the Rel protein to accumulate in the nucleus, IKKB cannot transform chicken lymphoid cells suggesting that these proteins are continually shuttling (Fan et al., 2003). between the cytoplasm and the nucleus in these cells v-Rel has multiple mutations that make it a more (Sachdev and Hannink, 1998; Starczynowski et al., effective transforming agent than its avian progenitor c- 2003). Third, the addition of a strong nuclear export Rel. The primary activating mutation in v-Rel is the sequence onto v-Rel reduces its ability to transform cells deletion of C-terminal c-Rel sequences that consist of a (Sachdev et al., 1997), suggesting that further increasing strong transactivation domain and, thus, chicken c-Rel v-Rel’s cytoplasmic presence or reducing its nuclear mutants missing these residues are more transforming residence impairs its transforming activity. Further- than wild-type chicken c-Rel (Kamens et al., 1991; more, shortly after infection of primary lymphoid cells Hrdlickova´ et al., 1994). Similarly, we have recently with v-Rel retroviral vector, v-Rel is in the nucleus, and shown that human REL can be rendered more only after a few weeks of culture does its localization transforming in chicken spleen cells by deletion of either change to primarily cytoplasmic (Hrdlickova´ et al., one of two C-terminal transactivation subdomains 1995b). Taken together, these results suggest the (Starczynowski et al., 2003). Moreover, the entire REL following model for Rel-mediated transformation of C-terminal transactivation domain can be functionally chicken lymphoid cells. When an oncogenic Rel protein replaced in lymphoid cell transformation assays by a is initially expressed in target lymphoid cells, it is mutant form of the viral VP16 transactivation domain primarily in the nucleus (Hrdlickova´ et al., 1995b). that activates at a level similar to REL, but cannot be However, as it progressively turns on kB site-containing replaced by the wild-type VP16 transactivation domain target genes, including p100, p105, and IkBa, the Rel that activates approximately 10 times more potently protein becomes sequestered largely in the cytoplasm than the REL transactivation domain (Starczynowski (Hrdlickova´ et al., 1995b). Nevertheless, even at these

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2279 later stages, Rel complexes continue to enter the nucleus In contrast, several lines of evidence suggest that c- to increase (and in some cases, probably decrease) the Rel has direct and complete oncogenic activity in expression of kB site target genes, and this chronic low- lymphoid cells. First, as discussed above, overexpression level induction of target gene expression is essential for of human REL (but not other human Rel/NF-kB family the maintenance of the transformed state (Sachdev and members) can malignantly transform chicken lymphoid Hannink, 1998; Starczynowski et al., 2003). cells in vitro (Gilmore et al., 2001; Starczynowski et al., Consistent with this model for B-cell transformation, 2003). Second, transgenic mice in which v-rel expression many v-Rel target genes in transformed chicken cells are is directed by a T cell-specific promoter develop T-cell involved in lymphoid cell growth control or survival tumors, albeit with a long latency of 6–10 months (antiapoptosis) (reviewed in Gilmore, 1999). Genes (Carrasco et al., 1996). Third, c-rel gene expression has showing increased expression in v-Rel-transformed cells been upregulated in B-cell lymphomas by genetic include proto-oncogenes (c-jun,c-rel), transcription mechanisms: that is, one chicken B-cell lymphoma cell factors (STAT1), cytokine receptors (IL-2Ra), growth- line has a retroviral integration upstream of c-rel inducing molecules (IRF-4), and antiapoptotic mole- (Kabrun et al., 1990); in one primary Hodgkin’s cules (IAP1). Moreover, a number of genes that show lymphoma the REL gene has been translocated near increased or decreased expression in v-Rel-induced the immunoglobulin light chain gene enhancer (Barth lymphoid cell tumors as compared to normal 2-week et al., 2003); and the REL gene is amplified in many bursa have recently been identified, including several human lymphomas (see below). involved in adhesion, B-cell receptor signaling (through PI3 kinase), and apoptosis (Neiman et al., 2003). It is not known which induced or repressed genes are REL gene rearrangements essential for transformation by v-Rel. Because the altered expression of multiple genes probably contri- REL gene rearrangements and deletions have been butes to v-Rel-induced oncogenesis, inhibition of only detected only rarely in human lymphomas. In the best- one or a few of these genes may not be sufficient to characterized case, the RC-K8 diffuse large B-cell block oncogenesis fully. lymphoma (DLBCL) cell line provides an interesting Although all of the molecular details of v-Rel- and glimpse into the genetic and molecular gymnastics that REL-induced transformation of avian lymphoid cells some lymphoma cells may go through to ensure optimal may not be identical to how misregulated REL is levels of REL signaling. The RC-K8 cell line has two involved in human B-cell cancers, it is likely that the known mutagenic events in the REL pathway: first, due general mechanism by which Rel proteins promote to a large deletion on chromosome 2, a chimeric REL avian lymphoid cell oncogenesis – by increasing the protein (REL–NRG) is made in which the REL DNA- expression of cell proliferation and survival genes – is binding/dimerization domain is fused to sequences of similar to what occurs in human cancers with mis- unknown function (Non-Rel-Gene); and second, there regulated REL. Nevertheless, it is important to point are inactivating mutations in the gene encoding IkBa, out that mutations analogous to those that enhance the such that no IkBa protein is produced (Lu et al., 1991; oncogenicity of c-Rel proteins in avian lymphoid cells Kalaitzidis and Gilmore, 2002; Kalaitzidis et al., 2002). have not yet been identified in any human cancers (see The REL–NRG fusion protein can bind to kB site below). DNA; however, because REL–NRG lacks a functional C-terminal transactivation domain, it does not activate REL and human B-cell cancers transcription from kB site-driven reporter genes and it cannot be inhibited by IkBa (Kalaitzidis et al., 2002). In Aberrant, constitutively active Rel/NF-kB activity has nuclear extracts of RC-K8 cells, one finds high levels of been implicated in human cancers in several settings kB site DNA-binding complexes that exclusively contain (Gilmore et al., 2002; Karin et al., 2002). Although p50– combinations of p50, wild-type REL, and REL–NRG; RelA NF-kB complexes are constitutively nuclear and furthermore, a number of known Rel/NF-kB target active in many human tumor cell types (Gilmore et al., genes, including BCL-x, TRAF1, Bfl-1/A1, ICAM-1, 2002), in most cases this activity probably contributes to and A20, show high levels of expression in RC-K8 cells the survival (i.e., antiapoptosis) of the tumor cells (Kalaitzidis et al., 2002). That wild-type REL activity is (Barkett and Gilmore, 1999; Baldwin, 2001). Thus, in required for the growth of RC-K8 cells is suggested by these settings, inhibition of NF-kB activity will likely be the finding that overexpression of wild-type or the valuable as adjuvant therapy: that is, to sensitize tumor super-repressor form of IkBa kills these cells (Kalaitzidis cells to the apoptotic effect of traditional chemo- or et al., 2002). Thus, one reasonable model for the role of radiotherapeutic agents. Similarly, although genetic REL signaling in RC-K8 cells is that the effect of alterations that lead to C-terminal trunction of p100 constitutive REL nuclear activity, caused by the lack of (in the NFKB2 gene) and overexpression of the p50/p52 IkBa protein, is reduced to a moderate and optimal coactivator BCL-3 have been found in human B-cell oncogenic level by competition with nonactivating leukemias/lymphomas (Gilmore et al., 2002; Karin et al., REL–NRG dimers. As described above for chicken 2002), these proteins have not been demonstrated to lymphoid cell transformation, it seems that REL-driven have oncogenic activity in lymphoid cells in vitro or in oncogenesis in RC-K8 cells is maintained by a tempered transgenic mice. level of constitutive transactivation.

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2280 Two additional REL gene rearrangements have Does REL gene amplification define a molecularly distinct recently been detected in classical Hodgkin’s lymphoma subclass of DLBCL? tumor samples: in one case the REL gene is translocated to a position near the light chain enhancer, and in a One cannot discuss the molecular basis of human B-cell second case there is an alteration near the 30 end of REL lymphomas without addressing the pioneering work of such that a C-terminally truncated REL protein is Standt and colleagues on the molecular profiling and synthesized that shows strong constitutively nuclear categorizing of DLBCLs (reviewed in Shaffer et al., staining (Barth et al., 2003). 2002), which may also have relevance to REL-induced oncogenesis. Based on cDNA microarray expression REL gene amplifications in B-cell lymphomas patterns, Staudt’s group has divided otherwise clinically homogeneous DLBCLs into three sub-types: ABC type REL gene amplifications have been found in a high (with expression patterns similar to Activated B cells), percentage of human B-cell lymphomas (Table 2), GCB type (with patterns similar to Germinal Center B including B15–20% of DLBCLs (Rosenwald et al., cells), and type III (with patterns distinct from the first 2002; Houldsworth et al., 2003), possibly 15–20% of two) (Alizadeh et al., 2000; Rosenwald et al., 2002). The follicular B-cell lymphomas, and 40–50% of classical ABC and GCB DLBCLs have been characterized in Hodgkin’s lymphomas. Surprisingly, however, there are most detail. The ABC DLBCL subtype, which was no well-characterized human B-cell lymphoma cell lines reported to have a poorer clinical prognosis by with a highly amplified REL locus, which has hindered Rosenwald et al. (2002), has a group of NF-kB target molecular studies on the role of REL gene amplification genes that have elevated expression. Paradoxically, these in human B-cell lymphomas. ‘NF-kB’ target genes are not highly expressed in the Despite the common amplification of the REL locus, GCB subtype, even though REL gene amplifications there has been scant analysis of REL protein expression were exclusively found in B15% of the GCB DLBCLs in primary human B-cell lymphomas. Namely, it is not (Rosenwald et al., 2002). By electrophoretic mobility clear whether REL gene amplification faithfully trans- shift assays, two ABC-like cell lines were reported to lates into increased REL protein expression. Three have more constitutively nuclear kB site binding activity reports have recently used immunohistochemistry to than two GCB-type cell lines, and the proliferation of analyse REL expression in DLBCL (Donnelly et al., the ABC cell lines was inhibited by introduction of a 2001; Houldsworth et al., 2003) and Hodgkin’s lym- vector encoding the IkBa super-repressor, whereas phoma (Barth et al., 2003) samples. In two of the reports proliferation of GCB cell lines was not inhibited by (Donnelly et al., 2001; Barth et al., 2003), lymphoma the IkBa super-repressor vector (Davis et al., 2001). samples with REL gene amplification showed increased Taken together, these results led to the suggestion that nuclear REL staining as compared to samples without constitutive activation of the Rel/NF-kB pathway REL gene amplification. In contrast, Houldsworth et al. contributes to the sustained oncogenicity of the ABC- (2003) did not find a correlation between nuclear REL like DLBCLs but not the GCB-like DLBCLs (Davis expression and REL gene amplification: that is, some et al., 2001). DLBCL samples with a high level of REL gene As provocative as the results of Staudt’s group may amplification had less nuclear REL staining than be, their experiments have some loose ends. First, a DLBCL samples with no REL gene amplification. recent analysis of a new panel of DLBCLs using the Nuclear REL staining has also recently been detected Staudt cDNA expression criteria found REL gene in five of six cases of mediastinal large B-cell lymphoma amplifications in about 20% of the samples in all three (Savage et al., 2003). molecular subtypes, and did not find that the ABC Instinctively, one might conclude that only lympho- subtype had a poorer clinical outcome (Houldsworth mas with nuclear REL staining have REL-driven et al., 2003). Second, we believe that it is an exaggeration oncogenic activity. However, it is important to remem- to say that the nuclear kB site binding activity in the ber that immunocytochemistry provides only a static GCB-like cell lines that Staudt’s group has characterized picture of REL. As discussed above, v-Rel and REL in as having ‘low’ nuclear NF-kB activity is really all that transformed chicken spleen cells also appear to be low. For example, as shown in Figure 2, the amount of primarily cytoplasmic proteins by immunofluorescence, active DNA-binding activity in the SUDHL-4 GCB-like and yet the evidence is quite convincing that these cell line (‘low NF-kB activity’) is similar to that in oncogenic proteins are continually entering the nucleus extracts from TNFa-stimulated 3T3 cells, which are to drive the gene transcription that causes transforma- commonly regarded as having ‘high’ or induced nuclear tion and immortalization of these cells. Thus, one kB site binding activity. However, the RC-K8 DLBCL cannot assume that cytoplasmic REL staining of cell line, which has no IkBa, has significantly more lymphoma samples mandates inactive REL complexes. nuclear c-Rel DNA-binding activity than either On the other hand, in that c-Rel complexes are normally SUDHL-4 cells or TNFa-stimulated mouse fibroblasts in the nucleus of mature B cells, one cannot determine (see Figure 2; Kalaitzidis et al., 2002). Third, the active by immunolocalization whether nuclear REL staining in Rel/NF-kB family proteins that were found in nuclear specific human B-cell lymphomas is driving malignancy extracts of ABC versus GCB cell lines are largely or is simply a marker for the developmental stage of the different: whereas the ABC lines had generally given tumor. p50-RelA, the GCB cell lines had REL-containing

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2281 Table 2 Amplification of the REL locus in human B-cell lymphomas B-cell lymphoma type Amplified REL/total no. analysed Comment Reference (% positive)

DLBCL 26/111 (23%) Houldsworth et al. (1996)

DLBCL 2/31 (6%) GI tract Barth et al. (1998)

DLBCL 21/91 (23%) Rao et al. (1998)

DLBCL 3/7 (43%) Goff et al. (2000)

DLBCL 1/34 (3%) GI tract Barth et al. (2001)

DLBCL 5/91 (6%) Palanisamy et al. (2002)

DLBCL 17/240 (7%) 17/115 (15%) of GCB subtype Rosenwald et al. (2002)

DLBCL 2/12 (17%) Martinez-Climent et al. (2003)

DLBCL 1/13 (8%) Wessendorf et al. (2003)

DLBCL 9/43 (21%) In all three molecular subtypes Houldsworth et al. (2003)

Follicular/diffuse 2/8 (25%) Lu et al. (1991)

Follicular 2/28 (7%) Bentz et al. (1996)

Follicular 3 /34 (9%) Avet-Loiseau et al. (1997)

Follicular 3/6 (50%) Goff et al. (2000)

Follicular 7/45 (16%) Neat et al. (2001)

Follicular 0/12(0%) Martinez-Climent et al. (2003)

Hodgkin’s 6/12(50%) Joos et al. (2002)

Hodgkin’s 24/44 (55%) Martin-Subero et al. (2002)

Hodgkin’s 12/25 (48%) Barth et al. (2003)

Hodgkin’s 22/39 (56%) 14/16 of NS type Joos et al. (2003)

Hodgkin’s 2/2 (100%) Martin-Subero et al. (2003)

Primary mediastinal 2/26 (8%) Joos et al. (1996)

Primary mediastinal 2/40 (5%) Palanisamy et al. (2002)

Aggressive B cell 3/61 (5%) Werner et al. (1997)

Extranodal marginal zone (MALT 2/18 (11%) Barth et al. (2001) type)

Cutaneous CD30+ anaplastic 6/8 (75%) Mao et al. (2003) large cell lymphoma

GI, gastrointestinal; GCB, germinal center B cell; NS, nodular sclerosis complexes (Davis et al., 2001; Liang et al., 2003). The et al., 2003), suggesting that overexpression of IkBa or significance and activity of different nuclear kB site the IkBa super-repressor may have effects in addition to DNA-binding complexes in the two DLBCL subtypes is inhibition of NF-kB. not clear. In addition, although the GCB-like SUDHL-4 How does one explain the apparent paradox of the cells were not killed by expression of the IkBa super- presence of REL gene amplifications in a relatively high repressor (Davis et al., 2001), we have recently shown percentage (15%) of the DLBCL subtype (GCB) that that we can readily kill GCB-like SUDHL-4 cells with a does not contain elevated NF-kB target gene expression synthetic NF-kB signaling inhibitor that correspond- and that is not killed by overexpression of the IkBa ingly reduces the REL DNA-binding complexes (Liang super-repressor (Rosenwald et al., 2002)? First, REL et al., 2003). Furthermore, IkBa has recently been gene amplifications may have nothing to do with shown to interact with tumor suppressor (Chang, DLBCL oncogenesis (for example, REL amplifications 2002; Zhou et al., 2003) and cell-cycle kinase CDK4 (Li may be a phenotype of this type of malignant cell or

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2282 neither. In addition, the BCL11a gene is translocated into the Ig heavy chain locus in a subset of B-cell chronic lymphocytic leukemias (Dyer, 2003), and is activated by retroviral insertions in some mouse myeloid leukemias (Nakamura et al., 2000; Suzuki et al., 2002). BCL11a can transform mouse 3T3 cells (Nakamura et al., 2000), but has not been shown to transform lymphoid cells. Bcl11a knockout mice die shortly after birth, and embryos from these mice have no B cells, due to a defect in the development of pre-pro-B cells from a lymphoid precursor cell (Liu et al., 2003). BCL11a encodes a zinc-finger Kruppel-like transcription repres- sor protein; however, its target genes are not known. It Figure 2 kB site DNA-binding activity in human diffuse B-cell lymphoma cell lines. Cell extracts from the indicated cell lines were is tempting to speculate that the balance between REL- analysed by an electrophoretic mobility shift assay using a induced gene activation and BCL11a-directed gene radiolabeled kB site probe as described (Liang et al., 2003). Mouse repression cooperates in certain B-cell lymphomas that 3T3 cells plus or minus tumor necrosis factor; SUDHL-4 have 2p13 amplicons, and certainly studies are war- lymphoma cells (NF-kB is mainly p50—REL; Liang et al., 2003); ranted to determine whether REL and BCL11a can RC-K8 lymphoma cells (NF-kB is comprised of complexes containing p50, REL, and REL—NRG; Kalaitzidis et al., 2002) cooperate in B-cell oncogenesis in either avian or mouse systems. The genes encoding interferon regulatory factor 4, a B amplification of genes near REL may be important for cell-specific transcriptional repressor, and the proto- malignancy (see Discussion below of BCL11a)). Second, oncogene c-myc are both transcriptional targets of c-Rel as suggested by Shaffer et al. (2002), increased REL and could also be REL cooperating genes in B-cell activity provided by REL gene amplification may not be lymphomagenesis. The IRF-4 gene is upregulated by relevant to final stage growth of GCB DLBCLs, but c-Rel in mitogen-stimulated B cells (Grumont and may be important at any earlier stage. Third, REL gene Gerondakis, 2000) and IRF4À/À B cells proliferate amplification may be important for GCB-like DLBCL poorly in response to mitogens (Mittrucker et al., 1997). cell growth in vivo but not in vitro. Indeed, it is Furthermore, the IRF4 gene is translocated into the important to note that the GCB cell lines that Davis immunoglobulin heavy chain locus in some human et al. (2001) analysed for in vitro growth inhibition by myelomas (Iida et al., 1997) and B-cell lymphomas the IkBa super-repressor were not those that contained (Tamura et al., 2001), and is highly expressed in many REL gene amplifications. GCB DLBCLs with REL gene DLBCLs (Tsuboi et al., 2000). Similarly, alterations in amplifications (and possibly REL overexpression) may c-myc gene structure and activity occur in many animal respond differently to overexpression of IkBa than those and human B-cell cancers (Popescu and Zimonjic, 2002). without REL gene amplification/overexpression. Thus, we believe that it is unresolved whether REL/NF-kB Proliferative activities of Rel that may not involve direct contributes to the malignant state in only a subset of regulation of transcription DLBCLs. The discussion above has made the tacit assumption that the proliferative and survival effects of c-Rel in B Genes that might cooperate with deregulated REL in cells are all manifested by direct effects of c-Rel on gene B-cell oncogenesis transcription. However, it is important to acknowledge Survival, developmental, or proliferation genes may that c-Rel may have effects that do not involve direct cooperate with REL in B-cell tumorigenesis. Two lines DNA binding and gene activation. Chen and Li (1998) of evidence point to the well-known antiapoptotic gene have shown that C-terminal sequences of REL can bind BCL-2 as one such REL-cooperating gene. First, BCL2 to cyclin E and Cdk2complexes, suggesting that c-Rel is rearranged in a high percentage of GCB-like could have nontranscriptional effects on cell-cycle DLBCLs, which constitute the same subclass of regulatory proteins. Mice in which sequences encoding DLBCLs in which REL gene amplifications are exclu- the C-terminal transactivation domain of c-Rel have sively found (Rosenwald et al., 2002). Second, coex- been knocked out show B-cell hyperplasia (Carrasco pression of Bcl-2can enhance the transforming activity et al., 1998). Finally, v-Rel can stimulate replication of of weakly transforming mutants of v-Rel and of human viral DNA (Ishikawa et al., 1993). REL in chicken lymphoid cells (White and Gilmore, 1996; Gilmore et al., 2001; Rayet et al., 2003). A second candidate gene for cooperating with REL is Conclusions and perspectives the BCL11a (aka Evi9) gene, which is located quite near to REL at chromosome 2p13 and is frequently As outlined in this review, there is reason to believe that coamplified with REL in B-cell lymphomas (Satterwhite enhanced REL activity plays a role in a large number of et al., 2001). Indeed, it is not clear whether the criminal human B-cell lymphomas. Nevertheless, the involve- gene in these amplifications is REL, BCL11a, both, or ment of REL in human lymphoid cell cancers is

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2283 unproven in that: (1) the REL gene amplifications seen tively, albeit with an extremely long latency in both cases in human diffuse B-cell lymphomas and Hodgkin’s (Carrasco et al., 1996; Romieu-Mourez et al., 2003). lymphomas still provide only circumstantial evidence Thus, the development of mammalian cell and animal that they are involved in the generation or maintenance systems for the study of REL-induced B-cell lympho- of these cancers; (2) there is still only a single model magenesis represents an important goal. In the mouse, system (the avian lymphoid cell system) in which human this may entail the coexpression of REL with other B- REL has clearly been shown to be oncogenic; and (3) cell proliferation or survival proteins in vitro in there are no specific inhibitors of REL that one can use appropriate cell types or the expression of REL in to test directly whether amplified or overexpressed REL transgenic mice at a specific stage of B-cell development is required for B-cell tumor growth. or in a suitable genetic background. In human cells, the overexpression of REL (and perhaps cooperating The missing mouse model for REL-induced B-cell proteins) in primary human lymphocytes may result in oncogenesis the establishment of long-term liquid cell cultures (as occurs with Epstein–Barr virus). In addition, the In spite of the acute transforming activity of v-Rel and establishment of B-cell lymphoma cell lines from chicken, mouse and human c-Rel in primary chicken primary human tumors with amplified REL will be spleen cells, there have been no reports of Rel-mediated useful for identifying REL target genes involved in transformation of mouse B lymphocytes. Indeed, early oncogenesis and for anti-REL therapy efforts. reports demonstrated that v-Rel expression directed by a retroviral vector was toxic in primary mouse bone Anti-REL therapy marrow cells (Schwartz and Witte, 1988). Despite a number of attempts, we have not been able to transform REL may be a particularly suitable therapeutic tran- primary mouse bone marrow or spleen cells by infection scription factor target for several reasons. First, gene with retroviral vectors for the expression of v-Rel or knockouts of c-rel do not affect development or viability human REL. Similarly, v-Rel cannot convert the pro-B in mice, and the activity of c-Rel appears to be primarily cell line BaF3 to growth factor independence (Gilmore required for the proliferation and survival of activated et al., 2003). Moreover, there have been no reports of mature B cells. Thus, anti-REL therapeutics may be transgenic mice that are capable of sustained B cell- generally tolerated. Second, REL gene amplifications specific expression of wild-type c-rel. Finally, c-rel was are now being identified with reasonable frequency in a not identified as a site for retroviral integration in mouse variety of somewhat common and therapeutically lymphomas in extensive screens for gene targets of accessible tumor types, including Hodgkin’s and non- insertional mutagenesis in mouse lymphomas (Lund Hodgkin’s B-cell lymphomas. Third, the structure of the et al., 2002; Suzuki et al., 2002). DNA-binding domain of c-Rel has been solved (Huang How then to reconcile the acute oncogenicity of Rel et al., 2001). Fourth, some Hodgkin’s tumor cells and proteins in avian B cells and the frequent amplification of DLBCLs have inactivating mutations in IkB genes REL in human B-cell lymphomas with the apparent (Wood et al., 1997; Cabannes et al., 1999; Emmerich inability of Rel proteins to transform murine B cells? et al., 1999; Krappmann et al., 1999; Jungnickel et al., There are three likely explanations. First, it may simply 2002; Kalaitzidis et al., 2002; Emmerich et al., 2003), be that for technical reasons, the oncogenicity of c-Rel and thus the Rel pathway is activated downstream of proteins has not been demonstrated in mouse B cells: IKK; as such, some REL-driven lymphoid cell cancers that is, that the Rel proteins need to be expressed under may be refractory to IKK inhibitors or may develop specific conditions in vitro or in vivo for their oncogeni- resistance to IKK inhibitors through mutations that city to be apparent. For example, sustained expression of inactivate IkBa. Therefore, direct inhibitors of REL, c-Rel might be required at a specific stage of B-cell rather than inhibitors of upstream Rel/NF-kB signaling development (e.g., only at the mature B-cell stage) for molecules, are likely to prove to be superior therapeutics transgenic mice to develop B-cell lymphomas. Second, it for many Hodgkin’s and DLBCLs. Nevertheless, how may be that avian and human B cells are susceptible to one might target c-Rel activity is not obvious. However, Rel-induced transformation, whereas mouse B cells are compounds that either specifically affect REL dimeriza- not. Indeed, with respect to expression of some tumor tion (with itself or p50) or REL’s interaction with suppressor proteins and telomerase, chicken cells and downstream activators (e.g., through protein–protein human cells are similar to one another and distinct from interactions mediated by the unique sequences of the mouse cells (Forsyth et al., 2002; Kim et al., 2002). REL transactivation domain) may provide safe and Third, avian lymphoid cells may be susceptible to single- effective therapies for a variety of human B-cell step transformation by Rel proteins (even from diverse lymphomas. species), while mammalian B cells require additional events. For example, chickens appear to lack the Ink4a Acknowledgements retinoblastoma inhibitor protein and a fully functional Research in our laboratory is supported by National Institutes Arf tumor suppressor (Kim et al., 2003). Mouse cells are of Health Grant CA47763 to TDG. M-CL was supported in not entirely refractory to transformation by Rel proteins part by a scholarship from the Ministry of Education, Taiwan; in that v-Rel and mouse c-Rel can induce tumors in DTS was supported in part by a Predoctoral Fellowship from transgenic mice in T cells and mammary cells, respec- the Natural Sciences & Engineering Research Council of Canada.

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2284 References

Agresti A and Vercelli D. (2002). Mol. Immunol., 38, Emmerich F, Theurich S, Hummel M, Haeffker A, Vry MS, 849–859. Do¨ hner K, Bommert K, Stein H and Do¨ rken B. (2003). J. Alizadeh AA, Eisen M, Davis RE, Ma C, Lossos IS, Pathol., 201, 413–420. Rosenwald A, Boldrick JC, Sabet H, Tran T, Yu X, Powell Fan Y, Rayet B and Ge´ linas C. (2003). Oncogene, (in press). JI, Yang L, Marti GE, Moore T, Hudson Jr J, Lu L, Lewis Fields ER, Seufzer BJ, Oltz EM and Miyamoto S. (2000). J. DB, Tibshirani G, Sherlock G, Chan WC, Greiner TC, Immunol., 164, 297–305. Weisenburger DE, Armitage JO, Warnke R, Levy R, Wilson Forsyth NR, Wright WE and Shay JW. (2002). Differentiation, W, Grever MR, Byrd JC, Botstein D, Brown PO and Staudt 69, 188–197. LM. (2000). Nature, 403, 503–511. Gerondakis S, Strasser A, Metcalf G, Grigoriadis G, Andjelic S, Hsia C, Suzuki H, Kadowaki T, Koyasu S and Scheerlinck JY and Grumont RJ. (1996). Proc. Natl. Acad. Liou H-C. (2000). J. Immunol., 165, 3860–3867. Sci. USA, 93, 3405–3409. Avet-Loiseau H, Vigier M, Moreau A, Mellerin MP, Gaillard Ghosh S and Karin M. (2002). Cell, 109, S81–S96. F, Harousseau JL, Bataille R and Milpied N. (1997). Br. J. Gilmore TD, Cormier C, Jean-Jacques J and Gapuzan M-E. Haematol., 97, 119–122. (2001). Oncogene, 20, 7098–7103. Baldwin AS. (2001). J. Clin. Invest., 107, 241–246. Gilmore TD, Gapuzan M-E, Kalaitzidis D and Starczynowski Barkett M and Gilmore TD. (1999). Oncogene, 18, D. (2002). Cancer Lett., 181, 1–9. 6910–6924. Gilmore TD, Jean-Jacques J, Richards R, Cormier C, Kim J Barth TF, Bentz M, Leithauser F, Stilgenbauer S, Siebert R, and Kalaitzidis D. (2003). Virology, 316, 9–16. Schlotter M, Schlenk RF, Do¨ hner H and Mo¨ ller P. (2001). Gilmore TD, White DW, Sarkar S and Sif S. (1995). The DNA Genes Chromosomes Cancer, 31, 316–325. Provirus: Howard Temin’s Scientific Legacy. Cooper GM, Barth TFE, Do¨ hner H, Werner CA, Stilgenbauer S, Schlotter Greenberg Temin R, Sugden B (eds). American Society for M, Pawlita M, Lichter P, Mo¨ ller P and Bentz M. (1998). Microbiology Press: Washington, DC, pp. 109–128. Blood, 91, 4321–4330. Gilmore TD. (1999). Oncogene, 18, 6925–6937. Barth TFE, Martin-Subero JI, Joos S, Menz CK, Hasel C, Gilmore TD. (2003). Rel/NF-k B transcription factors at Mechtersheimer G, Parwaresch RM, Lichter P, Siebert R www.nf-kb.org. and Mo¨ ller P. (2003). Blood, 101, 3681–3686. Glineur C, Davioud-Charvet E and Vandenbunder B. (2000). Bentz M, Werner CA, Do¨ hner H, Joos S, Barth TF, Siebert R, Biochem. J., 352, 583–591. Schroder M, Stilgenbauer S, Fischer K, Mo¨ ller P and Goff LK, Neat MJ, Crawley CR, Jones L, Jones E, Lichter P. (1996). Blood, 88, 1437–1444. Lister TA and Gupta RK. (2000). Br. J. Haematol., 111, Bureau F, Vanderplasschen A, Jaspar F, Minner F, Pastoret 618–625. PP, Merville MP, Bours V and Lekeux P. (2002). Blood, 99, Grossmann M, Metcalf D, Merryfull J, Beg A, Baltimore D 3683–3691. and Gerondakis S. (1999). Proc. Natl. Acad. Sci. USA, 96, Cabannes E, Khan G, Aillet F, Jarrett RF and Hay RT. 11848–11853. (1999). Oncogene, 18, 3063–3070. Grossmann M, O’Reilly LA, Gugasyan R, Strasser A, Cabatingan MS, Schmidt MR, Sen R and Woodland RT. Adams JM and Gerondakis S. (2000). EMBO J., 19, (2002). J. Immunol., 169, 6795–6805. 6351–6360. Cariappa A, Liou H-C, Horwitz BH and Pilai S. (2000). Grumont RJ and Gerondakis S. (1994). Cell Growth Differ., 5, J. Exp. Med., 192, 1175–1182. 1321–1331. Carrasco D, Cheng J, Lewin A, Warr G, Yang H, Rizzo C, Grumont RJ and Gerondakis S. (2000). J. Exp. Med., 191, Rosas F, Snapper C and Bravo R. (1998). J. Exp. Med., 187, 1281–1292. 973–984. Grumont RJ, Richardson IB, Gaff C and Gerondakis S. Carrasco D, Rizzo CA, Dorfman K and Bravo R. (1996). (1993). Cell Growth Differ., 4, 731–743. EMBO J., 15, 3640–3650. Grumont RJ, Rourke IJ and Gerondakis S. (1999). Genes Carrasco D, Weih F and Bravo R. (1994). Development, 120, Dev., 12, 400–411. 2991–3004. Grumont RJ, Strasser A and Gerondakis S. (2002). Mol. Cell, Chang NS. (2002). J. Biol. Chem., 277, 10323–10331. 10, 1283–1294. Chen C, Edelstein LC and Ge´ linas C. (2000). Mol. Cell. Biol., Harling-McNabb L, Deliyannis G, Jackson DC, Gerondakis 20, 2687–2695. S, Grigoriadis G and Brown LE. (1999). Int. Immunol., 11, Chen E and Li CC. (1998). Biochem. Biophys. Res. Commun., 1431–1439. 249, 728–734. Harnett M. (1996). Immunol. Today, 17, 4. Chen E, Hrdlickova´ R, Nehyba J, Longo DL, Bose Jr HR and Hayashi K, Nittono R, Okamoto N, Tsuji S, Hara Y, Li C-CH. (1998). J. Biol. Chem., 273, 35201–35207. Goitsuka R and Kitamura D. (2000). Proc. Natl. Acad. Cheng S, Hsia CY, Leone G and Liou H-C. (2003). Oncogene, Sci. USA, 97, 2755–2760. 22, 8472–8486. Heckman CA, Mehew JW and Boxer LM. (2002). Oncogene, Davis RE, Brown KD, Siebenlist U and Staudt LM. (2001). 21, 3898–3908. J. Exp. Med., 194, 1861–1874. Houldsworth J, Mathew S, Rao PH, Dyomina K, Louie DC, Doerre S and Corley RB. (1999). J. Immunol., 163, 269–277. Parsa N, Offit K and Chaganti RS. (1996). Blood, 87, Donnelly GB, Teruya-Feldstein J, Quin J, Zelenetz AD, 25–29. Houldsworth J and Chaganti RS. (2001). Blood, 98, Houldsworth J, Olshen AB, Cattoretti G, Donnelly GB, 302. Teruya-Feldstein J, Qin J, Palanisamy N, Shen Y, Dyomina Dyer MJ. (2003). Int. J. Hematol., 77, 315–320. K, Petlakh M, Pan Q, Zelenetz AD, Dalla-Favera R and Emmerich F, Meiser M, Hummel M, Demel G, Foss HD, Chaganti RS. (2003). Blood, (in press). Jundt F, Mathas S, Krappmann D, Scheidereit C, Stein H Hrdlickova´ R, Nehyba J and Bose Jr HR. (1995a). J. Virol., and Do¨ rken B. (1999). Blood, 94, 3129–3134. 69, 3369–3380.

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2285 Hrdlickova´ R, Nehyba J and Humphries EH. (1994). J. Virol., Lee H, Arsura M, Wu M, Duyao M, Buckler AJ and 68, 2371–2382. Sonenshein GE. (1995). J. Exp. Med., 181, 1169–1177. Hrdlickova´ R, Nehyba J, Roy A, Humphries EH and Bose Jr Li J, Joo SH and Tsai MD. (2003). Biochemistry, 42, HR. (1995b). J. Virol., 69, 403–413. 13476–13483. Hsia CY, Cheng S, Owyang AM, Dowdy SF and Liou H-C. Liang M-C, Bardhan S, Li C, Pace EA, Porco Jr JA and (2002). Int. Immunol., 14, 905–916. Gilmore TD. (2003). Mol. Pharmacol., 64, 123–131. Hu CJ, Rao S, Ramirez-Bergeron DL, Garrett-Sinha LA, Liou H-C and Hsia CY. (2003). BioEssays, 25, 767–780. Gerondakis S, Clark MR and Simon MC. (2001). Immunity, Liou H-C, Jin Z, Tumang J, Andjelic S, Smith KA and Liou 15, 545–555. ML. (1999). Int. Immunol., 11, 361–371. Huang D, Chen Y, Ruetsche M, Phelps CB and Ghosh G. Liou H-C, Sha WC, Scott ML and Baltimore D. (1994). Mol. (2001). Structure, 9, 669–678. Cell. Biol., 14, 5349–5359. Iida S, Rao PH, Butler M, Corradini P, Boccadoro M, Klein Liu J and Beller DJ. (2002). J. Immunol., 169, 581–586. B, Chaganti RS and Dalla-Favera R. (1997). Nat. Genet., 17, Liu P, Keller JR, Ortiz M, Tessarollo L, Rachel RA, 226–230. Nakamura T, Jenkins NA and Copeland NG. (2003). Nat. Ishikawa H, Asano M, Kanda T, Ge´ linas C and Ito Y. (1993). Immunol., 4, 525–532. Oncogene, 8, 2889–2896. Lu D, Thompson JD, Gorski GK, Rice NR, Mayer MG and Jin J and Harper JW. (2003). Cancer Cell, 3, 517–518. Yunis JJ. (1991). Oncogene, 6, 1235–1241. Joos S, Granzow M, Holtgreve-Grez H, Siebert R, Harder L, Lund AH, Turner G, Trubetskoy A, Verhoeven E, Wientjens Martin-Subero JI, Wolf J, Adamowicz M, Barth TF, Lichter E, Hulsman D, Russell R, DePinho RA, Lenz J and van P and Jauch A. (2003). Int. J. Cancer, 103, 489–495. Lohuizen M. (2002). Nat. Genet., 32, 160–165. Joos S, Menz CK, Wrobel G, Siebert R, Gesk S, Ohl S, Mao X, Orchard G, Lillington DM, Russell-Jones R, Young Mechtersheimer G, Trumper L, Mo¨ ller P, Lichter P and BD and Whittaker S. (2003). Genes Chromosomes Cancer, Barth TF. (2002). Blood, 99, 1381–1387. 37, 176–185. Joos S, Otan˜ o-Joos MI, Ziegler S, Bru¨ derlein S, du Manoir S, Martin AG and Fresno M. (2000). J. Biol. Chem., 275, Bentz M, Mo¨ ller P and Lichter P.. (1996). Blood, 87, 24383–24391. 1571–1578. Martin AG, San-Antonio B and Fresno M.. (2001). J. Biol. Jumaa H, Wollscheid B, Mitterer M, Wienands J, Reth M and Chem., 276, 15840–15849. Nielsen PJ. (1999). Immunity, 11, 547–554. Martinez-Climent JA, Alizadeh AA, Segraves R, Blesa D, Jun JE, Wilson LE, Vinuesa CG, Lesage S, Blery M, Miosge Rubio-Moscardo F, Albertson DG, Garcia-Conde J, Dyer LA, Cook MC, Kucharska EM, Hara H, Penninger JM, MJ, Levy R, Pinkel D and Lossos IS. (2003). Blood, 101, Domashenz H, Hong NA, Glynne RJ, Nelms KA and 3109–3117. Goodnow CC. (2003). Immunity, 18, 751–762. Martin-Subero JI, Gesk S, Harder L, Sonoki T, Tucker PW, Jungnickel B, Staratschek-Jox A, Bra¨ uninger A, Spieker T, Schlegelberger B, Grote W, Novo FJ, Calasanz MJ, Wolf J, Diehl V, Hansmann M-L, Rajewsky K and Ku¨ ppers Hansmann ML, Dyer MJ and Siebert R. (2002). Blood, 99, R. (2002). J. Exp. Med., 191, 395–401. 1474–1477. Kabrun N, Bumstead N, Hayman MJ and Enrietto PJ. (1990). Martin-Subero JI, Knippschild U, Harder L, Barth TF, Mol. Cell. Biol., 10, 4788–4794. Riemke J, Grohmann S, Gesk S, Hoppner J, Mo¨ ller P, Kaku H, Horikawa K, Obata Y, Kato I, Okamoto H, Parwaresch RM and Siebert R. (2003). Leukemia, 17, Sakaguchi N, Gerondakis S and Takatsu K. (2002). Int. 2214–2219. Immunol., 14, 1055–1064. Mittrucker HW, Matsuyama T, Grossman A, Kundig TM, Kalaitzidis D and Gilmore TD. (2002). Genes Chromosomes Potter J, Shahinian A, Wakeham A, Patterson B, Ohashi PS Cancer, 34, 129–135. and Mak TW. (1997). Science, 275, 540–543. Kalaitzidis D, Davis RE, Rosenwald A, Staudt LM and Miyamoto S, Chiao P and Verma IM. (1994a). Mol. Cell. Gilmore TD. (2002). Oncogene, 21, 8759–8768. Biol., 14, 3276–3282. Kamens J, Richardson P, Mosialos G, Brent R and Gilmore T. Miyamoto S, Chiao P and Verma IM. (1994b). Proc. Natl. (1991). Mol. Cell. Biol., 10, 2840–2847. Acad. Sci. USA, 91, 5056–5060. Karin M, Cao Y, Greten FR and Li ZW. (2002). Nat. Rev. Miyamoto S, Seufzer BJ and Shumway SD. (1998). Mol. Cell. Cancer, 2, 301–310. Biol., 18, 19–29. Kerner JD, Appleby MW, Mohr RN, Chien S, Rawlings DJ, Moscat J, Diaz-Meco MT and Rennert P. (2003). EMBO Rep., Maliszewski CR, Witte ON and Perlmutter RM. (1995). 4, 31–36. Immunity, 3, 301–312. Mosialos G, Hamer P, Capobianco AJ, Laursen R and Khan WN, Alt FW, Gerstein RM, Malynn BA, Larsson I, Gilmore TD. (1991). Mol. Cell. Biol., 11, 5867–5877. Rathbun G, Davidson L, Muller S, Kantor AB, Herzenberg Nakamura T, Yamazaki Y, Saiki Y, Moriyama M, Largae- LA, Rosen FS and Sideras P. (1995). Immunity, 3, 283–299. spada DA, Jenkins NA and Copeland NG. (2000). Mol. Kim H, You S, Farris J, Kong BW, Christman SA, Foster LK Cell. Biol., 20, 3178–3186. and Foster DN. (2002). Exp. Cell Res., 272, 199–208. Neat MJ, Foot N, Jenner M, Goff L, Ashcroft K, Burford D, Kim SH, Mitchell M, Fujii H, Llanos S and Peters G. (2003). Dunham A, Norton A, Lister TA and Fitzgibbon J. (2001). Proc. Natl. Acad. Sci. USA, 100, 211–216. Genes Chromosomes Cancer, 32, 236–243. Kistler B, Rolink R, Marienfeld R, Neumann M and Wirth T. Neiman PE, Grbic JJ, Polony TS, Kimmel R, (1998). J. Immunol., 160, 2308–2317. Bowers SJ, Delrow J and Beemon KL. (2003). Oncogene, Ko¨ ntgen F, Grumont RJ, Strasser A, Metcalf D, Li R, 22, 1073–1086. Tarlinton D and Gerondakis S. (1995). Genes Dev., 9, Neumann M, Tsapos K, Scheppler JA, Ross J and Franza Jr 1965–1977. BR. (1992). Oncogene, 7, 2095–2104. Krappmann D, Emmerich F, Kordes U, Scharschmidt E, Newton K and Dixit V. (2003). Curr. Biol., 13, 1247–1251. Do¨ rken B and Scheidereit C. (1999). Oncogene, 18, Okkenhaug K and Vanhaesebroeck B. (2003). Nat. Rev. 943–953. Immunol., 3, 317–330.

Oncogene c-Rel and B-cell proliferation/oncogenesis TD Gilmore et al 2286 Palanisamy N, Abou-Elella AA, Chaganti SR, Houldsworth J, Neuberg DS, Lander ES, Aster JC and Golub TR. (2002). Offit K, Louie DC, Terayu-Feldstein J, Cigudosa JC, Rao Nat. Med., 8, 68–74. PH, Sanger WG, Weisenburger DD and Chaganti RS. Shumway DS, Maki M and Miyamoto S. (1999). J. Biol. (2002). Genes Chromosomes Cancer, 33, 114–122. Chem., 274, 30874–30881. Pappu R, Chenng AM, Li B, Gong Q, Chiu C, Griffin N, Starczynowski DT, Reynolds JG and Gilmore TD. (2003). White M, Sleckman BP and Chan AC. (1999). Science, 286, Oncogene, 22, 6928–6936. 1949–1954. Suzuki H, Matsuda S, Terauchi Y, Fujiwara M, Ohteki T, Petro JB and Khan WN. (2001). J. Biol. Chem., 276, Asano T, Behrens TW, Kouro T, Takatsu K, Kadowaki T 1715–1719. and Koyasu S. (2003). Nat. Immunol., 4, 280–286. Petro JB, Rahman SM, Ballard DW and Khan WN. (2000). Suzuki T, Shen H, Akagi K, Morse HC, Malley JD, Naiman J. Exp. Med., 191, 1745–1753. DQ, Jenkins NA and Copeland NG. (2002). Nat. Genet., 32, Pohl T, Gugasyan R, Grumont TJ, Strasser A, Metcalf D, 166–174. Tarlinton D, Sha W, Baltimore D and Gerondakis S. (2002). Tam WF, Wang W and Sen R. (2001). Mol. Cell. Biol., 21, Proc. Natl. Acad. Sci. USA, 99, 4514–4519. 4837–4846. Popescu NC and Zimonjic DB. (2002). J. Cell Mol. Med., 6, Tamura A, Miura I, Iida S, Yokota S, Horiike S, Nishida K, 151–159. Fujii H, Nakamura S, Seto M, Ueda R and Taniwaki M. Rao PH, Houldsworth J, Dyomina K, Parsa NZ, Cigudosa (2001). Cancer Genet. Cytogenet., 129, 1–9. JC, Louie DC, Popplewell L, Offit K, Jhanwar SC and Tan JE, Wong SC, Gan SK, Xu S and Lam KP. (2001). J. Biol. Chaganti RS. (1998). Blood, 92, 234–240. Chem., 276, 20055–20063. Rayet B, Fan Y and Ge´ linas C. (2003). Mol. Cell. Biol., 23, Tarte K, Zhan F, De Vos J, Klein B and Shaughnessy J. 1520–1533. (2003). Blood, 102, 592–600. Romieu-Mourez R, Kim DW, Shin SM, Demicco EG, Tolnay M, Vereshchagina LA and Tsokos GC. (2002). J. Landesman-Bollag E, Seldin DC, Cardiff RD and Sonen- Immunol., 169, 6236–6243. shein GE. (2003). Mol. Cell. Biol., 23, 5738–5754. Tsuboi K, Iida S, Inagaki H, Kato M, Hayami Y, Hanamura Rosenwald A, Wright G, Chan WC, Connors JM, Campo E, I, Miura K, Harada S, Kikuchi M, Komatsu H, Banno S, Fisher RI, Gascoyne RD, Muller-Hermelink HK, Smeland Wakita A, Nakamura S, Eimoto T and Ueda R. (2000). EB, Giltnane JM, Hurt EM, Zhao H, Averett L, Yang L, Leukemia, 14, 449–456. Wilson WH, Jaffe ES, Simon R, Klausner RD, Powell J, Tumang JR, Hsia CY, Tian W, Bromberg JF and Liou H-C. Duffey PL, Longo DL, Greiner TC, Weisenburger DD, (2002). Cell Immunol., 217, 47–57. Sanger WG, Dave BJ, Lynch JC, Vose J, Armitage JO, Tumang JR, Owyang A, Andjelic S, Jin Z, Hardy RR, Montserrat E, Lopez-Guillermo A, Grogan TM, Miller TP, Liou ML and Liou H-C. (1998). Eur. J. Immunol., 28, LeBlanc M, Ott G, Kvaloy S, Delabie J, Holte H, Krajci P, 4299–4312. Stokke T and Staudt LM. (2002). N. Engl. J. Med., 346, Wang D, Feng J, Wen R, Marine JC, Sangster MY, Parganas 1937–1947. E, Hoffmeyer A, Jackson CW, Cleveland JL, Murray PJ and Sachdev S and Hannink M. (1998). Mol. Cell. Biol., 18, Ihle JN. (2000). Immunity, 13, 25–35. 5445–5456. Weih F, Carrasco D and Bravo R. (1994). Oncogene, 9, Sachdev S, Diehl JA, McLinsey TA, Hans A and Hannink M. 3289–3297. (1997). Oncogene, 14, 2585–2594. Werner CA, Dohner H, Joos S, Trumper LH, Baudis M, Barth Satterwhite E, Sonoki T, Willis TG, Harder L, Nowak R, TF, Ott G, Moller P, Lichter P and Bentz M. (1997). Am. J. Arriola EL, Liu H, Price HP, Gesk S, Steinemann D, Pathol., 151, 335–342. Schlegelberger B, Oscier DG, Siebert R, Tucker PW and Wessendorf S, Schwaenen C, Kohlhammer H, Kienle D, Dyer MJ. (2001). Blood, 98, 3413–3420. Wrobel G, Barth TF, Nessling M, Mo¨ ller P, Do¨ hner H, Savage KJ, Monti S, Kutok JL, Cattoretti G, Neuberg D, De Lichter P and Bentz M. (2003). Oncogene, 22, 1425–1429. Leval L, Kurtin P, Dal Cin P, Ladd C, Feuerhake F, White DW and Gilmore TD. (1996). Oncogene, 13, 891–899. Aguiar RC, Li S, Salles G, Berger F, Jing W, Pinkus GS, Wood KM, Roff M and Hay RT. (1997). Oncogene, 16, Habermann T, Dalla-Favera R, Harris NL, Aster JC, Golub 2131–2139. TR and Shipp MA. (2003). Blood, 102, 3871–3879. Xue L, Morris SW, Orihuela C, Tuomanen E, Cui X, Wen R Schmidt CS, Liu J, Zhang T, Song HY, Sandusky G, Mintze and Wang D. (2003). Nat. Immunol., 4, 857–865. K, Benschop RJ, Glasebrook A, Yang DD and Na S. (2003). Yamazaki T and Kurosaki T. (2003). Nat. Immunol., 4, J. Exp. Med., 197, 51–62. 857–865. Schwartz RC and Witte ON. (1988). Virology, 165, 182–190. Zhan F, Tian E, Bumm K, Smith R, Barlogie B and Shaffer AL, Rosenwald A and Staudt LM. (2002). Nat. Rev. Shaughnessy Jr J. (2003). Blood, 101, 1128–1140. Immunol., 2, 920–933. Zhou M, Gu L, Zhu N, Woods WG and Findley HW. (2003). Shipp MA, Ross KN, Tamayo P, Weng AP, Kutok JL, Aguiar Oncogene, 22, 8137–8144. RC, Gaasenbeek M, Angelo M, Reich M, Pinkus GS, Ray Zong WX, Edelstein LC, Chen C, Bash J and Ge´ linas C. TS, Koval MA, Last KW, Norton A, Lister TA, Mesirov J, (1999). Genes Dev., 13, 382–387.

Oncogene