Nfkb1 is dispensable for Myc-induced lymphomagenesis - Nature

Oncogene (2005) 24, 6231–6240

& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00 www.nature.com/onc

Nfkb1 is dispensable for Myc-induced lymphomagenesis Ulrich Keller1, Jonas A Nilsson1,2, Kirsteen H Maclean1, Jennifer B Old1 and John L Cleveland*,1 1

Department of Biochemistry, St Jude Children’s Research Hospital, 332 N. Lauderdale, Memphis, TN 38105, USA

Rel/NF-jB transcription factors are critical arbiters of immune responses, cell survival, and transformation, and are frequently deregulated in cancer. The p50 NF-jB1 component of Rel/NF-jB DNA-binding dimers regulates genes involved in both cell cycle traverse and apoptosis. Nfkb1 loss accelerates B cell growth and leads to increased B cell turnover in vivo, phenotypes akin to those manifested in B cells of El-Myc transgenic mice, a model of human Burkitt lymphoma. Interestingly, ElMyc B cells express reduced levels of cytoplasmic and nuclear NF-jB1 and have reduced Rel/NF-jB DNAbinding activity, suggesting that Myc-mediated repression of NF-jB1 might mediate its proliferative and apoptotic effects on B cells. Furthermore, Nfkb1 expression was reduced in the majority of El-Myc lymphomas and was also suppressed in human Burkitt lymphoma. Nonetheless, loss of Nfkb1 did not appreciably affect Myc’s proliferative or apoptotic responses in B cells and had no effect on lymphoma development in El-Myc mice. Therefore, Nfkb1 is dispensable for Myc-induced lymphomagenesis. Oncogene (2005) 24, 6231–6240. doi:10.1038/sj.onc.1208779; published online 6 June 2005 Keywords: NF-kB1; c-Myc; lymphomagenesis

Introduction Deregulated cell proliferation and apoptosis are hallmarks of cancer (Hanahan and Weinberg, 2000). Myc oncoproteins are basic helix–loop–helix–leucine zipper transcription factors, whose expression are normally tightly controlled by mitogens and are suppressed by growth inhibitory signaling pathways (Grandori et al., 2000; Nilsson and Cleveland, 2003). These controls on cMyc, N-Myc, and L-Myc expression are frequently lost in cancer, either directly by chromosomal amplifications or translocations, or indirectly by mutations in signaling pathways or tumor suppressors that hold Myc expression in check (Grandori et al., 2000; Evan and Vousden, 2001; Nilsson and Cleveland, 2003). The selection for Myc activation in cancer reflects, at least in part, its *Correspondence: JL Cleveland; E-mail: [email protected] 2 Current address: Department of Molecular Biology, Umea˚ University, SE-901 87 Umea˚, Sweden Received 25 February 2005; revised and accepted 15 April 2005; published online 6 June 2005

essential role in cell proliferation and/or growth (de Alboran et al., 2001; Trumpp et al., 2001), yet Myc overexpression also accelerates rates of cell cycle traverse (Bouchard et al., 1998), and is sufficient for driving quiescent cells into S phase (Cavallieri and Goldfarb, 1987), and blocks terminal differentiation (Coppola and Cole, 1986). In normal cells, Myc’s hyperproliferative response is held in check by the activation of apoptotic pathways that kill the insulted cell (Askew et al., 1991; Evan et al., 1992), through the agency of the Arf-p53 tumor suppressor pathway (Zindy et al., 1998), and through regulating anti- and proapoptotic members of the Bcl-2 family that regulate the cell death program (Eischen et al., 2001; Jeffers et al., 2003; Egle et al., 2004). These checkpoints guard against Myc-induced transformation and tumor development in vivo (Strasser et al., 1990; Eischen et al., 1999; Schmitt et al., 1999; Evan and Vousden, 2001), and disabling these apoptotic checkpoints is a common denominator of Myc-induced tumors (Eischen et al., 1999, 2001). In Em-Myc transgenic mice, a model of the MYC/Ig t(8:14) translocation in human Burkitt lymphoma where c-Myc is overexpressed in B lymphocytes by the immunoglobulin heavy chain enhancer (Em) (Adams et al., 1985), Myc-induced lymphoma is preceded by a prolonged precancerous phase in which the high proliferative rates of Em-Myc B cells are counterbalanced by a high apoptotic index (Maclean et al., 2003). With time, however, cell cycle and apoptotic regulators that hold Myc in check are disabled, and this leads to fulminant, aggressive clonal lymphomas that kill these mice by 3–6 months of age. Although p27Kip1 and the Arf-p53 pathway are key regulators of Myc’s proliferative and apoptotic responses in this system (Eischen et al., 1999; Schmitt et al., 1999; Martins and Berns, 2002), relatively little is known as to how Myc regulates these pathways. The Rel/NF-kB family of transcription factors (RelA [p65], RelB, c-Rel, Nfkb1 [p105/p50] and Nfkb2 [p100/ p52]) are compelling mediators of Myc’s responses in B cells, as many are key regulators of B cell proliferation and survival, and of B cell development. For example, loss of Nfkb1 augments B cell proliferation and provokes rapid B cell turnover in vivo (Sha et al., 1995; Grumont et al., 1998), whereas c-Rel is required for proper germline CH transcription, Ig class switching, and to protect B cells from antigen receptor-mediated apoptosis (Zelazowski et al., 1997; Owyang et al., 2001).

Regulation and role of NF-jB1 in Myc-induced lymphoma U Keller et al

6232

Furthermore, NF-kB2 (p100/p52) is required for normal splenic microarchitecture and B cell-mediated immune responses (Caamano et al., 1998). All Rel/NF-kB proteins contain an N-terminal B300-amino-acid Rel homology domain and bind to DNA as either homo- or heterodimers. Their activity is held in check by their interactions with IkB family proteins in the cytosol, which, in turn, are inactivated by their phosphorylations by IkB kinases, which target these proteins for destruction by the proteasome (reviewed in Ghosh and Karin, 2002; Hayden and Ghosh, 2004; Perkins, 2004). Several observations indicate a complex level of interplay between Myc and NF-kB. First, Rel/NF-kB upregulates the expression of c-Myc following treatment with anti-sIgM (Wu et al., 1996), and in Raji Burkitt lymphoma cells NF-kB regulates the expression of the translocated MYC gene, by binding to the Ig heavy chain enhancer (Kanda et al., 2000). Further, in primary B cells, NF-kB1 and c-Rel are required for mitogens to induce c-Myc and cell growth (Grumont et al., 2002). Conversely, Myc sensitizes cells to tumor necrosis factor-a (TNF-a)-mediated cell death by impairing Rel/NF-kB transactivation functions (Klefstrom et al., 1997; You et al., 2002). Given these connections we evaluated the regulation and potential role of NF-kB1 in Myc-mediated lymphomagenesis. Em-Myc transgenic B cells expressed reduced levels of Nfkb1 and NF-kB DNA binding activity, and Nfkb1 was suppressed in EmMyc lymphomas, and in human Burkitt lymphoma. Nonetheless, Nfkb1 loss has essentially no effect on Myc-induced lymphomagenesis.

Results Expression is suppressed in Em-Myc transgenic B cells Given the obvious interplay of Myc and Rel/NF-kB factors in mediating B cell proliferation and survival (Wu et al., 1996; Kanda et al., 2000; Grumont et al., 2002), we initially evaluated the expression of genes of the Rel/NF-kB pathway in Em-Myc transgenic versus normal B cells by expression profiling (Figure 1) and real-time RT–PCR of RNA isolated from B220 þ splenic B cells of precancerous 4-week-old Em-Myc transgenic mice (n ¼ 3) and their wild-type (n ¼ 3) littermates. Strikingly, there was a uniform downregulation of all mRNAs encoding rel/nfkb genes, nfkbia and nfkbib (encoding IkBa and IkBb) and, their upstream regulators, in Em-Myc transgenic B cells compared to normal B cells (Figures 1 and 2). Furthermore, the effects of the Myc transgene on the NF-kB network were manifested in both bone marrow (BM)- and spleen-derived B cells, and were evident in both immature (sIgM-) and mature (sIgM þ ) B cells (Figure 2 and data not shown). By contrast, consensus targets activated by Myc such as ornithine decarboxylase (Odc; BelloFernandez et al., 1993) and Rcl (Lewis et al., 1997) were highly elevated in Em-Myc B cells (Figures 1 and 2). Therefore, Myc selectively suppresses the expression of Oncogene

nearly all components of the NF-kB network in splenic B cells. Gene targeting has established that loss of Nfkb1, c-Rel or Nfkb2 all affect B cell homeostasis and/or functions (Sha et al., 1995; Zelazowski et al., 1997; Caamano et al., 1998; Gerondakis et al., 1999; Owyang et al., 2001). However, since the phenotypes manifested by Nfkb1 loss were most consistent with those of cells overexpressing c-Myc, with augmented B cell proliferation and higher rates of cell death (Grumont et al., 1998), we characterized the regulation and role of NFkB1 in Myc-induced responses in B cells in detail. First, to determine if reductions in nfkb1 transcripts in EmMyc transgenic B cells also led to appreciable changes in the levels of the protein, Western blots of NF-kB1 were performed on BM and splenic B220 þ B cells of precancerous Em-Myc transgenic mice and their wildtype littermates. p50 expression was dramatically reduced in BM-derived B cells from Em-Myc transgenic mice and reduced levels of p50 were also obvious in splenic B cells, while levels of p105 protein were reduced in splenic Em-Myc B cells (Figure 3a). The functions of Rel/NF-kB family proteins are largely regulated by their interactions with IkBa or IkBb, which sequester Rel and NF-kB in the cytoplasm (Hayden and Ghosh, 2004). We therefore also evaluated the subcellular localization of NF-kB1 by immunofluorescence of splenic and BM B220 þ B cells (Figure 3b), and the expression of IkBa or IkBb proteins by immunoblotting (Figure 3c). Consistent with immunoblotting analyses, the levels of p105/ p50 NF-kB1 protein were especially reduced in Em-Myc transgenic B cells from BM and reductions in NF-kB1 were also clearly evident in splenic Em-Myc B-cells (Figure 3b). There were, however, no obvious effects of the Myc transgene upon the overall ratio of cytosolic to nuclear NF-kB1 (Figure 3b). In accord with these findings, and despite changes in the levels of their transcripts (Figures 1 and 2), there were no significant changes in levels of IkBa or IkBb proteins in Em-Myc transgenic versus normal B cells (Figure 3c). Collectively, these findings suggest that Myc overexpression in B cells reduces Nfkb1 expression principally at the level of transcription. NF-kB DNA-binding activity is impaired in Em-Myc transgenic B cells The reductions in Nfkb1 expression in Em-Myc transgenic B cells suggested that this might also impair overall NF-kB DNA binding activity. We therefore compared the DNA-binding activity of Rel/NF-kB in Em-Myc versus normal B cells. NF-kB/Rel dimers are differentially expressed and activated in discrete B cell subsets during B cell development, where p50/p65 heterodimers are the predominant nuclear Rel/NF-kB activity in Pro- and Pre-B cells and where p50/c-Rel and to a lesser extent p50/p65 activity govern NF-kB activity in sIgM þ sIgD B cells (Liou et al., 1994; Lee et al., 1995; Grumont et al., 1998; and data not shown). We therefore evaluated NF-kB activity in precancerous Em-Myc B220 þ B cells sorted for expression of sIgM.


6233

Figure 1 Myc targets the NF-kB network in B cells. Hierarchical clustering of Rel/Nfkb family members, and of upstream kinases that regulate the NF-kB pathway, was performed using RNA prepared from B220 þ splenic B cells from three weanling-aged wild type (wt) and three Em-Myc transgenic (Em-Myc) mice. Probe set signals were normalized to the mean across mice, and values of each individual case are represented by a color, with green corresponding to expression below, and red corresponding to expression above the mean. As a control for a target gene induced by Myc the probe sets for Odc (Bello-Fernandez et al., 1993) are also shown

Notably, electromobility shift assays of nuclear extracts from both sIgM and sIgM þ B cells showed reduced levels of NF-kB DNA-binding activity in Em-Myc B cells compared to that present in normal B cells (Figure 4a and b). The most dramatic differences were manifest in splenic Em-Myc sIgM B cells, where p50/p65 NF-kB dimers were reduced to less than 10% of their activity seen in wild-type B cells (Figure 4a, compare lanes 10–12 to lanes 7–9). Accordingly, analysis of cytoplasmic and nuclear extracts showed that p50 protein was reduced in these Em-Myc B cells (Figure 4c). By contrast, the effects of Myc were more subtle in splenic sIgM þ Em-Myc B cells (lane 10 versus wt sIgM þ B cells, lane 7, Figure 4b), indicating that, as supported by immunoblot and immunofluorescence analyses of total p50 expression (Figure 3), there are cell context specific effects of Myc

on the activity and expression of NF-kB1 in B cells. Nonetheless, the data support the concept that Myc overexpression compromises the function of NF-kB family members throughout B cell development and particularly in immature B cells, and that this is specifically associated with the suppression of nuclear NF-kB1 p50 protein. Nfkb1 expression is suppressed in Em-Myc and human Burkitt lymphomas Since Nfkb1 expression was reduced in precancerous EmMyc B cells, we reasoned that Nfkb1 expression would also be reduced, or absent in Em-Myc lymphomas, and in human Burkitt lymphoma bearing MYC/Ig translocations. Interestingly, Nfkb1 RNA (Figure 5a) and Oncogene


6234

nfkb1

3

nfkb2

5

relA

3

4 2

relB

4 3

2

3

2 2

1

1 1

1 0

0

0

ikb

c-rel

6

1.5

0

ikk

ikb 3

1.5

1.0

2

1.0

0.5

1

0.5

0.0

0

0.0

5

Relative expression to ubiquitin

4 3 2 1 0

ikk

ikk 1.5

4

bcl-3

6

nik

4

5 3

3

4

1.0

3

2

2

2

0.5

1

1 0.0

1 0

0

0

BM odc 3

Spleen

BM

Spleen

rcl 10 8

2

wild type

6

Eµ-Myc

4

1

2 0

0

BM

Spleen

BM

Spleen

Figure 2 Real-time SYBR-green PCR analyses of mRNAs encoding components of the NF-kB network in B220 þ bone marrow (BM) and spleen B cells from 4-week-old wild type (wt) and Em-Myc transgenic littermates. The relative values were determined by comparing the expression of the indicated transcripts to that of ubiquitin (ub), which is not regulated by Myc. The bars represent the mean7s.e.m. Note: *indicates Po0.05. As a control for target genes induced by Myc the expression of Odc and Rcl (Lewis et al., 1997) was also analysed

protein (Figure 5b) levels were reduced in the majority of Em-Myc lymphomas when compared to their levels expressed in normal splenic B220 þ B cells. Furthermore, Nfkb1 expression was suppressed in all human Burkitt lymphoma samples tested as compared to CD19 þ B cells from healthy donors (Figure 5c). Thus, Nfkb1 expression is suppressed in Myc-driven lymphomas of both mice and man. Oncogene

Loss of Nfkb1 does not affect lymphoma development but augments tumor load in Em-Myc transgenic mice Nfkb1 loss alone accelerates rates of B cell proliferation (Grumont et al., 1998; Supplementary Figure 1). The reduced expression of NF-kB1 in Em-Myc transgenic B cells and lymphomas then suggested that total loss of Nfkb1 might accelerate lymphoma development in


6235

Figure 3 NF-kB1 p105 and p50 levels are reduced in B cells from precancerous Em-Myc transgenic mice. (a) Expression of NF-kB1 p105 and p50 was assessed by immunoblotting of whole-cell extracts of B220 þ B cells from bone marrow (BM) and spleen of 4-weekold wild-type (wt) and Em-Myc mice. (b) Total and nuclear NF-kB1 levels are reduced in Em-Myc B cells compared to wild-type B cells. Immunofluorescence analysis of p50/p105 NF-kB1 protein expressed in the bone marrow (BM) and spleen (Spl) B220 þ B cells from precancerous Em-Myc mice and their wild-type (wt) littermates is shown. Nuclei were stained with 40 ,60 -diamidino-2-phenylindole hydrochloride (DAPI). (c) B220 þ B cells from bone marrow (BM) and spleen of 4-week-old wild type (wt) and Em-Myc mice were subjected to immunoblot analysis using the indicated antibodies

Em-Myc transgenic mice as, for example, loss of the cyclin dependent kinase inhibitor p27Kip1 accelerates lymphoma development (Martins and Berns, 2002). To address this issue, Nfkb1 nullizygous mice were mated with Em-Myc mice, and F1 offspring were bred to obtain Nfkb1 þ / þ , Nfkb1 þ / and Nfkb1/ Em-Myc transgenic mice. The precancerous phase of Em-Myc transgenics is characterized by high rates of cell proliferation and an increased apoptotic index, and by lymphocytosis and splenomegaly. To initially assess the effects of loss of Nfkb1 upon Myc-driven proliferation and apoptosis, B cells were cultured from the BM of precancerous Nfkb1 þ / þ and Nfkb1/ Em-Myc transgenic mice. No differences were observed in the differentiation of Nfkb1/ Em-Myc cells, which like wild-type Em-Myc B cells were a mixture of pro-B and pre-B cells (data not shown). By contrast, Nfkb1/ Em-Myc cells displayed an obvious growth advantage during ex vivo culture (Figure 6a), suggesting that Nfkb1 loss cooperates with Myc to drive cell proliferation, but there were no effects of Nfkb1 loss on the high apoptotic index of Em-Myc transgenic B cells (Figure 6b). To determine whether similar effects were also evident in vivo, 4-week old Nfkb1 þ / þ and Nfkb1/ Em-Myc transgenic littermates were injected with BrdU and after 12 h B220 þ sIgM þ and sIgM- B cells from BM and spleen were assessed for their S phase fraction, and their apoptotic indices were

determined by staining these cells with Annexin V-FITC and propidium iodide. Notably, splenic sIgM þ Nfkb1/ Em-Myc B cells showed significant increases in BrdU þ cells, and more modest increases in proliferation were also evident in BM sIgM þ and splenic sIgM Nfkb1/ Em-Myc B cells (Figure 6c). By contrast, there were no significant differences in the apoptotic indices of BM or splenic Nfkb1/ Em-Myc B cells (Figure 6d). Although a failure to observe higher rates of apoptosis of Nfkb1/ Em-Myc B cells could reflect their effective clearance by phagocytes in vivo, the apoptotic index of Nfkb1/ and Nfkb1 þ / þ Em-Myc transgenic B cells was comparable in ex vivo culture (Figure 6b). Therefore, Nfkb1 loss selectively augments Myc’s proliferative response in B cells. Em-Myc transgenics typically succumb to aggressive lymphoma within 3–6 months of age. Hallmarks of disease include palpable lymph nodes and a marked lymphocytosis and splenomegaly. Assessment of tumor load at diagnosis established that Nfkb1/ Em-Myc transgenics had moderate increases in their spleen weights and aggravated leukemia compared to their Nfkb1 þ / and Nfkb1 þ / þ transgenic littermates (Figure 7a). However, despite the higher proliferative rates of Nfkb1/ Em-Myc B cells, the onset of lymphoma in Em-Myc transgenic mice of all Nfkb1 genotypes was virtually identical (median survival of 83, Oncogene


6236

a Ab:

no p50 p65 no p50 p65 no p50 p65 no p50 p65

sIgM -

1

2

3

4 5

6

7

Eµ-Myc

wt

9 10 11 12

Eµ-Myc wt Spleen

BM

b

8

Ab:

no p50 c-Rel no p50 c-Rel no p50 c-Rel no p50 c-Rel

sIgM +

1

2

wt

c

3

4 5

6

7

Eµ-Myc BM CE

8

9 10 11 12

wt

Eµ-Myc Spleen

NE

p50

BM sIgM-

p50

Spleen sIgM+

Actin w

t

-M Eµ

yc

w

t

-M Eµ

yc

Figure 4 Em-Myc B cells have reduced NF-kB DNA binding activity. Nuclear extracts from MACS-/FACS-sorted wild type (wt) and Em-Myc transgenic (a) B220 þ sIgM B cells and (b) B220 þ sIgM þ B cells were subjected to EMSA using a consensus NF-kB binding element. Composition of NF-kB dimers was determined by supershift analysis using the indicated antibodies (Ab). BM denotes bone marrow. (c) Immunoblot analysis of p50 NF-kB1 levels in cytosolic extracts (CE) and nuclear extracts (NE) of precancerous B220 þ B cells of the indicated sIgM phenotype is shown. BM denotes bone marrow

Figure 5 Analysis of Nfkb1 expression in Em-Myc lymphomas and human Burkitt lymphomas. (a) SYBR-green real-time PCR analysis of nfkb1 mRNA expression in 20 Em-Myc lymphomas was compared to levels of nfkb1 transcripts from pooled B220 þ splenic B cells from wild-type (B220 þ ) mice. The relative values were determined by comparing the expression of nfkb1 mRNA to the expression of ubiquitin (ub). (b) Immunoblot analysis of p105 and p50 NF-kB1 in six Em-Myc lymphomas (lanes 5–11) was compared to their levels in wild type (wt, lane 1) and precancerous Em-Myc B220 þ B cells. Nfkb1-deficient B cells were included as a control. (c) SYBR-green real-time PCR analysis of Nfkb1 mRNA expression in 17 human Burkitt lymphomas. CD19 þ B cells from healthy donors served as a control. The relative values were determined by comparing the expression of Nfkb1 mRNA to the expression of Ubiquitin (Ub)

Discussion 81 and 93 days for Nfkb1 þ / þ -, Nfkb1 þ /-, and Nfkb1/Em-Myc transgenics, respectively, Figure 7b). Furthermore, the phenotype of the lymphomas that arose in Nfkb1/-Em-Myc transgenics was identical to those that arose in wild-type transgenic littermates (pre-B and mature B cell lymphomas, data not shown). Thus, although loss of Nfkb1 increases proliferation of Myctransgenic B cells in vitro and in vivo, it does not affect Myc-induced lymphomagenesis. Oncogene

Myc overexpression in the B cells of Em-Myc transgenic mice leads to increased numbers of proliferating B cells in both BM and spleen (Adams et al., 1985; Baudino et al., 2003). Given that loss of Nfkb1 alone also leads to increased rates of B cell proliferation (Grumont et al., 1998), and that Myc overexpression in this compartment leads to reductions in NF-kB1 levels and to concomitant decreases in NF-kB DNA-binding activity, a response


6237

Figure 6 Nfkb1 loss augments the proliferation of Em-Myc B cells. (a) Bone marrow from 4-week-old precancerous Nfkb1/;Em-Myc and Nfkb1 þ / þ ;Em-Myc mice was cultured on S17 stromal cells for 10 days, and then transferred into IL-7 conditioned medium. Medium was changed every 2 days and the cells counted. The values shown represent the mean7s.e.m. of three independent experiments. (b) FACS analysis of ex vivo cultured Em-Myc B cells of different Nfkb1 genotypes for spontaneous apoptosis measured by Annexin V and PI staining. A representative experiment is shown. (c) Nfkb1 þ / þ and Nfkb1/ Em-Myc transgenic mice (4 weeks old) received one intraperitoneal injection of BrdU. After 12 h single-cell suspensions were prepared from bone marrow and spleen and analysed by FACS for BrdU incorporation. Bars show B220 þ B cells of the indicated sIgM phenotype in S-phase and represent the mean7s.e.m. from three independent experiments. (d) Apoptotic index of B cells from the indicated Em-Myc transgenic mice was analysed using an antibody dependent fluorescence assay. Annexin V þ B220 þ cells of sIgM- or sIgM þ phenotype are shown. The bars represent the mean7s.e.m. from three independent experiments

that was also manifested in the lymphomas that arise in these mice, a reasonable prediction was that total loss of Nfkb1 would lead to higher proliferative rates and accelerated rates of lymphoma development in Em-Myc mice. While Nfkb1 loss did indeed augment the proliferation of Em-Myc B cells both ex vivo and in vivo, lymphoma development was little affected by the loss of Nfkb1, other than modest increases in the overall tumor load in Nfkb1/ Em-Myc transgenics. Thus, at least in this cell context, Nfkb1 does not play a ratelimiting role in Myc-induced tumorigenesis.

Imbalances in Myc’s proliferative and apoptotic responses modify time to tumor onset in Em-Myc transgenic mice in rather profound ways. For example, loss of Arf or p53 disables Myc’s apoptotic response to dramatically accelerate lymphoma development (Eischen et al., 1999; Schmitt et al., 1999), whereas Mdm2 haploinsufficiency augments p53 dependent apoptosis and impairs tumor development (Alt et al., 2003). Furthermore, loss of p27Kip1 augments Myc’s proliferative response in Em-Myc B cells to accelerate lymphomagenesis (Martins and Berns, 2002), whereas Oncogene


6238

a

WBC

x 1000 / µl

200 150 100 50 0

+/Nfkb1

+/+

-/-

Spleen 600

mg

400 200 0 +/+

+/-

-/-

Nfkb1

b 100

% Survival

75 Nfkb1 +/+ Nfkb1 +/50

Nfkb1 -/-

25

0 0

50

100

150

200

250

Days Figure 7 Loss of Nfkb1 does not affect Myc-induced lymphomagenesis. (a) Tumor load at diagnosis. Peripheral blood and spleens of sick mice were harvested and their spleen weights and white blood counts (WBC) were determined. Bars show mean7s.e.m. (n ¼ 7 for Nfkb1/;Em-Myc and Nfkb1 þ / þ ;Em-Myc mice and n ¼ 20 for Nfkb1 þ /;Em-Myc mice). (b) Survival of Em-Myc transgenic mice of the indicated Nfkb1 genotypes was determined. The numbers of Em-Myc mice with the respective genotypes was: Nfkb1 þ / þ , n ¼ 38; Nfkb1 þ /, n ¼ 91; Nfkb1/, n ¼ 31. There were no statistically significant differences in the survival of these cohorts

E2f1 haploinsufficiency disables Myc’s ability to downregulate p27Kip1 and thus impairs disease (Baudino et al., 2003). It was therefore surprising that loss of Nfkb1, which clearly augments the proliferative responses of Em-Myc B cells, failed to appreciably affect lymphoma development. This perhaps reflects redundancy coming from other members of the Rel/NF-kB network, as c-Rel and NF-kB2 certainly play important roles in B cell development and survival (Zelazowski et al., 1997; Caamano et al., 1998; Owyang et al., 2001). Oncogene

The modest effects of Nfkb1 loss on Myc-induced lymphomagenesis could also reflect its failure to affect Myc’s apoptotic programs in Em-Myc B cells, which could effectively cancel the proliferative advantage of Nfkb1-deficient B cells. This was also rather surprising, as recent studies of others have suggested links between NF-kB and c-Myc- and E2f1-induced apoptosis of immortalized fibroblasts, where Myc induces E2f-1, which in turn binds to p65 and thus prevents the formation of the p50/p65 heterodimer that prevents cell death (Tanaka et al., 2002). Furthermore, Arf, which is induced in Em-Myc B cells (Zindy et al., 2003), inhibits the antiapoptotic functions of RelA by inducing its association with the histone deacetylase HDAC1, thus impairing its transactivation functions (Rocha et al., 2003). The increased rates of proliferation of Em-Myc Nfkb1-deficient B cells is perhaps consistent with increased E2f-1 activity in these B cells, yet if either of these models were operational one would have expected that suppressed levels of NF-kB in these cells, especially in the context of the Nfkb1 deficiency, would drastically increase apoptosis, and this is not the case. Thus, we propose that the apoptotic response in these cells is sufficient to cancel the proliferative advantage of Em-Myc Nfkb1/ B cells, which is really only manifested by modest increases in tumor load and is not rate limiting for lymphoma development. Precisely how Myc regulates NF-kB1 expression, and nearly all other Rel/NF-kB family members in B cells (Figures 1 and 2) is presently not clear, but at least for Nfkb1 this involves reductions in its transcripts. These findings suggest a potentially important regulatory feedback loop that controls NF-kB and Myc activity in a cell, whereby the induction of c-myc transcription by NF-kB, a well-known mechanism operational in B cells (Lee et al., 1995; Grumont et al., 1998, 2002), ultimately leads to increases in Myc that, in turn, dampen the activity of the NF-kB network. Indeed, such a model would allow the cell to fine-tune its Myc levels to provide balanced growth (cell mass) and proliferative signals. This feedback control is obviously lost in tumors where Myc is activated, and in such cancers reductions in NF-kB activity could sensitize these cells to specific chemotherapeutic regiments. Indeed, given the rather complex effects of the NF-kB network upon drug sensitivity and tumor maintenance (Baldwin, 2001; Orlowski and Baldwin, 2002), these remain important issues to test in the lymphomas arising in Nfkb1/EmMyc transgenic mice.

Material and methods Human tumor material With institutional review board approval and after informed consent, 17 tumor samples from Burkitt lymphoma patients were banked. As a control, pooled peripheral blood mononuclear cells from healthy donors were enriched using CD19MicroBeads, according to the manufacturer’s instruction (Miltenyi Biotech, Bergisch-Gladbach, Germany).


6239 Breeding of mice and tumor surveillance Nfkb1 null mice on a C57BL/6;129 mixed background (Sha et al., 1995), obtained from Jackson Laboratories (Bar Harbor, ME, USA), were interbred with Em-Myc transgenic mice (C57BL/6 background; Adams et al., 1985). F1 offspring were bred to obtain Nfkb1 þ / þ , Nfkb1 þ / and Nfkb1/ Em-Myc transgenic littermates. Animals were observed for signs of morbidity and tumor development. Cell culture Primary BM-derived pro-B/pre-B cell cultures were generated from 4 to 6 week old mice as described previously (Eischen et al., 1999). FACS analysis and magnetic-activated cell sorting (MACS) of B cells To assess potential differences in the maturation status of B cells in wild type versus Em-Myc transgenic mice, FACS analysis of BM and splenic B220 þ B cells from 4-week-old mice was performed for sIgM. Only moderate differences in sIgM þ and sIgM populations were seen in these analyses (wt BM: 76.775.3% B220 þ sIgM cells; Em-Myc BM: 87.076.0% B220 þ sIgM cells; wt spleen: 40.776.4% B220 þ sIgM- cells; Em-Myc spleen: 34.076.0% B220 þ sIgM cells, n ¼ 3 for each group). To obtain BM and spleen B cells, cell suspensions were incubated with B220 MicroBeads and enriched by magnetic cell sorting (MACS), according to the manufacturer’s instruction (Miltenyi Biotech). Rates of proliferation of B220 þ sIgM þ and B220 þ sIgM cells were then determined using a Flow Kit as described by the manufacturer (BD Biosciences Pharmingen, San Diego, CA, USA). Animals were injected intraperitoneally with 100 ml of 10 mg/ml BrdU in sterile PBS. Animals were killed 12 h following injection and BM and spleen were harvested. A red cell lysis was performed using ammonium chloride/potassium bicarbonate solution. Cells were then resuspended, incubated with antibody against B220 and sIgM (BD Biosciences Pharmingen, San Diego, CA, USA), washed and collected by centrifugation. One million cells were further processed and stained with antiBrdU-FITC antibody, and 5 105 cells were stained with propidium iodide and Annexin-V FITC antibody (Annexin-VFluor Kit, Roche Applied Sciences, Indianapolis, IN, USA). Following incubation, cells were washed and resuspended in PBS and were then analysed by FACS. The percentage of Annexin V-positive cells includes PI positive (dead) and negative (apoptotic) cells.

control sample set as 1. Sequences for primers are available upon request. Immunoblotting analyses Extracts from MACS-sorted B cells, lymphomas from Em-Myc mice, and human Burkitt lymphoma samples were prepared as previously described (Eischen et al., 1999). Protein (20 or 50 mg per lane) was electrophoretically separated on SDS–PAGE gels, transferred to membranes (Protran, Schleicher and Schuell, Dassel, Germany) and blotted with antibodies specific for c-Myc (Santa Cruz Inc., Santa Cruz, CA, USA), NF-kB1 (Santa Cruz) and b-Actin (Sigma Chemicals, St Louis, MO, USA). Immunofluorescence analyses Cytospins of 5 104 B220 þ B cells were derived from wild type or Em-Myc mice. Slides were fixed with 1 : 1 methanol–acetone for 30 min at 201C and air-dried. After blocking with 10% FBS/PBS, cells were incubated for 1 h with a polyclonal antibody for NF-kB1 (1 : 1000 dilution in 1% FBS/PBS; Santa Cruz, Inc.). Following 10 washes with PBS, primary antibody binding was visualized using an anti-rabbit Cy3-conjugate (Jackson Immunoresearch, MA, USA). DAPI (Molecular Probes, Eugene, OR, USA) was used to visualize nuclei. Mounted slides were analysed by confocal microscopy. Electrophoretic mobility shift assays (EMSA)

RNA preparation and analyses

MACS-sorted B220 þ B cells were FACS-sorted to obtain sIgM and sIgM þ B cells (FACSVantage, Becton Dickinson, San Jose, CA, USA; anti-sIgM antibody, BD Biosciences Pharmingen, San Diego, CA, USA). Nuclear and cytosolic extracts were isolated as described previously (Sun et al., 2000). Protein concentration was determined using the BioRad Protein Assay kit (Bio-Rad). In total, 3 mg of nuclear protein was subjected to bandshift analysis. EMSAs were performed with the following reagents: double-stranded 32 P-labeled NF-kB oligonucleotides: 50 -AGTTGAGGG GACTTTCCCAGC-30 (Promega, Madison, WI, USA); T4 Polynucleotide Kinase (Roche Applied Sciences, Indianapolis, IN, USA); NF-kB 5 binding buffer: 25 mM Tris-HCl pH 7.5, 5 mM EDTA, 250 mg/ml poly (dI:dC), 5 mM dithiothreitol and 1 mg/ml BSA. 106 cpm of labeled probe was used in each reaction, and bandshifts were resolved on 5% native polyacrylamide gels in 0.5 TBE running buffer. For supershift analysis, the following antibodies were used: p50, c-Rel and p65 (Santa Cruz). Dried gels were subjected to autoradiography.

RNA was prepared from MACS-sorted B cells using the RNeasy kit (Qiagen, Valencia, CA, USA). For Affymetrix analyses, cRNA was synthesized using the One-Cycle Target Labeling and Control Reagent package (Affymetrix Inc., Santa Clara, CA, USA) and the reaction was probed to the 430A mouse Affymetrix chip. The scanned data output was imported into the Spotfire software (Spotfire Inc., Somerville, MA, USA). Following normalization, selected probe sets for the genes indicated in Figure 1 were clustered using the Hierarchical Clustering function of Spotfire. For real-time PCR, cDNA was prepared from 1 mg RNA using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA). Real-time PCR was performed using an iCycler machine (Bio-Rad) and the iTaq SYBR green kit (Bio-Rad). Data analyses were performed by comparing Ct values with a

Acknowledgements We thank Sara Norton, Chunying Yang and Elsie White for expert technical assistance. We also thank Mihaela Onciu and John Sandlund for providing samples from Burkitt lymphoma patients, the Animal Resource Center, the Hartwell Center and the FACS Core Facility of St Jude Children’s Research Hospital. This work was supported by NIH Grant CA76379 (JLC), the Cancer Center (CORE) support Grant CA21765 and by the American Lebanese Syrian Associated Charities (ALSAC) of St Jude Children’s Research Hospital. UK was supported by the Deutsche Forschungsgemeinschaft (grant KE222/5-1). JBO was supported by NRSA Grant F32 CA099478, and JAN is the recipient of the George J Mitchell endowed fellowship from ALSAC. Oncogene


6240 References Adams JM, Harris AW, Pinkert CA, Corcoran LM, Alexander WS, Cory S, Palmiter RD and Brinster RL. (1985). Nature, 318, 533–538. Alt JR, Greiner TC, Cleveland JL and Eischen CM. (2003). EMBO J., 22, 1442–1450. Askew DS, Ashmun RA, Simmons BC and Cleveland JL. (1991). Oncogene, 6, 1915–1922. Baldwin AS. (2001). J. Clin. Invest., 107, 241–246. Baudino TA, Maclean KH, Brennan J, Parganas E, Yang C, Aslanian A, Lees JA, Sherr CJ, Roussel MF and Cleveland JL. (2003). Mol. Cell, 11, 905–914. Bello-Fernandez C, Packham G and Cleveland JL. (1993). Proc. Natl. Acad. Sci. USA, 90, 7804–7808. Bouchard C, Staller P and Eilers M. (1998). Trends Cell Biol., 8, 202–206. Caamano JH, Rizzo CA, Durham SK, Barton DS, RaventosSuarez C, Snapper CM and Bravo R. (1998). J. Exp. Med., 187, 185–196. Cavallieri F and Goldfarb M. (1987). Mol. Cell. Biol., 7, 3554–3560. Coppola JA and Cole MD. (1986). Nature, 320, 760–763. de Alboran IM, O’Hagan RC, Gartner F, Malynn B, Davidson L, Rickert R, Rajewsky K, DePinho RA and Alt FW. (2001). Immunity, 14, 45–55. Egle A, Harris AW, Bouillet P and Cory S. (2004). Proc. Natl. Acad. Sci. USA, 101, 6164–6169. Eischen CM, Weber JD, Roussel MF, Sherr CJ and Cleveland JL. (1999). Genes Dev., 13, 2658–2669. Eischen CM, Woo D, Roussel MF and Cleveland JL. (2001). Mol. Cell. Biol., 21, 5063–5070. Evan GI and Vousden KH. (2001). Nature, 411, 342–348. Evan GI, Wyllie AH, Gilbert CS, Littlewood TD, Land H, Brooks M, Waters CM, Penn LZ and Hancock DC. (1992). Cell, 69, 119–128. Gerondakis S, Grossmann M, Nakamura Y, Pohl T and Grumont R. (1999). Oncogene, 18, 6888–6895. Ghosh S and Karin M. (2002). Cell, 109, S81–S96. Grandori C, Cowley SM, James LP and Eisenman RN. (2000). Annu. Rev. Cell. Dev. Biol., 16, 653–699. Grumont RJ, Rourke IJ, O’Reilly LA, Strasser A, Miyake K, Sha W and Gerondakis S. (1998). J. Exp. Med., 187, 663–674. Grumont RJ, Strasser A and Gerondakis S. (2002). Mol. Cell, 10, 1283–1294. Hanahan D and Weinberg RA. (2000). Cell, 100, 57–70. Hayden MS and Ghosh S. (2004). Genes Dev., 18, 2195–2224. Jeffers JR, Parganas E, Lee Y, Yang C, Wang J, Brennan J, MacLean KH, Han J, Chittenden T, Ihle JN, McKinnon PJ, Cleveland JL and Zambetti GP. (2003). Cancer Cell, 4, 321–328.

Kanda K, Hu H-M, Zhang L, Grandchamps J and Boxer LM. (2000). J. Biol. Chem., 275, 32338–32346. Klefstrom J, Arighi E, Littlewood T, Jaattela M, Saksela E, Evan GI and Alitalo K. (1997). EMBO J., 16, 7382–7392. Lee H, Arsura M, Wu M, Duyao M, Buckler AJ and Sonenshein GE. (1995). J. Exp. Med., 181, 1169–1177. Lewis BC, Shim H, Li Q, Wu CS, Lee LA, Maity A and Dang CV. (1997). Mol. Cell. Biol., 17, 4967–4978. Liou HC, Sha WC, Scott ML and Baltimore D. (1994). Mol. Cell. Biol., 14, 5349–5359. Maclean KH, Keller UB, Rodriguez-Galindo C, Nilsson JA and Cleveland JL. (2003). Mol. Cell. Biol., 23, 7256–7270. Martins CP and Berns A. (2002). EMBO J., 21, 3739–3748. Nilsson JA and Cleveland JL. (2003). Oncogene, 22, 9007–9021. Orlowski RZ and Baldwin AS. (2002). Trends Mol Med., 8, 385–389. Owyang AM, Tumang JR, Schram BR, Hsia CY, Behrens TW, Rothstein TL and Liou HC. (2001). J. Immunol., 167, 4948–4956. Perkins ND. (2004). Trends Cell. Biol., 14, 64–69. Rocha S, Campbell KJ and Perkins ND. (2003). Mol. Cell, 12, 15–25. Schmitt CA, McCurrach ME, de Stanchina E, WallaceBrodeur RR and Lowe SW. (1999). Genes Dev., 13, 2670–2677. Sha WC, Liou HC, Tuomanen EI and Baltimore D. (1995). Cell, 80, 321–330. Strasser A, Harris AW, Bath ML and Cory S. (1990). Nature, 348, 331–333. Sun Z, Arendt CW, Ellmeier W, Schaeffer EM, Sunshine MJ, Gandhi L, Annes J, Petrzilka D, Kupfer A, Schwartzberg PL and Littman DR. (2000). Nature, 404, 402–407. Tanaka H, Matsumura I, Ezoe S, Satoh Y, Sakamaki T, Albanese C, Machii T, Pestell RG and Kanakura Y. (2002). Mol. Cell, 9, 1017–1029. Trumpp A, Refaeli Y, Oskarsson T, Gasser S, Murphy M, Martin GR and Bishop JM. (2001). Nature, 414, 768–773. Wu M, Lee H, Bellas RE, Schauer SL, Arsura M, Katz D, FitzGerald MJ, Rothstein TL, Sherr DH and Sonenshein GE. (1996). EMBO J., 15, 4682–4690. You Z, Madrid LV, Saims D, Sedivy J and Wang CY. (2002). J. Biol. Chem., 277, 36671–36677. Zelazowski P, Carrasco D, Rosas FR, Moorman MA, Bravo R and Snapper CM. (1997). J. Immunol., 159, 3133–3139. Zindy F, Eischen CM, Randle DH, Kamijo T, Cleveland JL, Sherr CJ and Roussel MF. (1998). Genes Dev., 12, 2424–2433. Zindy F, Williams RT, Baudino TA, Rehg JE, Skapek SX, Cleveland JL, Roussel MF and Sherr CJ. (2003). Proc. Natl. Acad. Sci. USA, 100, 15930–15935.

Supplementary Information accompanies the paper on Oncogene website (http://www.nature.com/onc)

Oncogene