Oncotarget

Research Papers:

The class-specific BCR tonic signal modulates lymphomagenesis in a c-myc deregulation transgenic model

PDF |  HTML  |  Supplementary Files  |  How to cite

Oncotarget. 2014; 5:8995-9006. https://doi.org/10.18632/oncotarget.2297

Metrics: PDF 2232 views  |   HTML 4045 views  |   ?  

Rada Amin, Abdelghafour Marfak, Céline Pangault, Christelle Oblet, Aurélie Chanut, Karin Tarte, Yves Denizot and Michel Cogné _

Abstract

Rada Amin1,2,3, Abdelghafour Marfak1 , Céline Pangault3, Christelle Oblet1,2, Aurélie Chanut1,2, Karin Tarte3, Yves Denizot1,2 and Michel Cogné1,2

1 Centre National de la Recherche Scientifique, Limoges, France 

2 Université de Limoges, Limoges, France 

3 INSERM UMR U917, Rennes, France

Correspondence:

Michel Cogné, email:

Keywords: BCR, c-myc, lymphoma, IgA, B cell

Received: June 09, 2014 Accepted: July 31, 2014 Published: July 31, 2014

Abstract

Deregulation of c-myc by translocation onto immunoglobulin (Ig) loci can promote B cell malignant proliferations with phenotypes as diverse as acute lymphoid leukemia, Burkitt lymphoma, diffuse large B cell lymphoma, myeloma... The B cell receptor (BCR) normally providing tonic signals for cell survival and mitogenic responses to antigens, can also contribute to lymphomagenesis upon sustained ligand binding or activating mutations. BCR signaling varies among cell compartments and BCR classes. For unknown reasons, some malignancies associate with expression of either IgM or class-switched Ig. We explored whether an IgA BCR, with strong tonic signaling, would affect lymphomagenesis in c-myc IgH 3’RR transgenic mice prone to lymphoproliferations. Breeding c-myc transgenics in a background where IgM expression was replaced with IgA delayed lymphomagenesis. By comparison to single c-myc transgenics, lymphomas from double mutant animals were more differentiated and less aggressive, with an altered transcriptional program. Larger tumor cells more often expressed CD43 and CD138, which culminated in a plasma cell phenotype in 10% of cases. BCR class-specific signals thus appear to modulate lymphomagenesis and may partly explain the observed association of specific Ig classes with human B cell malignancies of differential phenotype, progression and prognosis.


INTRODUCTION

In normal cells, c-myc oncogene expression is restricted to the early G1 phase of the cell cycle with a role in proliferation, differentiation, metabolism and apoptosis [1] . Deregulation of the c-myc gene is a constant feature of human Burkitt lymphoma (BL), with translocations linking c-myc to any of the immunoglobulin heavy or light chain (IgH, Igκ or Igλ) locus and with a phenotype that may vary from immature RAG expressing B cell lymphoma or leukemia to mature B cell lymphoma [2]. Translocation onto the IgH locus is also frequent in human myeloma and mouse plasmacytoma cases, suggesting that c-myc may participate to cell transformation at all stages of B cell differentiation. The frequency at which myc-driven malignancies occur clearly depends upon the cell microenvironment and the local production of cytokines. For example, inflammation-induced mouse plasmacytomas only develop in mice proficient for IL6 [3] and necessitates that these mice are kept in a conventional rather than specific pathogen free animal facility [4].

Beside oncogenes and the role of signals from the microenvironment, the phenotype of B cell malignancies also often involves signaling through the adaptive or innate immune receptors. The B cell receptor (BCR) is mandatory for survival of normal B cells by providing a ligand-independent tonic signal. It is also the main receptor controlling B cell activation during immune responses [5]. Expression of Ig of the various classes additionally has a clear influence on normal B cell fate [6-9].

In some cases of human lymphomas involving Epstein-Barr virus transformation, BCR-dependent activation or survival signals may be replaced by surrogate signals provided by the viral protein LMP2A [10]. Alternatively, in diffuse large B cell lymphomas (DLBCL), the activated B cell (ABC) phenotype features NF-κB activation due to somatic mutations at various levels of the BCR or TLR signaling cascades (Igα, Igβ, CARD11, Myd88...) and results in chronic BCR signaling in the absence of any BCR ligand [11, 12]. In other malignancies, ligand-dependent BCR signaling likely occurs. In follicular lymphoma, BCR molecules carrying high-mannose N-glycans may contribute to lymphomagenesis through their binding of C-type lectins in the tumor microenvironment [13]. In chronic lymphocytic leukemia (CLL), cell-autonomous signaling constitutively results from expression of self-binding BCRs [14]. Altogether, BCR signaling appears as strongly contributing to B cell transformation. As for untransformed B cells, the downstream effects of ligand-dependent or independent signals might be modulated according to the BCR class and to the stage at which oncogenic mutations immortalize a given cell. Interestingly, some B cell tumors overwhelmingly express IgM (as in CLL, follicular lymphoma, BL, lymphoplasmacytic lymphoma, ABC-type DLBCL, while others overwhelmingly express class-switched Ig (as in myeloma and GC-type DLBCL) [15]. It is thus questionable whether, similarly to their untransformed counterparts, malignant B cells could be drawn towards different phenotypes dependent about the Ig class produced.

We designed experiments in order to compare experimental B cell tumors induced by a similar oncogenic event, but in the context of either class-switched or IgM expression. We recently described a model of transgenic mice developing BL-like lymphomas with a mature B cell phenotype upon expression of a c-myc transgene. This transgene includes the four transcriptional enhancers (hs3a, hs1,2, hs3b and hs4) from the IgH locus 3′ regulatory region (3’RR), and is thus expressed at all stages of B cell differentiation, but with a higher activity at those stages undergoing terminal B cell differentiation and class switch recombination [16-21] . Transgenic animal carrying this c-myc3’RR transgene have been shown to develop BL in 75% of cases and diffuse anaplastic lymphomas in the remaining 25% [22]. Proliferations occurred with a still higher (100%) penetrance and were more heterogeneous in animal carrying both the c-myc3’RR transgene and p53 haploinsufficiency [23, 24]. Tumors from c-myc-3’RR mice were CD43 negative and had a mature B cell phenotype (B220+/IgM+/IgD+). In the present study, we introduced the c-myc3’RR transgene into a background carrying a homozygous mutation of the IgH locus (α1KI mice) [6-8]. This α1KI mutation corresponds to replacement of the Sµ region with a knock-in Cα IgH gene and replaces IgM/IgD expression by the premature expression of IgA. This forced expression of an IgA “class-switched” BCR has been shown to mediate constitutive pre-activation of B cells, even in the absence of any BCR ligation [6-8]. The present study was conducted by comparing single mutant (c-myc transgenics) with double mutant c-myc/α1KI animals. We thus monitored the occurrence of B cell malignancies upon the influence of both c-myc oncogene deregulation and class-switched-type constitutive BCR signaling and analyzed the phenotype of the observed malignancies.

RESULTS

Generation of double mutant α1KI c-myc3’RR mice

A colony of homozygous α1KI animals was generated, all bearing a single copy of the integrated c-myc cassette [22]. Single mutant α1KI homozygous mice were studied in parallel, as well as single c-myc transgenic animals.

B cell development and Ig secretion in young transgenic mice

We analyzed B cells in 6 weeks-old transgenic mice, before any manifestation of disease. Spleen and lymph nodes (LNs) from α1KI mice and α1KI c-myc3’RR mice were similar (not shown). Numbers of B cells did not differ between homozygous α1KI mice and the double transgenic α1KI c-myc3’RR, neither in spleen (6.7 ± 1.7% N=5 for α1KI versus 8.9 ± 2.1% N=5, i.e. 1.4 ± 0.4 106 cells vs 1.3 ± 0.19 106 cells), nor in LNs (5.1 ± 0.8%, n=3 vs 5.3 ± 1.2%, n=3, i.e. 0.62 ± 0.09 106 cells vs 0.44 ± 0.14 106 cells) (supplemental Fig 1A). By contrast, in vitro B cell proliferative responses to anti-CD40 plus IL4 were evaluated and proved higher in α1KI c-myc transgenic mice than in α1KI single mutant animals (supplemental Fig 1B).

Delayed lymphomas in α1KI c-myc3’RR mice vs c-myc3’RR single transgenics

Lymphoma development and mean lifespan were quoted for 61 double mutant α1KI c-myc3’RR transgenic mice, 34 single mutant α1KI homozygous mice and 22 single mutant c-myc3’RR animals. The overall tumor incidence in double mutant animals was 48% before 34 weeks of age. As expected by comparison to α1KI mice not susceptible to lymphoma, α1KI c-myc3’RR mice showed significantly increased mortality (Gehan Breslow Wilcoxon test, p <0.0001) that resulted from c-myc-driven tumor development, with the mean age of death at about 39 weeks (Fig 1A). This was significantly below the mortality observed in c-myc3’RR single transgenic mice (Gehan Breslow Wilcoxon test, p <0.001), among which 80% had developed tumors before 34 weeks of age (with mean age of death at 27 weeks).

At necropsy, the α1KI c-myc3’RR had enlarged spleens, LNs and eventually liver (Fig 1B). All mice with tumors showed leukemic peripheral blood involvement with large circulating nucleolated lymphoma cells (Fig 1C, up). Histological analysis of tumors showed a “starry-sky” pattern reminiscent of human BL (Fig 1C, down).

Generation of &#x3b1;1KI

Figure 1: Generation of α1KI c-myc3’ RR transgenic mice. (A) Survival curves of c-myc3’RR and α1KI c-myc3’RR mice. Twenty two c-myc3’RR mice (*), 61 α1KI c-myc3’RR transgenic mice () and 34 α1KI () mice were followed over a period of 60 weeks. (B) Enlarged organs from α1KI c-myc3’RR mice at the tumor development stage compared with organs from a control (*) α1KI c-myc3’RR at the pre-tumorous stage (Left: 2 representative tumorous spleens; right: 1 representative tumorous LN) (C) Histological features of tumors in double transgenic mice: (Up) Typical circulating tumor cells with a basophilic cytoplasm (original magnifications, x20). (Down) Sections of formalin-fixed, paraffin-embedded LNs were stained with hematoxylin-eosin.

Phenotype of B cell lymphomas in α1KI c-myc3’RR mice

FACS analyses of 29 tumors from α1KI c-myc3’RR transgenics revealed two kinds of tumors, either lymphocytic (type I) or plasmablastic (type II).

Lymphocytic (type I) tumors were the most frequent. They histologically resembled those tumors seen in c-myc single transgenics but could be further split in two subgroups according to CD43 expression: type Ia, IgA+/CD19+/B220+/CD43- (Fig 2, left, 9 cases) and type Ib, IgA+/CD19+/B220+/CD43+ (Fig 2, middle, 17 cases).

Plasmablastic (type II) tumors occurred less frequently (3 cases, i.e. roughly 10%). They strongly differed from those tumors reported in c-myc single transgenics and corresponded to malignant plasmablasts (with an IgA-/CD19-/B220-/CD43+/CD138+ phenotype) (Fig 2, right, 3 cases).

Analysis of lymphomas in &#x3b1;1KI

Figure 2: Analysis of lymphomas in α1KI c-myc3’ RR mice. Tumors from 29 mice were stained with different cell surface markers and felt into 3 tumor types, for each of which one representative case is shown: Ia: CD19+/B220+/human IgA+/CD43-/CD138- ; Ib: CD19+/human IgA+/B220+/CD43+/CD138- and II: B220-/CD138+.

By comparison to polyclonal B cells from pre-malignant animals, all lymphoma cases featured large B cells, with mean cell size increasing from type Ia to type Ib and to type II tumors (Fig 3A and B).

Lymphomas that developed in α1KI c-myc3’RR mice were checked for monoclonality by southern blot analysis of IgH gene configuration on tumor DNA samples taken at two different organ locations in each mouse. DNA from spleens of healthy α1KI controls showed the germline IgH band, whereas all tumors displayed rearranged bands with a monoclonal pattern (supplemental Fig 2).

The 29 α1KIc-myc3’RR tumors analyzed all overexpressed c-myc at the mRNA level (Fig 3C). We checked several α1KI c-myc3’RR tumors of either tumor subtype for potential mutations within c-myc and found it un-mutated in all cases analyzed.

Table 1: List of genes up-regulated and down-regulated in tumors of α1KI c-myc3’RR vs c-myc3’RR mice, according to their biological function.

Category of genes

Up

Down

Ribosome biogenesis and protein synthesis

24

9

Transcription factors and DNA-binding proteins

16

5

Highly expressed in immune system

9

20

Intracellular signal transduction modulators and effectors

10

28

metabolic enzyme-related proteins

2

16

Lipid metabolism

6

7

Cell cycle

2

4

Apoptose

4

0

Cell-adhesion-related proteins

2

6

Blood coagulation proteins

0

4

Mitochondrial-related protein

5

2

Others

17

15

TOTAL

97

116

Transcriptome analysis in α1KI c-myc3’RR lymphomas

The rare type II plasmablastic tumors arising in double transgenics obviously differed phenotypically from those seen in single c-myc transgenics and were not further explored at the mRNA level. Regarding type I lymphomas which were predominant and histologically reminiscent of those from c-myc single transgenic mice, we wondered whether subtle changes would be found by more in-depth molecular analyses. To thoroughly compare double mutant vs single c-myc transgenic mice tumors, we thus analyzed gene expression profiles of 8 IgA-expressing type I lymphomas (including 4 type Ia and 4 type Ib tumors) from double mutant mice compared to 8 IgM-expressing c-myc3’RR lymphomas (including 4 “BL-like” and 4 anaplastic lymphomas). Analysis was performed to identify differentially expressed genes, either significantly up-regulated or down-regulated in tumors from both groups of mice (Table 1 and supplemental Table 1). As previously described, lymphomas from c-myc single transgenics (whether BL-like or anaplastic) had roughly homogeneous profiles (Supplemental Fig 3) [22]. By contrast, the 8 α1KI tumors expressing an IgA BCR featured a clearly different gene expression profile, with 97 genes found up-regulated while 116 were down-regulated (both in type Ia and Ib tumors, whose profiles were roughly similar despite the differences regarding cell size and CD43 surface expression) (Table 1, Supplemental Fig. 3). Genes were classified according to their biological function (Table 1) and we used Ingenuity Pathway Analysis (IPA; Ingenuity Systems) to identify pathways that were enriched in the 2 lists reported in supplemental Table 2. The largest category of up-regulated genes involved ribosome biogenesis and protein synthesis (Table 1). These processes notably modulate cell growth and proliferation, and are frequently deregulated in cancer by influencing expression of oncogenes and tumor suppressors [28]. By itself, c-myc overexpression is known to stimulate ribosome assembly and result in increased cell size and protein synthesis [29].

Three genes negatively controlling cell proliferation were also strongly up-regulated in IgA-expressing tumors and may participate in delaying lymphomagenesis: the GADD34/ Ppp1r15a gene (up-regulated 12 fold) controls growth arrest during the unfolded protein response (UPR) [30]; Gas5 (overexpressed 12 fold) is transcribed into a noncoding RNA that induces growth arrest [31], while Btg1 (up-regulated 5 fold) is another inhibitor of cell proliferation [32]. By contrast, 4 down-regulated genes are tumor suppressors (Lxn, Rassf4, Dusp16, Ifitm2) [33-37].

Variations of another 3 genes could be connected with B cell activation: CD69 was overexpressed 4 fold in IgA tumors (similar to its overexpression polyclonal α1KI B cells) [6, 38] ); Trim12 (a repressor of lymphocyte activation) [39] was down-regulated 27 fold in IgA+ tumors, so as the gene for MARCH1, an E3 ubiquitin protein ligase targeting some membrane receptors for degradation and down-regulated by B cell activation [40].

Some other changes were related to membrane receptor expression or membrane trafficking, with notably a 3 fold increase in expression of the EHD2 ATP-ase, which stabilizes caveolae at the plasma membrane [41], and down-regulation of 15 other genes involved in related functions (Slc7a7, Slc39a4, Ggh, Sirpa, Rap1gap, H2afb3-ps, Slc37a2, Ctsd, Arhgap9, Gng12, Eml2, Ctsh Gpr18). Expression of Fcγr4 and of the γ chain of FcγR1 was also down-regulated.

Some changes were related to signal transduction and may notably increase tyrosine kinase and wnt signals in IgA-expressing tumors, with 5 fold increased Lgr5 (an activator of wnt signaling) [42] and a 3 fold down-regulation of two wnt inhibitors, Trim59 (a repressor of c-myc-dependent wnt induction) [43] and the CD9 tetraspanin which inhibits wnt/β-catenin signalization) [44]. Interestingly, two tyrosine kinase inhibitors are down regulated by 12 fold in IgA tumors: Lst1 a membrane adaptor recruiting phosphatases SHP1 and SHP2, and the DAP12 gene encoding tyrosine kinase binding protein TyroBP [45, 46]. NFATc2, a transcription factor involved in B cell anergy and whose defect has been correlated with B cell hyper-responsiveness is down-regulated 5 fold [47]. We also noticed a 4-fold increased expression of Dusp6, an inhibitory MAP kinase phosphatase, down-regulation (10-fold) of Nafm1, an ITAM receptor that modulates BCR signaling [48] and of Hck , a src family kinase expressed in B cell progenitors [49].

Another major gene category included additional transcription factors, DNA-binding proteins and histones, with strong up-regulation of the Fos-B transcription factor gene (FBJ osteosarcoma oncogene B) (P= 9.95E-05). Fos family proteins form heterodimers with jun family proteins and constitute AP-1 transcriptions factors. These factors regulate the expression of multiple genes modulating cell proliferation, activation, differentiation and eventually cell death [50-53].

Interestingly, Ingenuity Pathway Analysis of down-regulated genes pointed out that significantly enriched biologic functions were linked to cellular movement, immune cell trafficking and hematological system development and function (supplemental Table 2).

Finally, some variations might be correlated with the control of the tumor niche and of anti-tumor immunity, such as ebi3, a subunit of IL27 normally over-expressed in centrocytes (and down 16 fold in IgA tumors) [54], CCL9 (down 10 fold in IgA tumors), a chemokine known to attract CCR1-expressing stromal cells and also expressed in macrophages and myeloid cells [55, 56], the nab2 transcription factor (known to stimulate TRAIL expression) down-regulated 6 times [57], the 10-fold overexpressed pvrl2 (an inhibitor of NK cytotoxicity) [58], the 3-fold overexpressed cortistatin (an immunomodulatory anti-inflammatory peptide produced by immune cells) [59], and the 2.8-fold overexpressed CCL25, a chemokine normally attracting lymphocytes to the small bowel [60].

Tumor cell variations according to size and c-myc expression: (A) Total cells from &#x3b1;1KI

Figure 3: Tumor cell variations according to size and c-myc expression: (A) Total cells from α1KI c-myc3’RR tumors or normal B cells from α1KI and wt control mice were stained with PC5-conjugated anti-B220, PE-conjugated anti-CD19 (Type Ia and Ib) or PE-conjugated anti-CD138+ (type II) and analyzed by cell cytometry. Shown is the forward scatter (FSC, directly evaluating cell size) of cells gated as [B220+/CD19+] for wt, α1KI, type Ia and Ib tumor B cells, or gated as [CD138+] for type II tumors. (B) Histograms showed the X-median of FSC gated as [B220+ /CD19+] cells for control and type Ia and Ib tumors, and gated on [CD138+] cells for type II tumors (n=4). (C) Quantitative PCR analysis of c-myc expression for tumors of type Ia (n=5), type Ib (n=5) and type II (n=2) compared to control splenocytes from α1KI (n=3). The relative expression is shown as bar graph. (NS, not significant; *, p <0.05; ***, p <0.001, unpaired t-test)

DISCUSSION

Expression of the BCR is a major regulator of B cell fate, with signaling thresholds that vary between B1, follicular or marginal zone B cells. The tonic signal provided by BCR expression without any need for an extra-cellular ligand is mandatory for B cell survival and has also been reported as increased by class switching to the IgG or IgA class [6, 8, 61].

All along B lymphocyte development, cell survival is dependent from the production of Ig as membrane-anchored receptors (mIg). That even plasma cell survival might somehow benefit from soluble Ig production is still debated [62-64]. While of major importance in physiology, the role played by BCR signaling and Ig production with regard to the maintenance or growth of malignant B cells is still incompletely appreciated. Except for very immature proliferations, most cases of B cell malignancies feature expression of Ig, either in the form of a membrane-anchored pre-BCR / BCR for lymphomas or as secreted Ig for plasma cell proliferations. In some types of malignancies, the biased use of canonical V (D) J rearrangements suggests the role of chronic antigen/BCR stimulation [65]. Some other lymphoma cases involve deregulated signaling by mutated components of the BCR cascade [11]. The class of the BCR or of the secreted monoclonal Ig often correlates with the phenotype of human B cell malignancies, IgM expression being associated with less differentiated tumors [15, 66]. This raises the question of whether specific BCR properties might influence the tumor phenotype and supply different cell growth and/or survival signals. Beside the role of Ig-mediated tonic signals in cell survival, the ability of B cells to be activated and to differentiate relies on transient signals provided during BCR interaction with antigens. It is generally assumed that in immature B lymphocytes expressing IgM, early Ag contact leads to cell death or anergy while it promotes activation and differentiation of mature mIgM/mIgD-expressing B cells [67]. Also in class-switched mIgG+, mIgA+ or mIgE+ B cells and memory cells, BCR activation by antigen usually induces proliferation and differentiation into plasma cells. While all membrane Ig associate with the Ig-α / Ig-β complex to constitute the BCR, several studies have suggested that cells with a class-switched BCR hereby undergo some phenotypic changes. In particular, signaling from an IgG or IgA class BCR seems to induce more efficient B cell activation and short-term memory, while long-term accumulation as memory B cells might be a preferential feature of IgM expressing cells [8, 9, 61].

Among lymphomas, BL is characterized by the translocation of c-myc onto one of the Ig loci and the resulting c-myc deregulation then stands as the initial and driver hit [2]. It was shown in mice that the IgH locus 3’RR is instrumental for oncogenicity of c-myc translocations [68], which can also be artificially mimicked by associating a c-myc transgene with the IgH 3’RR [22]. The latter model induced two kinds of B cell neoplasias with different kinetics. Prevalent tumors overexpressed c-myc and arose at the mean age of 12 weeks with features resembling human BL: high proliferation, high level of apoptosis and a typical “starry sky” histological morphology, involving medium-sized mature B cells (B220+CD19+IgM+IgD+) with a basophilic cytoplasm. A second type of neoplasms arose at the mean age of 20 weeks, mostly involved liver and spleen, presented medium to large B220neg, CD138neg cells and were classified as anaplastic lymphomas [22, 23, 69].

In the present study, we backcrossed the c-myc3’RR transgene in the α1KI genetic background expressing a modified BCR of the IgA class. Such an IgA BCR was previously shown to yield high constitutive signaling in the absence of any BCR ligation. The present study shows that in double α1KI c-myc3’RR transgenic animals, the frequency of c-myc-driven tumors is decreased, their progression is lowered and the incurring lymphomas appear as more differentiated or even committed (in 10% of cases) into terminal BCR-, CD43+, CD138high plasma cell differentiation.

The slowed kinetics in the onset and progression of tumors is in agreement with the involvement of more differentiated cells. While all the tumors from the single transgenic mice were negative for both CD43 and CD138, most of double transgenic tumors have an activated phenotype with large CD43+ cells.

The c-myc oncogene has been described as a global regulator of various cell functions, and its overexpression usually results in tumorigenesis through a global deregulation of processes including increased cell proliferation, increased apoptosis, increased cell size, and marked changes in the total rate of protein synthesis [29, 70, 71]. In that regard, c-myc increases transcription of ribosomal RNA genes, recruitment of rRNA processing factors, ribosomal subunit export and recruitment of translation initiation factors [72]. In addition, it was shown that the c-myc3’RR transgene yielded more diverse tumor phenotypes (including mantle cell lymphoma, BL and plasma cell lymphoma) upon association to other deregulated oncogenic processes in cdk4 [24, 73] or p53 [23] mutant mice.

In the present study, the IgA class BCR modifies c-myc–driven tumorigenesis by modulating genes involved in distinct biology function, then stimulating cell activation and differentiation while inhibiting cell growth. and further increasing the ribosome biogenesis and protein synthesis already targeted by c-myc. Interestingly, 3 of the genes up-regulated in IgA+ mouse lymphomas have already been underlined in human lymphomas as among the 6 genes suggested as sufficient to distinguish GC-type from ABC-type DLBCLs [74]. These genes are GADD34 (up-regulated 12-fold), Myosin1 (up-regulated 5.9 fold) and fibronectin-1 (up-regulated 3.9 fold). Such genes up-regulated both in GC-type vs ABC-type DLBCL and in IgA+ vs IgM+ mouse lymphomas, are reminiscent of the observation that GC-type DLBCLs mostly express class-switched BCRs, while ABC-type DLBCLs mostly express IgM [14].

Altogether, IgA tumors in α1KI c-myc3’RR double mutant animals have a modified phenotype by comparison to IgM tumors from single c-myc transgenic mice, showing that a class-switched BCR signal resulted in cell transformation with a more differentiated or activated lymphoid phenotype or even increased occurrence of plasma cell phenotypes, then reminiscent of human plasma cell proliferations that overwhelmingly produce class-switched Ig. This could also be pertinent to the observations made in patients, where IgM expressing malignancies are usually more aggressive than related class-switched proliferations, as noticed for the rare IgM myelomas, with lymphoplasmacytic rather than plasmacytic proliferations or as noticed for DLBCL, where the worse prognosis associates with the IgM-expressing ABC-type.

MATERIALS AND METHODS

Mice

All mice were in a C57Bl6 background previously shown to allow a nearly constant occurrence of c-myc-driven lymphomas within the first 9 months of life [22]. The c-myc transgene includes the murine c-myc as a 7-kb SmaI-KpnI genomic fragment including P1 and P2 promoters, followed with the four IgH 3’RR enhancers: hs1,2 as a 0.6-kb StuI-EcoRV fragment flanked on both sides with hs3a and hs3b as two 2.1-kb EcoRI-HindIII fragments cloned with symmetric orientations, and hs4 as a 1.38-kb PstI-HindIII fragment [25]. This c-myc expression cassette is flanked by β-globin insulators. The α1KI mutation corresponds to insertion of a 6 kb IgH Cα1 human genomic fragment (from upstream of CH1α to a BamHI site downstream of the membrane-form transcript polyadenylation signal) in replacement of the mouse IgH Sµ region [6]. The project was approved by our local ethics committee review board. Mice were followed daily for clinical examination and macroscopic tumor development. Animal exhibiting important tumors or presenting severe signs of anemia or illness were sacrificed.

Southern blot analysis

Genomic DNA was extracted from mice tails. Ten µg of DNA were digested with EcoRI, loaded on 0.7% agarose gels, transferred on nylon sheets (Amersham, Buckinghamshire, UK), hybridized with a 32P-labeled hs4 probe (a 1.3 kb PstI fragment) and autoradiographed to ensure the presence of the c-myc3’RR transgene as previously reported [22] .

Clonality of tumors was also assayed by southern blots of EcoRI digested tumor DNAs, with an IgH probe located around the JH4 segment (a 0.5kb SacI-NaeI fragment).

In vitro stimulation assays.

Splenic B cells from α1KI c-myc3’RR and α1KI homozygous control mice were stimulated for 3 days with 20 µg/mL of LPS from Salmonella typhimurium (Sigma, St-Louis, MO) or with 5 μg/mL anti-CD40 (R&D systems, Lille, France) plus 40 ng/mL murine  IL4 (PeproTech) in RPMI supplemented with 10% heat-inactivated fetal calf serum. The number of viable proliferating cells was assessed using MTS solution [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium], and a colorimetric assay (Promega) following the manufacturer’s recommendations.

mRNA expression

Total RNA was extracted from fresh tissues and cell suspensions using tri-reagent (Ambion Austin, TX). RNA was reverse-transcribed into cDNA by addition of reverse transcriptase to 2 µg total RNA in a final volume of 20 µl. Real-time PCR was performed in duplicate by using TaqMan assay reagents and analyzed on an ABI Prism 7000 system (Applied Biosystems Foster City, CA). Transcripts analyzed were c-myc (Mm00487803-m1) and 18S RNA (Hs99999901-s1) used for gene expression level normalization (Applied Biosystems).

Microarray preparations

RNA was extracted from hyperplasic lymph nodes (LNs). Experiments were done in quadruplicate and 0.2µg of RNA was used for each microarray analysis. cDNA labeling and dual color microarray hybridization were done in the Nice Sophia-Antipolis Microarray Facility (Nice Sophia Antipolis, France).

Gene Data Expression Analyses

Cluster analysis.  GenBank accession number - GSE 32229. Log 2 signals were extracted after interslide quantile normalization and performed with R (Bioconductor). Probe sets showing minimal variation across all samples were filtered out, only retaining those 3217 with a coefficient of variation over 0.1, and using them to cluster the samples through the hierarchical agglomerative algorithm. Average linkage and Euclidian distance were used as linkage and similarity methods, respectively. Supervised classification was carried out using Cluster and TreeView software.

Detection of differentially expressed genes. Statistical analyses used the R and Bioconductor software. To identify over- or under-expressed genes, we used the empirical Bayesian method for differential expression calculation [26] implemented in the Linear Models for Microarray Analysis (LIMMA) package. For each pairwise comparison, p-values were adjusted after multiple testing corrections using the Benjamini and Hochberg method [27]. Probes with a corrected value below 0.05 were considered as differentially expressed.

Futhermore, gene lists of enriched pathways were generated using Ingenuity Pathway Analysis.

Flow cytometry analysis

Cell suspensions from spleen, liver and LNs were stained with various antibodies (Southern Biotechnologies or eBiosciences): SpectralRed (PC5)-conjugated anti-B220, phycoerythrin-labeled (PE) anti-CD19, anti-CD138, FITC-conjugated anti-human IgA and PE anti-mouse CD43. Cells were analyzed on a Coulter XL apparatus (Beckman Coulter, Fullerton, CA).

Blood sampling

Blood samples were collected from transgenic and wild-type (wt) mice with heparinised needles. Blood smears allowed differential counting of blood cells after staining by the May-Grünwald-Giemsa method.

ACKNOWLEDGMENTS

We thank Dr. Franck Trimoreau for his help in reading blood smears. This work was supported by grants from Association pour la Recherche sur le Cancer (grant ARC SL220100601332), Ligue Nationale Contre le Cancer, Institut National du Cancer (InCA) and Région Limousin.

References

1. Pelengaris S, Khan M and Evan G. c-MYC: more than just a matter of life and death. Nature reviews Cancer. 2002; 2(10):764-776.

2. Blum KA, Lozanski G and Byrd JC. Adult Burkitt leukemia and lymphoma. Blood. 2004; 104(10):3009-3020.

3. Lattanzio G, Libert C, Aquilina M, Cappelletti M, Ciliberto G, Musiani P and Poli V. Defective development of pristane-oil-induced plasmacytomas in interleukin-6-deficient BALB/c mice. The American journal of pathology. 1997; 151(3):689-696.

4. Byrd LG, McDonald AH, Gold LG and Potter M. Specific pathogen-free BALB/cAn mice are refractory to plasmacytoma induction by pristane. J Immunol. 1991; 147(10):3632-3637.

5. Carey GB, Donjerkovic D, Mueller CM, Liu S, Hinshaw JA, Tonnetti L, Davidson W and Scott DW. B-cell receptor and Fas-mediated signals for life and death. Immunological reviews. 2000; 176:105-115.

6. Duchez S, Amin R, Cogne N, Delpy L, Sirac C, Pascal V, Corthesy B and Cogne M. Premature replacement of mu with alpha immunoglobulin chains impairs lymphopoiesis and mucosal homing but promotes plasma cell maturation. Proceedings of the National Academy of Sciences of the United States of America. 2010; 107(7):3064-3069.

7. Martin SW and Goodnow CC. Burst-enhancing role of the IgG membrane tail as a molecular determinant of memory. Nature immunology. 2002; 3(2):182-188.

8. Pogue SL and Goodnow CC. Gene dose-dependent maturation and receptor editing of B cells expressing immunoglobulin (Ig)G1 or IgM/IgG1 tail antigen receptors. The Journal of experimental medicine. 2000; 191(6):1031-1044.

9. Laffleur B, Denis-Lagache N, Péron S, Sirac C, Moreau J and Cogné M. Control of B cell fate after AID-induced remodeling of immunoglobulin genes. Oncotarget. 2014:1118-1131.

10. Caldwell RG, Wilson JB, Anderson SJ and Longnecker R. Epstein-Barr virus LMP2A drives B cell development and survival in the absence of normal B cell receptor signals. Immunity. 1998; 9(3):405-411.

11. Compagno M, Lim WK, Grunn A, Nandula SV, Brahmachary M, Shen Q, Bertoni F, Ponzoni M, Scandurra M, Califano A, Bhagat G, Chadburn A, Dalla-Favera R and Pasqualucci L. Mutations of multiple genes cause deregulation of NF-kappaB in diffuse large B-cell lymphoma. Nature. 2009; 459(7247):717-721.

12. Ngo VN, Young RM, Schmitz R, Jhavar S, Xiao W, Lim KH, Kohlhammer H, Xu W, Yang Y, Zhao H, Shaffer AL, Romesser P, Wright G, Powell J, Rosenwald A, Muller-Hermelink HK, et al. Oncogenically active MYD88 mutations in human lymphoma. Nature. 2011; 470(7332):115-119.

13. Coelho V, Krysov S, Ghaemmaghami AM, Emara M, Potter KN, Johnson P, Packham G, Martinez-Pomares L and Stevenson FK. Glycosylation of surface Ig creates a functional bridge between human follicular lymphoma and microenvironmental lectins. Proceedings of the National Academy of Sciences of the United States of America. 2010; 107(43):18587-18592.

14. Eschbach C, Bach MP, Fidler I, Pelanda R, Kohler F, Rajewsky K and Jumaa H. Efficient generation of B lymphocytes by recognition of self-antigens. European journal of immunology. 2011; 41(8):2397-2403.

15. Ruminy P, Etancelin P, Couronne L, Parmentier F, Rainville V, Mareschal S, Bohers E, Burgot C, Cornic M, Bertrand P, Lenormand B, Picquenot JM, Jardin F, Tilly H and Bastard C. The isotype of the BCR as a surrogate for the GCB and ABC molecular subtypes in diffuse large B-cell lymphoma. Leukemia. 2011; 25(4):681-688.

16. Chauveau C and Cogne M. Palindromic structure of the IgH 3’locus control region. Nature genetics. 1996; 14(1):15-16.

17. Guglielmi L, Le Bert M, Truffinet V, Cogne M and Denizot Y. Insulators to improve expression of a 3(‘)IgH LCR-driven reporter gene in transgenic mouse models. Biochemical and biophysical research communications. 2003; 307(3):466-471.

18. Rouaud P, Vincent-Fabert C, Saintamand A, Fiancette R, Marquet M, Robert I, Reina-San-Martin B, Pinaud E, Cogne M and Denizot Y. The IgH 3’ regulatory region controls somatic hypermutation in germinal center B cells. The Journal of experimental medicine. 2013; 210(8):1501-1507.

19. Madisen L and Groudine M. Identification of a locus control region in the immunoglobulin heavy-chain locus that deregulates c-myc expression in plasmacytoma and Burkitt’s lymphoma cells. Genes & development. 1994; 8(18):2212-2226.

20. Morvan CL, Pinaud E, Decourt C, Cuvillier A and Cogne M. The immunoglobulin heavy-chain locus hs3b and hs4 3’ enhancers are dispensable for VDJ assembly and somatic hypermutation. Blood. 2003; 102(4):1421-1427.

21. Pinaud E, Khamlichi AA, Le Morvan C, Drouet M, Nalesso V, Le Bert M and Cogne M. Localization of the 3’ IgH locus elements that effect long-distance regulation of class switch recombination. Immunity. 2001; 15(2):187-199.

22. Truffinet V, Pinaud E, Cogne N, Petit B, Guglielmi L, Cogne M and Denizot Y. The 3’ IgH locus control region is sufficient to deregulate a c-myc transgene and promote mature B cell malignancies with a predominant Burkitt-like phenotype. J Immunol. 2007; 179(9):6033-6042.

23. Fiancette R, Rouaud P, Vincent-Fabert C, Laffleur B, Magnone V, Cogne M and Denizot Y. A p53 defect sensitizes various stages of B cell development to lymphomagenesis in mice carrying an IgH 3’ regulatory region-driven c-myc transgene. J Immunol. 2011; 187(11):5772-5782.

24. Rouaud P, Fiancette R, Vincent-Fabert C, Magnone V, Cogne M, Dubus P and Denizot Y. Mantle cell lymphoma-like lymphomas in c-myc-3’RR/p53+/- mice and c-myc-3’RR/Cdk4R24C mice: differential oncogenic mechanisms but similar cellular origin. Oncotarget. 2012; 3(5):586-593.

25. Michaelson JS, Giannini SL and Birshtein BK. Identification of 3’ alpha-hs4, a novel Ig heavy chain enhancer element regulated at multiple stages of B cell differentiation. Nucleic acids research. 1995; 23(6):975-981.

26. Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Statistical applications in genetics and molecular biology. 2004; 3:Article3.

27. Reiner A, Yekutieli D and Benjamini Y. Identifying differentially expressed genes using false discovery rate controlling procedures. Bioinformatics. 2003; 19(3):368-375.

28. Ruggero D and Pandolfi PP. Does the ribosome translate cancer? Nature reviews Cancer. 2003; 3(3):179-192.

29. Iritani BM and Eisenman RN. c-Myc enhances protein synthesis and cell size during B lymphocyte development. Proceedings of the National Academy of Sciences of the United States of America. 1999; 96(23):13180-13185.

30. Novoa I, Zeng H, Harding HP and Ron D. Feedback inhibition of the unfolded protein response by GADD34-mediated dephosphorylation of eIF2alpha. The Journal of cell biology. 2001; 153(5):1011-1022.

31. Williams GT, Mourtada-Maarabouni M and Farzaneh F. A critical role for non-coding RNA GAS5 in growth arrest and rapamycin inhibition in human T-lymphocytes. Biochemical Society transactions. 2011; 39(2):482-486.

32. Rouault JP, Rimokh R, Tessa C, Paranhos G, Ffrench M, Duret L, Garoccio M, Germain D, Samarut J and Magaud JP. BTG1, a member of a new family of antiproliferative genes. The EMBO journal. 1992; 11(4):1663-1670.

33. Daniel-Carmi V, Makovitzki-Avraham E, Reuven EM, Goldstein I, Zilkha N, Rotter V, Tzehoval E and Eisenbach L. The human 1-8D gene (IFITM2) is a novel p53 independent pro-apoptotic gene. International journal of cancer Journal international du cancer. 2009; 125(12):2810-2819.

34. Eckfeld K, Hesson L, Vos MD, Bieche I, Latif F and Clark GJ. RASSF4/AD037 is a potential ras effector/tumor suppressor of the RASSF family. Cancer research. 2004; 64(23):8688-8693.

35. Hoornaert I, Marynen P, Goris J, Sciot R and Baens M. MAPK phosphatase DUSP16/MKP-7, a candidate tumor suppressor for chromosome region 12p12-13, reduces BCR-ABL-induced transformation. Oncogene. 2003; 22(49):7728-7736.

36. Muthusamy V, Premi S, Soper C, Platt J and Bosenberg M. The hematopoietic stem cell regulatory gene latexin has tumor-suppressive properties in malignant melanoma. The Journal of investigative dermatology. 2013; 133(7):1827-1833.

37. Siegrist F, Ebeling M and Certa U. The small interferon-induced transmembrane genes and proteins. Journal of interferon & cytokine research : the official journal of the International Society for Interferon and Cytokine Research. 2011; 31(1):183-197.

38. Sancho D, Gomez M and Sanchez-Madrid F. CD69 is an immunoregulatory molecule induced following activation. Trends in immunology. 2005; 26(3):136-140.

39. Liston A, Hardy K, Pittelkow Y, Wilson SR, Makaroff LE, Fahrer AM and Goodnow CC. Impairment of organ-specific T cell negative selection by diabetes susceptibility genes: genomic analysis by mRNA profiling. Genome biology. 2007; 8(1):R12.

40. Galbas T, Steimle V, Lapointe R, Ishido S and Thibodeau J. MARCH1 down-regulation in IL-10-activated B cells increases MHC class II expression. Cytokine. 2012; 59(1):27-30.

41. Stoeber M, Stoeck IK, Hanni C, Bleck CK, Balistreri G and Helenius A. Oligomers of the ATPase EHD2 confine caveolae to the plasma membrane through association with actin. The EMBO journal. 2012; 31(10):2350-2364.

42. Carmon KS, Lin Q, Gong X, Thomas A and Liu Q. LGR5 interacts and cointernalizes with Wnt receptors to modulate Wnt/beta-catenin signaling. Molecular and cellular biology. 2012; 32(11):2054-2064.

43. Licchesi JD, Van Neste L, Tiwari VK, Cope L, Lin X, Baylin SB and Herman JG. Transcriptional regulation of Wnt inhibitory factor-1 by Miz-1/c-Myc. Oncogene. 2010; 29(44):5923-5934.

44. Chairoungdua A, Smith DL, Pochard P, Hull M and Caplan MJ. Exosome release of beta-catenin: a novel mechanism that antagonizes Wnt signaling. The Journal of cell biology. 2010; 190(6):1079-1091.

45. Draber P, Stepanek O, Hrdinka M, Drobek A, Chmatal L, Mala L, Ormsby T, Angelisova P, Horejsi V and Brdicka T. LST1/A is a myeloid leukocyte-specific transmembrane adaptor protein recruiting protein tyrosine phosphatases SHP-1 and SHP-2 to the plasma membrane. The Journal of biological chemistry. 2012; 287(27):22812-22821.

46. Nakano-Yokomizo T, Tahara-Hanaoka S, Nakahashi-Oda C, Nabekura T, Tchao NK, Kadosaki M, Totsuka N, Kurita N, Nakamagoe K, Tamaoka A, Takai T, Yasui T, Kikutani H, Honda S, Shibuya K, Lanier LL, et al. The immunoreceptor adapter protein DAP12 suppresses B lymphocyte-driven adaptive immune responses. The Journal of experimental medicine. 2011; 208(8):1661-1671.

47. Samanta DN, Palmetshofer A, Marinkovic D, Wirth T, Serfling E and Nitschke L. B cell hyperresponsiveness and expansion of mature follicular B cells but not of marginal zone B cells in NFATc2/c3 double-deficient mice. J Immunol. 2005; 174(8):4797-4802.

48. Ohtsuka M, Arase H, Takeuchi A, Yamasaki S, Shiina R, Suenaga T, Sakurai D, Yokosuka T, Arase N, Iwashima M, Kitamura T, Moriya H and Saito T. NFAM1, an immunoreceptor tyrosine-based activation motif-bearing molecule that regulates B cell development and signaling. Proceedings of the National Academy of Sciences of the United States of America. 2004; 101(21):8126-8131.

49. Taguchi T, Kiyokawa N, Sato N, Saito M and Fujimoto J. Characteristic expression of Hck in human B-cell precursors. Experimental hematology. 2000; 28(1):55-64.

50. Jochum W, Passegue E and Wagner EF. AP-1 in mouse development and tumorigenesis. Oncogene. 2001; 20(19):2401-2412.

51. Karin M. The regulation of AP-1 activity by mitogen-activated protein kinases. The Journal of biological chemistry. 1995; 270(28):16483-16486.

52. Nakabeppu Y and Nathans D. A naturally occurring truncated form of FosB that inhibits Fos/Jun transcriptional activity. Cell. 1991; 64(4):751-759.

53. Nakabeppu Y, Ryder K and Nathans D. DNA binding activities of three murine Jun proteins: stimulation by Fos. Cell. 1988; 55(5):907-915.

54. Larousserie F, Bardel E, Coulomb L’Hermine A, Canioni D, Brousse N, Kastelein RA and Devergne O. Variable expression of Epstein-Barr virus-induced gene 3 during normal B-cell differentiation and among B-cell lymphomas. The Journal of pathology. 2006; 209(3):360-368.

55. Lean JM, Murphy C, Fuller K and Chambers TJ. CCL9/MIP-1gamma and its receptor CCR1 are the major chemokine ligand/receptor species expressed by osteoclasts. Journal of cellular biochemistry. 2002; 87(4):386-393.

56. Youn BS, Jang IK, Broxmeyer HE, Cooper S, Jenkins NA, Gilbert DJ, Copeland NG, Elick TA, Fraser MJ, Jr. and Kwon BS. A novel chemokine, macrophage inflammatory protein-related protein-2, inhibits colony formation of bone marrow myeloid progenitors. J Immunol. 1995; 155(5):2661-2667.

57. Balzarolo M, Karrich JJ, Engels S, Blom B, Medema JP and Wolkers MC. The transcriptional regulator NAB2 reveals a two-step induction of TRAIL in activated plasmacytoid DCs. European journal of immunology. 2012; 42(11):3019-3027.

58. Stanietsky N, Simic H, Arapovic J, Toporik A, Levy O, Novik A, Levine Z, Beiman M, Dassa L, Achdout H, Stern-Ginossar N, Tsukerman P, Jonjic S and Mandelboim O. The interaction of TIGIT with PVR and PVRL2 inhibits human NK cell cytotoxicity. Proceedings of the National Academy of Sciences of the United States of America. 2009; 106(42):17858-17863.

59. van Hagen PM, Dalm VA, Staal F and Hofland LJ. The role of cortistatin in the human immune system. Molecular and cellular endocrinology. 2008; 286(1-2):141-147.

60. Campbell DJ and Butcher EC. Intestinal attraction: CCL25 functions in effector lymphocyte recruitment to the small intestine. The Journal of clinical investigation. 2002; 110(8):1079-1081.

61. Wakabayashi C, Adachi T, Wienands J and Tsubata T. A distinct signaling pathway used by the IgG-containing B cell antigen receptor. Science. 2002; 298(5602):2392-2395.

62. Decourt C, Galea HR, Sirac C and Cogne M. Immunologic basis for the rare occurrence of true nonsecretory plasma cell dyscrasias. Journal of leukocyte biology. 2004; 76(3):528-536.

63. Kumazaki K, Tirosh B, Maehr R, Boes M, Honjo T and Ploegh HL. AID-/-mus-/- mice are agammaglobulinemic and fail to maintain B220-CD138+ plasma cells. J Immunol. 2007; 178(4):2192-2203.

64. Lechouane F, Bonaud A, Delpy L, Casola S, Oruc Z, Chemin G, Cogne M and Sirac C. B-cell receptor signal strength influences terminal differentiation. European journal of immunology. 2013; 43(3):619-628.

65. Sutton LA, Kostareli E, Hadzidimitriou A, Darzentas N, Tsaftaris A, Anagnostopoulos A, Rosenquist R and Stamatopoulos K. Extensive intraclonal diversification in a subgroup of chronic lymphocytic leukemia patients with stereotyped IGHV4-34 receptors: implications for ongoing interactions with antigen. Blood. 2009; 114(20):4460-4468.

66. Avet-Loiseau H, Garand R, Lode L, Harousseau JL and Bataille R. Translocation t(11;14)(q13;q32) is the hallmark of IgM, IgE, and nonsecretory multiple myeloma variants. Blood. 2003; 101(4):1570-1571.

67. Eeva J and Pelkonen J. Mechanisms of B cell receptor induced apoptosis. Apoptosis : an international journal on programmed cell death. 2004; 9(5):525-531.

68. Gostissa M, Yan CT, Bianco JM, Cogne M, Pinaud E and Alt FW. Long-range oncogenic activation of Igh-c-myc translocations by the Igh 3’ regulatory region. Nature. 2009; 462(7274):803-807.

69. Morse HC, 3rd, Anver MR, Fredrickson TN, Haines DC, Harris AW, Harris NL, Jaffe ES, Kogan SC, MacLennan IC, Pattengale PK and Ward JM. Bethesda proposals for classification of lymphoid neoplasms in mice. Blood. 2002; 100(1):246-258.

70. Schuhmacher M, Staege MS, Pajic A, Polack A, Weidle UH, Bornkamm GW, Eick D and Kohlhuber F. Control of cell growth by c-Myc in the absence of cell division. Current biology : CB. 1999; 9(21):1255-1258.

71. Nie Z, Hu G, Wei G, Cui K, Yamane A, Resch W, Wang R, Green DR, Tessarollo L, Casellas R, Zhao K and Levens D. c-Myc is a universal amplifier of expressed genes in lymphocytes and embryonic stem cells. Cell. 2012; 151(1):68-79.

72. van Riggelen J, Yetil A and Felsher DW. MYC as a regulator of ribosome biogenesis and protein synthesis. Nature reviews Cancer. 2010; 10(4):301-309.

73. Vincent-Fabert C, Fiancette R, Rouaud P, Baudet C, Truffinet V, Magnone V, Guillaudeau A, Cogne M, Dubus P and Denizot Y. A defect of the INK4-Cdk4 checkpoint and Myc collaborate in blastoid mantle cell lymphoma-like lymphoma formation in mice. The American journal of pathology. 2012; 180(4):1688-1701.

74. Lossos IS, Czerwinski DK, Alizadeh AA, Wechser MA, Tibshirani R, Botstein D and Levy R. Prediction of survival in diffuse large-B-cell lymphoma based on the expression of six genes. The New England journal of medicine. 2004; 350(18):1828-1837.


Creative Commons License All site content, except where otherwise noted, is licensed under a Creative Commons Attribution 4.0 License.
PII: 2297