Essential Role of Phospholipase Cγ2 in Early B-Cell Development and Myc-Mediated Lymphomagenesis

Abstract

Phospholipase Cγ2 (PLCγ2) is a critical signaling effector of the B-cell receptor (BCR). Here we show that PLCγ2 deficiency impedes early B-cell development, resulting in an increase of B220⁺ CD43⁺ BP-1⁺ CD24^hi pre-BCR⁺ large pre-B cells. PLCγ2 deficiency impairs pre-BCR-mediated functions, leading to enhanced interleukin-7 (IL-7) signaling and elevated levels of RAGs in the selected large pre-B cells. Consequently, PLCγ2 deficiency renders large pre-B cells susceptible to transformation, resulting in dramatic acceleration of Myc-induced lymphomagenesis. PLCγ2^−/− Eμ-Myc transgenic mice mainly develop lymphomas of B220⁺ CD43⁺ BP-1⁺ CD24^hi pre-BCR⁺ large pre-B-cell origin, which are uncommon in wild-type Eμ-Myc transgenics. Furthermore, lymphomas from PLCγ2^−/− Eμ-Myc transgenic mice exhibited a loss of p27^Kip1 and often displayed alterations in Arf or p53. Thus, PLCγ2 plays an important role in pre-BCR-mediated early B-cell development, and its deficiency leads to markedly increased pools of the most at-risk large pre-B cells, which display hyperresponsiveness to IL-7 and express high levels of RAGs, making them prone to secondary mutations and Myc-induced malignancy.

B-cell development is orchestrated by complex signaling networks, including those emanating from the pre-B-cell receptor (pre-BCR) and the BCR (10, 15). B-cell development follows an ordered series of events that relies on the sequential and proper rearrangements of the immunoglobulin heavy (IgH) and light (IgL) chain genes and upon the controlled expression of cell surface markers and transcription factors (10, 15). IgH chain gene rearrangement initiates in pro-B cells, and its successful rearrangement leads to the formation of the pre-BCR, which consists of the newly generated H chain in complex with the VpreB/λ5 surrogate light chain (10, 15). Signals emanating from the pre-BCR then provoke the expansion of pre-B cells and direct IgL chain gene rearrangements. Finally, the successfully rearranged L chain complexes with the H chain to generate a surface IgM form of the BCR, a hallmark of immature B cells (10, 15), and signaling from the BCR then orchestrates further B-cell maturation and directs B-cell function (33, 35).

The pre-BCR and BCR complexes have common signal transduction pathway components, including the Ig(α) and Ig(β) transmembrane subunits (18, 24, 62). Their signaling relies on the sequential activation of three cytoplasmic tyrosine kinases, Lyn, Syk, and Btk, and upon the recruitment, tyrosine phosphorylation, and activation of the adapter protein SLP-65/BLNK and of the lipid kinase phosphatidylinositol 3-kinase (28, 40, 46). In turn, these events activate phospholipase Cγ2 (PLCγ2), which hydrolyzes phosphatidylinositol 4,5-bisphosphate to generate diacylglycerol and inositol 1,4,5-trisphosphate, which are required second messengers for diverse cellular responses (47, 48). Underscoring its essential role as a signaling effector, PLCγ2-deficient mice have profound defects in the transition from immature to mature B cells, and PLCγ2-deficient B cells fail to respond to mitogens and lack characteristic Ca²⁺ fluxes that follow engagement of the BCR (13, 64).

Due to the requirements for somatic antibody diversity and proper rearrangements of Ig genes, the B cell is an inherently hypermutable environment. Thus, chromosomal lesions can often occur that disrupt B-cell proliferation, apoptosis, and/or differentiation, and these changes ultimately result in B-cell leukemia or lymphoma (55). The t(8;14) chromosomal translocation, which involves the c-Myc oncogene and the regulatory regions of the Ig loci, is the underlying genetic event that gives rise to human Burkitt lymphoma (4, 61). The role of Myc in this disease was established by the creation of the Eμ-Myc transgenic mouse, where c-Myc is overexpressed in the B-cell compartment by virtue of the IgH chain enhancer (Eμ) (1). B cells from these mice display high proliferative rates that are initially offset by Myc's ability to trigger the apoptotic program (42, 52). Ultimately, however, secondary changes occur that bypass Myc's apoptotic program, and these mice generally succumb to lethal lymphoma by approximately 4 months of age (1, 5, 12, 20, 21, 29, 36, 54).

Genetic studies have established that Myc's ability to accelerate cell growth and trigger apoptosis are both rate limiting for lymphoma development in Eμ-Myc transgenics. For example, Myc triggers apoptosis through the agency of the Arf-p53 tumor suppressor pathway, and mutations in Arf and p53 are a hallmark of lymphomas in Eμ-Myc transgenic mice (5, 56) and Burkitt lymphomas (51). Furthermore, Myc's ability to accelerate cell proliferation is linked to its capacity to downregulate the expression of the cyclin-dependent kinase (Cdk) inhibitor p27^Kip1, and loss of p27^Kip1 accelerates Myc-induced lymphomagenesis (2, 36, 39, 41, 45, 63). The Btk and SLP-65 BCR signaling effectors have been suggested to function as tumor suppressors in B-cell transformation (6, 25). Here we report that the PLCγ2 deficiency impairs pre-BCR signaling and results in an increase of large pre-B cells, their elevated expression of RAGs and interleukin-7 (IL-7) receptor, and their susceptibilities to Myc-induced transformation.

MATERIALS AND METHODS

Interbreeding of mice and tumor surveillance.

PLCγ2^+/− mice (C57Bl/6 × 129/svj, backcrossed to C57BL/6) were interbred with Eμ-Myc transgenic mice (congenic C57Bl/6) (1, 58). The F₁ offspring were crossed to PLCγ2^+/− mice to generate PLCγ2^+/+, PLCγ2^+/−, and PLCγ2^−/− Eμ-Myc transgenic littermates. These mice were monitored daily for signs of morbidity and tumor development. A log-rank test was performed to determine the statistical significance of the survival between the different genotypes of Eμ-Myc transgenic mice. Tumors that arose were harvested immediately after sacrifice of animals. Single-cell suspensions were obtained from parts of the tumors and were subjected to fluorescent-activated cell sorter (FACS) analysis. The remainder of the tumor samples was snap frozen in liquid nitrogen for DNA, RNA, and protein analyses.

Flow cytometry.

Single-cell suspensions of spleen, bone marrow (BM), lymph node, or tumor tissue were treated with Gey's solution to remove red blood cells and resuspended in phosphate-buffered saline (PBS) with 2% bovine serum albumin (BSA). The cells were then stained with a combination of fluorescence-conjugated antibodies. Cy-Chrome-conjugated anti-B220 (15-0452), allophycocyanin (APC)-conjugated anti-B220 (17-0452), phycoerythrin (PE)-conjugated c-Kit (12-1171), and fluorescein isothiocyanate (FITC)-conjugated anti-Thy1.2 (11-0902) were purchased from eBioscience. Both FITC (1140-02)- and PE-conjugated (1140-09) anti-μ were purchased from Southern Biotechnology. PE-conjugated anti-B220 (553090), PE-conjugated anti-CD43 (553271), biotin-conjugated anti-pre-BCR (clone SL156,551863), biotin-conjugated anti-CD179β (λ5) (clone LM34,551865), APC-conjugated anti-CD19 (550992), biotin-conjugated anti-CD24 (555296), biotin-conjugated IL-7Rα (555288), FITC-conjugated anti-BP-1 (01284D), PE-conjugated CD25 (09985B), FITC-conjugated anti-mouse Igκ light chain (550003), FITC-conjugated anti-mouse Igλ light chain (553434), PE-conjugated CD25 (09985B), PE-conjugated streptavidin (554061), Cy-Chrome-conjugated streptavidin (554062), and biotin-conjugated Rat IgG2a isotype control (553997) were purchased from BD Biosciences Pharmingen. Anti-mouse CD179α (VpreB) (59) was conjugated with biotin following the manufacturer's recommendations (Pierce). All antibodies were monoclonal antibodies. Samples were applied to FACS analysis.

Thymidine and bromodeoxyuridine (BrdU) incorporation assays.

Freshly isolated bone marrow B cells, bone marrow-derived B cells, or FACS-sorted bone marrow B cells were cultured at 2 × 10⁴/well in round-bottom 96-well plates in RPMI-1640 medium with 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, 5 × 10⁻⁵ M 2-mercaptoethanol, 100 U/ml penicillin, 100 μg/ml streptomycin, 2 mM glutamine (all from Invitrogen Life Technologies), 10% heat-inactivated fetal bovine serum (HyClone), and the indicated concentrations of IL-7 (0, 0.02, 0.2, 2 ng/ml) for 2 days (IL-7 culture-derived BM B cells) or 4 days (freshly isolated BM B cells). For thymidine incorporation assays, the culture was then pulsed with 1 μCi/well of tritiated deoxythymidine (³H-dT; DuPont) for 18 h and harvested. Incorporated ³H-dT was determined by a Betaplate Liquid Scintillation Counter. For BrdU incorporation assays, FACS-sorted BM B cells were cultured with IL-7 (2 ng/ml) in 48-well plates for 3 days. Subsequently, the cells were incubated with BrdU (10 μM) for 1 h followed by staining with FITC-conjugated anti-BrdU antibodies (347583; BD Bioscience Pharmingen). After staining, the cells were resuspended in 300 μl PBS containing 2% BSA and 1 μg/ml 7-amino-actinomycin D and were immediately analyzed by FACS.

Semiquantitative and real-time RT-PCR.

Total RNA were prepared from IL-7 BM culture-derived cells by RNA STAT-60 (Tel-Test, Inc.), and first-strand cDNA was synthesized from total RNA with Omniscript (QIAGEN) according to the manufacturer's instructions. For semiquantitative reverse transcription-PCR (RT-PCR), the specific primers are as follows: RAG1, TGCAGACATTCTAGCACTCTGG (5′ primer) and ACATCTGCCTTCACGTCGAT(3′ primer); RAG2, GCTATGTCAGAAGCATTCTATTTC (5′ primer) and CTTGGCAGGAGTCAAGACTTTCCC (3′ primer); β-actin, ACTCCTATGTGGGTGACGAG and CAGGTCCAGACGCAGGATGGC (3′ primer). RAG1 was amplified by 34 cycles of 94°C for 45 s, 55°C for 30 s, and 72°C for 1 min. RAG2 was amplified by 45 cycles of 94°C for 45 s, 60°C for 45 s, and 72°C for 1 min. The β-actin gene was used as an internal RNA level control. Real-time RT-PCR was performed as previously described (19). Briefly, the specific primers are as follows: RAG1, CATTCTAGCACTCTGGCCGG (5′ primer) and TCATCGGGTGCAGAACTGAA (3′ primer); RAG2, TTAATTCCTGGCTTGGCCG and TTCCTGCTTGTGGATGTGAAAT (3′ primer); CD19, AATCCACGCATTCAAGTCCAG (5′ primer) and GAGCCCTCCTCGCTGTCTG (3′ primer). The real-time PCR was performed with an iCycler iQ (Bio-Rad) and was carried out in duplicate or triplicate at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min in 25-μl reaction volumes. SYBR-Green dye was used to detect amplified DNA. RAG1 or RAG2 transcript was normalized to CD19 expression using standard curves generated for each sample by a series of four consecutive 10-fold dilutions of the cDNA template with iCycler iQ analyzing software.

Western blotting.

Protein was extracted from normal primary B cells or B-cell tumors arising in Eμ-Myc transgenic mice, as previously described, with modifications (5). Briefly, cells or tumors were sonicated seven times for 1 s each in ice-cold lysis buffer (50 mM Tris, pH 8.0, 150 mM NaCl, 100 mM Na₃VO₄, 50 mM NaF, 0.1 mM EDTA, 1% Triton X-100, 1 mM dithiothreitol, 0.4 mM phenylmethylsulfonyl fluoride, 3 μg/ml aprotinin, 2 μg/ml pepstain A, 1 μg/ml leupeptin). Lysates were cleared of debris at 12,000 × g for 15 min, and protein in the supernatant was quantified. Protein (100 to 150 μg/lane) was separated in sodium dodecyl sulfate-10% polyacrylamide gel electrophoresis (SDS-10% PAGE), transferred to nitrocellulose membranes, and blotted with antibodies specific for Arf (Ab80; Abcam), p53 (Ab7; Calbiochem), p27^Kip1 (K25020; BD Transduction Labs), PLCγ2 (sc-407; Santa Cruz Biotechnology), RAG1 (sc-363; Santa Cruz Biotechnology), and actin (1378996; Boehringer Mannheim).

Southern blotting.

Genomic DNA was isolated from normal tissue or tumors arising in PLCγ2^+/+ Eμ-Myc, PLCγ2^+/− Eμ-Myc, or PLCγ2^−/− Eμ-Myc transgenic mice. Genomic DNAs (20 μg) were digested with AflII or BamHI, separated in 0.8% agarose gels, transferred to Nytran membranes (Schleicher & Schuell), and probed with the cDNAs coding for Arf (exon 1β) (AflII digested) or p53 (exons 2 to 10) (BamHI digested).

RESULTS

PLCγ2 deficiency results in an increase of large pre-B cells.

PLCγ2 deficiency impedes BCR-mediated B-cell maturation (13, 64). Our recent studies have indicated that PLCγ2 might play a role in early B-cell development (67). To further study the role of PLCγ2 in pre-BCR-mediated early B-cell development, we analyzed in detail the development of bone marrow (BM) B cells in PLCγ2-deficient mice. Based on the expression of the cell surface markers BP-1 and CD24, B220⁺ CD43⁺ B-cell progenitors can be divided into pre-pro-B (fraction A) (B220⁺ CD43⁺ BP-1⁻ CD24⁻), early pro-B (fraction B) (B220⁺ CD43⁺ BP-1⁻ CD24⁺), late pro-B (fraction C) (B220⁺ CD43⁺ BP-1⁺ CD24⁺⁾, and large pre-B (also termed early pre-B) (fraction C′) (B220⁺ CD43⁺ BP-1⁺ CD24^hi) cells (9, 10). FACS analyses of BM cells demonstrated that PLCγ2-deficient mice had marked reductions in late-stage B cells (B220⁺ CD43⁻), which should be due to drastic reductions in mature B (B220^hi IgM⁺) cells (13, 64), and increased proportions of B220⁺ CD43⁺ B-cell progenitors (Fig. 1A). Similar to CD43, c-Kit is also expressed on B-cell progenitors (10). Staining of BM cells with B220 and c-Kit also demonstrated that proportions of B220⁺ c-Kit⁺ B-cell progenitors were increased in PLCγ2^−/− relative to wild-type mice, although the increases were less dramatic than those of B220⁺ CD43⁺ progenitors (Fig. 1B). Within the B220⁺ CD43⁺ cells, PLCγ2^−/− mice displayed marked increases in large pre-B cells (fraction C′; B220⁺ CD43⁺ BP-1⁺ CD24^hi) and decreases in early pro-B cells (fraction B) (Fig. 1C). In contrast, the pre-pro-B (fraction A) and late pro-B (fraction C) populations were comparable in wild-type and PLCγ2^−/− mice (Fig. 1C). In addition, CD25 is only expressed on large and small pre-B cells during B-cell development (10, 50). Although the total population of B220⁺ CD25⁺ B cells was decreased in PLCγ2^−/− relative to wild-type mice (data not shown), the proportions of large pre-B versus small pre-B among this population were increased in mutant mice compared with wild-type mice (Fig. 1D). Therefore, PLCγ2 deficiency impairs early B-cell development, resulting in an increase of large pre-B cells (fraction C′).

FIG. 1.

PLCγ2-deficient mice have a marked increase in the proportions of large pre-B cells. (A) PLCγ2^−/− mice have an increase in B220⁺ CD43⁺ B-cell progenitor populations. Bone marrow cells from the indicated mice were stained with anti-B220 and anti-CD43 antibodies. Percentages indicate cells in the gated live BM B220⁺ populations. Data are representative of at least 12 mice per genotype. (B) PLCγ2^−/− mice have an increase in B220⁺ c-Kit⁺ B-cell progenitor populations. BM cells from the indicated mice were stained with anti-B220 and anti-c-Kit antibodies. Percentages indicate cells in the gated live BM B220⁺ populations. Data are representative of six mice per genotype. (C) PLCγ2^−/− mice have an increase of fraction C′, the large pre-B-cell subpopulation. BM cells from mice of the indicated genotypes were stained with anti-B220, anti-CD43, anti-BP-1, and anti-CD24 antibodies. FACS analysis with BP-1 and CD24 staining of B220⁺ CD43⁺ gated cells is shown. Percentages indicate cells in the gated B220⁺ CD43⁺ populations. Data are representative of 12 mice per genotype. (D) PLCγ2^−/− mice have increased proportions of large pre-B versus small pre-B cells among B220⁺ CD25⁺ cells. BM cells from the indicated mice were stained with anti-B220 and anti-CD25 antibodies and examined by forward light-scatter analysis. Percentages indicate cells in the gated B220⁺ CD25⁺ populations. Data are representative of six mice per genotype. (E) PLCγ2^−/− BM B cells display augmented rates of proliferation in response to IL-7. B220⁺ B cells were purified from BM derived from PLCγ2^+/+ and PLCγ2^−/− mice, and their rates of proliferation in response to IL-7 were determined by the incorporation of [³H]thymidine. Data are representative of three independent experiments. (F) PLCγ2^−/− large pre-B cells display augmented rates of [³H]thymidine incorporation in response to IL-7. B220⁺ CD43⁺ BP-1⁺ CD24^hi large pre-B cells were sorted from BM derived from PLCγ2^+/+ and PLCγ2^−/− mice, and their rates of proliferation in response to IL-7 were determined by the incorporation of [³H]thymidine. Data are representative of three independent experiments. (G) PLCγ2^−/− large pre-B cells display augmented rates of BrdU incorporation in response to IL-7. B220⁺ CD43⁺ BP-1⁺ CD24^hi large pre-B cells were sorted from BM derived from PLCγ2^+/+ and PLCγ2^−/− mice, and their rates of proliferation in response to IL-7 were determined by the incorporation of BrdU. Data are representative of two independent experiments.

Early pro-B, late pro-B, and large pre-B progenitors proliferate in response to IL-7 stimulation (9, 10). In accord with increased proportions of large pre-B cells in PLCγ2^−/− mice, B cells purified from BM of PLCγ2^−/− mice proliferated at a faster rate (a three- to fourfold increase in [³H]thymidine incorporation) than wild-type total BM B cells in response to IL-7 (Fig. 1E). These results are consistent with the previous finding that there is an ∼10-fold increase in the numbers of bone marrow-derived IL-7-responsive B-cell colonies in PLCγ2^−/− mice (64). Furthermore, PLCγ2^−/− large pre-B cells displayed increased IL-7-induced proliferation relative to wild-type cells (Fig. 1F). Thus, increases in [³H]thymidine incorporation of PLCγ2^−/− BM B cells are due to increases in both the number of fraction C′ cells and their responsiveness to IL-7. In addition, BrdU incorporation experiments demonstrated that PLCγ2^−/− large pre-B cells displayed markedly increased proliferation relative to wild-type cells in response to IL-7 (Fig. 1G). PLCγ2^−/− early pro-B cells (fraction B) also exhibited increased, albeit to a lesser extent, IL-7-induced proliferation relative to wild-type cells (data not shown). Therefore, PLCγ2-deficient bone marrow B cells harbor increased numbers of large pre-B cells that are hyperresponsive to IL-7.

PLCγ2 deficiency impairs pre-BCR-mediated functions.

Signals emanating from functional pre-BCR receptors downregulate IL-7 signaling following the transition of pre-BCR-negative late pro-B cells (fraction C) to pre-BCR-positive large pre-B cells (fraction C′) (9, 10, 34, 66). To initially assess whether the IL-7 hyperresponsiveness of fraction C′ PLCγ2-deficient B cells was due to alterations of this regulatory loop, we determined the growth potential of B-cell progenitor subsets derived from the in vitro culture of BM from wild-type and PLCγ2^−/− mice. Interestingly, the in vitro culture of wild-type BM predominately gave rise to IL-7-responsive B220⁺ CD43⁺ κ⁻ λ⁻ B-cell progenitors (Fig. 2A to C). These progenitors were BP-1⁺ CD24^hi large pre-B cells (Fig. 2D) that were pre-BCR⁻ (pre-BCR is detected with a monoclonal antibody [SL156] that recognizes the surrogate light chain component λ5 in association with μ protein) (69) (Fig. 2E), indicating a selection for the growth of progenitors that were at an intermediate developmental stage between late pro-B (fraction C) and large pre-B (fraction C′) cells. In contrast, the IL-7 culture-derived progenitors from PLCγ2^−/− BM were pre-BCR⁺ (Fig. 2E), indicating a selection for a distinct intermediate developmental stage that again lies between fraction C and C′ cells.

FIG. 2.

PLCγ2-deficient BM cells emerge as highly IL-7-responsive large pre-B cells following in vitro IL-7 culture. BM cells from the indicated mice were cultured in IL-7-containing media for 8 days and then stained with a combination of antibodies to B220, κ, and λ, to B220, CD43, BP-1, and CD24, and to B220 and pre-BCR (SL156). (A) FACS analysis of B220 and CD43 expression. Percentages indicate cells in B220⁺ gated cells. (B) FACS analysis of B220 and κ expression. Percentages indicate cells in the gated live cells. (C) FACS analysis of B220 and λ expression. Percentages indicate cells in the gated live cells. (D) FACS analysis of BP-1 and CD24 expression in B220⁺ CD43⁺ cells. Percentages indicate cells in the gated B220⁺ CD43⁺ populations. (E) FACS analysis of B220 and pre-BCR (SL156) expression. Percentages indicate cells in the gated live cells. Data are representative of three independent experiments.

These findings suggested that the pre-BCR fails to downregulate IL-7 signaling in the absence of PLCγ2 and that this allows for the generation of an IL-7-hyperresponsive/pre-BCR⁺ population of large pre-B progenitors. Thus, we addressed whether the absence of PLCγ2 impairs pre-BCR signaling and subsequently affects the downregulation of IL-7 signaling. BM cells from wild-type and PLCγ2^−/− mice were cultured in vitro with IL-7. After 8 days in culture, the levels of IL-7 receptor expression on the emerged B-cell progenitors were examined by FACS analysis. As expected, IL-7 receptors were detected in wild-type IL-7-responsive pre-BCR-negative late pro-B progenitors, yet their expression was increased in pre-BCR-positive large pre-B PLCγ2^−/− progenitors (Fig. 3A). Furthermore, the proliferative response of the pre-BCR-positive large pre-B PLCγ2^−/− progenitors to IL-7 was augmented (approximately two- to threefold increase) in limiting dosages of IL-7 (Fig. 3B). Therefore, in the absence of PLCγ2, the pre-BCR fails to downregulate IL-7 receptors and their signaling, which renders these cells hyperresponsive to IL-7.

FIG. 3.

PLCγ2 deficiency impairs pre-BCR-mediated functions. (A) PLCγ2 deficiency impairs pre-BCR-mediated downregulation of IL-7 receptors. BM cells from indicated mice were cultured in vitro with IL-7 for 8 days as described in the legend to Fig. 2. The levels of IL-7 receptor expression on the emerged B-cell progenitors were then determined by FACS analysis with anti-IL-7Rα antibodies. (B) BM culture-derived PLCγ2^−/− large pre-B cells are hyperresponsive to IL-7. BM IL-7 culture-derived B-cell progenitors were obtained as described above. Their rates of proliferation in response to IL-7 were then determined by the incorporation of [³H]thymidine. (C) PLCγ2 deficiency impairs pre-BCR-mediated Ca²⁺ flux. BM cells derived from the indicated mice were stained with antibodies to B220 as well as to κ and λ chains, followed by incubation with indo-1^AM. Cells were then washed and stimulated with anti-μ antibodies. Pre-BCR-induced Ca²⁺ flux was determined in B220⁺ κ⁻ λ⁻ gated B-cell progenitors by flow cytometry. Anti-μ antibodies were added at times indicated by the arrows. (D) PLCγ2 deficiency impairs Ig^HEL-mediated downregulation of mRNA of RAGs in IL-7-responsive B-cell progenitors. BM cells from PLCγ2^+/+ Ig^HEL and PLCγ2^−/−Ig^HEL transgenic mice were cultured in IL-7-containing media for 5 days. Total mRNA extracted from the cells was subjected to semiquantitative RT-PCR using primers designed to detect RAG1 or RAG2. RT-PCR products from the β-actin gene served as controls for the quantity of the mRNA. (E) Impairment of Ig^HEL-mediated downregulation of mRNA of RAGs in IL-7-responsive B-cell progenitors by PLCγ2 deficiency is detected by real-time RT-PCR. The cDNAs derived in experiments depicted in panel D were subjected to real-time PCR using primers designed to detect RAG1 or RAG2. The data are expressed as the ratio of RAG1 or RAG2 transcript to CD19 transcript. Data are representative of four independent experiments. (F) PLCγ2 deficiency impairs Ig^HEL-mediated downregulation of RAG1 proteins in IL-7-responsive B-cell progenitors. Total cell lysates from the cells derived in the experiments depicted in panel D were subjected to SDS-PAGE and direct Western blot analysis with anti-RAG1 or anti-α-actin antibodies. (G) PLCγ2 deficiency impairs pre-BCR-mediated downregulation of RAG. BM IL-7 culture-derived B-cell progenitors from PLCγ2^+/+ and PLCγ2^−/− mice were obtained as described in the legend to Fig. 2. Total cell lysates were subjected to SDS-PAGE and direct Western blot analysis with anti-RAG1 or anti-α-actin antibodies. (H) PLCγ2 deficiency impairs Ig^HEL-mediated downregulation of VpreB and λ5 in BM B cells. BM cells from PLCγ2^+/+ Ig^HEL and PLCγ2^−/− Ig^HEL transgenic mice were stained with a combination of antibodies to CD19, VpreB, and λ5. Percentages indicate cells in CD19⁺ gated cells. Data are representative of two mice per genotype.

Another hallmark of pre-BCR ligation is the induction of intracellular Ca²⁺ flux (17, 53). We therefore also compared pre-BCR-induced Ca²⁺ flux in B-cell progenitors derived from the BM cells of wild-type and PLCγ2^−/− mice. B220⁺ cells from wild-type and PLCγ2^−/− mice were gated on their κ and λ expression status, and the pre-BCR of B220⁺ κ⁻ λ⁻ gated (pre-B) B-cell progenitors was engaged with antibody to μ chains, a component of the pre-BCR receptor (6). In wild-type pre-B-cell progenitors, pre-BCR engagement induced the expected flux in intracellular Ca²⁺, whereas PLCγ2 deficiency dramatically reduced this response (Fig. 3C). Therefore, PLCγ2 deficiency impairs pre-BCR-mediated Ca²⁺ mobilization.

During B-cell development, rearrangements of the Ig heavy and light chain loci are mediated by recombination activating enzymes RAG1 and RAG2. The expression of these two proteins is also tightly controlled during B-cell development, as they are highly expressed in early and late pro-B cells, are low in pre-BCR-positive large pre-B cells, and are then elevated again in small pre-B cells (8, 31). Signals from the pre-BCR mediates downregulation of RAGs in pre-BCR-positive early pre-B cells (8). Thus, we examined whether PLCγ2 deficiency also impaired pre-BCR-mediated downregulation of RAGs. The mixed subpopulations of BM B cells in both wild-type and PLCγ2^−/− mice hindered these analyses. To overcome this difficulty, we employed Ig^HEL transgenic mice, in which BM B cells are largely a single population and uniformly express the hen egg lysozyme (HEL)-specific BCR (7). BM cells from wild-type Ig^HEL and PLCγ2^−/− Ig^HEL transgenic mice were cultured in vitro with IL-7 for 5 days, and the levels of RAG1 and RAG2 expression on the emerging B-cell progenitors were determined by semiquantitative RT-PCR. Expression of both RAG1 and RAG2 was low in wild-type IL-7-responsive Ig^HEL-positive B-cell progenitors, yet it was markedly increased in PLCγ2^−/− IL-7-responsive Ig^HEL-positive progenitors (Fig. 3D). The elevation of RAG1 and RAG2 in PLCγ2^−/− IL-7-responsive Ig^HEL-positive progenitors was confirmed by real-time PCR (Fig. 3E). In addition, the elevation of RAG1 in PLCγ2^−/− IL-7-responsive Ig^HEL-positive progenitors was also confirmed by immunoblot analysis (Fig. 3F). Thus, in the absence of PLCγ2, HEL-specific BCR fails to downregulate RAGs in B-cell progenitors. Moreover, failure of pre-BCR to downregulate RAG without PLCγ2 was confirmed in PLCγ2^−/− progenitors. BM cells from wild-type and PLCγ2^−/− mice were cultured in vitro with IL-7 for 8 days, and the levels of RAG1 expression in the emerging B-cell progenitors were determined by immunoblot analysis. Expression of RAG1 was markedly increased in pre-BCR-positive large pre-B PLCγ2^−/− progenitors relative to wild-type IL-7-responsive pre-BCR-negative late pro-B progenitors (Fig. 3G). Therefore, PLCγ2 deficiency leads to a failure of the pre-BCR to downregulate RAGs.

During B-cell development, signals from the pre-BCR also mediate downregulation of VpreB and λ5 (44, 65). Thus, we examined whether PLCγ2 deficiency also impaired pre-BCR-mediated downregulation of VpreB and λ5 by examining BM cells from PLCγ2^−/− Ig^HEL transgenic mice. HEL-specific BCR expression is initiated in pro-B cells, and nearly all BM B cells from wild-type Ig^HEL and PLCγ2^−/− Ig^HEL transgenic mice were Ig^HEL positive (data not shown). As expected, few PLCγ2^+/+ Ig^HEL BM B cells expressed VpreB or λ5 on the cell surface, at least by conventional staining techniques (65) (Fig. 3H). In contrast, the vast majority of PLCγ2^−/− Ig^HEL BM B cells expressed VpreB or λ5 (Fig. 3H). PLCγ2^−/− Ig^HEL BM B cells also expressed higher levels of VpreB and λ5 protein in their cytoplasm (data not shown). Thus, PLCγ2 deficiency also impairs pre-BCR-mediated downregulation of VpreB or λ5.

PLCγ2 deficiency accelerates Myc-induced lymphomagenesis.

PLCγ2 deficiency results in an increase of large pre-B cells with hyperresponsiveness to IL-7 and high recombinogenic activities. These PLCγ2-deficient large pre-B cells are likely at risk for mutations that lead to transformation. Although spontaneous B-cell lymphomas have not been detected in PLCγ2^−/− mice, we reasoned that the susceptibility of PLCγ2-deficient large pre-B cells to transformation might be revealed in the context of other lesions that promote lymphomagenesis. To test this hypothesis, we examined the contribution of PLCγ2 to B-cell lymphomagenesis in Eμ-Myc transgenic mice, which overexpress c-Myc in B cells by virtue of the IgH Eμ enhancer and die of clonal pre-B- and B-cell lymphoma beginning at 4 months of age (1).

Congenic C57Bl/6 Eμ-Myc transgenic mice were bred to C57Bl/6 PLCγ2^+/− mice, and the PLCγ2^+/− Eμ-Myc F₁ offspring were bred with PLCγ2^+/− mice to obtain PLCγ2^+/+, PLCγ2^+/−, and PLCγ2^−/− Eμ-Myc mice. These littermates were then monitored for their course of disease. Wild-type Eμ-Myc littermates displayed a mortality curve typical for these mice, with a mean mortality of 19 weeks (Fig. 4A) (1). Notably, there were no effects of PLCγ2 heterozygous deficiency on tumor development, as PLCγ2^+/− Eμ-Myc mice died at rates comparable to those of wild-type transgenics (Fig. 4A), suggesting PLCγ2 did not behave as a classic tumor suppressor. Nonetheless, PLCγ2^−/− Eμ-Myc transgenics displayed a greatly accelerated course of disease, with a mean mortality of 10 weeks (Fig. 4A). By 8 weeks of age, most PLCγ2^−/− Eμ-Myc transgenics had massively enlarged lymph nodes, a hallmark of Eμ-Myc-induced lymphoma (1), whereas this was not evident in wild-type Eμ-Myc and PLCγ2^+/− Eμ-Myc littermates (Fig. 4B and data not shown). Further, FACS analyses showed that cells from the lymph nodes of control wild-type, PLCγ2^−/−, and precancerous wild-type Eμ-Myc mice were small lymphocytes as expected, whereas cells from the lymph nodes of 8-week-old diseased PLCγ2^−/− Eμ-Myc transgenics were mainly large B cells (Fig. 4C and D). Therefore, loss of PLCγ2 accelerates Myc-induced lymphomagenesis.

FIG. 4.

PLCγ2 deficiency accelerates Myc-induced lymphomagenesis. (A) Kaplan-Meier survival curves of PLCγ2^+/+, PLCγ2^−/−, PLCγ2^+/+ Eμ-Myc, PLCγ2^+/− Eμ-myc, and PLCγ2^−/− Eμ-Myc mice. The genotypes of the mice are indicated next to the survival curves, and the numbers of mice in each group are indicated by the n values. Vertical lines indicate ages of surviving mice. Although PLCγ2 deficiency alone led to some mortality, this appears due to immune deficiency and bleeding (64), and the majority of PLCγ2^−/− mice survived more than 40 weeks when supplemented with the antibiotic sulfatrim (a condition under which all mice were maintained). (B) PLCγ2^−/− Eμ-Myc transgenic mice develop rapid lymphoma. A substantial portion of PLCγ2^−/− Eμ-Myc mice (8 weeks old) had enlarged lymph nodes, whereas PLCγ2^+/+ Eμ-Myc mice (8 weeks old) had normal lymph nodes (upper). Examination of these same mice displayed massive enlargement of axillary and inguinal lymph nodes of PLCγ2^−/− Eμ-Myc mice (lower). (C) Lymphoma arising in the lymph nodes derived from PLCγ2^−/− Eμ-Myc mice consists of large cells. Lymph node cells from 8-week-old PLCγ2^+/+, PLCγ2^−/−, PLCγ2^+/+ Eμ-Myc, and PLCγ2^−/− Eμ-Myc mice were examined by forward and side light-scatter analysis. The very large lymphoma cells arising in the lymph nodes are typical of PLCγ2^−/− Eμ-Myc mice. (D) Lymphoma arising in the lymph nodes derived from PLCγ2^−/− Eμ-Myc mice is a B-cell type. Lymph node cells from the indicated mice were stained with anti-B220 and anti-Thy1.2. Percentages indicate cells in the gated live cell populations. Data are representative of at least 12 mice per genotype.

PLCγ2 deficiency selects for transformation of large pre-B cells.

If increases of at-risk PLCγ2-deficient large pre-B cells account for the acceleration of Myc-induced lymphomagenesis by PLCγ2 deficiency, the lymphomas arising in PLCγ2^−/− Eμ-Myc mice should be mainly large pre-B-cell types. Previous studies have suggested that lymphomas arising in C57Bl/6 wild-type Eμ-Myc mice are at various stages of B-cell maturation (13, 64). Here, we determined the developmental origins of B-cell lymphomas by detailed FACS analyses. Indeed, FACS analyses of lymphomas arising in 18 wild-type Eμ-Myc littermates demonstrated that 10 of them represented later B-cell types that were B220⁺ CD43⁻ (Fig. 5A, first panel; Table 1).Among the 10 lymphomas, 5 were μ⁺ κ⁺, an immature/mature type, and 5 were μ⁻ κ⁻, an aberrant type (Table 1). On the other hand, 8 of the 18 lymphomas were of B220⁺ CD43⁺ early pro-B-cell lineage (Fig. 5A, second panel), as 3 of them were B220⁺ CD43⁺ BP-1⁻ CD24⁺ SL156⁻ μ⁻ κ⁻, a true early pro-B-cell type (Table 1), 4 of them were B220⁺ CD43⁺ BP-1⁻ CD24⁺ SL156⁻ μ⁺ κ⁺, an aberrant early pro B-cell type (Table 1), and 1 was B220⁺ CD43⁺ BP-1⁺ CD24⁺ SL156⁺ μ⁻ κ⁻, a large pre-B-cell type (Table 1).

FIG. 5.

PLCγ2-deficient Eμ-Myc mice exclusively develop lymphomas of large pre-B-cell origin. (A) Lymphoma cells from PLCγ2^−/− Eμ-Myc mice are exclusively B220⁺ CD43⁺ B-cell progenitors. Lymphoma cells from the indicated mice were stained with anti-B220 and anti-CD43. Percentages indicate cells in the gated live cell populations. Data are representative of 18 mice per genotype. (B) Lymphoma cells from the vast majority of PLCγ2^−/− Eμ-Myc mice are of B220⁺ CD43⁺ BP-1⁺ CD24^hi large pre-B-cell origin. Lymphoma cells from PLCγ2^−/− Eμ-Myc mice were stained with anti-B220, anti-CD43, anti-BP-1, and anti-CD24 antibodies. FACS analysis in the gated B220⁺ CD43⁺ populations is shown. Percentages indicate cells in the gated B220⁺ CD43⁺ populations. Data are representative of 15 B220⁺ CD43⁺ PLCγ2^−/− Eμ-Myc lymphomas. (C) Lymphoma cells from PLCγ2^−/− Eμ-Myc mice are pre-BCR⁺. Lymphoma cells from PLCγ2^−/− Eμ-Myc mice were stained with anti-B220 and anti-pre-BCR (SL156). Percentages indicate cells in the gated live cell populations. Data are representative of 18 PLCγ2^−/− Eμ-Myc lymphomas. (D) Lymphoma cells from PLCγ2^−/− Eμ-Myc mice are μ⁺ κ⁺. Lymphoma cells from PLCγ2^−/− Eμ-Myc mice were stained with anti-B220, anti-μ, and anti-κ. Percentages indicate cells in the gated B220⁺ populations. Data are representative of 18 PLCγ2^−/− Eμ-Myc lymphomas. (E) Lymphoma cells from PLCγ2^−/− Eμ-Myc mice respond to IL-7. The rates of proliferation of lymphoma cells from the indicated individual mouse in the absence (−) or presence (IL-7) of IL-7 were determined by the incorporation of [³H]thymidine. Data are representative of four PLCγ2^+/+ Eμ-Myc and four PLCγ2^−/− Eμ-Myc lymphomas from individual mice.

TABLE 1.

Lymphomas arising in PLCγ2^+/+ Eμ-Myc and PLCγ2^−/− Eμ-Myc mice^a

Genotype	No.	Phenotype	Cell type
PLCγ2+/+ Eμ-Myc	5	B220+ CD43− μ+ κ+	Immature/mature
	5	B220+ CD43− μ− κ−	Aberrant
	2	B220+ CD43+ BP-1− CD24+ SL156− μ− κ−	Early pro-B
	4	B220+ CD43+ BP-1− CD24+ SL156− μ+ κ+	Early pro-B (aberrant)
	1	B220+ CD43+ BP-1+ CD24+ SL156− μ− κ−	Late pro-B
	1	B220+ CD43+ BP-1+ CD24+ SL156+ μ− κ−	Large pre-B
PLCγ2−/− Eμ-Myc	15	B220+ CD43+ BP-1+ CD24+ SL156+ μ+ κ+	Large pre-B origin
	3	B220+ CD43+ BP-1− CD24+ SL156+ μ+ κ+	Large pre-B (aberrant)

^aLymphomas arising in 18 PLCγ2^+/+ Eμ-Myc mice and 18 PLCγ2^−/− Eμ-Myc mice were stained with different combinations of anti-B220, anti-CD43, anti-BP-1, anti-CD24, anti-pre-BCR (SL156), anti-μ, and anti-κ antibodies.

Analysis of the same panel of markers for B cells demonstrated that the lymphoma cells derived from all of the 18 examined PLCγ2^−/− Eμ-Myc mice were of B220⁺ CD43⁺ B-cell progenitor origin (Fig. 5A, third panel). The vast majority of the lymphomas (15 of 18) were defined as BP-1⁺ CD24^hi large pre-B cells (Fig. 5B; Table 1). The large pre-B-cell origin of the lymphomas arising in PLCγ2^−/− Eμ-Myc mice was confirmed by the fact that these lymphoma cells were largely pre-BCR⁺ (Fig. 5C; Table 1). Of note, 3 of the 18 lymphomas were BP-1⁻ CD24⁺ pre-BCR⁺, an aberrant large pre-B-cell type (Table 1). In addition, all of the early pre-B-cell-origin lymphoma cells from PLCγ2^−/− Eμ-Myc transgenics were not only μ⁺ but were also κ⁺ (Fig. 5D), demonstrating a continuous rearrangement of light chain loci. This finding is consistent with the observation that PLCγ2-deficient large pre-B cells expressed high levels of RAGs (Fig. 3D to G). Therefore, although the lymphomas from PLCγ2-deficient Eμ-Myc transgenic mice were of large pre-B-cell origin, they were unusual in that they also expressed κ⁺ light chains.

Moreover, lymphoma cells arising in most wild-type Eμ-Myc mice (four of five examined) proliferated comparably in the absence and presence of IL-7 (Fig. 5E). In contrast, IL-7 further markedly enhanced proliferation of lymphoma cells arising in all PLCγ2^−/− Eμ-Myc mice we examined (Fig. 5E), demonstrating expression of IL-7 receptors on the PLCγ2^−/− Eμ-Myc lymphoma cells. Taken together, the PLCγ2 deficiency selectively leads to transformation of IL-7-responsive B220⁺ CD43⁺ BP-1⁺ CD24^hi pre-BCR⁺ large pre-B cells, and these are highly prone to unscheduled immunoglobulin gene rearrangements.

Lymphomas from PLCγ2^−/− Eμ-Myc transgenic mice exhibited a loss of p27^Kip1 and often displayed alterations in Arf or p53.

Myc accelerates cell proliferation at least in part through its ability to promote degradation of the cyclin-dependent kinase (Cdk) inhibitor p27^Kip1 (2, 39, 41, 45, 63), and loss of p27^Kip1 accelerates Myc-induced lymphomagenesis (36). We therefore evaluated whether the effects of the PLCγ2 deficiency in accelerating Myc-induced lymphoma development could be attributed to its effects upon p27^Kip1. As expected, p27^Kip1 levels were reduced in precancerous Eμ-Myc B cells compared to those present in wild-type B cells, but loss of PLCγ2 alone had no effect on p27^Kip1 levels or on the reduced levels of p27^Kip1 expressed in Eμ-Myc B cells )(Fig. 6A.Nevertheless, the Myc-to-p27^Kip1 pathway was targeted in lymphomas, as levels of p27^Kip1 were essentially undetectable in lymphomas derived from both wild-type Eμ-Myc and PLCγ2^−/− Eμ-Myc transgenic mice (Fig. 6A). Therefore, suppression of p27^Kip1 is a hallmark of lymphoma development in Eμ-Myc transgenic mice, and in the absence of PLCγ2, Myc still induces lymphoma development by targeting p27^Kip1. B-cell lymphomas lacking p27^Kip1 expression arise within 1 to 2 months in PLCγ2^−/− Eμ-Myc mice versus 4 to 6 months in wild-type Eμ-Myc mice.

FIG. 6.

PLCγ2 deficiency accelerates the rate of loss of p27^Kip1 and mutations in the Arf-p53 pathway. (A) Expression of p27^Kip1, p53, Arf, and PLCγ2 in lymphomas arising in PLCγ2^+/+ Eμ-Myc, PLCγ2^+/− Eμ-Myc, and PLCγ2^−/− Eμ-Myc mice. Total cell lysates of pro/pre and lymph node (LN) B cells and of lymphomas from the indicated mice were subjected to SDS-PAGE and Western blot analysis with the indicated antibodies. Lymphomas harboring p53 mutations and/or overexpressing Arf protein are indicated by an asterisk. (B) Southern blot analysis of the lymphomas. Genomic DNAs were extracted from the indicated lymphomas, digested with AflII, and hybridized with an Arf probe that detects a fragment containing Arf exon 1β. BamHI-digested DNAs were hybridized with a p53 probe that detects a fragment containing p53 exons 2 to 10. Samples with deletion of p53 or Arf alleles are indicated by an asterisk. Lymphomas from 14 PLCγ2^+/+ Eμ-Myc or PLCγ2^+/− Eμ-Myc mice and from 21 PLCγ2^−/− Eμ-Myc mice were analyzed.

Mutations that disable p53 or Arf functions are another hallmark of B-cell lymphoma in Eμ-Myc transgenic mice (5) and in human Burkitt lymphoma (16, 32, 68). We therefore evaluated whether the effects of the PLCγ2 deficiency in accelerating Myc-induced lymphoma development could also be attributed to its effects upon p53 or Arf. p53's functions are usually disrupted by missense hot-spot point mutations of a single allele, which give rise to dominant-negative forms of the p53 protein, whereas Arf is inactivated by biallelic deletions (26, 30). Mutant p53 proteins are unable to transcriptionally induce Mdm2 to initiate their own destruction, and this results in high levels of mutant p53 protein (14, 27). Indeed, Western blot analyses demonstrated high levels of p53 protein in 2 of 21 tumors (∼10%) arising in PLCγ2-deficient Eμ-Myc mice (Fig. 6A), and sequencing of p53 cDNAs amplified from these tumors confirmed missense mutations of p53 (R172H and R277H), which correspond to hot-spot mutations of p53 found in human tumors (26). Arf is transcriptionally repressed by p53, and thus loss of p53 function results in dramatic increases in the levels of Arf protein (23, 49, 60). Accordingly, levels of Arf were very high in tumors having mutant p53, but another lymphoma (C1192) also displayed high levels of Arf protein (Fig. 6A). Sequencing analysis established that this was associated with a mutant splice variant of p53 that deleted its 90 C-terminal residues due to a premature termination. Thus, p53 mutants were observed in about 15% of Myc-induced lymphomas in PLCγ2-deficient mice, a frequency slightly lower than that (24 to 38%) observed in wild-type Eμ-Myc lymphomas (5). Southern blot analyses revealed that 4 of 21 tumors from PLCγ2-deficient Eμ-Myc mice had deletions of both Arf alleles (Fig. 6B), a frequency comparable to that (20 to 25%) which occurs in wild-type Eμ-Myc lymphomas (Fig. 6B) (5). Of note, PLCγ2 expression was sustained in lymphomas arising in PLCγ2^+/− Eμ-Myc mice (Fig. 6A), indicating that PLCγ2 did not behave as a classic tumor suppressor. Collectively these findings indicate that the PLCγ2 deficiency has little effect on the frequency of disruption of the Arf-p53 suppressor pathway, and B-cell lymphomas with disrupted Arf or p53 arise within 1 to 2 months in PLCγ2^−/− Eμ-Myc mice, versus 4 to 6 months in wild-type Eμ-Myc mice.

DISCUSSION

PLCγ2 is predominantly expressed in B cells and is activated following ligation of the pre-BCR or BCR (3, 17). The PLCγ2 deficiency severely impairs the late stages of B-cell development and abolishes the mitogenic response following BCR engagement (13, 64), yet the studies reported herein revealed that the PLCγ2 deficiency also has profound effects on early B-cell development, as there are substantial increases in the B220⁺ CD43⁺ BP-1⁺ CD24^hi pre-BCR⁺ fraction C′ cells, the large recycling pre-B cells (11). Increases in this compartment are indicative of an impaired transition of large recycling CD43⁺ pre-B cells to small resting CD43⁻ pre-B cells (11). Previous studies have demonstrated that functions of IL-7 receptors are not impaired by the absence of PLCγ2 (67), and indeed here we have shown that the IL-7 response is augmented in pre-BCR-positive large pre-B PLCγ2^−/− progenitors. Thus, the pre-BCR requires PLCγ2 to direct this developmental transition, and in the absence of PLCγ2 there is an expansion of the IL-7-responsive large pre-B cells. Interestingly, a blockade in the transition of large recycling CD43⁺ pre-B cells to small resting CD43⁻ pre-B cells is also observed in mice lacking the Btk or SLP-65 signaling effector, which functions immediately upstream of PLCγ2 in directing pre-BCR and BCR signals (22, 37, 43, 70), and Btk- or SLP-65-deficient large pre-B cells are also hyperresponsive to IL-7 (6, 25).

During the progression of pre-BCR⁻ late pro-B (fraction C) cells to pre-BCR⁺ large pre-B (fraction C′) cells, signals emanating from newly formed pre-BCR not only promote B-cell differentiation but also rapidly downregulate IL-7 receptor signaling (9, 10, 34, 66). PLCγ2 deficiency impairs pre-BCR signaling, and this results in a failure to downregulate IL-7 receptors such that the in vitro culture of BM progenitors from PLCγ2^−/− mice in IL-7 generates B220⁺ CD43⁺ BP-1⁺ CD24^hi pre-BCR⁺ large pre-B-cell progenitors that have elevated, rather than reduced, levels of IL-7 receptors. These IL-7-responsive/pre-BCR⁺ B-cell progenitors may represent an intermediate developmental stage between late pro-B (fraction C) and large pre-B (fraction C′) cells, which might exist only momentarily and are thus not readily detected in wild-type mice. Finally, these PLCγ2^−/− large pre-B cells have dramatically increased capacities in IL-7-mediated proliferation, and this is associated with marked increases in B220⁺ CD43⁺ BP-1⁺ CD24^hi pre-BCR⁺ large pre-B-cell progenitors in PLCγ2^−/− mice (64). Similarly, Btk- and SLP-65-deficient bone marrow B cells show an enhanced proliferative expansion in response to IL-7 in vitro, consistent with increases in large pre-B cells observed in these knockout mice (6, 37). Given these similarities, we predict that the Btk and SLP-65 deficiencies also generate increased numbers of B220⁺ CD43⁺ BP-1⁺ CD24^hi pre-BCR⁺ large pre-B-cell progenitors that have augmented levels of IL-7 receptors.

In vivo, large recycling pre-B cells are the most rapidly proliferating subset of B-cell progenitors (11) and are inherently susceptible to genetic insults. The high susceptibility of this compartment is caused by the recombination machinery driven by the expression of the RAG recombinases in these cells (19, 31, 71). Normally, the expression of RAG proteins is tightly controlled and undergoes three waves of expression, being high in early and late pro-B cells, low in pre-BCR⁺ large pre-B cells, and then high again in small pre-B cells (8, 31). Signals from the pre-BCR downregulate the expression of RAGs in pre-BCR⁺ large pre-B cells (8). Interestingly, PLCγ2 is here revealed to be required for suppressing RAG expression in pre-BCR⁺ large pre-B cells, and this finding is in accord with the rapidly arising lymphomas of PLCγ2^−/− Eμ-Myc transgenic mice, which originate exclusively from large pre-B cells (B220⁺ CD43⁺ BP-1⁺ CD24^hi pre-BCR⁺) and that have unscheduled rearrangements of κ light chain loci. RAG activity has been linked to an increased tendency for chromosomal translocations and for tumor development (72). Despite the evidence that high levels of RAG activities might play a role in the PLCγ2 deficiency-caused acceleration of Myc-induced lymphomagenesis, we couldn't directly examine the role of RAG in this process by deletion of RAG in PLCγ2^−/− Eμ-Myc mice, as the RAG deficiency results in a complete absence of pre-B cells (38, 57), which would prevent large pre-B-cell lymphoma formation in PLCγ2^−/− Eμ-Myc mice. However, introduction of the Ig^HEL transgene into RAG^−/− PLCγ2^−/− Eμ-Myc and RAG^+/+ PLCγ2^−/− Eμ-Myc mice and subsequent studies of the kinetics of development of lethal B-cell lymphomas in RAG^−/− PLCγ2^−/− Eμ-Myc Ig^HEL and RAG^+/+ PLCγ2^−/− Eμ-Myc Ig^HEL mice would address the role of RAG in the acceleration of Myc-induced lymphomagenesis by PLCγ2 deficiency and would be warranted. Regardless, our findings are consistent with a model where highly proliferating and genetically unstable large pre-B cells are at a very high risk for progression to frank lymphoma once oncogenic insults arise. Further, these findings suggest that mice lacking Btk and/or SLP-65 would also display accelerated rates of large pre-B-cell lymphoma development in the face of an oncogenic insult such as Myc.

Despite these findings, we fail to observe spontaneous B-cell lymphomas in PLCγ2-deficient mice (although splenomegaly is frequently evident). At face value these findings would appear to contrast with those reported for SLP-65-deficient mice, as these mice have been reported to develop pre-B-cell lymphomas (6). However, the reported frequency of these tumors in SLP-65-deficient mice is also low, and differences in tumor incidence could be easily attributed to different strain backgrounds (the PLCγ2-deficient mice used here are on a C57Bl/6 background, whereas SLP-65-deficient mice were on a BALB/c background (6, 22). Furthermore, Btk-deficient mice are not prone to the development of B-cell lymphomas (25). Alternatively, the SLP-65 deficiency could be more severe in impairing the transition from large cycling pre-B cells to small resting pre-B cells than that observed in PLCγ2- and Btk-deficient mice, a scenario that would increase the numbers of at-risk cells that are prone to transformation. In support of this notion, mice lacking both Btk and SLP-65, which have arrested early B-cell development at the large cycling pre-B-cell stage, are highly prone to lymphoma development (25). Collectively, these findings strongly support the simple concept that the larger the compartment of proliferating large pre-B cells, the higher the risk of developing lymphoma.

Myc's ability to provoke tumorigenesis is held in check by the p27^Kip1 Cdk inhibitor that restricts G₁- to S-phase cell cycle progression (2, 39, 41, 45, 63) and by the Arf-p53 pathway that provokes apoptosis (5, 56). Loss of PLCγ2 has essentially no effect on the overall frequency of mutations in the Arf-p53 pathway. Nonetheless, B-cell lymphomas with disrupted Arf or p53 arise within 1 to 2 months in PLCγ2^−/− Eμ-Myc mice versus 4 to 6 months in wild-type Eμ-Myc mice. It seems that PLCγ2 deficiency might accelerate the rate of disruption of this tumor suppressor pathway during in vivo lymphomagenesis. However, it is equally possible that the rate of Arf or p53 disruption is unchanged, but the overall probabilities increase due to enlarged pools and/or enhanced proliferation rates of target cells in the absence of PLCγ2. In addition, loss of p27^Kip1 is also revealed here as a hallmark of lymphoma development, but this occurs regardless of PLCγ2 status. B-cell lymphomas lacking p27^Kip1 expression arise within 1 to 2 months in PLCγ2^−/− Eμ-Myc mice versus 4 to 6 months in wild-type Eμ-Myc mice. Again, the rate of p27^Kip1 suppression could be hastened by PLCγ2 deficiency, or, equally possible, the probabilities but not the rate of p27^Kip1 disappearance increase as pools and/or proliferation rates of target cells increase in the absence of PLCγ2.

Most importantly, these findings support the notion that signals orchestrated by the pre-BCR serve as guardians against B-cell transformation by holding proliferation and genome stability in check and by limiting the numbers of cells that are at highest risk for transformation. In this light PLCγ2 could be cast as a tumor suppressor, as suggested for SLP-65 (25). However, cells heterozygous for mutations in classic tumor suppressors such as Rb, p53, and Arf nearly always display loss of the wild-type allele or silencing of the gene, and as a consequence there are always significant effects of heterozygous deficiency on tumor formation. Such is not the case for Eμ-Myc mice heterozygous deficient for PLCγ2, which develop lymphomas at the same pace as wild-type transgenics, do not display loss of the wild-type PLCγ2 allele, and which sustain PLCγ2 expression. In accord, B-cell tumors have also never been reported in Btk^+/− or SLP-65^+/− mice, as would be expected for mice bearing lesions in bona fide tumor suppressors (6, 25). Therefore, we propose that PLCγ2, and by inference SLP-65 and Btk, behave as guardians that prevent B-cell transformation by coordinating developmental cues with cell proliferation.