| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Pathology and Laboratory Medicine and
Medicine, Weill Medical College, and
Immunology Graduate Program, Weill Graduate School of Medical Sciences, Cornell University, New York, NY 10021
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B, down-regulates the antiapoptotic proteins Bcl-2 and Bcl-xL, and up-regulates the proapoptotic protein Bax. Conversely, exposure of NHL B cells to recombinant or myeloid cell-derived BAFF and APRIL attenuates apoptosis, increases NF-B activation, up-regulates Bcl-2 and Bcl-xL, and down-regulates Bax. In some NHLs, exogenous BAFF and APRIL up-regulate c-Myc, an inducer of cell proliferation; down-regulate p53, an inhibitor of cell proliferation; and increase Bcl-6, an inhibitor of B cell differentiation. By showing that nonmalignant B cells up-regulate BAFF and APRIL upon stimulation by T cell CD40 ligand, our findings indicate that NHL B cells deregulate an otherwise physiological autocrine survival pathway to evade apoptosis. Thus, neutralization of BAFF and APRIL by soluble TACI and BCMA decoy receptors could be useful to dampen the accumulation of malignant B cells in NHL patients. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
GC B cells undergo programmed cell death or apoptosis unless positively selected by Ag, CD40L, or IL-4 (5). These stimuli rescue GC B cells from death signals transmitted through Fas, a CD40-inducible receptor that engages FasL on Ag-activated CD4+ T cells (6). Ultimately, B cell survival results from up-regulation of antiapoptotic and concomitant down-regulation of proapoptotic intracellular proteins. In general, the antiapoptotic proteins Bcl-2 and Bcl-xL play a key role in the regulation of B cell survival (7). Another crucial prosurvival protein is NF-B, a transcription factor required for the up-regulation of Bcl-2 and Bcl-xL (8, 9). The Bcl-2 pathway must be tightly regulated, because abnormal Bcl-2 expression favors the survival of autoreactive and aberrant B cells otherwise doomed to die. Consistent with this, Bcl-2 transgenic mice develop autoimmune disorders and B cell lymphomas (10, 11).
Lymphomas comprise the fifth most common cancer type in the U.S., with 
55,000 cases of non-Hodgkin’s lymphoma (NHL) each year. NHLs include multiple clinical entities that derive from the transformation and clonal expansion of mature B cells at distinct stages of differentiation (12, 13, 14). Mantle cell lymphoma (MCL), chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL), and marginal zone lymphoma (MZL) derive from mantle zone CD5+ naive B cells, CD5+ memory B cells, and marginal zone (MZ) CD5- B cells, respectively (15, 16, 17). In contrast, Burkitt’s lymphoma (BL) and follicular lymphoma (FL) originate from follicular GC B cells, including centroblasts (CBs) and centrocytes (CCs) (15). Finally, most, but not all, diffuse large cell B cell lymphoma (DLCL) tumors emerge from either GC B cells or activated B cells (18).
NHLs are characterized by diverse molecular abnormalities that promote the proliferation, enhance the survival, block the differentiation, or alter the genome integrity of malignant B cells (14). Nonrandom reciprocal translocations are frequent NHL-associated oncogenic events that originate from aberrant targeting of non-Ig loci by the SHM and CSR machineries (19). The resulting illegitimate gene recombination alters the transcription of key growth-, survival-, or differentiation-associated proto-oncogenes, such as MYC, BCL-2, and BCL-6, by placing them under the control of powerful Ig enhancer/promoter DNA regions (11, 12, 13, 20, 21, 22). Nevertheless, NHL B cells often overexpress c-Myc, Bcl-2, Bcl-6, and other crucial regulatory proteins, including NF-B (23), in the absence of translocations. Although aberrant SHM of promoter regions 5' of key genes plays an important role (24), the mechanisms underlying the overexpression of growth and survival proteins in tumoral B cells remain unclear.
NHL patients display increased circulating levels of B cell-activating factor of the TNF family (BAFF; also known as BLyS) (25), a molecule that promotes peripheral B cell survival as well as CD40-independent CSR and Ab production (26, 27, 28, 29). BAFF is produced by myeloid cells (28, 30) and binds to transmembrane activator and calcium modulator and cyclophilin ligand interactor (TACI), B cell maturation Ag (BCMA), and BAFF receptor (BAFF-R; also known as BR3), three receptors preferentially expressed by B cells (31, 32). Like CD40 (1), TACI, BCMA, and BAFF-R recruit TNF receptor-associated factors (TRAFs) (33, 34, 35) and elicit nuclear translocation of NF-B-Rel proteins by activating NF-B-inducing kinase (NIK) and IB kinase (IKK) (36, 37, 38). Once in the nucleus, BAFF-induced NF-B would activate key survival-associated genes, including Bcl-2 (39). Interestingly, TACI and BCMA also bind a proliferation-inducing ligand (APRIL), a BAFF-related molecule produced by myeloid cells that induces B cell proliferation, CSR, and plasma cell survival (28, 40, 41).
The following considerations prompted us to hypothesize that BAFF and APRIL play important roles in NHL. First, like Bcl-2 transgenic mice (10), BAFF-transgenic mice develop follicular and MZ B cell hyperplasia as well as autoimmunity (31). Second, the 13q32–34 locus encompassing the BAFF gene is often amplified in NHLs (42), and soluble BAFF (sBAFF) is increased in patients with NHL (25) and Sjoegren’s syndrome (43), an autoimmune disorder associated with MZL. Third, BAFF and APRIL have been recently shown to enhance the survival and proliferation of several tumors, including CLL and multiple myeloma (MM) (44, 45, 46, 47). We show in this study that autocrine and paracrine BAFF and APRIL promote the accumulation of malignant B cells from different types of NHL by attenuating apoptosis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Normal B cells were isolated from the peripheral blood (PB) and tonsils of healthy subjects. PB CD19+ or IgD+ B cells were purified through a magnetic cell-sorting device (Miltenyi Biotec, Auburn, CA) after incubating PBMCs with biotin-conjugated Abs to CD19 (BD PharMingen, San Diego, CA) or IgD (Southern Biotechnology Associates, Birmingham, AL) and streptavidin (SA)-conjugated MicroBeads (Miltenyi Biotec). A similar strategy was used to segregate tonsillar IgD+CD38-, IgD+CD38+, IgD-CD38+, IgD-CD38-, CD38+CD77+, and CD38+CD77- B cells. Briefly, tonsillar B cells were enriched by SRBC rosetting and magnetically sorted into CD38+ and CD38- fractions with a PE-conjugated mAb to CD38 (BD PharMingen) and anti-PE-MicroBeads. Then, CD38+ B cells were treated with MultiSort Release Reagent (Miltenyi Biotec) and divided into two aliquots. One aliquot was separated into IgD+CD38+ and IgD-CD38+ fractions using an FITC-conjugated mAb to IgD and anti-FITC-MicroBeads. Another aliquot was separated into CD38+CD77+ and CD38+CD77- fractions using an FITC-conjugated mAb to CD77 (BD PharMingen) and anti-FITC-MicroBeads. CD38- B cells were separated into IgD+CD38- and IgD-CD38- fractions using an FITC-conjugated mAb to IgD and anti-FITC MicroBeads. Finally, a biotinylated mAb to CD14 (BD PharMingen) and SA-conjugated MicroBeads were used to sort monocytes from PBMCs. Macrophages obtained by incubating monocytes with 100 U/ml IFN-
(Sigma-Aldrich, St. Louis, MO) for 4 days were cocultured with B cells at a 1:5 ratio.
Malignant cells
NHL subtypes were diagnosed according to the criteria established by the World Health Organization (48), and specimens were obtained after obtaining informed consent. Mononuclear cells from PB or lymph node specimens of patients with CLL, MCL, FL, BL, or DLCL were enriched in CD19+ B cells by magnetically removing T, NK, and myeloid cells with biotinylated mAbs to CD3, CD11c, CD14, and CD16 (BD PharMingen) and SA-conjugated MicroBeads. After this procedure, >90% of CD19+ B cells were clonal and expressed either Ig or Ig
(BD PharMingen). Cells were used immediately or were frozen in DMSO. BL2, BL16, BL30, BL74, Ramos, Bjab, HBL-1, and HBL-3 BL cell lines and the acute myeloid leukemia (AML) HL-60 cell line were obtained from American Type Culture Collection (Manassas, VA) and R. Dalla-Favera (Columbia University, New York, NY).
Cell cultures and reagents
Cells were cultured in complete RPMI 1640 medium supplemented with 10% FCS unless otherwise specified. Recombinant BAFF (Research Diagnostics, Flanders, NJ), APRIL MegaLigand (Alexis Biochemicals, San Diego, CA), CD40L (Immunex, Seattle, WA), and IL-4 (Schering-Plough, Kenilworth, NJ) were used at 0.1 µg/ml, 0.1 µg/ml, 0.5 µg/ml, and 500 U/ml, respectively. MOPC-21 control Ig (Sigma-Aldrich), TACI-Ig (Alexis Biochemicals), BCMA-Ig (Cell Sciences, Norwood, MA), and CD40-Ig (Ancell, Bayport, MN) were used at 30 µg/ml. BCR (or surface Ig) was engaged with 2 µg/ml Immunobead reagent (Irvine Scientifics, Camarillo, CA), which binds both H and L chains of human Igs. Cycloheximide (Sigma-Aldrich) was used at 100 µg/ml.
Flow cytometry
IgD, CD14, CD38, and CD77 were labeled with FITC- or PE-conjugated Abs. Membrane-bound BAFF (mBAFF) was labeled with a mouse Buffy-2 Ab to BAFF (Alexis Biochemicals) and a PE-conjugated anti-mouse Ab (BD PharMingen). Biotinylated IgD was labeled with SA-PerCP and used in triple-fluorescence studies together with an FITC-conjugated mAb to CD38 or CD77 and a PE-labeled Ab to BAFF. Surface BAFF-binding activity was measured with a human BAFF protein fused to murine CD8 (Ancell) and a PE-conjugated Ab to mouse CD8 (BD PharMingen). Cells were acquired using a FACSCalibur analyzer and mean fluorescence intensity was corrected for isotype control Ab staining.
Fluorescence microscopy
Cells were cytospun, fixed, and washed as previously described (28). Nuclei were visualized with 4',6-diamidine-2'-phenylindole dihydrochloride (Roche, Indianapolis, IN). Rabbit IgG to BAFF (Upstate Biotechnology, Lake Placid, NY) and goat IgG to APRIL (Alexis Biochemicals) were labeled with Cy3- and fluorescein-conjugated secondary Abs (Jackson ImmunoResearch Laboratories, West Grove, PA), respectively. Slides were analyzed and photographed using a fluorescence microscope (Zeiss Axioplan 2; Atto Instruments, Rockville, MD).
Confocal microscopy
Mouse Ab to BAFF (Alexis Biochemicals) was labeled with a Cy5-conjugated (blue) anti-mouse Ab (Jackson ImmunoResearch Laboratories). Goat Abs to TACI and BCMA (Santa Cruz Biotechnology, Santa Cruz, CA) were labeled with an Alexa 456 (red)-conjugated anti-goat Ab (Molecular Probes, Eugene, OR). Rabbit Abs to TRAF2 and TRAF6 (Santa Cruz Biotechnology) were labeled with an Alexa 488 (green)-conjugated anti-rabbit Ab (Molecular Probes). Primary and secondary Abs were used at 1/200 and 1/1000 dilutions, respectively. Coverslips were applied with Slow Fade reagent (Molecular Probes). Cells were visualized using an LSM510 laser-scanning microscope (Carl Zeiss, Jena, Germany).
Proliferation and apoptosis assays
To measure DNA synthesis, 104 cells/200 µl were seeded in 96-well plates and pulsed with 1 µCi of [3H]TdR on day 2 of culture. After 18 h, cells were harvested to measure [3H]TdR uptake. Alternatively, 5 x 104 cells were incubated in a final volume of 100 µl. Formazan release was measured after 2 days through the Celltiter 96 AQ assay following the manufacturer’s instructions (Promega). Cell viability was evaluated by seeding 5 x 106 cells in six-well plates at a final volume of 5 ml. After 2 or 4 days, viable cells were counted with a trypan blue exclusion test. Cell cycle analysis was performed on 3 x 106 cells fixed in 70% ethanol at 4°C overnight and incubated in 1 ml of PBS containing 50 µg/ml propidium iodide (PI) and 10 U/ml RNase A (Sigma-Aldrich) for 30 min at room temperature in the dark. Apoptosis was assayed using an Annexin VFITC apoptosis detection kit (Oncogene Research Products, San Diego, CA). Cells stained with PI and annexin V were analyzed by flow cytometry.
RT-PCRs
Activation-induced cytidine deaminase (AID; 382 bp), APRIL (417 bp), BAFF (398 bp), CD40L (450 bp), 
-actin (593 bp), TACI (323 bp), BCMA (326 bp), and BAFF-R (260 bp) were RT-PCR-amplified for 25 cycles (28, 49). BAFF transcripts were quantified through real-time PCR as previously reported (45). Briefly, cDNA synthesis and RT-PCR were performed in a single 50-µl reaction containing 25 µl of Onestep RT-PCR Master Mix with AmpliTaq Gold DNA polymerase, 1.25 µl of 40x MultiScribe reverse transcriptase, 125 mM double fluorescently labeled probe for BAFF (5'-CCACCAGCTCCAGGAGAAGGCAACTC-3') or cyclophilin B (5'-AGCATCTACGGTGAGCGCTTCCCC-3'), 300 nM forward and reverse primers for BAFF (5'-ACCGCGGGACTGAAAATCT-3' and 5'-CACGCTTATTTCTGCTGTTCTGA-3', respectively) and cyclophilin B (5'-GGAGATGGCACAGGAGGAAA-3' and 5'-CGTAGTGCTTCAGTTTGAAGTTCTCA-3', respectively), and 5 µg of RNA. BAFF was measured with an ABI PRISM 7000 sequence detection system (PE Applied Biosystems, Foster City, CA) and normalized for cyclophilin B.
Plasmids and luciferase reporter assays
The -839/+232 genomic DNA segment encompassing the BAFF promoter was cloned into a promoterless pGL3-Basic luciferase (Luc) reporter vector as previously reported (49). Cells (20 x 106/ml) were transfected with 40 µl of DNA-TE solution containing 20 µg of BAFF-luciferase (BAFF-Luc) or a minimal NF-B-driven reporter vector encompassing two NF-B-binding B sites (B-Luc) and 200 ng of pRL-TK control vector (Promega). In some experiments cells were cotransfected with 10 µg/ml empty pcDNA3 or IB
-pcDNA3 expression vector. Electroporation was performed at 625 V/cm (BL16) or 525 V/cm (Bjab) and 950 µF using a Gene Pulser II apparatus (Bio-Rad, Hercules, CA). Transfected cells (1 x 106 x ml) were cultured for 48 h. Firefly and Renilla activities were measured with the Dual-Luciferase Assay System (Promega) to assess promoter activity and transfection efficiency, respectively.
Immunoblots
Membrane proteins were purified from 15 x 106 cells with a ProteoPrep Membrane Extraction Kit (Sigma-Aldrich). Total proteins were obtained by treating 20 x 106 cells with a standard lysis buffer. Proteins (20 µg) were fractionated onto a 10% SDS-PAGE and transferred to nylon membranes. After blocking, membranes were probed with Abs to BAFF (Upstate Biotechnology), APRIL, TACI, BCMA, IB, Bcl-2, Bcl-xL, Bcl-6, c-Myc, p53, actin (Santa Cruz Biotechnology), p38, phospho (Thr180/Tyr182)-p38, phospho (Ser32/Ser36)-IB (Cell Signaling Technology, Beverly, MA), and poly ADP-ribose polymerase (PARP; BD PharMingen). Membranes were then washed and incubated with an appropriate secondary Ab (Santa Cruz Biotechnology). Proteins were detected with an ECL detection system (Amersham Pharmacia Biotech, Little Chalfont, U.K.).
Immunoprecipitations
Cells (20 x 106) were lysed on ice in 1% Nonidet P-40 lysis buffer containing protease inhibitors for 30 min. Cell lysates (500 µg) were centrifuged for 15 min at 14,000 x g and precleared with protein G- or A-Sepharose beads (Santa Cruz Biotechnology) for 2 h. Cleared lysates were first incubated with 5 µg of a goat Ab to an irrelevant Ag, TACI, or BCMA (Santa Cruz Biotechnology) overnight at 4°C and then incubated with 2 µl of protein G- or A-Sepharose beads for 1 h at 4°C. Beads were washed three times, resuspended in sample buffer, and boiled. After centrifugation, the immunoprecipitated material was immunoblotted with Abs to different proteins, including CD80 (BD PharMingen). To detect sBAFF, fluids from 1 x 106/ml cell cultures were concentrated with centrifugal filter devices (Millipore, Bedford, MA), incubated with a mAb to BAFF (Upstate Biotechnology), immunoprecipitated with protein A-Sepharose beads, and probed with an appropriate secondary Ab.
EMSAs
Nuclear proteins were extracted from 5 x 106 cells as previously reported (28, 50). A double-stranded oligonucleotide probe encompassing a consensus B site from the Ig gene promoter was end-labeled with [-P32]ATP by T4 kinase and used at 30,000 cpm in each EMSA reaction. Reaction samples were prepared as previously described (28, 50), incubated at room temperature for 15 min, and electrophoresed through a 6% nondenaturing polyacrylamide gel. The composition of DNA-bound nuclear NF-B-Rel complexes was determined by incubating reaction mixtures with 1 µl of Ab to p65, c-Rel, Rel-B, p50, or p52 (Santa Cruz Biotechnology) for 15 min at room temperature before adding the radiolabeled probe (28, 50).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In healthy subjects, PB B lymphocytes include IgD+CD38- naive and IgD-CD38- memory B cells, whereas B lymphocytes from secondary lymphoid organs encompass IgD+CD38- naive, IgD+CD38+ founder GC, IgD-CD38+ GC, and IgD-CD38- memory B cells. GC B cells can be further distinguished into CD77+ CBs and CD77- CCs. Normal PB CD19+ B cells as well as tonsillar naive, founder GC, and memory B cell fractions expressed surface BAFF-binding activity, but no or low mBAFF (Fig. 1A and Table I). Similar B cells contained TACI, BCMA, and BAFF-R transcripts, but lacked BAFF and APRIL transcripts as well as transcripts for AID (Fig. 1B and Table I), a GC B cell-specific enzyme that plays a key role in CSR and SHM (3). In contrast, total GC B cells as well as purified CBs and CCs expressed BAFF, APRIL, and AID in addition to TACI, BCMA, and BAFF-R transcripts. All mature B cell subsets lacked transcripts for CD40L and (not shown) CD68, which are expressed by T cells and macrophages, respectively. Unlike B cells, PB or tonsillar CD14+ myeloid cells expressed mBAFF, but not BAFF-binding activity (Fig. 1A). Consistent with this, CD14+ myeloid cells expressed BAFF and APRIL, but not TACI, BCMA, or BAFF-R transcripts (Fig. 1B). Similar myeloid cells lacked AID and CD40L transcripts. Thus, all normal B cell subsets express TACI, BCMA, and BAFF-R, whereas BAFF and APRIL are confined to normal GC B cells and myeloid cells.
|
|
The expression of TACI, BCMA, BAFF-R, BAFF, and APRIL in NHL remains unclear. Unlike normal B cells, fresh malignant CD19+ B cells from CLL/SLL, FL, BL, or DLCL expressed surface BAFF-binding activity as well as mBAFF (Fig. 2A and Table I). In general, DLCL B cells expressed more mBAFF than CLL/SLL, FL, or BL B cells (Table I). CD14+ myeloid cells from all NHL patients lacked BAFF-binding activity, but expressed more mBAFF than CD14+ myeloid cells from healthy subjects. Malignant CLL/SLL, MCL, BL, FCL, and DLCL B cells also expressed TACI, BCMA, and BAFF-R as well as BAFF, APRIL, and AID transcripts. Finally, malignant B cells from some CLL/SLL and DLCL cases contained CD40L transcripts, although less than BAFF and APRIL transcripts. Thus, NHL B cells express TACI, BCMA, and BAFF-R and are exposed to both autocrine and paracrine BAFF and APRIL.
|
BAFF and APRIL expression was further evaluated in NHL B cell lines. Total lysates from normal PB IgM+ (IgD+) B cells contained TACI and BCMA, but lacked BAFF and APRIL proteins (Fig. 3A). In contrast, total lysates from IgM+ BL B cell lines contained BAFF and APRIL in addition to TACI and BCMA proteins. As expected, a control AML cell line contained BAFF and APRIL, but not TACI and BCMA, proteins. Purified membrane proteins from normal IgM+ B cells included BCMA, but not BAFF (Fig. 3B). In contrast, purified membrane proteins from IgM+ BL B cell lines included BCMA, BAFF, as well as a 17-kDa protein corresponding to sBAFF, a BAFF cleavage byproduct that retains the ability to bind and activate membrane-anchored receptors, including BCMA (26). Although including BAFF, purified membrane proteins from a control AML cell line lacked BCMA as well as sBAFF. Unlike normal IgM+ B cells, IgM+ BL B cells and control AML cells released sBAFF in the culture fluids (Fig. 3B). Finally, BAFF and APRIL proteins could be visualized in the cytoplasm of IgM+ BL B cells and control AML cells, but not in that of normal IgM+ B cells (Fig. 3C). Thus, NHL B cell lines contain more BAFF and APRIL, express more mBAFF, and release more sBAFF than normal B cells.
|
The following experiments were performed to assess whether BAFF and APRIL enhance NHL B cell survival. The number of viable cells spontaneously declined upon incubation with FCL, DLCL, and BL B lymphocytes with medium alone for 2 days (Fig. 4A). This decline was not affected by adding control Ig or CD40-Ig, a soluble decoy receptor that blocks the binding of CD40 on B cells by CD40L. In contrast, the number of viable NHL B cells was dramatically decreased by BCMA-Ig, a soluble decoy receptor that prevents the binding of BAFF and APRIL to B cell-anchored TACI, BCMA, and BAFF-R. A similar decrease was induced by dexamethasone, an NF-B-inhibiting steroid often used to treat NHLs. Conversely, soluble BAFF, soluble APRIL, or macrophages increased the viability of NHL B cells. The prosurvival activity of macrophages was dependent upon BAFF and APRIL, but not CD40L, because it was inhibited by BCMA-Ig, but not CD40-Ig.
|
The prosurvival activity of autocrine BAFF was further investigated in IgM+ BL B cell lines. Some BL lines contained an amount of BAFF transcripts comparable to that detected in the control AML cell line, HL60. Other BL cell lines contained lesser amounts of BAFF transcripts (Fig. 5A). BL B cell lines expressing more BAFF transcripts were more resistant to serum starvation- or BCR-induced apoptosis than BL B cell lines expressing less BAFF transcripts. A positive correlation between BAFF expression and NHL B cell resistance to apoptosis was also found in fresh NHL B cells. Large FCL B cells expressing more mBAFF were more resistant to serum starvation-induced apoptosis than clonally related small FCL B cells expressing less mBAFF (Fig. 5B). Exogenous BAFF partially rescued both mBAFFhigh and mBAFFlow FCL B cells from apoptosis. Collectively, our results indicate that BAFF and APRIL attenuate NHL B cell apoptosis through both autocrine and paracrine pathways.
|
Bcl-2 and Bcl-xL are NF-B-inducible Bcl-2 family members that enhance cell survival by interfering with the release of cytochrome c from mitochondria (7, 8, 9). Cytochrome c initiates apoptosis by activating caspases, a family of intracellular proteases that degrade intracellular proteins such as PARP (7). Unlike Bcl-2 and Bcl-xL, the Bcl-2 family member Bax enhances apoptosis by facilitating cytochrome c release (7). Compared with DLCL B cells incubated with control Ig, freshly isolated DLCL B cells incubated for 2 days with BCMA-Ig or TACI-Ig increased spontaneous PARP cleavage, down-regulated Bcl-2 and Bcl-xL, and up-regulated Bax (Fig. 6A). Conversely, DLCL B cells attenuated spontaneous PARP cleavage, up-regulated Bcl-2 and Bcl-xL, and down-regulated Bax upon exposure to BAFF or APRIL. Similar results were obtained with fresh FL and BL B cells (not shown) as well as with BL B cell lines, including BL16 (Fig. 7A). Thus, both autocrine and paracrine BAFF and APRIL shift the Bcl-2 family ratio to favor NHL B cell survival.
|
|
B in NHL B cells
In resting B cells cytoplasmic p50, p65 (RelA), and c-Rel NF-B-Rel proteins occur as inactive dimers bound to IB (51, 52). After stimulation, B cells phosphorylate IB through an IKK complex that includes two catalytic subunits, and , as well as a regulatory subunit, . In this classical pathway, degradation of phosphorylated IB (p-IB) is followed by translocation of NF-B-Rel dimers from the cytoplasm to the nucleus (52). In addition to inducing proliferation, differentiation, and survival genes, such as Bcl-2 and Bcl-xL (8, 9), NF-B turns on the IB gene. This negative feedback limits further NF-B activation. Unstimulated DLCL B cells incubated with a control Ab constitutively expressed p-IB and contained DNA-binding nuclear complexes corresponding to p50-c-Rel and p50-p65 (Fig. 6, B and C). In the presence of TACI-Ig or BCMA-Ig, DLCL B cells up-regulated IB, but down-regulated p-IB and NF-B-Rel binding to DNA. Conversely, DLCL B cells incubated with BAFF or APRIL down-regulated IB, but up-regulated p-IB and NF-B-Rel binding to DNA.
Similar results were obtained with fresh FL and BL B cells (not shown) and with BL cell lines, including BL16 (Fig. 7B). BL16 B cells down-regulated IB as early as 10 min after stimulation with BAFF or APRIL (Fig. 7C). By 6 h, stimulated BL16 B cells re-up-regulated IB unless exposed to the protein synthesis inhibitor cycloheximide. Additional experiments verified whether NF-B induced by BAFF and APRIL turns on gene transcription. In the presence of BAFF or APRIL, BL16 B cells activated a luciferase reporter plasmid containing an NF-B-dependent minimal promoter with two B sites (B-Luc; Fig. 7D). BCMA-Ig, but not control Ig, inhibited the activation of B-Luc by BAFF or APRIL, and this inhibition was comparable to that induced by overexpressing IB. Thus, BAFF and APRIL activate NF-B in NHL B cells through both autocrine and paracrine pathways.
TACI and BCMA constitutively signal in NHL B cells
The following experiments were performed to verify whether TACI and BCMA constitutively signal in NHL B cells. TACI and BCMA colocalized with BAFF, TRAF-2, and TRAF-6 in discrete areas of unstimulated BL16 B cells (Fig. 8A). Consistent with this, TACI and BCMA immunoprecipitated with BAFF, APRIL, TRAF-2, TRAF-6, IB, and p-IB in BL16 B cells (Fig. 8B). In similar B cells, TACI immunoprecipitated with BCMA, and BCMA with TACI. In contrast, TACI and BCMA did not coimmunoprecipitate with CD80, a costimulatory molecule expressed by BL16 B cells. Furthermore, a control mAb did not immunoprecipitate TACI and BCMA in BL cells. Conversely, anti-TACI and anti-BCMA mAbs did not immunoprecipitate TACI and BCMA in control AML cells. Exposure of BL16 cells to BAFF up-regulated the amount of TRAF2 coimmunoprecipitating with TACI and that of TRAF6 coimmunoprecipitating with BCMA. In similar BL16 cells, BAFF up-regulated the coimmunoprecipitation of p-IB, but down-regulated that of IB with TACI and BCMA. These findings suggest that autocrine BAFF and APRIL recruit a TACI-BCMA-TRAF2-TRAF6 complex that continuously activates NF-B.
|
Additional experiments were performed to verify whether BAFF and APRIL modulate NHL B cell proliferation. Fresh FL, DLCL, and BL B cells up-regulated the synthesis of DNA upon exposure to APRIL, but not BAFF (Fig. 9A). Similarly, BL B cells increased the release of formazan, a metabolite produced by actively proliferating cells, upon exposure to APRIL, whereas BAFF or BCR engagement did not affect and inhibited formazan release, respectively (Fig. 9B). Notably, NHL B cells exposed to both BAFF and APRIL released more formazan than NHL B cells exposed to BAFF or APRIL alone. Unlike BAFF, APRIL increased the proportion of actively cycling BL B cells in the S and G2/M phases of the cell cycle (Fig. 9C). Similar BL B cells down-regulated p53 (Fig. 9D), an inhibitor of cell proliferation (53), and up-regulated c-Myc, an inducer of cell proliferation (54), after exposure to BAFF or APRIL. Furthermore, BAFF and, to a larger extent, APRIL up-regulated the expression of p-p38, a mitogen-activated protein kinase that facilitates c-Myc-induced cell proliferation (55) as well as that of Bcl-6, a proto-oncogene that inhibits GC B cell differentiation and favors lymphomagenesis (20, 22, 56, 57). Thus, BAFF and APRIL deliver distinct signals that cooperatively stimulate NHL B cell growth.
|
B cells might up-regulate BAFF and APRIL upon activation by canonical GC differentiation-inducing stimuli, including CD40L. PB naive B cells from healthy subjects up-regulated BAFF and APRIL transcripts as early as 6 h after exposure to CD40L and as early as 24 h after BCR cross-linking (Fig. 10A). In addition, naive B cells up-regulated BAFF and APRIL proteins upon stimulation through BCR engagement, CD40L, and/or IL-4 for 4 days (Fig. 10B). Neutralization of autocrine BAFF and APRIL by TACI-Ig or BCMA-Ig, but not control Ig, attenuated the viability (Fig. 10C) and IgM-secreting activity (Fig. 8D) of naive B cells incubated with CD40L and IL-4 for 7 days. In a BL B cell line, the up-regulation of mBAFF by CD40L and IL-4 was associated with increased activation of BAFF-Luc, a luciferase reporter plasmid driven by the BAFF gene promoter (Fig. 8, E and F). NF-B was crucial for the transcriptional activation of the BAFF gene promoter, because BL B cells overexpressing IB failed to activate BAFF-Luc upon stimulation with CD40L and IL-4. Thus, normal B cells physiologically up-regulate active BAFF and APRIL upon BCR engagement or exposure to CD40L and IL-4.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
By showing that macrophages attenuate the apoptosis of NHL B cells through a paracrine pathway that is inhibited by soluble BCMA and TACI decoy receptors, our findings indicate that BAFF and APRIL expressed on and released by bystander myeloid cells play a key role in malignant B cell survival. Intriguingly, macrophages from NHL patients express more mBAFF than macrophages from healthy subjects. This might stem from the expression and release of BAFF-inducing factors, including lymphotoxin-/ (58), by the progressively expanding malignant B cell clone. BAFF and APRIL would further stimulate malignant B cell survival through an autocrine pathway. Consistent with this, NHL B cells as well as CLL and MM B cells (45, 46, 47) express both BAFF and APRIL. Notably, BAFF is expressed on the surface of some, but not all, elements of a given NHL clone. These mBAFF+ cells would accumulate independently from external survival signals, whereas the remaining mBAFF- cells probably require exogenous BAFF from myeloid cells to survive.
Our data raise the question of whether autocrine BAFF and APRIL expression is confined to malignant B cells or extends to normal B cells. We show that normal naive B cells rapidly up-regulate BAFF and APRIL upon exposure to GC differentiation-inducing stimuli, including CD40L. The up-regulation of BAFF by CD40L requires trans-activation of the BAFF gene promoter and may account for the ability of CD40L to turn on a noncanonical NF-B pathway in B cells that is usually induced by BAFF-R (37, 38, 59). By enhancing the survival of CD40-activated B cells, including GC CBs and CCs, autocrine BAFF and APRIL might optimize the Ab response to T cell-dependent Ags. This would explain why BAFF and APRIL are more abundant in GC-derived B cell tumors. Notably, our and previous (45, 46, 47) studies indicate that tumoral B cells express more APRIL and BAFF, namely mBAFF and sBAFF, than normal B cells. In addition, tumoral B cells constantly express BAFF and APRIL, whereas GC B cells down-regulate BAFF and APRIL upon differentiation to memory B cells. These observations imply that malignant B cells deregulate an otherwise physiological prosurvival autocrine pathway to gain growth advantage.
Neoplastic B cells would deregulate BAFF and APRIL through a number of mechanisms that probably take place after the initial transforming event. In most NHL B cells, SHM is actively ongoing (50, 60, 61) and might alter BAFF and APRIL gene transcription by aberrantly targeting 5' gene promoters (24). Abnormal BAFF expression might also stem from amplification of the 13q32–34 locus, an abnormality often detected in FL and DLCL (42). In DLCL and MZL, alterations of genes encoding the IKK-activating Bcl-10 and Malt proteins (14) might deregulate the nuclear translocation of NF-B, a key trans-activator of the BAFF gene promoter (49). Abnormal NF-B activation might also up-regulate BAFF in AIDS-associated and post-transplant NHLs (62). These B cell neoplasias often express latent membrane protein 1 and/or 2, two EBV-encoded NF-B-activating proteins that up-regulate BAFF and APRIL in B cells (49). Finally, chronic immune stimulation by self or foreign Ags might deregulate BAFF and APRIL expression in MZLs associated with Sjoegren’s syndrome and Helicobacter pylori infection (14, 31, 63).
In NHL B cells, autocrine and paracrine BAFF and APRIL attenuate apoptosis by shifting the balance of intracellular Bcl-2 family members toward survival. Consistent with this, BAFF up-regulates antiapoptotic Bcl-2 and Bcl-xL, but down-regulates proapoptotic Bax. Conversely, TACI-Ig and BCMA-Ig down-regulate Bcl-2 and Bcl-xL, but up-regulate Bax and, as shown by others (45), activate apoptosis-inducing caspases. Although unable to bind BAFF-R, APRIL rescues NHL B cells from apoptosis almost as efficiently as BAFF. This might reflect the activation of BCMA and/or an additional, as yet unknown, prosurvival receptor by APRIL (41). Intriguingly, BL B cell-bound BCMA is associated with TACI, suggesting that TACI, BCMA, and perhaps BAFF-R form promiscuous receptor complexes with abnormal signaling activity. BAFF and APRIL signaling could be further deregulated by chronic ligand-induced internalization of TACI, BCMA, and BAFF-R, which would continuously stimulate new TACI, BCMA, and BAFF-R synthesis. This might explain the diffuse cytoplasmic expression pattern of TACI and BCMA in NHL B cells.
NF-B is crucial to up-regulate Bcl-2 and Bcl-xL and attenuate B cell apoptosis (8, 9). In normal mouse B cells, BAFF-R typically turns on an alternative NF-B pathway in which phosphorylation of cytoplasmic p100 (also known as NF-B2) by TRAF-activated NIK and IKK is followed by p100 ubiquitination and processing to p52, which then translocates to the nucleus in association with RelB (37, 38, 39). In similar B cells, BAFF-R as well as TACI and BCMA also activate a canonical NF-B pathway (33, 34, 35, 36), in which IB phosphorylation by a TRAF-activated IKK-- complex is followed by IB degradation and subsequent p50-c-Rel and p50-p65 nuclear translocation (51). Although unable to assess the role of BAFF-R, our experiments suggest that BAFF and APRIL produced by NHL B cells constitutively activate the canonical NF-B pathway through BCMA and perhaps TACI. Consistent with this, BCMA and TACI on NHL B cells are constitutively associated with autocrine BAFF and APRIL as well as TRAF2, TRAF6, and p-IB. That autocrine BAFF and APRIL contribute to the deregulation of the canonical NF-B pathway in NHL B cells is also shown by the ability of these malignant B cells to attenuate constitutive IB phosphorylation, IB degradation, and p50-c-Rel and p50-p65 nuclear translocation upon exposure to soluble TACI and BCMA decoy receptors. Conversely, NHL B cells up-regulate IB phosphorylation, IB degradation, and p50-c-Rel and p50-p65 nuclear translocation upon exposure to exogenous BAFF or APRIL. Finally, preliminary data (not shown) suggest that NHL B cells activate the alternative NF-B pathway upon exposure to BAFF, but not APRIL. Thus, it is likely that BAFF and APRIL cooperatively enhance NHL B cell survival by activating both classical and alternative NF-B pathways through distinct receptors.
In addition to activating NF-B (33), TACI limits the survival and expansion of activated B cells by triggering apoptosis (27, 64). The paradoxical ability of TACI to activate NF-B and yet induce cell death is similar to that observed for other TNF receptor family members, which trigger either survival or death depending on the cell context (65). Thus, it is conceivable that TACI triggers apoptosis in normal B cells, but enhances survival in NHL B cells. The outcome of TACI signaling might also depend on the relative expression of BCMA and BAFF-R. Should this be the case, certain NHL B cells might become more sensitive to the prosurvival activity of BCMA and BAFF-R by down-regulating TACI.
We have recently shown that B cells exposed to BAFF and APRIL up-regulate AID (28). This enzyme is thought to initiate SHM and CSR by favoring the introduction of DNA breaks within the Ig V(D)J and switch regions (3). Because of its potential oncogenic activity (66), AID is only transiently expressed by activated B cells, including GC B cells (3, 50). As shown by our data, B cell NHLs constitutively express AID (50, 67, 68) and actively undergo SHM and CSR in a variable fraction of the clone (50, 61). Ongoing SHM might select increasingly dangerous clonal variants by aberrantly targeting non-Ig loci (24), whereas ongoing CSR might favor the outgrowth of class-switched IgD-IgM- elements with conserved responsiveness to environmental Ags (60, 69). In this regard, IgG-BCR is less sensitive than IgD-BCR and IgM-BCR to CD22, an inhibitory coreceptor that negatively modulates Ag-induced B cell growth by turning off the Ig-Ig signaling unit of the BCR complex (70). It is tempting to speculate that autocrine and paracrine BAFF and APRIL contribute to the deregulation of SHM and CSR by up-regulating AID and increasing the survival of AID-expressing NHL B cells.
By showing that soluble TACI and BCMA decoy receptors attenuate spontaneous NHL B cell DNA synthesis and that APRIL increases the proportion of NHL B cells entering the S and G2-M phases of the cell cycle, our data suggest that APRIL stimulates malignant B cell growth through both autocrine and paracrine pathways. In BL B cells, APRIL-induced proliferation is associated with up-regulation of the growth-inducing proteins c-Myc and p-p38, down-regulation of the growth-inhibiting protein p53, and up-regulation of the GC B cell differentiation-inhibiting protein Bcl-6. Although inducing similar changes in c-Myc, p38, p53, and Bcl-6 expression, BAFF does not increase the proportion of NHL B cells in the S and G2-M phases of the cell cycle. Yet, BAFF enhances the proliferation of NHL B cells exposed to APRIL. These observations and recent studies showing that BAFF augments the proliferation of MM B cells (47) suggest that BAFF and APRIL cooperatively stimulate the growth of tumoral B cells by delivering qualitatively distinct signals through different receptors.
Our finding that engagement of CD40 on tumoral B cells by CD40L up-regulates BAFF extends previous studies showing that certain NHLs express CD40L in addition to CD40 (71, 72, 73) and raises the possibility that multiple TNF-related ligands cooperatively enhance the accumulation of NHL B cells. This would be in line with a recent report showing that BAFF increases the proliferation and survival of normal B cells exposed to CD40L (39). In addition to attenuating apoptosis, BAFF and APRIL might increase the proliferation and limit the terminal differentiation of CD40L-induced NHL B cells by up-regulating Bcl-6. This transcriptional repressor retains actively proliferating B cells in the GC (56), and its down-regulation by CD40L (74) induces cell cycle arrest as well as maturation of both normal and malignant GC B cells to terminally differentiated plasma cells (57).
NHLs constitute a heterogeneous collection of diseases, and successful therapy will ultimately depend on detailed knowledge of the pathogenetic pathways involved in each lymphoma. Our findings outline a key survival pathway that is shared by a vast spectrum of B cell tumors. This implies that BAFF and APRIL inhibitors, such as soluble TACI and BCMA decoy receptors, could be used as adjuvants in the context of specific antilymphoma therapies. Given the ability of CD40L to cooperate with BAFF and APRIL, a combination of CD40L blockers and BAFF and APRIL inhibitors might be beneficial in selected NHL cases with increased CD40L expression.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Andrea Cerutti, Weill Medical College, Cornell University, 1300 York Avenue, New York, NY 10021. E-mail address: acerutti{at}med.cornell.edu
3 Abbreviations used in this paper: CD40L, CD40 ligand; AID, activation-induced deaminase; AML, acute myeloid leukemia; APRIL, a proliferation-inducing ligand; BAFF, B cell-activating factor belonging to the TNF family; BAFF-R, BAFF receptor; BCMA, B cell maturation Ag; BCR, B cell Ag receptor; BL, Burkitt’s lymphoma; CB, centroblast; CC, centrocyte; CLL/SLL, chronic lymphocytic leukemia/small lymphocytic lymphoma; CSR, Ig class switch DNA recombination; DLCL, diffuse large cell lymphoma; FL, follicular lymphoma; GC, germinal center; IKK, IB kinase; Luc, luciferase reporter plasmid; mBAFF, membrane-bound BAFF; MCL, mantle cell lymphoma; MM, multiple myeloma; MZL, marginal zone lymphoma; NHL, non-Hodgkin’s lymphoma; NIK, NF-B-inducing kinase; p, phospho; PARP, poly-ADP-ribose-polymerase; PB, peripheral blood; PI, propidium iodide; p-IB, phosphorylated IB; SA, streptavidin; sBAFF, soluble BAFF; SHM, Ig V(D)J gene somatic hypermutation; TACI, transmembrane activator and calcium modulator and cyclophilin ligand interactor; TRAF, TNFR-associated factor.
Received for publication August 8, 2003. Accepted for publication December 23, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B-mediated up-regulation of Bcl-x and Bfl-1/A1 is required for CD40 survival signaling in B lymphocytes. Proc. Natl. Acad. Sci. USA 96:9136.B activity is required for survival of activated B cell-like diffuse large B cell lymphoma cells. J. Exp. Med. 194:1861.
in the cytograms indicate B cells and CD14+ monocytes/macrophages. Right, IgD+CD38- naive B cells (N), IgD+CD38+ founder GC B cells (FGC), IgD-CD38+ GC B cells (GC), and IgD-CD38- memory B cells (M) were isolated after tonsillar B cell staining for CD38 and IgD. Purified CD38+ B cells were further fractionated into CD38+CD77+ CBs and CD38+CD77- CCs. mBAFF was analyzed on all these B cell subsets and M
) or CD40L plus IL-4 (
). Luciferase activity was measured after 2 days and is expressed as the fold induction. F, mBAFF (open profiles) on Bjab BL B cells incubated with medium alone (control) or CD40L plus IL-4 for 4 days. The shaded profiles depict cell staining with an isotype-matched control Ab. The data in A–F represent one of three similar experiments.