Constitutive Activation of Extracellular Signal-Regulated Kinase Predisposes Diffuse Large B-Cell Lymphoma Cell Lines to

CD40 promotes survival, proliferation, and differentiation of normal B cells but can cause activation-induced cell death in malignant B lymphocytes. CD40 ligand and anti-CD40 antibodies have been used successfully to induce apoptosis in lymphoma lines both in vitro and in xenograft tumor models. Although this makes CD40 an attractive target for antitumor therapies, the response of malignant B cells to CD40 signaling is variable, and CD40 stimulation can enhance proliferation and can increase chemoresistance in some cell lines. It would therefore be useful to identify markers that predict whether a specific cell line or tumor will undergo apoptosis when stimulated with CD40 and to identify targets downstream of CD40 that affect only the apoptotic arm of CD40 signaling. We have analyzed gene expression patterns in CD40-sensitive and CD40-resistant diffuse large B-cell lymphoma (DLBCL) cell lines to identify signaling pathways that are involved in CD40-mediated apoptosis. CD40-resistant lines expressed pre-B-cell markers, including RAG and VPREB, whereas CD40-sensitive cells resembled mature B cells and expressed higher levels of transcripts encoding several members of the CD40 signaling pathway, including LCK and VAV. In addition, CD40-sensitive DLBCL cell lines also displayed constitutive activation of extracellular signal-regulated kinase (ERK) and failed to undergo apoptosis when ERK phosphorylation was inhibited. In contrast, CD40-resistant lines showed no constitutive activation of ERK and no increase in ERK activity in response to CD40 stimulation. Our results suggest that constitutive activation of ERK may be required for death signaling by CD40.


Introduction
CD40 is a member of the tumor necrosis factor receptor superfamily that is expressed on the surface of B lymphocytes as well as a variety of nonlymphoid cells (1).Stimulation of CD40 by CD40L, which is expressed primarily on T cells, is essential for normal B lymphocyte development and function (1).The importance of CD40 in maintaining normal immune function is evident in individuals with inactivating mutations in CD40 or CD40L who suffer from severe immunodeficiency characterized by susceptibility to bacterial infections and accumulation of IgM + B cells that are unable to switch their immunoglobulin class (1).
CD40 is first expressed before the rearrangement of immunoglobulin heavy chain genes in early B-cell development; its expression is maintained through all subsequent stages of B-cell development and is not lost during malignant transformation (2).Malignant B lymphocytes and other CD40-positive tumor cells differ from normal B cells in that they undergo apoptosis following CD40 stimulation (3).CD40 stimulation shows promise as an antitumor therapy in murine models of B-cell lymphoma and breast cancer (4,5).CD40 ligand has also been tested in phase I clinical trials (6); however, the use of CD40-directed therapy remains controversial because CD40 signaling can enhance cell proliferation and survival as well as induce resistance to chemotherapeutic agents in some B-cell malignancies (7,8).Consequently, elucidation of the mechanisms involved in CD40mediated apoptosis may help to identify markers for susceptibility to CD40-mediated cell death and reveal proteins that specifically control the apoptotic arm of CD40 signaling.
CD40 stimulation has been shown to activate both the nuclear factor-nB (NF-nB) and the mitogen-activated protein kinase (MAPK) signaling pathways (1).NF-nB activation on CD40 stimulation has been extensively studied and is generally associated with cell survival in both normal and malignant B lymphocytes (9).In contrast, MAPK activation can have proliferative or apoptotic effects depending on the stimulus and cell type.In general, activation of extracellular signal-regulated kinase (ERK) is associated with survival, whereas activation of c-Jun NH 2 -terminal kinase (JNK) and p38 tends to promote apoptosis (10).In B lymphocytes, p38 activation has been shown to promote CD40-induced cell proliferation (11), whereas ERK activation can be associated with both apoptotic and antiapoptotic effects (12).The biological effects of CD40 stimulation in B cells are also strongly dependent on maturation stage, which may reflect availability of different signal transduction pathways (1).
We hypothesize that cells that differ in their response to CD40 ligation will show differences in gene expression before stimulation, resulting in the activation of alternate signaling pathways.We therefore studied expression profiles of CD40-sensitive and CD40resistant diffuse large B-cell lymphoma (DLBCL) cell lines and correlated gene expression profiles of unstimulated cell lines with their sensitivity to CD40-mediated apoptosis.Resistance to CD40mediated cell death was associated with expression of pre-B-cell markers, whereas susceptibility was correlated with a mature B-cell phenotype and increased transcript levels for several proteins involved in CD40 signal transduction.Based on these gene expression patterns, we investigated the activity of the LCK-VAV-MAPK
CD40 stimulation and assessment of viability.Cells were resuspended at 1 Â 10 5 /mL in RPMI 1640 supplemented as described above.Anti-CD40 antibody G28.5 and secondary cross-linking antibody HB58 were added to a final concentration of 10 Ag/mL each, and the cells were incubated at 37jC.At 24-hour intervals, aliquots of cells were harvested by centrifugation, washed once in PBS, and resuspended in fluorescence-activated cell sorting buffer.Nonpermeabilized cells were stained by addition of 1 Ag/mL propidium iodide, and the density of viable cells was determined using a FACScan flow cytometer and CellQuest software (Becton Dickinson, Mississauga, Ontario, Canada) set to count for a fixed 30-second time interval.To determine the fraction of cells that had undergone apoptosis, total intracellular DNA content was measured by propidium iodide staining of ethanol-permeabilized cells as described previously (16).Expression of CD40 on the cell surface was confirmed by staining nonpermeabilized cells with a phycoerythrin-linked anti-CD40 antibody (clone 5C3, Becton Dickinson) as per manufacturer's directions and followed by flow cytometry.
Growth inhibition by kinase inhibitors.The Src family kinase inhibitor PP1 and the MAPK/ERK kinase (MEK) inhibitor U0126 were purchased from Biomol (Plymouth Meeting, PA) and Cell Signaling Technology (Beverly, MA), respectively.Kinase inhibition assays were done using cells seeded at a density of 5 Â 10 4 /mL in 96-well plates, to which kinase inhibitors were added to final concentrations of 10 À4 to 10 À8 mol/L.The treated cells were incubated at 37jC for 96 hours, and cell viability was quantitated by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay as described previously (17).For inhibition of CD40-mediated apoptosis, the MEK inhibitors U0126 (10 Amol/L) and PD98059 (50 Amol/L) as well as the p38 inhibitor SB203580 (10 Amol/L; Cell Signaling Technology) were added 30 minutes before CD40 stimulation.The dose of kinase inhibitors used in this study has been shown previously to mediate targetspecific effects in lymphocytes (18).
Detection of intracellular phosphorylated ERK by flow cytometry.Cells were permeabilized in Cytofix/Cytoperm (Becton Dickinson) for 20 minutes and stained with anti-phosphorylated ERK antibody (Cell Signaling Technology) and FITC-conjugated anti-rabbit secondary antibody (Cedarlane Laboratories, Hornby, Ontario, Canada) following the manufacturers' instructions.The stained cells were analyzed by flow cytometry with a FACScan flow cytometer and CellQuest software.
Immunoprecipitation and Western blot.LCK was immunoprecipitated from nondenatured cell lysate with a polyclonal rabbit antibody (Cell Signaling Technology) as recommended by the manufacturer.Proteins were separated by SDS-PAGE on a 14% acrylamide gel at 180 V for 1 hour and transferred to a nitrocellulose membrane by electrophoretic transfer at 100 V for 1 hour.Western blots were done as described previously (16).Primary antibodies against total and phosphorylated ERK, JNK, and p38 (Cell Signaling Technology) were used at a dilution of 1:100, primary antibodies against LCK and activated Src family kinases (Cell Signaling Technology) were diluted 1:500, and peroxidase-conjugated goat anti-mouse or donkey anti-rabbit secondary antibodies (BioCan Scientific, Mississauga, Ontario, Canada) were used at a dilution of 1:10,000.The compositions of blocking and antibody dilution solutions were specified by the manufacturer.Blots were immersed in Femto West enhanced chemiluminescence staining solution (Pierce) for 30 seconds and photographed with a Syngene chemiluminescence imaging system and GeneSnap software.Image size and contrast were adjusted using Canvas software.
Expression microarrays.Total RNA was prepared from cultured cells using Trizol reagent according to the manufacturer's instructions (Invitrogen, Carlsbad, CA).The integrity of the purified RNA was verified using an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA).Probes for microarray analysis were prepared using total RNA (10 Ag) and hybridized to Affymetrix HG-U113A GeneChips (Affymetrix, Santa Clara, CA) as described previously (19).To minimize technical variability, RNA processing steps (RNA extraction, probe labeling, and chip hybridization) were done in parallel for each set of four RNA samples.
The hybridized arrays were scanned, and raw data were extracted using the Microarray Analysis Suite 5.0 (MAS5; Affymetrix).The raw data were normalized using RMAExpress (http://stat-www.berkeley.edu/users/bolstad/RMAExpress/RMAExpress.html;ref. 20) and filtered to exclude genes that MAS5 did not identify as ''present'' in any expression profile.Differentially expressed genes were identified by doing a t test between each pair of CD40-sensitive and CD40-resistant lines.False-positive error correction was done to maintain a 10% false discovery rate (21).Genes whose expression differed significantly and showed the same direction of change in all four comparisons were considered to be differentially regulated between sensitive and resistant lines.Functional assessment of these expression differences was done using contingency table analysis based on hand-curated lists of NF-nB targets (references obtained from http://people.bu.edu/gilmore/nf-kb; ref. 22), genes involved in CD40 and BCR signal transduction (see Supplementary Table S2 for gene lists and references), B-cell maturation markers (Supplementary Table S2), and the gene ontology classification of proteins implicated in apoptosis (Affymetrix).Figures illustrating relevant expression profiles were prepared using Genesis (http://genome.tugraz.at/Software/GenesisCenter.html;ref. 23).

Results
Susceptibility of DLBCL cell lines to CD40-mediated cell death.DLBCL cell lines were screened for susceptibility to CD40mediated cell death.The number of viable OCI-Ly7 and Su-DHL4 cells was reduced after 72 hours of exposure to cross-linked anti-CD40 antibody, whereas the viability of OCI-Ly1 and OCI-Ly8 cells was unaffected (Fig. 1A).CD40 stimulation also increased the proportion of cells with sub-G 1 DNA content at 48 hours in OCI-Ly7 and Su-DHL4 but not OCI-Ly1 or OCI-Ly8 (Fig. 1B).Expression of CD40 on the cell surface was confirmed by flow cytometry for all four cell lines (Fig. 1C).
Gene expression profiling of CD40-sensitive and CD40resistant DLBCL lines.To identify preexisting gene expression patterns that may predict susceptibility to CD40-mediated cell death, RNA from unstimulated OCI-Ly1, OCI-Ly8, OCI-Ly7, and Su-DHL4 cells was analyzed on Affymetrix U133A oligonucleotide arrays.In total, 304 genes were differentially expressed among CD40-sensitive and CD40-resistant cells (Supplementary Table S1).Further analysis was restricted to subsets of genes likely to be involved in CD40 signaling, including regulators of apoptosis, B-cell-specific genes, and NF-nB-regulated genes.Susceptibility to CD40-mediated apoptosis was not associated with differences in expression of proapoptotic or antiapoptotic transcripts, and there was no statistically significant difference in expression of NF-nB-regulated genes (Table 1).Differences in expression of  transcripts encoding B-cell maturation-specific genes and members of the CD40 signaling pathway were statistically significant (Table 1).CD40-resistant OCI-Ly1 and OCI-Ly8 cells expressed significantly higher levels of pre-B-cell genes, including RAG1, RAG2, IGLL1, CD9, and VPREB1, whereas CD40-sensitive OCI-Ly7 and Su-DHL4 cells expressed higher levels of CD22 and CD38 (Fig. 2).CD40-sensitive cells also expressed higher levels of several genes in the CD40 signaling pathway, including Bruton's tyrosine kinase, VAV, LYN, LCK, and MEK1/MAP2K1 (Fig. 2).Differential expression of several genes was confirmed by reverse transcription-PCR (RT-PCR; Fig. 3).Interestingly, transcripts for two of these genes could only be detected in one group of cell lines: RAG1 was easily detectable in CD40-resistant OCI-Ly1 and OCI-Ly8 cells but absent in CD40-sensitive OCI-Ly7 and Su-DHL4 cells, and VAV1 was present in CD40-sensitive cells but undetectable in CD40-resistant cells.
Signaling in the LCK-VAV-MAPK pathway differs in CD40sensitive and CD40-resistant cell lines.In lymphoma cells, CD40 signaling is mediated by three interacting pathways.Current evidence suggests that NF-nB pathway is the most important of these pathways for controlling cell survival.Surprisingly, our microarray studies showed that CD40 sensitivity was not associated with differences in expression of transcripts encoding members of this pathway but rather with changes in the expression level of LCK and VAV1, which activate the Ras-Raf-ERK pathway.To further characterize the association between the MAPK pathway and CD40 sensitivity, we investigated activation of LCK (which activates VAV1) and MAPKs (which are activated by VAV1).LCK is a member of the Src protein family of kinases, which can be phosphorylated at two sites, one of which activates the kinase, whereas the other is inhibitory.Western blots showed that activated LCK was present in all four cell lines in the absence of CD40 stimulation (Fig. 4B).This is consistent with previous results that have identified constitutively active LCK in the majority of B-cell malignancies (27).In contrast, only CD40-sensitive OCI-Ly7 and Su-DHL4 cells (but not CD40-resistant OCI-Ly1 or OCI-Ly8 cells) contained constitutively phosphorylated ERK p42/p44 (Fig. 4B).As p38 and JNK are activated by CD40 (28) and are downstream targets of VAV (29,30), we investigated the phosphorylation state of these MAPK family proteins.p38 was constitutively phosphorylated in all cell lines, and JNK was present but not phosphorylated in any of the cell lines (data not shown).
LCK transgenic mice have been shown to develop thymic lymphomas containing constitutively activated VAV and ERK, and cell lines derived from such tumors were dependent on tyrosine phosphorylation and Raf-dependent ERK activation for survival (31).CD40-sensitive cell lines but not CD40-resistant cell lines .Functional analysis of differentially expressed CD40 signaling proteins.A, two genes involved in CD40-mediated ERK signaling are differentially expressed among CD40-sensitive and CD40-resistant cell lines.Differences in expression levels based on oligonucleotide array analysis.Fold difference in expression could not be calculated for VAV, as this mRNA was not detectable in CD40-resistant cells.Sites of action of two kinase inhibitors.B, activity of LCK and ERK was analyzed with phosphorylation-specific antibodies.LCK was immunoprecipitated from nondenatured cell lysates and detected by Western blotting with phosphorylated Src family antibody (phospho-SRC ).The blot was stripped and reprobed with an antibody against LCK.A commercially available Jurkat cell lysate was used as a positive control.This sample was used without prior immunoprecipitation, allowing distinction of LCK from the slightly lower mobility band representing the antibody used in immunoprecipitation.ERK was detected by immunoblotting with an antibody recognizing phosphorylated ERK1/2 (phospho ), and the blot was reprobed with a pan-specific anti-ERK1/2 antibody (total ).C, sensitivity of the cell lines to inhibition of LCK and ERK activity was tested by addition of the Src family kinase inhibitor PP1 or the MEK1/2 inhibitor U0126.Cells were incubated for 96 hours in inhibitor concentrations ranging from 10 À4 to 10 À8 mol/L, and viability was assessed by MTT assay.Viability is expressed as a percentage of the absorbance compared with untreated samples.Results of two experiments with quadruplicate samples each were combined.Bars, SD.
Figure 3. RT-PCR analysis of expression levels of genes in the CD40 signaling pathway.Expression of several genes, which have shown differential expression by oligonucleotide array analysis, was verified by RT-PCR.PCR products were analyzed by gel electrophoresis and photographed with a Syngene gel documentation system.A, representative gels from three experiments.B, relative levels of expression were quantitated with GeneTools software by comparison with a standard curve generated by amplification of PCR product from serially diluted cDNA.Results from triplicate samples.Bars, SD; where these are not visible, variations between samples were very small.N.D., not detected.
express VAV, a central part of the LCK-ERK signal transduction pathway; therefore, we would expect growth of CD40-sensitive cells but not CD40-resistant cells to be inhibited by blocking LCK or ERK activity.Cells were exposed to a range of concentrations of the Src family kinase inhibitor PP1 (32).As predicted by our model, growth of CD40-sensitive OCI-Ly7 and Su-DHL4 cells was inhibited more efficiently than growth of CD40-resistant OCI-Ly1 and OCI-Ly8 cells (Fig. 4C).In contrast, similar experiments with the MEK1/2 inhibitor U0126, which prevents phosphorylation of ERK (33), did not show differential effects on growth of CD40-sensitive and CD40-resistant DLBCL lines (Fig. 4C).These observations suggest that LCK but not ERK is involved in maintaining growth of CD40-sensitive DLBCL cell lines.
Phosphorylated ERK is required for CD40-mediated cell death.CD40 stimulation has been shown to activate both ERK and p38; however, inhibition of ERK but not p38 phosphorylation has been reported to enhance CD40-mediated apoptosis in a carcinoma cell line (34).We therefore determined the effect of CD40 stimulation on ERK activation by Western blotting for phosphorylated and total ERK.ERK was constitutively phosphorylated in OCI-Ly7 and Su-DHL4 cells and not phosphorylated before or after CD40 ligation in OCI-Ly1 and OCI-Ly8 cells (Fig. 5A).This suggests that, unlike carcinoma cell lines, DLBCL cell lines do not up-regulate ERK activity on CD40 signaling; however, constitutively active ERK may influence the outcome of CD40 signaling.For example, if ERK protects DLBCL against CD40mediated apoptosis, inhibition of ERK before CD40 stimulation should enhance cell death in OCI-Ly7 and Su-DHL4 cells by reducing constitutive ERK activity, whereas OCI-Ly1 and OCI-Ly8 cells that lack active ERK should be unaffected by an ERK inhibitor.This hypothesis was tested using pharmacologic inhibitors of MEK.Phosphorylation of ERK was reduced after addition of the MEK inhibitor U0126 (Fig. 5B).To determine if ERK is involved in CD40mediated cell death, cells were treated with the MEK1/2 inhibitors U0126 or PD98058 before CD40 stimulation.The p38 inhibitor SB230580, which has been shown previously to have no effect on activation-induced cell death (11), was used as a negative control.Viable cell counts were measured 96 hours after CD40 ligation.The MEK1/2 inhibitors U0126 and PD98058 but not the p38 inhibitor SB230580 reduced CD40-mediated cell death in OCI-Ly7 and Su-DHL4 cells (Fig. 5C), suggesting that ERK promotes rather than opposes activation-induced cell death in DLBCL cell lines.

Discussion
Differences in the effects elicited by CD40 stimulation in B-cell lines suggest that different signal transduction pathways may be expressed in CD40-sensitive and CD40-resistant cells.To assess this, we did expression microarray studies on CD40-sensitive and CD40-resistant DLBCL cell lines.Our observation that CD40sensitive and CD40-resistant cells display different gene expression profiles suggests that these lines may be derived from two distinct B-cell populations.CD40-sensitive cells overexpressed CD22 and CD38, which are found on mature, activated B cells (35).CD40resistant cells expressed high levels of CD9, the recombination activating genes RAG1 and RAG2, and the pre-B-cell receptor genes IGLL1 and VPREB1, which are characteristics of pre-B cells at the stage of immunoglobulin rearrangement (36)(37)(38)(39) and have also been detected in B lymphocytes in germinal centers (40).This suggests that the CD40-sensitive OCI-Ly7 and Su-DHL4 cell lines may be derived from mature, activated B cells, whereas the CD40-resistant OCI-Ly1 and OCI-Ly8 cell lines resemble immature B cells.We investigated expression of members of the CD40 signaling pathway to determine the underlying mechanism of CD40mediated cell death.Failure of OCI-Ly1 and OCI-Ly8 cells to undergo apoptosis on CD40 stimulation may be the result of defects in the CD40 signal transduction pathway.These two cell lines showed reduced expression of several genes involved in CD40 signaling, including LCK, VAV, and MEK1.Analysis of LCK by RT-PCR and Western blot revealed that although RNA levels differed among CD40-sensitive and CD40-resistant lines, all four cell lines expressed significant amounts of active LCK.This suggests that differences in LCK mRNA levels may not have any functional consequences.However, differences in VAV, a phosphorylation target of LCK, were striking, with strong mRNA expression in CD40-sensitive lines and undetectable levels in CD40resistant lines.LCK and VAV have been shown previously to maintain constitutive activation of ERK via stimulation of the Ras pathway (31), which is consistent with our observation that ERK was constitutively phosphorylated in CD40-sensitive DLBCL cell lines but permanently inactive in VAV-deficient CD40-resistant lines.Although all four cell lines expressed active LCK, the Src family inhibitor PP1 was more effective at reducing proliferation of OCI-Ly7 and Su-DHL4 cells.Differential sensitivity to PP1 could result from inhibition of other Src family kinases or from the differential function of downstream effectors, such as VAV, Ras, and ERK.Lack of VAV and inactive ERK in OCI-Ly1 and OCI-Ly8 cells is consistent with a dead end in the Src signal transduction pathway.This could render the cells resistant to Src family kinase inhibitors even if the targeted kinase itself is active.
ERK activation has often been reported to be antiapoptotic in both lymphoid and nonlymphoid malignancies (34,41); however, the effect of ERK phosphorylation varies even among different stimuli in the same cell line (12).We found that two different ERK inhibitors did not affect the growth of DLBCL cell lines containing activated ERK, which suggests that these lines are not dependent on ERK signaling for survival or proliferation.Addition of ERK inhibitors before CD40 stimulation blocked activation-induced cell death, which indicates that overexpression or aberrant activation of a protein in the ERK signaling cascade may sensitize DLBCL cell lines to CD40.Our observation that ERK is constitutively active in these cell lines suggests that the actual death signal must be initiated by a second pathway when CD40 signaling occurs.ERK has been implicated in apoptosis in other systems of activationinduced cell death both directly and as a predisposing factor.T-cell receptor (TCR)-mediated activation-induced cell death in a TCR hybridoma cell line was shown to be mediated by activation of VAV and ERK (42).Transfection of Rat-1 fibroblasts or MCF7 human breast cancer cells with the ERK-regulated transcription factor Elk-1 did not induce apoptosis directly but rendered the cells susceptible to killing by a calcium ionophore (38).ERK may function in a similar way in DLBCL cell lines, rendering cells susceptible to a second signal caused by CD40 stimulation.However, the signal is unlikely to be calcium mediated, as treatment of the cells with the calcium ionophore ionomycin induced cell death in Su-DHL4 cells but not OCI-Ly7 cells. 3 Our observations on two pairs of cell lines are insufficient to determine whether differential expression of maturation-specific genes outside of the LCK-ERK pathway plays a role in determining susceptibility to CD40-mediated cell death; however, they are consistent with a previous report that B cells at different stages of development differ in the protein phosphorylation patterns elicited by CD40 stimulation (43).As CD40 signaling can promote proliferation and resistance to apoptosis in normal B cells (44), it is possible that constitutive activation of this pathway in a tumor may lead to a more aggressive phenotype.DLBCL has been shown to segregate into two subtypes with distinct gene expression patterns, one resembling germinal center cells (germinal center type) and the other similar to mature B cells that have been subjected to CD40 and B-cell receptor stimulation (activated B-cell type; ref. 45).The prognosis was shown to differ significantly, with 5-year survival being 76% for germinal center and 34% for activated-type DLBCL (46).Neither the prevalence of constitutive ERK activation in either subtype of DLBCL nor the correlation with CD40 sensitivity has yet been investigated in tumor tissue.This area may be fruitful for future investigation based not only on our observations but also on recent development of inhibitors that modulate relevant signal transduction pathways.The Src-VAV-ERK signal transduction pathway is activated by both CD40 and B-cell receptor signaling (12) and has been implicated in proliferation of B-cell malignancies (41).Several proteins in this pathway, including CD40, Src family kinases, Ras, and MEK, are being investigated as targets for antitumor therapies (47)(48)(49)(50).
Our studies show that increased expression of genes involved CD40/BCR signal transduction coincided with constitutive ERK activation and susceptibility to CD40-mediated cell death in DLBCL cell lines.A constitutively active phenotype has been previously associated with reduced survival in DLBCL (45).This raises the question whether the poor prognosis of activated type DLBCL may be the result of one or more overactive proteins in the CD40 signal transduction pathway.An interesting but as yet untested hypothesis is that the mechanism underlying the aggressive nature of activated DLBCL may also be its Achilles heel, leaving it vulnerable to therapies targeting the CD40 signal transduction pathway.

Figure 1 .
Figure1.Antiproliferative and apoptotic effects of CD40 signaling differ among DLBCL lines.Cells were seeded at 1 Â 10 5 /mL in supplemented RPMI 1640 containing 10 Ag/mL anti-CD40 and 10 Ag/mL cross-linking secondary antibody (anti-CD40 ) or secondary antibody alone (control ).A, for assessment of cell viability, aliquots were removed at 24-hour intervals, stained with propidium iodide without prior permeabilization, and analyzed by flow cytometry using a fixed-time setting of 30 seconds to quantitate the number of viable cells per unit volume.Results of two experiments of three replicates each were combined.Bars, SD; where these are not visible, they are covered by the overlying symbol.Growth of OCI-Ly7 and Su-DHL4 cells was inhibited by cross-linked anti-CD40.For quantitation of apoptosis, cells were stimulated for 48 hours with anti-CD40 or control antibody as described above and then permeabilized with ethanol before staining with propidium iodide.Intracellular DNA content was analyzed by flow cytometry.B, cells containing less DNA than the G 1 fraction, representing apoptotic cells, were quantitated.Results from two experiments of three replicates each.Bars, SD. *, P < 0.01, statistically significant differences between control and anti-CD40-treated cells.The fraction of cells with sub-G 1 DNA content increased in OCI-Ly7 and Su-DHL4 cells.C, CD40 could be detected on all four DLBCL cell lines by flow cytometry.Open histograms, fluorescence from phycoerythrin-conjugated anti-CD40 antibody; shaded histograms, similarly conjugated isotype control antibody.Cells were not permeabilized before staining, thereby restricting staining to antigens exposed on the cell surface.

Figure 2 .
Figure 2. Expression profiles of CD40-sensitive and CD40-resistant DLBCL: B-cell markers and CD40 signaling.mRNA from unstimulated DLBCL lines was analyzed on Affymetrix HG-U113A v 2.0 GeneChips as described in Materials and Methods.B-cell differentiation markers and genes in the CD40/BCR signaling pathway were compared in triplicate samples of each of the four cell lines.Normalized log 2 expression values.Brackets, hierarchical clustering done with Genesis software (23); horizontal bar, group of cosegregating pre-B-cell markers (pre-B ).

Figure 4
Figure 4. Functional analysis of differentially expressed CD40 signaling proteins.A, two genes involved in CD40-mediated ERK signaling are differentially expressed among CD40-sensitive and CD40-resistant cell lines.Differences in expression levels based on oligonucleotide array analysis.Fold difference in expression could not be calculated for VAV, as this mRNA was not detectable in CD40-resistant cells.Sites of action of two kinase inhibitors.B, activity of LCK and ERK was analyzed with phosphorylation-specific antibodies.LCK was immunoprecipitated from nondenatured cell lysates and detected by Western blotting with phosphorylated Src family antibody (phospho-SRC ).The blot was stripped and reprobed with an antibody against LCK.A commercially available Jurkat cell lysate was used as a positive control.This sample was used without prior immunoprecipitation, allowing distinction of LCK from the slightly lower mobility band representing the antibody used in immunoprecipitation.ERK was detected by immunoblotting with an antibody recognizing phosphorylated ERK1/2 (phospho ), and the blot was reprobed with a pan-specific anti-ERK1/2 antibody (total ).C, sensitivity of the cell lines to inhibition of LCK and ERK activity was tested by addition of the Src family kinase inhibitor PP1 or the MEK1/2 inhibitor U0126.Cells were incubated for 96 hours in inhibitor concentrations ranging from 10 À4 to 10 À8 mol/L, and viability was assessed by MTT assay.Viability is expressed as a percentage of the absorbance compared with untreated samples.Results of two experiments with quadruplicate samples each were combined.Bars, SD.

Figure 5 .
Figure 5. Constitutive ERK phosphorylation correlates with CD40-mediated cell death.A, assessment of the phosphorylation state of ERK.Cytoplasmic cell extracts were prepared from cells stimulated with cross-linked anti-CD40 antibody for 0 to 180 minutes.Total and phosphorylated ERK was detected by Western blot.Blots were first probed with the phosphospecific antibody, and the same membrane was reprobed for total protein with the pan-specific antibody.B, reduction of ERK phosphorylation by U0126.Su-DHL4 cells were incubated in the presence of 10 Amol/L U0126 for 48 hours, permeabilized and stained with an anti-phosphorylated ERK antibody as described in Materials and Methods, and analyzed by flow cytometry.C, effects of ERK inhibition on CD40-mediated apoptosis.Cells were subjected to CD40 stimulation for 96 hours in the presence of the MEK1/2 inhibitors U0126 (10 Amol/L) or PD98059 (50 Amol/L) or the p38 inhibitor SB203580 (10 Amol/L).Cell viability was quantitated by staining nonpermeabilized cells with propidium iodide and counting the number of unstained (live) cells per unit volume using a Becton Dickinson FACScan flow cytometer set to count for 30 seconds.The number of viable cells in anti-CD40-treated samples was expressed as a percentage of the number of viable cells incubated in secondary antibody only.Results from three experiments of four replicates each.Bars, SD.In the presence of DMSO (control) or the p38 inhibitor SB203580, the number of viable OCI-Ly7 and Su-DHL4 cells was reduced by CD40 ligation (P < 0.01).The MEK1/2 inhibitors U0126 and PD98059 prevented CD40-mediated reduction in viability.