Nurselike cells express BAFF and APRIL, which can promote survival of chronic lymphocytic leukemia cells via a paracrine pathway distinct from that of SDF-1alpha

Abstract

We examined expression of B cell-activating factor of the tumor necrosis factor (TNF) family (BAFF) and a proliferation-inducing ligand (APRIL) on chronic lymphocytic leukemia (CLL) B cells and nurselike cells (NLCs), which differentiate from CD14+ cells when cultured with CLL B cells. NLCs expressed significantly higher levels of APRIL than monocytes and significantly higher levels of BAFF and APRIL than CLL B cells. Also, the viability of CLL B cells cultured with NLCs was significantly reduced when CLL B cells were cultured with decoy receptor of B-cell maturation antigen (BCMA), which can bind both BAFF and APRIL, but not with BAFF receptor:Fc (BAFF-R:Fc), which binds only to BAFF. The effect(s) of BAFF or APRIL on leukemia cell survival appeared additive and distinct from that of stromal cell-derived factor-1alpha (SDF-1alpha), which in contrast to BAFF or APRIL induced leukemia cell phosphorylation of p44/42 mitogen-activated protein kinase (extracellular signal-regulated kinase-1/2 [ERK1/2]) and AKT. Conversely, BAFF and APRIL, but not SDF-1alpha, induced CLL-cell activation of the nuclear factor-kappaB1 (NF-kappaB1) and enhanced CLL-cell expression of the antiapoptotic protein Mcl-1. However, BAFF, but not APRIL, also induced CLL-cell activation of NF-kappaB2. We conclude that BAFF and APRIL from NLCs can function in a paracrine manner to support leukemia cell survival via mechanisms that are distinct from those of SDF-1alpha, indicating that NLCs use multiple distinct pathways to support CLL-cell survival.

Figure 1.

Expression of BAFF mRNA and protein. (A) Quantitative real-time RT-PCR was performed on RNA samples isolated from the blood mononuclear cells of individual patients with CLL before (left) and after (right) depletion of CD2⁺ and CD14⁺ cells. The lines connect the preisolation and postisolation levels of BAFF mRNA detected in each sample. The amount of BAFF mRNA detected is indicated in arbitrary units. The amount of BAFF mRNA detected in an equivalent number of U937 cells is 1000 units (data not shown). (B) Quantitative real-time RT-PCR measurement of the average amount of BAFF mRNA detected in CD14⁺ cells (n = 4), NLCs (n = 12), purified CLL B cells (n = 12), and isolated CD19⁺ blood B cells of healthy donors (n = 2), as indicated at the bottom of the panel, ± SD (^**the level of BAFF mRNA detected in NLCs was significantly greater than that found in isolated CLL B cells, P < .001). (C) Reconstitution experiments in which small numbers of CD14⁺ blood mononuclear cells are added to 5 × 10⁶ isolated CLL B cells that subsequently were evaluated for BAFF mRNA in 2 representative patients. On the x-axis is the percent of CD14⁺ cells detected by FACS in the reconstituted cell population prior to extraction of RNA. The y-axis indicates the level of BAFF mRNA detected in units as defined in the description for panel A. (D) Representative histograms depicting surface BAFF detected by flow cytometry on CD14⁺ cells, NLCs, CD19⁺ CLL B cells, or CD19⁺ blood B cells of healthy donors, as indicated at the top of each graph. Shaded histograms represent the fluorescence of cells stained with a fluorochrome-labeled anti-BAFF mAb, whereas the open histograms depict the fluorescence of cells stained with an isotype control mAb. (E) An immunofluorescence picture of NLCs and CLL cells stained with fluorescein-labeled anti-CD19 mAb (green) and a PE-labeled anti-BAFF mAb (red). The nuclei are labeled blue with Hoechst 33342.

Figure 2.

Expression of APRIL mRNA and protein. (A) Quantitative real-time RT-PCR was performed on RNA samples isolated from the blood mononuclear cells of patients with CLL before (left) and after (right) depletion of CD2⁺ and CD14⁺ cells. The lines connect the preisolation and postisolation levels of APRIL mRNA in each sample. The amount of APRIL mRNA detected is indicated in arbitrary units. The amount of APRIL mRNA detected in an equivalent number of U937 cells is 30 units (data not shown). (B) Quantitative real-time RT-PCR measurement of the average amount of APRIL mRNA detected in CD14⁺ cells (n = 4), NLCs (n = 11), purified CLL B cells (n = 11), or isolated CD19⁺ blood B cells of healthy donors (n = 3), as indicated at the bottom of the histogram, ± SD (^**the mean level of APRIL mRNA detected in NLCs was significantly greater than that found in isolated CLL B cells, P < .01). (C) Representative immunoblot data showing the expression of APRIL by NLCs, CD14⁺ blood mononuclear cells, CLL B cells, or isolated CD19⁺ blood B cells of healthy donors. Whole cell lysates were prepared as described in “Materials and methods.” The protein content was normalized to 20 μg and subjected to immunoblot analysis with antibodies specific for APRIL or β-actin using ECL-based detection. (D) An immunofluorescence picture of NLCs and CLL cells stained with phycoerythrin-labeled anti-CD19 mAb (red) and goat IgG anti-APRIL polyclonal antibody that was detected using a fluorescein-labeled anti–goat IgG (green). The nuclei are labeled blue with Hoechst 33342.

Figure 3.

CLL-cell survival with or without NLCs. (A) Inhibition of CLL-cell survival on NLCs by BCMA-Fc but not BAFF-R:Fc. CLL B cells were cultured with (□) or without (▪) NLCs and 1 μg/mL control Ig. BCMA-Fc (▴) or BAFF-R:Fc (•) at 1 μg/mL was added to the wells of CLL B cells cultured with NLCs at day 0. Viability was subsequently determined for each time point, as indicated on the horizontal axis. Displayed is the mean percent viability ± SD (error bars) of samples from each of 5 patients. The percent viability of BCMA-Fc–treated cultures was significantly less than that of control Ig–treated cultures (^*P < .05; ^**P < .01; Bonferroni t test). (B) Enhanced CLL-cell survival with NLCs or rhBAFF or rhAPRIL. A total of 1 × 10⁶/mL isolated CD19⁺ CLL B cells was cultured alone (□), with 50 ng/mL rhBAFF (▴), 500 ng/mL rhAPRIL (•), both rhBAFF and rhAPRIL (○), or NLCs (▪) and evaluated over time. Displayed is the mean percent viability ± SD of samples from each of 3 patients. The percent viability of rhBAFF-treated CLL cells or rhAPRIL-treated CLL cells was significantly greater than that of control-treated CLL cells (^*P < .05; ^**P < .01; Bonferroni t test).

Figure 4.

Effect of rhBAFF, rhAPRIL, and/or SDF-1α on CLL-cell survival. CLL B cells were cultured with (□) or without (▪) NLCs. SDF-1α (•) rhAPRIL (♦) at 500 ng/mL, rhBAFF (▴) at 50 ng/mL, or both (⋄) were added to wells without NLCs at day 0. Also SDF-1α and rhBAFF (⋄) or SDF-1α and rhAPRIL (○) were added to the cultures without NLCs. The mean viability ± SE of replicate wells was determined for each time point indicated on the horizontal axis. A representative example of 3 different CLL patients is presented.

Figure 5.

Activation of NF-κB in CLL B cells by rhBAFF or rhAPRIL. (A) Processing of p100 and nuclear translocation of p52 or p65. CLL B cells were cultured with or without SDF-1α (500 ng/mL), rhBAFF (50 ng/mL), and rhAPRIL (500 ng/mL) for 24 hours. Cytoplasmic and nuclear extracts were prepared as described in “Materials and methods” for immunoblot analysis with anti-p100 or anti-p65 antibodies as indicated on the left side of each panel. The agent used to treat the CLL cells is indicated at the top of each panel under the label indicating whether the extract was derived from cytoplasmic (left panel) or nuclear (right panel) cell fractions. We evaluated for equal loading in each lane by stripping the blot and probing it again with antibodies specific for β-actin (for cytoplasmic extracts) or SP-1 (for nuclear extracts), as indicated on the left side of each panel. (B) Degradation of IκBα. Extracts of CLL cells were prepared for immunoblot analysis prior to treatment (Pre-Tx) or after a 30-minute incubation with culture medium alone (Medium) or medium supplemented with SDF-1α (500 ng/mL), rhBAFF (50 ng/mL), rhAPRIL (500 ng/mL), or TNFα (50 ng/mL), as indicated at the top of each lane. The immunoblot was probed with antibodies specific for IκBα (top blot). We evaluated for equal loading in each lane by stripping the blot and probing it again with antibodies specific for β-actin (bottom blot).

Figure 6.

Activation of MAPK (ERK1/2) and AKT in CLL cells. (A) CLL B cells were cultured for 3 or 10 minutes with SDF-1α (200 ng/mL), rhBAFF (50 ng/mL), or media, as indicated above the sample lanes. Cell lysates were prepared and analyzed by immunoblot using antibodies specific for phosphorylated ERK1/2 (P-ERK1/2), ERK1/2, phosphorylated AKT (P-AKT^Ser473), or AKT, as indicated on the left margin. Equal loading in the lanes was evaluated by stripping the blot and probing again with anti-ERK1/2 and an anti-AKT antibody. Five different CLL B cells gave similar results. (B) The CLL cells were stimulated for 3 minutes with either media (far left lane) or SDF-1α (200 ng/mL) (right 3 lanes). For samples treated with SDF-1α we included the CXCR4 antagonist 4F-benzoyl-TE14011 (4F) at 0 nM, 50 nM, or 500 nM. The samples were analyzed and the results presented as noted in panel A.

Figure 7.

Expression of Mcl-1 in CLL B cells by immunoblot analysis. Representative immunoblot data showing up-regulation of Mcl-1 by NLCs or rhBAFF. CLL B cells were cultured with or without NLCs or with SDF-1α (500 ng/mL) or rhBAFF (50 ng/mL) for 24 hours. Whole cell lysates were then prepared. The protein content was normalized to 12.5 μg and analyzed by immunoblot analysis with antibodies specific for Mcl-1, Bax, Bcl-2, or β-actin using ECL-based detection.