Ligation of CD40 receptor in human B lymphocytes triggers the 5-lipoxygenase pathway to produce reactive oxygen species and activate p38 MAPK

Abstract

Previously, we reported that CD40-induced production of reactive oxygen species (ROS) by NADPH oxidase requires the TNF receptor-associated factor (TRAF) 3, as well as the activities of phosphatidylinositol 3-kinase (PI3K) and Rac1. Here we investigated the possible mechanisms of the production of ROS after CD40 ligation in B cells. We describe an alternative ROS production pathway that is triggered by CD40 ligation, involves 5-lipoxygenase (5-LO), and results in activation of p38 MAPK. Our studies in Raji human B lymphomas revealed that CD40-induced ROS production by 5-LO also requires the activities of PI3K and Rac1. In contrast to the NADPH oxidase pathway, however, TRAF molecules are not required for the CD40-induced ROS production by 5-LO. The association of CD40 with 5-LO is dependent on CD40 ligation in Raji B cells, and co-immunoprecipitation experiments using epitope- tagged proteins transiently expressed in human embryonic kidney 293T cells revealed the role of the regulatory subunit of PI3K, p85, in this association. Collectively, these data suggest a separate pathway for the CD40-induced ROS production in B cells and demonstrate that this pathway requires 5-LO via direct association of p85 with both CD40 and 5-LO.

Figure 1

ROS production after CD40 ligation in Raji human B cells. (A) The cells (1 × 10⁶) that were incubated with 20 µM DCFDA for 15 min were stimulated with either control Ig or anti-CD40 monoclonal Ab (10 µg/ml). When pretreated with the inhibitors, the cells were preincubated in the presence of inhibitors for 30 min before the addition of DCFDA. Representative microscopic fields of 2', 7'-dichlorofluorescein (DCF) fluorescence from the Raji cells after a 20-min stimulation with anti-CD40 in the absence or presence of treatment with several inhibitors (15 µM DPI, 35 µM ETYA, 0.5 µM MK-886, 10 µM LY294002, or 0.1 µM wortmannin). Fluorescence of the cells after stimulation with control Ig or CD40 ligation is shown. Data represent three independent experiments. (B) The DCF fluorescence intensity in Raji B cells after a 20-min stimulation with anti-CD40 is shown. The different treatments are indicated. The numbers of cells (n = 35) on the fluorescence image from all three experiments are given. The data are presented as mean ± SD. Statistical significance was determined by two-way analysis of variance, and the data was subjected to paired t-tests as post-hoc analysis. ^***, P < 0.001 indicates significantly lower values for cells under CD40 ligation than values for cells either untreated or treated with DPI.

Figure 2

Correlation of ROS production and p38 MAPK activation after CD40 ligation. (A) The cells (5 × 10⁶) were pretreated with 30 mM NAC for 1 h or were untreated, and then the cells were stimulated with medium containing control Ig (control) or 10 µg/ml anti-CD40 for various times. Some of the untreated cells (5 × 10⁶) were also incubated with 300 µM H₂O₂ for the indicated times as a positive control for ROS-mediated p38 activation. The lysates from 1 × 10⁶ cells were subjected to SDS-PAGE and a phospho-p38 (P-p38) WB analysis. The same blot was stripped and re-probed with anti-p38 Ab to insure equal loading of the cell lysates in each lane. (B-D) The cells were pretreated with either 15 µM DPI (B), 35 µM ETYA, or 0.5 µM MK886 (C), 0.1 µM wortmannin, or 10 µM LY294002 (D) or left untreated before CD40 ligation. Cells were also treated with 0.6 M D-sorbitol (S) for 30 min as a positive control for p38 activation. The data shown represent three separate experiments. The intensities of the bands in the phospho-p38 and p38 WB were quantitated, and normalized values (P-p38/p38) were calculated. The data shown are the average of the normalized values from three separate WB for each inhibitor treatment. The data are presented as mean ± SD. Statistical significance was determined by two-way analysis of variance, and paired t-tests were used as post-hoc analysis. ^*, P < 0.05 and ^**, P < 0.01 indicate values significantly lower than control at the same time point.

Figure 3

Effects of N17Rac1, DNTRAF 2, 3, or 6 on CD40-induced ROS production. The cells were transiently transfected with either empty vector or one of the DN expression plasmids, along with a red fluorescent protein, DsRed. Intracellular ROS production after a 20-min stimulation with anti-CD40 was measured (described in Figure 1). (A) Changes in the fluorescence to yellow, due to the green fluorescence from CD40-induced ROS production, are clearly shown in cells that were doubly transfected with the vector and DsRed plasmids. In the case of the concurrent expression of DsRed and a DN mutant of Rac1 (N17Rac1), however, not all of the cells responded to the CD40 stimulation that led to fluorescence changes. The fluorescence intensity after CD40 ligation relative to the control is shown in cells that were transfected either with a vector or a DN mutant of Rac1 (N17Rac1). The average fluorescence intensity from at least 50 cells from three experiments are presented as the average ± SD; ^*, P < 0.001. (B) The relative fluorescence intensity from ROS production after CD40 ligation for 20 min in cells that were transfected with either vector alone or with each DNTRAF is given as a ratio to control fluorescence. Data obtained as in (A) are presented as the average ± S.D.

Figure 4

Molecular associations in CD40 signaling complexes involved in ROS production. (A) HEK 293T cells were transfected transiently with combinations of plasmids: human CD40 plasmid with myc epitope, hCD40(myc) and either empty vector, p85(myc), or 5-LO(flag); 5-LO plasmid with either empty vector or hCD40; and p85 plasmid with either empty vector or hCD40. Analyses of the anti-CD40 immune complexes show specific associations of CD40 with p85 but not with 5-LO. In reciprocal experiments, CD40 was detected in p85 but not in 5-LO immunoprecipitates. (B) HEK 293T cells were transfected transiently with a combination of p85(myc) and either empty vector or 5-LO(flag) or vice versa. Analyses of the immune complexes of the cell lysates with anti-myc Ab show a selective association of p85 with 5-LO. Reciprocal experiments also showed associations of these molecules. The expression of the plasmids was determined by WB of the lysates using Abs against epitope tags. The data shown represent three separate experiments.

Figure 5

Recruitment of 5-LO to the CD40 receptor complex through association with p85 after stimulation in Raji B cells. (A) The lysates from 1 × 10⁷ Raji B cells that were stimulated either with anti-mouse (control) or anti-CD40 Ig for 10 min were tumbled with GammaBind G sepharose beads. The immune complexes for each stimulation condition were then subjected to SDS-PAGE, which was followed by WB with anti-p85, anti-5-LO, and anti-CD40 Ab. The data represent three independent experiments. (B, C) The Raji B cells, either unstimulated (control) or stimulated with anti-CD40 for 15 min, were labeled for endogenous p85 (green) or CD40 (green) along with 5-LO (red) using anti-p85, anti-CD40, and anti-5-LO, followed by FITC- and TRITC-conjugated secondary Ab, respectively. Cell nuclei were labeled using DAPI (blue). Fluorescence was visualized under a confocal microscope. The merged staining patterns of p85-5-LO and CD40-5-LO are shown in the middle panels. The merged staining patterns of all three colors are shown in the far right panels. Representative images from three separate experiments with similar results are shown.

Figure 6

A schematic diagram showing the signaling pathways in the CD40-induced ROS production. The solid lines represent direct protein: protein associations and the dashed line represent the presence of intermediary steps. As shown, the NADPH oxidase pathway is presented in gray color.