Engagement of the BCR with Ags triggers signaling pathways for commitment of B lymphocyte responses that can be regulated, in part, by reactive oxygen species. To investigate the functional relevance of reactive oxygen species produced in primary B cells, we focused on the role of the hydrogen peroxide generator Duox1 in stimulated splenic B cells under the influence of the TH2 cytokine IL-4. We found that H2O2 production in wild type (WT) and Nox2-deficient CD19+ B cells was boosted concomitantly with enhanced expression of Duox1 following costimulation with BCR agonists together with IL-4, whereas stimulated Duox1−/− cells showed attenuated H2O2 release. We examined whether Duox1-derived H2O2 contributes to proliferative activity and Ig isotype production in CD19+ cells upon BCR stimulation. Duox1−/− CD19+ B cells showed normal responses of Ig production but a higher rate of proliferation than WT or Nox2-deficient cells. Furthermore, we demonstrated that the H2O2 scavenger catalase mimics the effect of Duox1 deficiency by enhancing proliferation of WT CD19+ B cells in vitro. Results from immunized mice reflected the in vitro observations: T cell–independent Ag induced increased B cell expansion in germinal centers from Duox1−/− mice relative to WT and Nox2−/− mice, whereas immunization with T cell–dependent or –independent Ag elicited normal Ig isotype secretion in the Duox1 mutant mice. These observations, obtained both by in vitro and in vivo approaches, strongly suggest that Duox1-derived hydrogen peroxide negatively regulates proliferative activity but not Ig isotype production in primary splenic CD19+ B cells.

The strength of signaling pathways triggered by engagement of BCR is critical for the commitment to B cell proliferation and differentiation for adaptive immune responses (1, 2). Cross-linking of BCR by Ags initiates activation of receptor proximal protein tyrosine kinases, such as Syk and Lyn, and rapid downstream phosphorylation of several tyrosine kinase substrate proteins, including Btk, phospholipase-Cγ2 (PLCγ2), and Akt kinase (3). In contrast, BCR stimulation also recruits inhibitory regulators of tyrosine phosphorylation (protein tyrosine phosphatases [PTPs]) such as Src homology region 2 domain–containing phosphatase 1 (SHP-1), which binds to ITIMs on CD22 for negative regulation of BCR signaling (4). Therefore, transmission of BCR signaling is strictly controlled by antagonistic feedback systems.

Reactive oxygen species (ROS) such as hydrogen peroxide (H2O2) have been considered potent inhibitors of PTPs in lymphocytes because cysteine residues in their catalytic sites can be readily and reversibly oxidized (3). ROS generation has been shown in B lymphocytes. Over 20 years ago, Kanegasaki’s group (5) first established that one of the NADPH oxidase family members, Nox2 (originally designated the cytochrome b558 β-subunit, or gp91phox), is constitutively expressed on the human B cell surface and that it generates superoxide anion (O2·) following cell stimulation, similar to the respiratory burst of phagocytic cells required for microbial killing. Interestingly, recent work has shown that phagocytic CD19+ B cells isolated from the mouse peritoneal cavity also exhibit Nox2-dependent microbicidal activity (6). It was suggested initially that O2· produced by Nox2 is not essential for B cell development and differentiation because chronic granulomatous disease (CGD) patients lacking Nox2 activity were not recognized as having any B cell dysfunction. However, more recent work indicated that CGD patients have reduced memory and increased naive B cell counts, which may affect secondary Ab responses (79). Studies in mice examining the relationship of Nox2-derived O2· to primary B cell responses used splenic B cells from Nox2-deficient or neutrophil cytosolic factor 1 (Ncf1/p47phox)–deficient mice that are functionally Nox2-deficient (10, 11). These studies indicated that Nox2 contributes to an early phase of superoxide O2· generation in response to BCR stimulation. However, the Nox2 defects in these models do not appear to affect normal B cell development and maturation significantly.

Dual oxidase (Duox) molecules are among the Nox family NADPH oxidase members that have been associated with signaling pathways in lymphocytes. Duox1 and Duox2 exhibit calcium-dependent generation of H2O2 produced directly from a superoxide intermediate because of activity of their unique extracellular peroxidase-like domains (12, 13). In T cells, H2O2 derived from Duox1 following TCR stimulation inactivates PTP activity of SHP-2, which enhances proximal TCR signaling through enhanced phosphorylation and activation of ZAP-70 molecules (14). A similar BCR signaling feedback system involving Duox1-derived H2O2 was proposed in B lymphocytes. Using A20 murine B lymphoma cells, this study suggested that H2O2 produced by Duox1 following BCR stimulation enhances intracellular calcium signaling and a positive feedback loop of Lyn phosphorylation by negatively regulating PTP activity of SHP-1 (15). In contrast, it has not been clear from these studies whether Duox1-derived H2O2 contributes to later phase intrinsic B cell activities, including cell proliferation or IgG production.

The function of Duox1 is boosted by TH2-type cytokines in several cell types. IL-4 and IL-13 have the ability to promote production of Duox1 and increase H2O2 generation for innate immune defense in human primary respiratory epithelial cells and pulmonary carcinoma cell lines (16, 17). Furthermore, airway TH2-based innate immune responses to allergic asthmatic challenges are diminished in mice deficient in Duox1 or Duoxa (18, 19). An enhancement of the Duox1–H2O2 axis by IL-4 or IL-13 was also observed in human primary keratinocytes (20). In this study, knocking down Duox1 alters expression of transcription factors and phosphorylation of STAT6 induced by TH2-type cytokines because of decreased H2O2 production. Together, these observations indicate that Duox1 and the H2O2 it generates participate in a positive feedback loop of TH2-type cytokine signaling in keratinocytes and respiratory epithelium.

TH2-type cytokines are well-known essential mediators of B cell function and development. The relationship between Duox1 and TH2-type cytokines has not yet been examined in the primary B cells but should be approached for a better understanding of signaling in developing B cells. In this study, we determined that Duox1 affects BCR stimulation of primary splenic B cells specifically under the influence of the TH2-type cytokine IL-4. Following this concept, ROS production mediated by Duox1 in the B cells was examined by comparing cells from wild type (WT) and Duox1 mutant mice to determine whether ROS play a regulatory role in B cell signaling. We further aimed to uncover downstream functional consequences of Duox1-derived ROS in primary B cell activities.

Mice between the ages of 8 and 12 wk were used for experiments. B6 mice (000664; C57BL/6J) and B6-background gp91phox-deficient mice (002365; B6.129S-Cybbtm1Din/J) were purchased from The Jackson Laboratory (Bar Harbor, ME). The Duox1-deficient mice were established on a B6 background by retroviral-based gene-trapping methodology (21, 22) and were kindly provided by Dr. M. Geiszt (Semmelweis University, Budapest, Hungary). Mice were housed with a standard diet and given water in a specific pathogen-free facility according to National Institutes of Health guidelines. Animal use was approved by the National Institute of Allergy and Infectious Diseases Animal Care and Use Committee.

Fluorophore-conjugated Abs directed against the following surface markers were used: B220 (RA3-6B2), CD23 (B3B4), CD69 (H1.2F3), CD93 (AA4.1), and CD86 (GL-1) from eBioscience (San Diego, CA); CD19 (1D3) from BD Pharmingen (San Diego, CA); and IgM (goat polyclonal F(ab′)2, μ-chain specific) from Jackson ImmunoResearch (West Grove, PA). B cell populations from the spleen and bone marrow (BM) were stained for 30 min on ice with anti-CD23-FITC, anti-IgM-PE-Cy7, anti-CD93-allophycocyanin, and anti-B220-allophycocyanin-eFlour 780, and B cell subpopulations from each mouse strain were investigated on an LSR II flow cytometer (BD Biosciences, San Jose, CA) by the protocol of Allman et al. (23). Dead cells were identified and excluded by propidium iodide staining (BD Pharmingen). Flow cytometry data were analyzed with FlowJo software version 10 (Tree Star, Ashland, OR).

Total splenic cells were collected from spleens by homogenization, and the single-cell suspensions were treated with lysis buffer (Lonza, Basel, Switzerland) to eliminate erythrocytes. CD19+ B cells were positively selected by MACS of total splenic cell suspensions using microbeads and MACS LS separation columns (Miltenyi Biotec, San Diego, CA), according to the manufacturer’s instructions.

ROS production in the vicinity of BCR was detected by use of anti-IgM F(ab′)2 conjugated to OxyBURST Green H2DCFDA succinimidyl ester (Invitrogen), as described in previous reports (11, 24). One million CD19+ cells were resuspended in phenol red-free HBSS (Ca/Mg salts plus) supplemented with 10 mM HEPES (Life Technologies) and 1% FBS. Cells were incubated with 20 μg of OxyBURST Green–conjugated anti-IgM F(ab′)2 for 30 min on ice, then were washed and activated at 37°C for 10 min. ROS production was measured by detecting 488 nm fluorescence emission by flow cytometry using an LSR II and BD FACSDIVA software.

The purified CD19+ cells were stimulated by goat anti-mouse IgM F(ab′)2 fragment (SouthernBiotech, Birmingham, AL), mouse IL-4 (R&D Systems, Emeryville, CA), or IL-13 (R&D Systems) in complete RPMI 1640 medium supplemented with 10% FBS (Atlanta Biologicals, Flowery Branch, GA), 1% penicillin/streptomycin (Life Technologies, Carlsbad, CA), 5 mM HEPES (Life Technologies), 1% MEM nonessential amino acids (Life Technologies), 1 mM sodium pyruvate (Life Technologies), 2 mM l-glutamine (Life Technologies), and 50 μM 2-ME for 24 h at 37°C under 5% CO2. After the incubation, the cells were transferred into microcentrifuge tubes, then immediately double-stained in phenol red-free HBSS (Ca/Mg salts plus, 10 mM HEPES, and 1% FBS) containing 5 μM OxyBURST Green H2DCFDA succinimidyl ester (Invitrogen) and CellRox (Invitrogen) for 30 min at 37°C. The cells were washed and subjected to flow cytometry analysis using an LSR II and BD FACSDIVA software.

Stimulant-induced B cell proliferation in vitro was analyzed by CFSE staining methodology (Thermo Fisher Scientific, Asheville, NC). Purified splenic CD19+ B cells were labeled with 5 mM CFSE for 10 min at 37°C. After washing, the labeled cells were resuspended at a concentration of 1 × 106 cells/ml in complete RPMI 1640 medium and were cultured with 5 μg/ml anti-mouse IgM F(ab′)2 in the presence or absence of IL-4 (20 ng/ml), either with or without 1 U/μl catalase, for 72 h at 37°C. Cells were collected, and diluted fluorescence peaks of CFSE detected with each successive cell division were analyzed by flow cytometry using an LSR II and BD FACSDIVA software.

Total RNA was prepared from CD19+ B cells using the RNeasy kit (QIAGEN, Venlo, the Netherlands) by treatment with RNase-free DNase (QIAGEN). The cDNA was generated with a ThermoScript reverse transcriptase (Invitrogen, Carlsbad, CA). Real-time PCR was performed in a 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA) using Power SYBR Green PCR Master Mix (Applied Biosystems) reagents with specific oligonucleotide primer pairs (Supplemental Table I). The PCR conditions were 50°C for 2 min and then 95°C for 15 s and 60°C for 1 min, repeated for 40 cycles, with a hot start at 95°C for 10 min. The expression levels of each gene were normalized to those of an internal control gene, eukaryotic translation initiation factor 3, subunit F (EIF3F).

Preimmunized sera were obtained from WT, Duox1−/−, or Nox2−/− mice before immunization. For T cell–independent (TI) immunizations, several groups of three to four mice per strain (8–10 wk old) were injected i.p. with 50 μg of nitrophenyl-LPS (NP-LPS; Biosearch Technologies) in 200 μl of PBS. Other mice were injected with 100 μg of NP–keyhole limpet hemocyanin precipitated with alum (ImjectR; Thermo Scientific). Immune sera were collected from mice on 7 or 8 d (TI) and 14 d (T cell–dependent [TD]) after immunization.

To examine Ig isotype production in vitro, purified splenic CD19+ B cells were resuspended at a concentration of 1 × 106 cells/ml in complete RPMI 1640 medium and were cultured with IL-4 alone (20 ng/ml), 5 μg/ml anti-mouse IgM F(ab′)2, or anti-mouse CD40 Ab (SouthernBiotech) in the presence or absence of IL-4 (20 ng/ml) for 72 h at 37°C. After coculture with stimulants, culture supernatants were collected for ELISAs.

To measure production of each Ig isotype in the culture supernatants, a Mouse Ig Isotyping ELISA Ready-SET-Go! kit (eBioscience, San Diego, CA) was used according to the manufacturer’s instructions using standards of mouse IgM, IgG1, IgG2a, IgG2b, IgG3, and IgA (eBioscience). ELISAs were performed in 96-well plates (Greiner Bio-One, Monroe, NC) from OD (490–520 nm) measurements recorded with a microplate reader (Victor2 1420 Multilabel counter; Wallac).

Whole spleen tissues collected from immunized and nonimmunized mice were fixed in 10% buffered formalin and embedded in paraffin. Tissue sections were either stained with H&E or processed for immunostaining according to standard protocols (25). Immunostaining was performed by the avidin-biotin peroxidase or alkaline phosphatase complex methods using the VECTASTAIN Elite ABC kits (Vector Laboratories, Burlingame, CA) according to the manufacturer’s instruction. Abs and other reagents were obtained from the following sources: rat anti-CD3 (no. MCA1477, 1:100; Bio-Rad Laboratories), anti-B220 (no. 553086, 1:200; BD Biosciences), rabbit rat anti-Ki67 (no. 16667, 1:100; Abcam), peanut agglutinin (PNA)-biotin conjugate (no. B-1075, 1:1000; Vector Laboratories), anti-IgM (no. ba2020, 1:100; Vector Laboratories), and goat anti-PAX5 (no. sc-1974, 1:750; Santa Cruz Biotechnology).

For quantitative analyses of microscopic images, 20 high-power (×40) fields were taken in each case for proliferating cells with Ki67 positivity of blast cells in the germinal centers (GCs) of spleens (three mice for each group to take the average number of Ki67 positives) and at low power (×5) fields to count all GCs with PNA reaction positivity (three mice for each group to obtain average GC numbers within spleen sections). All immunohistochemical data were analyzed by two pathologists.

CD19+ B cells were washed in PBS and were suspended in RIPA buffer (Sigma-Aldrich, St. Louis, MO) containing protease inhibitor and phosphatase inhibitor mixtures (Sigma-Aldrich). After incubation on ice for 15 min, supernatants were collected by centrifugation as whole-cell protein extracts and were quantified by Bradford assays using Coomassie Plus Protein Assay reagents (Thermo Fisher Scientific, Pittsburgh, PA). Twenty to thirty micrograms of the whole-cell extracts was resolved electrophoretically on 4–12% Bis-Tris NuPAGE gels (Invitrogen) then transferred on polyvinylidene difluoride membranes (Life Technologies). Immunoblotting was performed with commercially available Abs: mouse anti-phosphotyrosine mAb (MilliporeSigma, Billerica, MA), mouse anti-human/mouse regulator of G-protein signaling 16 (RGS16) mAb (4E5; Abcam), rabbit anti-mouse Akt polyclonal Ab (Cell Signaling Technology, Danvers, MA), anti-mouse phospho-Akt (Ser473) polyclonal Ab (D9E; Cell Signaling Technology), goat anti-human/mouse B-cell adapter for PI3K (BCAP)/PIK3AP1 polyclonal Ab (Novus Biologicals, Littleton, CO), and rabbit anti-human/mouse GAPDH polyclonal Ab (Trevigen, Gaithersburg, MD). Some membranes were stripped using Restore PLUS Western Blot Stripping Buffer (Thermo Fisher Scientific) and reprobed.

All data sets in each experiment were subjected to a normality test by Shapiro–Wilk test. Based on the results, statistical differences were assessed by Student t test, Mann–Whitney U test, Kruskal–Wallis test, Friedman test, or two-way ANOVA using GraphPad Prism software version 5 (GraphPad Software, La Jolla, CA).

In initial experiments, we assessed the influence of Duox1 deficiency on B cell homeostasis by analyzing B cell populations isolated from BM and spleen. The absence of Duox1 in mice resulted in a normal distribution of each B cell subset detected in populations isolated from BM and spleen when compared with those from WT mice (Fig. 1A).

FIGURE 1.

Duox1 deficiency does not alter B cell homeostasis or early oxidative responses to BCR stimulation. (A) Development and maturation of each B cell subset in BM and spleen. BM cells and splenocytes collected from 8- to 12-wk-old mice were stained with fluorescence-conjugated anti-CD23, anti-IgM, anti-CD93, and anti-B220, and the distribution of each B cell subset was analyzed by flow cytometry. Frequencies of each B cell subtype in BM and spleen are shown as percentages of B220+ cells from WT and Duox1−/− mice. Data are expressed as mean ± SE of results from four to six mice of each strain. n.s., not significant (unpaired Student t test). (B) Gene expression patterns of Nox/Duox family members in freshly isolated nonstimulated CD19+ B cells. The expression of each gene was normalized to the levels of EIF3F mRNA. Duox1 mRNAs from Duox1−/− mice were detected as a frame-shifted transcript (22). Data are expressed as mean ± SE of results from three mice (WT: n = 3; Duox1−/−: n = 3). n.s., not significant (unpaired Student t test). (C) Early phase ROS production by splenic CD19+ B cells following BCR stimulation in the presence or absence of HRP. H2O2 was detected in the vicinity of BCR with DCFDA-conjugated anti-IgM F(ab′)2 by flow cytometry. Data are shown as MFI of FITC channel fluorescence from two mice in each strain (WT, Duox1−/−, or Nox2−/−).

FIGURE 1.

Duox1 deficiency does not alter B cell homeostasis or early oxidative responses to BCR stimulation. (A) Development and maturation of each B cell subset in BM and spleen. BM cells and splenocytes collected from 8- to 12-wk-old mice were stained with fluorescence-conjugated anti-CD23, anti-IgM, anti-CD93, and anti-B220, and the distribution of each B cell subset was analyzed by flow cytometry. Frequencies of each B cell subtype in BM and spleen are shown as percentages of B220+ cells from WT and Duox1−/− mice. Data are expressed as mean ± SE of results from four to six mice of each strain. n.s., not significant (unpaired Student t test). (B) Gene expression patterns of Nox/Duox family members in freshly isolated nonstimulated CD19+ B cells. The expression of each gene was normalized to the levels of EIF3F mRNA. Duox1 mRNAs from Duox1−/− mice were detected as a frame-shifted transcript (22). Data are expressed as mean ± SE of results from three mice (WT: n = 3; Duox1−/−: n = 3). n.s., not significant (unpaired Student t test). (C) Early phase ROS production by splenic CD19+ B cells following BCR stimulation in the presence or absence of HRP. H2O2 was detected in the vicinity of BCR with DCFDA-conjugated anti-IgM F(ab′)2 by flow cytometry. Data are shown as MFI of FITC channel fluorescence from two mice in each strain (WT, Duox1−/−, or Nox2−/−).

Close modal

We investigated the expression patterns of all Nox/Duox family member transcripts in freshly isolated splenic CD19+ B cells handled under nonstimulatory conditions. The magnetically isolated CD19+ cell populations from spleen were confirmed for CD19 expression on their cell surface, showing 97.5% CD19 positivity within this sorted population (data not shown). No differences in the relative expression of Nox1, 2, 3, 4, and Duox2 transcripts in splenic CD19+ B cells were detected between WT and Duox1-deficient mice (Fig. 1B). Although Duox1 transcripts were detected in the cells of mutant mice at normal levels (Fig. 1B), a previous report characterizing this mutant strain demonstrated that the Duox1-targeted frame-shifted transcript is not translated into a detectable or functional protein (22). These observations indicate that the absence of Duox1 does not influence B cell development or expression levels of other Nox family transcripts, at least under healthy, unstimulated conditions.

We detected an early phase (at 10 min) of ROS production in the vicinity of stimulated BCR using OxyBURST Green succinimidyl ester (DCFDA-SE) conjugated to anti-IgM F(ab′)2 as both an agonist and redox-sensing probe. BCR stimulation increased the mean fluorescence intensity (MFI) detected by OxyBURST Green on anti-IgM F(ab′)2 in WT CD19+ B cells (Fig. 1C). Duox1-deficient cells showed OxyBURST MFI values comparable to those of WT cells, whereas Nox2-deficent (gp91phox-deficient) cells showed remarkably lower signals following BCR stimulation (Fig. 1C), consistent with earlier studies on Nox2-deficient mice (11). Addition of HRP catalyzed enhanced oxidation of DCFDA on the anti-IgM F(ab′)2 by H2O2 produced by stimulated CD19+ B cells, resulting in greatly enhanced fluorescence signals (>15-fold relative to the assays lacking HRP) from cells from WT and Duox1−/− mice (Fig. 1C). These findings indicated that this early-phase signal is in large part attributable to extracellular H2O2 release from Nox2 in the vicinity of stimulated BCR.

To explore factors that could affect gene expression of Duox1, we investigated whether the expression of Duox1 in splenic CD19+ B cells is changed by stimulation with anti-IgM F(ab′)2 in combination with TH2-type cytokines. Stimulation with anti-IgM F(ab′)2 or IL-4 or costimulation with anti-IgM F(ab′)2 and IL-4 for 6 h did not alter Duox1 expression (Supplemental Fig. 1A), whereas costimulation with anti-IgM F(ab′)2 and IL-4 for 24 h induced significantly higher expression of Duox1 (Fig. 2A). To consider other possible ROS sources, we examined expression patterns of all other NADPH isoforms at this time point. Gene transcript levels of Nox1, Nox2, Nox3, Nox4, and Duox2 were not influenced by any stimulants for 24 h (Fig. 2A). Treatment with another TH2-type cytokine, IL-13, did not enhance Duox1 gene expression in CD19+ B cells after 24 h of costimulation (Supplemental Fig. 1B). The enhanced Duox1 expression by costimulation with BCR plus IL-4 was not maintained at 48 or 72 h poststimulation (Supplemental Fig. 1A).

FIGURE 2.

IL-4 induces Duox1-dependent H2O2 production by CD19+ B cells. (A) Relative expression of Nox/Duox genes in CD19+ B cells stimulated for 24 h by IL-4 (20 ng/ml), anti-IgM F(ab′)2 (5 μg/ml), or a combination of both. Expression of each NADPH oxidase was examined by quantitative RT-PCR and normalized to unstimulated cell levels relative to EIF3F mRNA. Data are expressed as mean ± SE of results from four WT mice. *p < 0.05 (paired Student t test). (B) ROS production in CD19+ B cells from spleens of WT, Duox1-deficient, and Nox2-deficient mice stimulated for 24 h. Representative histograms of ROS detection in stimulated CD19+ B cells by cell surface–conjugated OxyBURST Green succinimidyl ester fluorescence (upper left panel) or intracellular CellRox fluorescence (lower left panel). Cells were costimulated by anti-IgM F(ab′)2 and IL-4 for 24 h, collected and labeled or loaded with both dye reagents in the presence of HRP, and then analyzed by flow cytometry. Right panels summarize data on percent increase of MFI ± SE (WT: n = 8; Duox1−/−: n = 5; and Nox2−/−: n = 3). *p < 0.05, **p < 0.01 (unpaired Student t test).

FIGURE 2.

IL-4 induces Duox1-dependent H2O2 production by CD19+ B cells. (A) Relative expression of Nox/Duox genes in CD19+ B cells stimulated for 24 h by IL-4 (20 ng/ml), anti-IgM F(ab′)2 (5 μg/ml), or a combination of both. Expression of each NADPH oxidase was examined by quantitative RT-PCR and normalized to unstimulated cell levels relative to EIF3F mRNA. Data are expressed as mean ± SE of results from four WT mice. *p < 0.05 (paired Student t test). (B) ROS production in CD19+ B cells from spleens of WT, Duox1-deficient, and Nox2-deficient mice stimulated for 24 h. Representative histograms of ROS detection in stimulated CD19+ B cells by cell surface–conjugated OxyBURST Green succinimidyl ester fluorescence (upper left panel) or intracellular CellRox fluorescence (lower left panel). Cells were costimulated by anti-IgM F(ab′)2 and IL-4 for 24 h, collected and labeled or loaded with both dye reagents in the presence of HRP, and then analyzed by flow cytometry. Right panels summarize data on percent increase of MFI ± SE (WT: n = 8; Duox1−/−: n = 5; and Nox2−/−: n = 3). *p < 0.05, **p < 0.01 (unpaired Student t test).

Close modal

We designed strategies for simultaneous detection of H2O2 and superoxide (O2·) in cells at 24 h poststimulation, when Duox1 expression is maximally enhanced. For these parallel ROS assays, we used OxyBURST Green succinimidyl ester and CellRox dye reagents for double staining because cell surface labeling with OxyBURST Green detects Duox1-derived H2O2, whereas CellRox preferentially reacts with intracellular O2· produced by Nox2. In comparison with unstimulated cells, 24 h of stimulation with anti-IgM F(ab′)2 alone increased the MFI of OxyBURST Green on WT CD19+ B cells (Fig. 2B, upper panel). Costimulation with anti-IgM F(ab′)2 and IL-4 remarkably induced the OxyBURST fluorescence in WT and Nox2 mutant B cells, whereas the signal detected from Duox1-deficient cells was considerably lower (Fig. 2B, upper panel). Meanwhile, production of O2· sensed by CellRox also increased following stimulation with anti-IgM F(ab′)2, and the signal was further enhanced by additional IL-4 in both WT and Duox1 mutant CD19+ B cells (Fig. 2B, lower panel). CellRox fluorescence was comparable between WT and Duox1−/− CD19+ B cells but was attenuated in Nox2 mutant cells costimulated with anti-IgM F(ab′)2 and IL-4 (Fig. 2B, lower panel). Similar trends in ROS production were detected using OxyBURST Orange succinimidyl ester and CellRox in both WT and Duox1 mutant B cells following costimulation (Supplemental Fig. 2), whereas treatment with IL-13 in place of IL-4 did not increase H2O2 signals (Supplemental Fig. 1C). These observations indicate that the H2O2 signal increases following 24 h of costimulation by IL-4 and BCR are Duox1-dependent, whereas the O2· signal is unaffected by Duox1 deficiency.

We examined whether there are any detectable signaling pathway differences between WT and mutant CD19+ B cells within 24 h of stimulation (Fig. 3). We compared the total cellular tyrosine phosphorylation status of WT, Duox1−/−, and Nox2−/− CD19+ B cells at 1 min, 5 min, and 24 h poststimulation by Western blotting (Fig. 3A). The phosphorylation was greatly reduced at 5 min in the B cells during early responses to stimulation with anti-IgM F(ab′)2 alone. This supports findings reported by Singh and collaborators (15) that phosphorylation of more than 60% among phosphoproteins was transiently attenuated at 5 min during fragment anti-Ig stimulation. Costimulation with anti-IgM F(ab′)2 and IL-4 for 24 h led to several differences in phosphorylation responses in the B cells from WT and mutant mice, particularly in the Mr ∼60 kDa range. Surface expression of an early B cell activation marker, CD69, was enhanced in Duox1−/− CD19+ B cells by costimulation with anti-IgM F(ab′)2 and IL-4 for 6 h but was comparable with WT cells up to 12 h poststimulation (Fig. 3B). Another activation marker, CD86, showed comparable expression levels between WT and Duox1-deficient cells at 12 and 18 h.

FIGURE 3.

Duox1 affects expression of B cell activation markers and BCR signal mediators. (A) Total phosphotyrosine (p-Tyr) immunoblot (fingerprint) analysis of unstimulated or stimulated splenic CD19+ B cells from WT, Nox2−/−, and Duox1−/− mice. Differential phosphorylation patterns between WT and mutant CD19+ B cells are indicated by arrows on the right. (B) Cell surface expression of CD69 and CD86 in stimulated CD19+ B cells. A representative histogram showing expression at 6 or 18 h poststimulation (left), which was analyzed by flow cytometry comparing MFI between WT and Duox1−/− CD19+ B cells (right). Data are shown as MFI ± SE of results from four mice each at multiple time points poststimulation. n.s., not significant. *p < 0.05 (paired Student t test). (C) Expression of BCR signaling mediators. Relative expression after 24 h of stimulation was normalized to EIF3F and compared between WT and Duox1−/− CD19+ B cells. Data are expressed as mean ± SE (WT: n = 5; Duox1−/−: n = 6). *p < 0.05 (paired Student t test). (D) Immunoblotting of BCAP and RGS16 in CD19+ B cells after 24 h of stimulation with anti-IgM F(ab′)2 and IL-4. (E) Akt protein production and phosphorylation. Left, Representative immunoblots of total Akt and p-Akt (Ser473) following 24 h of stimulation of CD19+ B cells. Right, Induction of Akt phosphorylation following 24 h of stimulation, determined by blot image densitometry normalized relative to GAPDH. Data represent the average ± SE of three to four separate experiments.

FIGURE 3.

Duox1 affects expression of B cell activation markers and BCR signal mediators. (A) Total phosphotyrosine (p-Tyr) immunoblot (fingerprint) analysis of unstimulated or stimulated splenic CD19+ B cells from WT, Nox2−/−, and Duox1−/− mice. Differential phosphorylation patterns between WT and mutant CD19+ B cells are indicated by arrows on the right. (B) Cell surface expression of CD69 and CD86 in stimulated CD19+ B cells. A representative histogram showing expression at 6 or 18 h poststimulation (left), which was analyzed by flow cytometry comparing MFI between WT and Duox1−/− CD19+ B cells (right). Data are shown as MFI ± SE of results from four mice each at multiple time points poststimulation. n.s., not significant. *p < 0.05 (paired Student t test). (C) Expression of BCR signaling mediators. Relative expression after 24 h of stimulation was normalized to EIF3F and compared between WT and Duox1−/− CD19+ B cells. Data are expressed as mean ± SE (WT: n = 5; Duox1−/−: n = 6). *p < 0.05 (paired Student t test). (D) Immunoblotting of BCAP and RGS16 in CD19+ B cells after 24 h of stimulation with anti-IgM F(ab′)2 and IL-4. (E) Akt protein production and phosphorylation. Left, Representative immunoblots of total Akt and p-Akt (Ser473) following 24 h of stimulation of CD19+ B cells. Right, Induction of Akt phosphorylation following 24 h of stimulation, determined by blot image densitometry normalized relative to GAPDH. Data represent the average ± SE of three to four separate experiments.

Close modal

At 24 h poststimulation, we observed no significant changes in expression of the BLNK adaptor molecule in Duox1−/− versus WT D19+ B cells by treatment with IL-4 or anti-IgM F(ab′)2 alone or in combination (Fig. 3C). In contrast, BCAP, another adaptor molecule for BCR signaling, was relatively increased by each stimulus. Interestingly, the RGS16 gene transcript level was increased in Duox1−/− CD19+ B cells by stimulation with either anti-IgM F(ab′)2 alone or the combination of anti-IgM F(ab′)2 and IL-4 (Fig. 3C). The increase of BCAP and RGS16 transcripts was reflected in the levels of protein production by the combination (Fig. 3D). Basal production and phosphorylation of Akt were further examined; however, we did not observe any differences in total Akt production levels or its phosphorylation at residue Ser473 in Duox1−/− cells in comparison with that from WT or Nox2−/− mice (Fig. 3E). In all three mouse strains, we observed enhanced p-Akt levels either in response to anti-IgM F(ab′)2 stimulation alone or in combination with IL-4 treatment.

To approach other functional consequences of H2O2 generated by Duox1 in splenic CD19+ B cells, we next focused on intrinsic B cell functions. Duox1-deficient CD19+ B cells stimulated by anti-IgM F(ab′)2 alone for 3 d showed higher proliferative activity, and additional IL-4 treatment further enhanced this effect in WT cells (Fig. 4A). We quantified actual cell numbers in each proliferative cycle peak within whole divided CD19+ B cells using counting microbeads as a standard in the CFSE-based assay. Duox1-deficient cells showed higher counts than WT cells in each division cycle (1, 2, or 3) in response to a single anti-IgM F(ab′)2 stimulus (Fig. 4B, left panel). Following costimulation with anti-IgM F(ab′)2 and IL-4, both Duox1−/− and Nox2−/− CD19+ B cells showed significantly increased counts within division cycles 2 and 3; however, the counts of Nox2 mutants were lower than those of Duox1 mutant cells (Fig. 4B, right panel). In terms of total divided cells, costimulation with anti-IgM F(ab′)2 and IL-4 led to higher proliferative rates of CD19+ B cells than with single stimulation with anti-IgM F(ab′)2 in WT cells (Fig. 4C). The proliferative capacity of Duox1-deficient cells greatly exceeded that of WT cells, whereas Nox2 mutant cells showed the second-largest proliferation increases following costimulation (Fig. 4C). With each stimulus, significant differences in dead cells, detected as propidium iodide–positive (PI+) events, were not seen between WT and Duox1−/− CD19+ B cells. However, PI+ events were statistically increased in Nox2−/− cells following stimulant-induced proliferation (Fig. 4D).

FIGURE 4.

Duox1 deficiency affects proliferative activity of stimulated splenic CD19+ B cells. (A) Representative CSFE flow cytometric histograms of proliferating CD19+ B cells isolated from WT and Duox1−/− mouse spleen. The cells were cultured with or without 5 μg/ml of anti-IgM F(ab′)2 and/or 20 ng/ml of IL-4 for 72 h, and their proliferative status was analyzed by dilution of CFSE fluorescence. Each cell division cycle is specified by numbers given below peaks: 0, undivided cell population; 1, division cycle 1; 2, division cycle 2; and 3, division cycle 3. (B) Comparison of cell numbers in each cell cycle division of WT, Duox1−/−, and Nox2−/− CD19+ B cells. Precise cell numbers were calculated by normalizing with counting microbeads. Each plot is expressed as median of results from individual numbers of each strain. **p < 0.01, ***p < 0.001 for the test of WT versus Duox1−/−. p < 0.05, ††p < 0.01 for the test of WT versus Nox2−/− (two-way ANOVA). (C) Total number of divided CD19+ B cells of WT, Duox1−/−, or Nox2−/− mice. Actual cell numbers were calculated by normalizing with counting microbeads. Data are expressed as box-and-whisker plots with median and quartile deviation. Individual numbers of three strains for the analyses and statistical p values are shown. n.s., not significant (Kruskal–Wallis test). (D) Actual number of PI+ events in stimulated cultures. Data are shown as mean ± SE of results with statistical p values from individual numbers of each strain. n.s., not significant (Kruskal–Wallis test).

FIGURE 4.

Duox1 deficiency affects proliferative activity of stimulated splenic CD19+ B cells. (A) Representative CSFE flow cytometric histograms of proliferating CD19+ B cells isolated from WT and Duox1−/− mouse spleen. The cells were cultured with or without 5 μg/ml of anti-IgM F(ab′)2 and/or 20 ng/ml of IL-4 for 72 h, and their proliferative status was analyzed by dilution of CFSE fluorescence. Each cell division cycle is specified by numbers given below peaks: 0, undivided cell population; 1, division cycle 1; 2, division cycle 2; and 3, division cycle 3. (B) Comparison of cell numbers in each cell cycle division of WT, Duox1−/−, and Nox2−/− CD19+ B cells. Precise cell numbers were calculated by normalizing with counting microbeads. Each plot is expressed as median of results from individual numbers of each strain. **p < 0.01, ***p < 0.001 for the test of WT versus Duox1−/−. p < 0.05, ††p < 0.01 for the test of WT versus Nox2−/− (two-way ANOVA). (C) Total number of divided CD19+ B cells of WT, Duox1−/−, or Nox2−/− mice. Actual cell numbers were calculated by normalizing with counting microbeads. Data are expressed as box-and-whisker plots with median and quartile deviation. Individual numbers of three strains for the analyses and statistical p values are shown. n.s., not significant (Kruskal–Wallis test). (D) Actual number of PI+ events in stimulated cultures. Data are shown as mean ± SE of results with statistical p values from individual numbers of each strain. n.s., not significant (Kruskal–Wallis test).

Close modal

Based on the above cell proliferation findings, we asked whether catalase, acting as an exogenous H2O2 scavenger, would mimic the effects of Duox1 deficiency, thereby leading to enhanced proliferation of WT CD19+ B cells. Cultivation with catalase in the media for 72 h led to increased actual numbers of proliferating WT cells following stimulation with either anti-IgM F(ab′)2 alone or with anti-IgM F(ab′)2 and IL-4 together (Fig. 5). Meanwhile, PI+ cell numbers were unaffected by addition of catalase to these proliferating WT CD19+ B cell populations.

FIGURE 5.

Enhanced proliferation of WT CD19+ B cells by scavenging of H2O2. CD19+ B cells isolated from WT mice were cocultured with 5 μg/ml of anti-IgM F(ab′)2 in the presence (A) or absence (B) of IL-4 (20 ng/ml), either with or without 1 U/ml of catalase, for 72 h. Proliferative activity was compared by CFSE staining methodology. Actual cell numbers of total divided CD19+ B cells are shown with medians, along with lines tracing changes in individual cultures following treatments. PI+ staining events during the proliferation are expressed as mean ± SE. n.s., not significant (Friedman test).

FIGURE 5.

Enhanced proliferation of WT CD19+ B cells by scavenging of H2O2. CD19+ B cells isolated from WT mice were cocultured with 5 μg/ml of anti-IgM F(ab′)2 in the presence (A) or absence (B) of IL-4 (20 ng/ml), either with or without 1 U/ml of catalase, for 72 h. Proliferative activity was compared by CFSE staining methodology. Actual cell numbers of total divided CD19+ B cells are shown with medians, along with lines tracing changes in individual cultures following treatments. PI+ staining events during the proliferation are expressed as mean ± SE. n.s., not significant (Friedman test).

Close modal

We investigated whether B cell Ig production is influenced by Duox1 by comparing both in vitro and in vivo production of each Ig isotype (Fig. 6, Supplemental Fig. 3). Several stimulants tested under in vitro conditions induced production of IgM, IgG1, IgG2a, IgG2b, IgG3, and IgA by CD19+ B cells, but their levels were comparable in WT and Duox1-deficient cell cultures (Fig. 6). Nox2-deficient cells showed lower production of IgM, IgG1, IgG2a, IgG2b, and IgG3 following several kinds of costimulation in comparison with WT cells.

FIGURE 6.

Duox1-deficient splenic CD19+ B cells secrete normal levels of Ig isotypes in vitro. CD19+ B cells isolated from WT, Duox1−/−, or Nox2−/− mice were cultured with each stimulus as indicated in the graph for 3 d. Secretion of six major Ig isotypes into the culture media was measured with ELISA. Data are expressed as median values from 6 to 21 mice of each strain. p < 0.05 considered statistically significant (Kruskal–Wallis test).

FIGURE 6.

Duox1-deficient splenic CD19+ B cells secrete normal levels of Ig isotypes in vitro. CD19+ B cells isolated from WT, Duox1−/−, or Nox2−/− mice were cultured with each stimulus as indicated in the graph for 3 d. Secretion of six major Ig isotypes into the culture media was measured with ELISA. Data are expressed as median values from 6 to 21 mice of each strain. p < 0.05 considered statistically significant (Kruskal–Wallis test).

Close modal

We subsequently examined production levels of Ag-specific Ab in vivo by immunization with TI Ag or TD Ag (Supplemental Fig. 3A–C). To compare early-phase Ig production responses between WT and Duox1−/− mice, we detected serum IgM levels in immunized mice with TI Ag. Duox1 deficiency showed comparable levels of serum IgM at 8 d postimmunization with TI Ag in comparison with its levels in WT mice (Supplemental Fig. 3A). To focus on later responses of Ig isotype production, serum IgM and IgG1 levels in WT and Duox1−/− mice were monitored at 14 d postimmunization with TD Ags. Both mouse strains did not exhibit any differences in IgM and IgG1 production levels at 14 d postimmunization by TD Ag (Supplemental Fig. 3C). Meanwhile, Nox2-deficient mice showed comparable levels of serum IgM with those of WT or Duox1−/− mice at 7 d postimmunization with TI Ag (Supplemental Fig. 3B).

We investigated proliferation of B cells within GCs in the spleen from immunized mice by immunohistological evaluation of several markers (Fig. 7, Table I). At 8 d postimmunization with TI Ag (NP-LPS), histological examination of Duox1-deficient spleens revealed greatly enhanced signals of the B cell proliferation marker Ki67 within follicular regions identified by the B cell lineage marker Pax5 in comparison with WT and Nox2 mutant mice (Fig. 7A, Table I). Increased numbers of GCs were also observed, along with the increased GC size and reactivity detected by GC marker PNA positivity, in the spleens from Duox1−/− mice relative to the other strains following TI immunization (Fig. 7B). In contrast, we detected no remarkable immunohistochemical staining pattern differences among the three mouse strains following immunization with TD Ag (NP-KLH, 14 d; Supplemental Fig. 3D). Together, the above findings suggest that Duox1-derived H2O2 plays some role in restricting splenic CD19+ B cell proliferation but does not affect Ig isotype production.

FIGURE 7.

Immunohistological detection of B cell responses in vivo in spleens following immunization. WT, Duox1−/−, and Nox2−/− mice were injected with NP-LPS, and then spleens were harvested at 8 d after the immunization. Immunohistochemistry was performed on spleen sections by immune-peroxidase and -phosphatase labeling of (A) anti-Ki67 (brown) and anti-Pax5 (blue), respectively, or (B) PNA (brown) and anti-IgM (blue), respectively. Ki67+ proliferating GC cell clusters are indicated with white arrows in (A), whereas PNA-reactive GCs are marked with white arrows in (B). Data shown are representative of sections analyzed from three mice of each genotype.

FIGURE 7.

Immunohistological detection of B cell responses in vivo in spleens following immunization. WT, Duox1−/−, and Nox2−/− mice were injected with NP-LPS, and then spleens were harvested at 8 d after the immunization. Immunohistochemistry was performed on spleen sections by immune-peroxidase and -phosphatase labeling of (A) anti-Ki67 (brown) and anti-Pax5 (blue), respectively, or (B) PNA (brown) and anti-IgM (blue), respectively. Ki67+ proliferating GC cell clusters are indicated with white arrows in (A), whereas PNA-reactive GCs are marked with white arrows in (B). Data shown are representative of sections analyzed from three mice of each genotype.

Close modal
Table I.
Histological evaluation of lymphoid tissue (GCs)
StrainsFollicleNumber of GCsGC ReactionGC ProliferationT Cell ZoneMarginal Zone
Nonimmunized (n = 3 each)       
 WT 
 Duox1−/− 
 Nox2−/− 
NP-LPS–immunized (n = 3 each)       
 WT ++ 21 ++ ++ 
 Duox1−/− ++ 32 +++ +++ 
 Nox2−/− ++ 20 ++ ++ 
StrainsFollicleNumber of GCsGC ReactionGC ProliferationT Cell ZoneMarginal Zone
Nonimmunized (n = 3 each)       
 WT 
 Duox1−/− 
 Nox2−/− 
NP-LPS–immunized (n = 3 each)       
 WT ++ 21 ++ ++ 
 Duox1−/− ++ 32 +++ +++ 
 Nox2−/− ++ 20 ++ ++ 

Histological marker scoring criteria: follicle (IgM, B200): + = normal, ++ = mild proliferation, and +++ = moderate; number of GCs (PNA): + = 1–15, ++ = 16–30, and +++ = 30 or more; GC reaction (PNA): + = normal, ++ = moderate, and +++ = strong reaction; GC proliferation (Ki67): + = normal, ++ = mild proliferation, and +++ = moderate; T cell zone (CD-4): + = normal, ++ = mild activity, and +++ = moderate; and marginal zone (B220): + = normal, ++ = mild expansion, and +++ = moderate.

Deliberate and robust production of ROS by phagocytic blood cells has long been recognized as a unique feature of these cells and referred to as the “respiratory burst,” which is activated by phagocytic stimuli for oxidant-based microbial killing (26, 27). This activity is attributed to a multicomponent Nox2-based NADPH oxidase in phagocytes, although this enzyme is also detected in B lymphocytes. Studies on the role of ROS in primary B cells have revealed that Nox2-derived O2· influences B cell functions, including by altering production of Ig isotypes from several B cell subsets in mice and humans (8, 10, 11). Our current studies, using both in vitro and in vivo approaches, confirmed a role for Nox2 in the early phase of ROS generation poststimulation and then focused on another TH2 cytokine-inducible ROS generator, Duox1, to explore its capacity for H2O2 production and downstream functional consequences in comparison with Nox2-derived O2· in primary B cells.

Splenic CD19+ B cells lacking Duox1 showed normal basal expression of Nox1, Nox2, Nox3, Nox4, and Duox2 transcripts under nonstimulated healthy conditions. Early-phase H2O2 levels detected in the vicinity of stimulated BCR by OxyBURST-conjugated anti-IgM F(ab′)2 were comparable between WT and Duox1−/− CD19+ cells, whereas Nox2-deficient cells showed a remarkable reduction of detectable H2O2 in our experiments. Nox2 is an abundant NADPH oxidase in peripheral and BM-derived Ig-expressing B cells that produces O2· through mitogenic stimulation of human B cells (5). In mouse splenic resting B cells, the Nox2 transcript is the predominant NADPH oxidase detected in either stimulated or unstimulated cells. O2· produced following BCR stimulation is converted into H2O2, which can be detected in the vicinity of BCR with OxyBURST Green conjugated to anti-IgM F(ab′)2 (Fig. 1C, Ref. 11). These findings indicate that H2O2 localized within the vicinity of BCR is mainly supplied by Nox2-derived superoxide intermediate in the early phase of BCR stimulation, rather than Duox1-derived H2O2. Previous studies by Singh et al. (15) using RNA silencing methodology in A20 lymphoma cells suggested that Duox1 is a source of early-phase ROS elicited by anti-IgG F(ab′)2 stimulation alone, although effects of Nox2 silencing in this cell line were not examined.

We showed that costimulation with anti-IgM F(ab′)2 and IL-4 enhanced expression of Duox1 among all the NADPH oxidase gene family members surveyed in splenic CD19+ B cells. In human keratinocytes and airway epithelial cells, treatment with IL-4 induces upregulation of Duox1 production concomitant with enhanced H2O2 generation in these cells (1620). Our studies found that primary CD19+ B lymphocytes exhibit enhanced Duox1 expression and H2O2 production 24 h following costimulation with IL-4 (Fig. 2A, Supplemental Fig. 1), whereas H2O2 production was greatly diminished in Duox1-deficient cells. Moreover, the H2O2 levels detected in Nox2-deficient cells were comparable to those in WT cells, although intracellular O2· production was attenuated in these mutant cells (Fig. 2B). These results suggest that Duox1-derived H2O2 release at 24 h by BCR-stimulated cells under the influence of IL-4 costimulation becomes predominant relative to the H2O2 converted from Nox2-derived O2·. In other experiments, we found that IL-13 does not induce enhanced expression of Duox1 or H2O2 production in CD19+ B cells (Supplemental Fig. 1), consistent with observations that mouse B cells express little or no IL-13Rα1 but high levels of a decoy receptor, IL-13Rα2, which lacks a signaling domain (2831).

The observation of enhanced proliferation of B cells that lack Duox1-derived H2O2 raises a question because PTPases, such as SHP-1, are thought to inhibit BCR signaling. Negative regulation of SHP-1 by Duox1-derived H2O2 could potentially enhance B cell activity, including proliferation through oxidation of SHP-1, as suggested by the findings in A20 lymphoma cells (15). However, that study (15) never examined the prolonged effects of Duox1 silencing on cell proliferation. Our observations did not support a role for Duox1 in early-phase ROS generation or proximal BCR signaling, and the loss of H2O2 due to Duox1 knockout increases proliferation of splenic CD19+ B cells. In our in vivo approaches using whole immunized mice, Duox1 deficiency indeed leads to enhanced B cell proliferation within GCs in spleens of whole immunized mice, further supporting our proposal that Duox1 restricts B cell proliferation. Another study using CD19-cre-conditional SHP-1−/− mice demonstrated that Ca2+ influx induced by BCR stimulation with anti-IgM F(ab′)2 is downregulated by SHP-1 in the B-1a cells, whereas the Ca2+ response is not influenced at all by the PTPase in B-2 cells among total CD19+ splenic cells (32). These findings indicate that not all primary B cell subsets are regulated by SHP-1. Thus, the issue of whether the popularly held SHP-1 oxidation–based paradigm applies to BCR signaling remains debatable, particularly in late-phase responses.

In our experiments with primary CD19+ B cells costimulated with anti-IgM F(ab′)2 and IL-4 for 24 h, the absence of Duox1 did not affect BLNK gene expression; this led instead to upregulation of another adaptor molecule, BCAP. BCAP mediates BCR signaling via Src family kinases by phosphorylating itself and serving as an intermediary in the downstream Akt pathway, thereby positively regulating proliferative activity of B cells (33, 34). In the current study, BCAP expression was increased by Duox1 deficiency but does not appear to influence the phosphorylation and production of Akt (Fig. 3C–E). Nonetheless, Duox1-deficient CD19+ B cells showed enhanced proliferation by BCR stimulation with anti-IgM F(ab′)2 or additional IL-4 costimulation, which upregulates Duox1 expression in WT cells. Thus, it appears that Duox1-derived H2O2 restricts proliferative activity independent of the Akt pathway. We further showed that the increased proliferation observed with Duox1-deficient cells could be mimicked with WT cells by scavenging H2O2 with catalase (Fig. 5). Catalase is naturally deployed in intracellular peroxisomes (35), and this enzyme converts H2O2 into less reactive stable substances, H2O and oxygen, to regulate excessive cell signaling (36) as well as to prevent cytotoxic damage (37). Supplementation with catalase promotes decomposition of Duox1-derived H2O2, which thereby diminishes the suppressive effects of Duox1 on proliferation of CD19+ B cells, further supporting our hypothesis that Duox1 plays a negative role in proliferation.

Enhanced proliferative activity of Nox2 (gp91phox)-deficient CD19+ cells costimulated with anti-IgM F(ab′)2 and IL-4 was also observed in our experiments, albeit to a lesser extent than in Duox1−/− cells. Similar effects of enhanced proliferation associated with gp91phox deficiency in total splenic B cells were reported previously (10). In this study, Nox2-derived O2· production following BCR stimulation appears to upregulate p27kip1, which acts as a brake on the G0 to G1 transition of the cell cycle to promote B cell proliferation. Thus, the enhanced proliferation of Nox2-deficient (gp91phox−/−) CD19+ B cells that we observed likely reflects relief of inhibitory effects of p27kip1. In contrast, it was reported that resting splenic B cells from Ncf1/p47phox mutant mice that are functionally Nox2-deficient show normal proliferative responses to mitogenic stimuli (11). Neutrophils from CGD patients with p47phox deficiency show some residual NOX2 activity (38), which may explain the discrepancy in proliferation responses between Ncf1- and Nox2-deficient B cells.

Duox1 deficiency also significantly increased RGS16 expression in CD19+ B cells following costimulation. Upregulation of RGS16 expression was observed in human primary keratinocytes in which Duox1 was silenced by RNA interference, suggesting that RGS16 expression is negatively regulated by Duox1-derived H2O2 (20). It appears that RGS16 retains B cells within GCs to suppress B cell migration induced by the chemokine CXCR4 (39, 40). Our in vivo TI Ag immune challenge experiments detected increased B cell proliferation in GCs of spleens of Duox1−/− mice. Therefore, enhanced RGS16 expression caused in the absence of Duox1 could favor localization of CD19+ B cells within GCs to further induce robust proliferative activity.

Typically, TI immunization induces rapid but transient GC formation, as class-switching and somatic hypermutation does not occur without T cell help, such that basal production of Ag-specific IgM is elevated through expansion of single or limited numbers of germline-derived, low-affinity Ab B cell clones (41, 42). Our study showed that Duox1 deficiency in CD19+ B cells resulted in normal in vitro production of six Ig isotypes following stimulation with anti-IgM F(ab′)2 or anti-CD40 in the presence or absence of IL-4 (Fig. 6). Likewise, Duox1-deficient mice showed comparable serum IgM production relative to WT mice following a single immunization with the TI Ag NP-LPS (Supplemental Fig. 3A, 3B). The absence of Duox1 in mice also did not significantly influence serum IgM and IgG isotype production following single TD Ag immunizations (Supplemental Fig. 3C). Although Duox1-derived H2O2 does not appear to play a critical role in the regulation of Ig isotype production in response to single TI or TD Ag imunizations, our observations on CD19+ B cell proliferation in vitro and in vivo suggest that Duox1 could have a role in restricting clonal B cell expansion in response to certain antigenic stimuli. TI immunization not only induces transient GC formation (41, 42, Fig. 7A) but also generates and maintains long-lived plasma cells in both spleen and BM (43). Long-term IgM-dependent protection against specific pathogens that induce TI-type immunity, such as Streptococcus pneumoniae and Entamoeba muris, has been attributed to TI-induced memory B and long-lived plasma cells (43, 44). Thus, Duox1 may influence immune defense reactions by regulating B cell proliferation during repeated or sustained microbial infections, particularly by specific pathogens that evade TD immune responses, and may affect the generation of long-lived plasma cells that contribute to host immune homeostasis.

In contrast, Nox2 deficiency led to decreases of IgM, IgG1, IgG2a, and IgG2b production in CD19+ B cells following stimulation with anti-IgM F(ab′)2 in the presence of IL-4 in vitro. In contrast, specific IgM levels in sera from Nox2−/− mice were not lower than those of WT postimmunization with TI Ag (Supplemental Fig. 3B). Several studies reported that immunization of Nox2−/− animal models with TD or TI Ag enhances production of Ag-specific IgM and IgG isotypes in vivo (10, 11). How Nox2 deficiency causes such disparate effects on Ig production in vitro versus in vivo is not entirely clear. We did observe significant increases in PI+ events in Nox2-deficient CD19+ B cells in culture following BCR stimulation, explaining how the lack of Nox2 could adversely affect production of several Ig isotypes in vitro. In several nonphagocytic cell types, such as pancreatic cancer, vascular endothelial cells, or cardiomyocytes, suppression of Nox2 attenuates O2· production and causes decreased cell viability (4547). Silencing of Nox2 activates the caspase pathway in vascular endothelial cells, suggesting Nox2 serves antiapoptotic or prosurvival roles in nonphagocytic cells (45). Furthermore, Nox2 or Ncf1 knockout cardiomyocytes showed decreased viability (46). A Nox2-induced cell survival mechanism is supported by our current findings in the primary CD19+ B cells. In contrast, loss of H2O2 with Duox1 deficiency did not increase the proportion of PI+ cells among CD19+ B cells relative to WT cells treated under the same proliferating conditions. Moreover, supplementation with the H2O2 scavenger catalase during cultivation with stimuli caused enhanced cell proliferation but had no effects on PI+ events in WT CD19+ B cells. These observations suggest that Duox1-derived H2O2, unlike Nox2-derived O2·, does not influence survival or death responses of primary CD19+ B cells.

Our current studies focused primarily on murine splenic CD19+ B cells, which would encompass most B cell subsets in the spleen (48, 49), to elucidate roles of Duox1-derived H2O2 in primary B cells. Using this approach, we found that Duox1, under the influence of IL-4, negatively regulates stimulus-induced proliferation through production of H2O2. Consistent with these observations, we found that scavenging H2O2 with catalase in vitro supports the expansion of WT CD19+ B cells without influencing cell viability, suggesting a potential means for development of B cell modifications ex vivo for clinical applications. Other experiments in whole animals immunized with TI Ag confirmed that the antiproliferative effects of Duox1 are also manifested in vivo, where we observed rapid increases in the size and number of GCs in spleens of Duox1−/− mice. In contrast, we observed no effects of Duox1 deficiency on GC development or immune responses to TD Ag immunization. Our findings suggest that Duox1 may impact B cell proliferation in the reaction to microbial infections, particularly by pathogens that induce TI responses. Thus, future studies should investigate whether Duox1 regulates long-term humoral immune defense against such pathogens by affecting the expansion of B cell populations in sites beyond splenic GCs or the generation and longevity of long-lived plasma cells.

We thank Dr. Miklos Geiszt (Semmelweis University, Budapest, Hungary) for providing the Duox1 knockout mice. We also thank members of the National Institute of Allergy and Infectious Diseases Research Technologies Branch for assistance with flow cytometry and Donna Butcher (National Cancer Institute at Frederick) and Jawara Allen for providing immunohistochemistry technical support and advice.

This work was supported by the Intramural Research Program of the National Institute of Allergy and Infectious Diseases, National Institutes of Health, Fund ZIA AI001060-11.

The online version of this article contains supplemental material.

Abbreviations used in this article:

BCAP

B-cell adapter for PI3K

BM

bone marrow

CGD

chronic granulomatous disease

Duox

dual oxidase

EIF3F

eukaryotic translation initiation factor 3, subunit F

GC

germinal center

MFI

mean fluorescence intensity

Ncf1

p47phox, neutrophil cytosolic factor 1

NP-LPS

nitrophenyl-LPS

PI+

propidium iodide–positive

PNA

peanut agglutinin

PTP

protein tyrosine phosphatase

RGS16

regulator of G-protein signaling 16

ROS

reactive oxygen species

SHP-1

Src homology region 2 domain–containing phosphatase 1

TD

T cell–dependent

TI

T cell–independent

WT

wild type.

1
Kurosaki
,
T.
,
H.
Shinohara
,
Y.
Baba
.
2010
.
B cell signaling and fate decision.
Annu. Rev. Immunol.
28
:
21
55
.
2
Paus
,
D.
,
T. G.
Phan
,
T. D.
Chan
,
S.
Gardam
,
A.
Basten
,
R.
Brink
.
2006
.
Antigen recognition strength regulates the choice between extrafollicular plasma cell and germinal center B cell differentiation.
J. Exp. Med.
203
:
1081
1091
.
3
Reth
,
M.
,
T.
Brummer
.
2004
.
Feedback regulation of lymphocyte signalling.
Nat. Rev. Immunol.
4
:
269
277
.
4
Smith
,
K. G.
,
D. M.
Tarlinton
,
G. M.
Doody
,
M. L.
Hibbs
,
D. T.
Fearon
.
1998
.
Inhibition of the B cell by CD22: a requirement for Lyn.
J. Exp. Med.
187
:
807
811
.
5
Kobayashi
,
S.
,
S.
Imajoh-Ohmi
,
M.
Nakamura
,
S.
Kanegasaki
.
1990
.
Occurrence of cytochrome b558 in B-cell lineage of human lymphocytes.
Blood
75
:
458
461
.
6
Kovács
,
I.
,
M.
Horváth
,
Á.
Lányi
,
G. L.
Petheő
,
M.
Geiszt
.
2015
.
Reactive oxygen species-mediated bacterial killing by B lymphocytes.
J. Leukoc. Biol.
97
:
1133
1137
.
7
Bleesing
,
J. J.
,
M. M.
Souto-Carneiro
,
W. J.
Savage
,
M. R.
Brown
,
C.
Martinez
,
S.
Yavuz
,
S.
Brenner
,
R. M.
Siegel
,
M. E.
Horwitz
,
P. E.
Lipsky
, et al
.
2006
.
Patients with chronic granulomatous disease have a reduced peripheral blood memory B cell compartment.
J. Immunol.
176
:
7096
7103
.
8
Moir
,
S.
,
S. S.
De Ravin
,
B. H.
Santich
,
J. Y.
Kim
,
J. G.
Posada
,
J.
Ho
,
C. M.
Buckner
,
W.
Wang
,
L.
Kardava
,
M.
Garofalo
, et al
.
2012
.
Humans with chronic granulomatous disease maintain humoral immunologic memory despite low frequencies of circulating memory B cells.
Blood
120
:
4850
4858
.
9
Cotugno
,
N.
,
A.
Finocchi
,
A.
Cagigi
,
G.
Di Matteo
,
M.
Chiriaco
,
S.
Di Cesare
,
P.
Rossi
,
A.
Aiuti
,
P.
Palma
,
I.
Douagi
.
2015
.
Defective B-cell proliferation and maintenance of long-term memory in patients with chronic granulomatous disease.
J. Allergy Clin. Immunol.
135
:
753
61.e2
.
10
Richards
,
S. M.
,
E. A.
Clark
.
2009
.
BCR-induced superoxide negatively regulates B-cell proliferation and T-cell-independent type 2 Ab responses.
Eur. J. Immunol.
39
:
3395
3403
.
11
Wheeler
,
M. L.
,
A. L.
Defranco
.
2012
.
Prolonged production of reactive oxygen species in response to B cell receptor stimulation promotes B cell activation and proliferation.
J. Immunol.
189
:
4405
4416
.
12
Morand
,
S.
,
T.
Ueyama
,
S.
Tsujibe
,
N.
Saito
,
A.
Korzeniowska
,
T. L.
Leto
.
2009
.
Duox maturation factors form cell surface complexes with Duox affecting the specificity of reactive oxygen species generation.
FASEB J.
23
:
1205
1218
.
13
Rada
,
B.
,
T. L.
Leto
.
2008
.
Oxidative innate immune defenses by Nox/Duox family NADPH oxidases.
Contrib. Microbiol.
15
:
164
187
.
14
Kwon
,
J.
,
K. E.
Shatynski
,
H.
Chen
,
S.
Morand
,
X.
de Deken
,
F.
Miot
,
T. L.
Leto
,
M. S.
Williams
.
2010
.
The nonphagocytic NADPH oxidase DUOX1 mediates a positive feedback loop during T cell receptor signaling.
Sci. Signal.
3
:
ra59
. doi: 10.1126/scisignal.2000976.
15
Singh
,
D. K.
,
D.
Kumar
,
Z.
Siddiqui
,
S. K.
Basu
,
V.
Kumar
,
K. V. S.
Rao
.
2005
.
The strength of receptor signaling is centrally controlled through a cooperative loop between Ca2+ and an oxidant signal.
Cell
121
:
281
293
.
16
Harper
,
R. W.
,
C.
Xu
,
J. P.
Eiserich
,
Y.
Chen
,
C. Y.
Kao
,
P.
Thai
,
H.
Setiadi
,
R.
Wu
.
2005
.
Differential regulation of dual NADPH oxidases/peroxidases, Duox1 and Duox2, by Th1 and Th2 cytokines in respiratory tract epithelium.
FEBS Lett.
579
:
4911
4917
.
17
Rada
,
B.
,
K.
Lekstrom
,
S.
Damian
,
C.
Dupuy
,
T. L.
Leto
.
2008
.
The Pseudomonas toxin pyocyanin inhibits the dual oxidase-based antimicrobial system as it imposes oxidative stress on airway epithelial cells.
J. Immunol.
181
:
4883
4893
.
18
Hristova
,
M.
,
A.
Habibovic
,
C.
Veith
,
Y. M. W.
Janssen-Heininger
,
A. E.
Dixon
,
M.
Geiszt
,
A.
van der Vliet
.
2016
.
Airway epithelial dual oxidase 1 mediates allergen-induced IL-33 secretion and activation of type 2 immune responses.
J. Allergy Clin. Immunol.
137
:
1545
1556.e11
.
19
Chang
,
S.
,
A.
Linderholm
,
L.
Franzi
,
N.
Kenyon
,
H.
Grasberger
,
R.
Harper
.
2013
.
Dual oxidase regulates neutrophil recruitment in allergic airways.
Free Radic. Biol. Med.
65
:
38
46
.
20
Hirakawa
,
S.
,
R.
Saito
,
H.
Ohara
,
R.
Okuyama
,
S.
Aiba
.
2011
.
Dual oxidase 1 induced by Th2 cytokines promotes STAT6 phosphorylation via oxidative inactivation of protein tyrosine phosphatase 1B in human epidermal keratinocytes.
J. Immunol.
186
:
4762
4770
.
21
Zambrowicz
,
B. P.
,
G. A.
Friedrich
,
E. C.
Buxton
,
S. L.
Lilleberg
,
C.
Person
,
A. T.
Sands
.
1998
.
Disruption and sequence identification of 2,000 genes in mouse embryonic stem cells.
Nature
392
:
608
611
.
22
Donkó
,
A.
,
E.
Ruisanchez
,
A.
Orient
,
B.
Enyedi
,
R.
Kapui
,
Z.
Péterfi
,
X.
de Deken
,
Z.
Benyó
,
M.
Geiszt
.
2010
.
Urothelial cells produce hydrogen peroxide through the activation of Duox1.
Free Radic. Biol. Med.
49
:
2040
2048
.
23
Allman
,
D.
,
R. C.
Lindsley
,
W.
DeMuth
,
K.
Rudd
,
S. A.
Shinton
,
R. R.
Hardy
.
2001
.
Resolution of three nonproliferative immature splenic B cell subsets reveals multiple selection points during peripheral B cell maturation.
J. Immunol.
167
:
6834
6840
.
24
Capasso
,
M.
,
M. K.
Bhamrah
,
T.
Henley
,
R. S.
Boyd
,
C.
Langlais
,
K.
Cain
,
D.
Dinsdale
,
K.
Pulford
,
M.
Khan
,
B.
Musset
, et al
.
2010
.
HVCN1 modulates BCR signal strength via regulation of BCR-dependent generation of reactive oxygen species.
Nat. Immunol.
11
:
265
272
.
25
Qi
,
C. F.
,
M.
Hori
,
A. E.
Coleman
,
T. A.
Torrey
,
L.
Taddesse-Heath
,
B. H.
Ye
,
S. K.
Chattopadhyay
,
J. W.
Hartley
,
H. C.
Morse
III
.
2000
.
Genomic organisation and expression of BCL6 in murine B-cell lymphomas.
Leuk. Res.
24
:
719
732
.
26
Morel
,
F.
,
J.
Doussiere
,
P. V.
Vignais
.
1991
.
The superoxide-generating oxidase of phagocytic cells. Physiological, molecular and pathological aspects.
Eur. J. Biochem.
201
:
523
546
.
27
Bellavite
,
P.
1988
.
The superoxide-forming enzymatic system of phagocytes.
Free Radic. Biol. Med.
4
:
225
261
.
28
Wills-Karp
,
M.
,
F. D.
Finkelman
.
2008
.
Untangling the complex web of IL-4- and IL-13-mediated signaling pathways.
Sci. Signal.
1
:
pe55
.
29
Donaldson
,
D. D.
,
M. J.
Whitters
,
L. J.
Fitz
,
T. Y.
Neben
,
H.
Finnerty
,
S. L.
Henderson
,
R. M.
O’Hara
Jr.
,
D. R.
Beier
,
K. J.
Turner
,
C. R.
Wood
,
M.
Collins
.
1998
.
The murine IL-13 receptor alpha 2: molecular cloning, characterization, and comparison with murine IL-13 receptor alpha 1.
J. Immunol.
161
:
2317
2324
.
30
de Vries
,
J. E.
1998
.
The role of IL-13 and its receptor in allergy and inflammatory responses.
J. Allergy Clin. Immunol.
102
:
165
169
.
31
Andrews
,
R.
,
L.
Rosa
,
M.
Daines
,
G.
Khurana Hershey
.
2001
.
Reconstitution of a functional human type II IL-4/IL-13 receptor in mouse B cells: demonstration of species specificity.
J. Immunol.
166
:
1716
1722
.
32
Pao
,
L. I.
,
K. P.
Lam
,
J. M.
Henderson
,
J. L.
Kutok
,
M.
Alimzhanov
,
L.
Nitschke
,
M. L.
Thomas
,
B. G.
Neel
,
K.
Rajewsky
.
2007
.
B cell-specific deletion of protein-tyrosine phosphatase Shp1 promotes B-1a cell development and causes systemic autoimmunity.
Immunity
27
:
35
48
.
33
Okada
,
T.
,
A.
Maeda
,
A.
Iwamatsu
,
K.
Gotoh
,
T.
Kurosaki
.
2000
.
BCAP: the tyrosine kinase substrate that connects B cell receptor to phosphoinositide 3-kinase activation.
Immunity
13
:
817
827
.
34
Yamazaki
,
T.
,
T.
Kurosaki
.
2003
.
Contribution of BCAP to maintenance of mature B cells through c-Rel.
Nat. Immunol.
4
:
780
786
.
35
Schrader
,
M.
,
H. D.
Fahimi
.
2006
.
Peroxisomes and oxidative stress.
Biochim. Biophys. Acta
1763
:
1755
1766
.
36
Yano
,
S.
,
N.
Yano
.
2002
.
Regulation of catalase enzyme activity by cell signaling molecules.
Mol. Cell. Biochem.
240
:
119
130
.
37
Siraki
,
A. G.
,
J.
Pourahmad
,
T. S.
Chan
,
S.
Khan
,
P. J.
O’Brien
.
2002
.
Endogenous and endobiotic induced reactive oxygen species formation by isolated hepatocytes.
Free Radic. Biol. Med.
32
:
2
10
.
38
Kuhns
,
D. B.
,
W. G.
Alvord
,
T.
Heller
,
J. J.
Feld
,
K. M.
Pike
,
B. E.
Marciano
,
G.
Uzel
,
S. S.
DeRavin
,
D. A.
Priel
,
B. P.
Soule
, et al
.
2010
.
Residual NADPH oxidase and survival in chronic granulomatous disease.
N. Engl. J. Med.
363
:
2600
2610
.
39
Hsu
,
H. C.
,
P.
Yang
,
J.
Wang
,
Q.
Wu
,
R.
Myers
,
J.
Chen
,
J.
Yi
,
T.
Guentert
,
A.
Tousson
,
A. L.
Stanus
, et al
.
2008
.
Interleukin 17-producing T helper cells and interleukin 17 orchestrate autoreactive germinal center development in autoimmune BXD2 mice.
Nat. Immunol.
9
:
166
175
.
40
Xie
,
S.
,
J.
Li
,
J. H.
Wang
,
Q.
Wu
,
P.
Yang
,
H. C.
Hsu
,
L. E.
Smythies
,
J. D.
Mountz
.
2010
.
IL-17 activates the canonical NF-kappaB signaling pathway in autoimmune B cells of BXD2 mice to upregulate the expression of regulators of G-protein signaling 16.
J. Immunol.
184
:
2289
2296
.
41
Lentz
,
V. M.
,
T.
Manser
.
2001
.
Cutting edge: germinal centers can be induced in the absence of T cells.
J. Immunol.
167
:
15
20
.
42
Toellner
,
K. M.
,
W. E.
Jenkinson
,
D. R.
Taylor
,
M.
Khan
,
D. M. Y.
Sze
,
D. M.
Sansom
,
C. G.
Vinuesa
,
I. C.
MacLennan
.
2002
.
Low-level hypermutation in T cell-independent germinal centers compared with high mutation rates associated with T cell-dependent germinal centers.
J. Exp. Med.
195
:
383
389
.
43
Bortnick
,
A.
,
I.
Chernova
,
W. J.
Quinn
III
,
M.
Mugnier
,
M. P.
Cancro
,
D.
Allman
.
2012
.
Long-lived bone marrow plasma cells are induced early in response to T cell-independent or T cell-dependent antigens.
J. Immunol.
188
:
5389
5396
.
44
Obukhanych
,
T. V.
,
M. C.
Nussenzweig
.
2006
.
T-independent type II immune responses generate memory B cells.
J. Exp. Med.
203
:
305
310
.
45
Peshavariya
,
H.
,
G. J.
Dusting
,
F.
Jiang
,
L. R.
Halmos
,
C. G.
Sobey
,
G. R.
Drummond
,
S.
Selemidis
.
2009
.
NADPH oxidase isoform selective regulation of endothelial cell proliferation and survival.
Naunyn Schmiedebergs Arch. Pharmacol.
380
:
193
204
.
46
Du
,
J.
,
E. S.
Nelson
,
A. L.
Simons
,
K. E.
Olney
,
J. C.
Moser
,
H. E.
Schrock
,
B. A.
Wagner
,
G. R.
Buettner
,
B. J.
Smith
,
M. L.
Teoh
, et al
.
2013
.
Regulation of pancreatic cancer growth by superoxide.
Mol. Carcinog.
52
:
555
567
.
47
Rosc-Schlüter
,
B. I.
,
S. P.
Häuselmann
,
V.
Lorenz
,
M.
Mochizuki
,
F.
Facciotti
,
O.
Pfister
,
G. M.
Kuster
.
2012
.
NOX2-derived reactive oxygen species are crucial for CD29-induced pro-survival signalling in cardiomyocytes.
Cardiovasc. Res.
93
:
454
462
.
48
Hardy
,
R. R.
,
S. A.
Shinton
.
2004
.
Characterization of B lymphopoiesis in mouse bone marrow and spleen.
Methods Mol. Biol.
271
:
1
24
.
49
Vugmeyster
,
Y.
,
K.
Howell
,
A.
Bakshi
,
C.
Flores
,
O.
Hwang
,
K.
McKeever
.
2004
.
B-cell subsets in blood and lymphoid organs in Macaca fascicularis.
Cytometry A
61
:
69
75
.

The authors have no financial conflicts of interest.

Supplementary data