Dock2^−/− mice on C57BL/6 background (CD45.2⁺) have been described previously (26–28). HyHEL10 mice were generated by crossing VDJ9 HyHEL10 heavy-chain knock-in mice with Vκ5 HyHEL10 light chain transgenic mice (30) and bred to congenic C57BL/6 mice carrying CD45.1⁺ allele. For adoptive transfer experiments, Dock2^+/− or Dock2^−/− HyHEL10 mice carrying CD45.1⁺ allele were generated. For development of DOCK2 conditional KO mice, ES cell harboring loxP-flanked exon 3 of Dock2 allele (EUCOMM consortium) were microinjected into C57BL/6 blastocysts, and the male chimeras were crossed with C57BL/6 mice to obtain Dock2^lox/lox mice (for details, see Figure S1 in Supplementary Material). Dock2^lox/lox mice were crossed with CD19-Cre mice (CD19-Cre^+/−), in which a Cre recombinase gene is inserted heterozygously into the first exon of the CD19 (31).

B cells were purified from the peripheral LN cells with B cell Isolation kit (Miltenyi Biotec) and cultured in RPMI 1640 medium (Wako) containing 10% heat-inactivated fetal calf serum (Nichirei Biosciences), 50 µM 2-mercaptoethanol (Nacalai tesque), 2 mM l-glutamine (Life Technologies), 100 U/ml penicillin (Life Technologies), 100 µg/ml streptomycin (Life Technologies), 1 mM sodium pyruvate (Life Technologies), and MEM non-essential amino acids (Life Technologies) (designated complete RPMI medium). For proliferation assay, LN B cells (5 × 10⁴/well) were stimulated in complete RPMI medium with the specified concentrations of anti-IgM F(ab′)₂ antibody (Jackson ImmunoResearch Laboratories), anti-CD40 antibody (BD Biosciences), or lipopolysaccharide (LPS; Sigma-Aldrich) in the presence or absence of IL-4 (8 ng/ml; PeproTech) or IL-5 (10 ng/ml; PeproTech) for 48 h, and [³H]-thymidine (37 kBq) was added during the final 18 h of the culture. To assess PC differentiation in vitro, LN B cells (1 × 10⁵/well) were stimulated in complete RPMI medium with anti-IgM F(ab′)₂ antibody (33 µg/ml; Jackson ImmunoResearch Laboratories), anti-CD40 antibody (5 µg/ml; BD Biosciences) or LPS (10 µg/ml; Sigma-Aldrich) in the presence or absence of IL-4 and IL-5 (both from PeproTech; 10 ng/ml) for 96 h, as described previously (32). In some experiments, CPYPP (33), a small-molecule inhibitor of DOCK2, was added to the culture at 12.5 µM.

The following antibodies and reagents were used at the indicated concentrations. Allophycocyanin (APC)-conjugated or phycoerythrin (PE)-conjugated anti-mouse CD45R/B220 (RA3-6B2; 2 µg/ml), fluorescein isothiocyanate (FITC)- or PE-conjugated anti-IgM (R6-60.2; 5 µg/ml), APC-anti-CD19 (1D3; 2 µg/ml), FITC-anti-IgD (11-26c.2a; 5 µg/ml), FITC-anti-CD21/CD35 (7G6; 10 µg/ml), biotinylated anti-heat stable antigen (HSA; M1/69; 0.1 µg/ml), FITC-anti-CD45.1 (A20; 2 µg/ml), PE-anti-CD38 (90; 0.7 µg/ml), PE-anti-IgG1 (A85-1; 1 µg/ml), PE-anti-CD138 (281-1; 2 µg/ml), PE-conjugated streptavidin (1 µg/ml) or PerCP-5.5cyanine-conjugated streptavidin (0.5 µg/ml) were from BD Bioscience. Biotinylated anti-GL7 (GL7; 2 µg/ml) purchased from eBioscience. Alexa Fluor 647-labeled HEL was prepared with an Alexa Fluor647 antibody-labeling kit (Invitrogen). Before staining with the antibodies, cells were incubated for 10 min on ice with anti-Fcγ III/II receptor (2.4G2; 0.5 µg/ml; BD Bioscience) to block Fc receptors. In some experiments, mice were injected intraperitoneally with 300 µl of BrdU (Invitrogen) and sacrificed 5 h later. Splenocytes were then stained with APC BrdU flow kit (Becton-Dickinson). Flow cytometric analyses were done on FACS Calibur (BD Bioscience).

Ninety six-well polystyrene plates (Thermo 3855) were coated overnight at 4°C with OVA (0.5 µg), HEL (2 µg), or goat anti-mouse Ig (IgM + IgG + IgA, H + L; #1010-01) antibody (Southern Biotech). After the wells were blocked with 150 µl phosphate-buffered saline (PBS) containing 1% sodium casein and 0.1% Tween-20, serial dilutions of sera were added. Alkaline phosphatase-conjugated isotype-specific antibodies (IgM, IgG1, IgG2b; Southern Biotech) were used to detect bound antibody. The reactions were visualized with the substrate p-nitrophenyl phosphate (Sigma-Aldrich) and detected at 405 nm.

The cDNA encoding HEL was amplified by PCR using the pET-22b HEL (amino acid residues 19–147) as a template (34). The following primers were used: 5′-ATGAGGTCTTTGCTAATCTTGGTGCTTTGCTTCCTGCCCCTGGCTGCTCTGGGGAAAGTCTTTGGACGATGTGAG-3′ and 5′-TCACAGCCGGCAGCCTCTGA-3′. The cDNAs encoding enhanced green fluorescent protein (EGFP) and the GPI anchor domain were prepared as described previously (35). The HEL-, EGFP-, and GPI anchor-coding cDNAs were cloned into the EcoR I-Pst I, Pst I-BamH I, and BamH I-Not I sites of the pBSSK vector, respectively, which was then cloned into the Xho I-Not I site of the pBJ1 vector. The pBJ1-HEL-GFP-GPI construct was linearized with Sal I and electroporated into the baby hamster kidney (BHK) cells expressing ICAM-1-GPI (36), together with pTRE2-puro vector. Cells were cultured in the presence of puromycin (0.3 µg/ml) and clones stably expressing HEL-GFP-GPI were selected.

For Rac activation assays, LN B cells (1.25 × 10⁷ per sample) were stimulated with anti-IgM F(ab′)₂ antibody (33 µg/ml; Jackson ImmunoResearch Laboratories) at 37°C for the specified times. Cells were then lysed by adding 1× MLB [Mg²⁺ Lysis Buffer: 25 mM Hepes (pH 7.5), 150 mM NaCl, 1% lgepal CA-630, 10 mM MgCl₂, 1 mM EDTA, 10% glycerol; Millipore], followed by centrifugation at 20,000 × g for 1 min at 4°C. Aliquots were saved for total cell lysate controls, and the remaining lysates were incubated with agarose beads containing the GST-fusion Rac binding domain of PAK1 (#14-325; Millipore) at 4°C for 1 h. The beads were washed twice with 1× MLB buffer and suspended in 1× SDS-PAGE sample buffer [62.5 mM Tris–HCl (pH 6.8), 0.005% bromophenol blue, 2% SDS, 10% glycerol, 100 mM dithiothreitol]. The bound proteins and the same amount of total lysates were analyzed by SDS-PAGE, and blots were probed with the anti-Rac1 (23A8; Millipore) or Rac2 (3B10-2D9; Sigma-Aldrich) antibody.

To examine expression and activation of each signaling molecule, cells were lysed on ice in 20 mM Tris–HCl buffer (pH 7.5) containing 1% Triton X-100, 150 mM NaCl, 1 mM β-glycerophosphate, 1 mM Na₃VO₄, and complete™ protease inhibitors (Roche). After centrifugation, the supernatants were mixed with an equal volume of 2× SDS-PAGE sample buffer (125 mM Tris–HCl, 0.01% bromophenol blue, 4% SDS, 20% glycerol, and 200 mM dithiothreitol). Samples were boiled for 5 min and analyzed by immunoblotting with the following antibodies: rabbit anti-DOCK2 (09-454; 1:1,000 dilution; Millipore); rabbit anti-Syk (1:1,000 dilution; Cell Signaling Technology), rabbit anti-phospho-Syk (Tyr323; 1:1,000 dilution; Cell Signaling Technology), rabbit anti-p44/42 MAPK (Erk1/2) (1:1,000 dilution; Cell Signaling Technology), rabbit anti- phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204; 1:1,000 dilution; Cell Signaling Technology), rabbit anti-Akt (1:1,000 dilution; Cell Signaling Technology), rabbit anti- phospho-Akt (Thr308; 1:1,000 dilution; Cell Signaling Technology), rabbit anti-BLNK (1:1,000 dilution; Cell Signaling Technology), rabbit anti- phospho-BLNK (Tyr96; 1:1,000 dilution; Cell Signaling Technology), rabbit anti-CD19 (1:1,000 dilution; Cell Signaling Technology), rabbit anti- phospho-CD19 (Tyr513; 1:1,000 dilution; Cell Signaling Technology). To analyze tyrosine phosphorylation of Vav or PLCγ2, cell extracts were incubated with protein G sepharose conjugated with anti-Vav (C-14; 1:1,000 dilution; Santa Cruz Biotechnology) or anti-PLCγ2 (Q-20; 1:1,000 dilution; Santa Cruz Biotechnology) antibody. The precipitates were subjected to immunoblotting using anti-phosphotyrosine antibody (pY99; 1:1,000 dilution; Santa Cruz Biotechnology).

Lymph node B cells (1 × 10⁶) were loaded with 3 µM Fura 2-AM (Wako Chemicals) for 30 min at 37°C. Cells were then resuspended in Hank’s buffered salt solution containing calcium and magnesium, and were stimulated with anti-IgM F(ab′)₂ antibody (33 µg/ml). Fluorescence intensities were monitored at an excitation wavelength of 340 or 380 nm and emission wavelength of 510 nm using a Flex Station3 (Molecular Devices). Ionomycin (10 µM; Sigma-Aldrich) was used as a positive control.

To analyze IS formation, LN B cells (3 × 10⁵) from HyHEL10 mice were stained with PKH26 (Sigma-Aldrich) or biotinylated anti-LFA-1 antibody (2D7; BD Biosciences) followed by Alexa546-conjugated streptavidin (Invitrogen) before assays. After LN B cells were incubated on a monolayer of BHK cells expressing ICAM1-GPI and HEL-GFP-GPI (designated BHK-ICAM-HEL cells) at 37°C for the specified times, cells were fixed with 4% paraformaldehyde for 12 min. All images were taken with FV3000 laser scanning confocal microscopy (Olympus).

HEL was covalently conjugated to sheep red blood cell (SRBC; KOHJIN BIO) with 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (Sigma-Aldrich) as described previously (37). CD45.1⁺ LN B cells (1 × 10⁵ B cells from Dock2^+/− HyHEL10 mice or 2 × 10⁵ B cells from Dock2^−/− HyHEL10 mice) were adoptively transferred into 6- to 7-week-old C57BL/6 mice (CD45.2⁺) together with 2 × 10⁸ HEL-SRBC or SRBC alone. Mice were sacrificed at the specified time points, and spleen cells were analyzed by flow cytometry. In some experiments, mice were injected intraperitoneally with 300 µl of BrdU (Invitrogen) to assess cell proliferation.

To examine antigen-specific antibody responses, mice were immunized by intraperitoneal injection of ovalbumin (OVA; 50 µg per mouse; Sigma-Aldrich) emulsified in complete Freund’s adjuvant (CFA; Difco Laboratories). Fourteen days later, the serum levels of anti-OVA antibody were determined by ELISA.

Freshly prepared spleens were embedded in Tissue-Tek OCT compound (Sakura Finetechnical), and frozen at −80°C. Cryosections (10 µm) were fixed with 4% (w/v) paraformaldehyde for 10 min at 37°C. After being blocked with 10% horse serum (Sigma-Aldrich), samples were stained with FITC-conjugated anti-B220 (RA3-6B2; BD Biosciences), PE-conjugated anti-CD3 (17A2; BioLegend), anti-metallophillic macrophages (MOMA1; BMA Biomedicals) followed by Alexa 647-conjugated goat anti-rat antibody (Invitrogen). All images were obtained with a laser-scanning confocal microscope (LSM510 META, Carl Zeiss).

Total RNA was isolated using ISOGEN (Nippon Gene). After treatment with RNase-free DNase I (Life Technologies), RNA samples were reverse-transcribed with oligo(dT) primers (Life Technologies) and SuperScript III reverse transcriptase (Life Technologies) for amplification by PCR. The following PCR primers were used: Prdm1; 5′-GACTGGGTGGACATGAGAGAG-3′ and 5′-CCATCAATGAAGTGGTGGAAC-3′. Gapdh; 5′-ACCACAGTCCATGCCATCAC-3′ and 5′-TCCACCACCCTGTTGCTGTA-3′.

B cells were purified from the LNs from Dock2^+/+ and Dock2^−/− mice and labeled with PKH26 fluorescent cell linkers (Sigma-Aldrich) or CMTMR (Life Technologies), respectively. After intravenous injection of LN B cells (1–2 × 10⁷) into C57BL/6 mice, the ratio of transferred B cells in the white pulp was compared at 48 h later.

Statistical analyses were performed using GraphPad Prism. The data was initially tested with a Kolmogorov–Smirnov test for normal distribution. Parametric data were analyzed using a two-tailed unpaired Student’s t-test when two groups were compared. Nonparametric data were analyzed with a two-tailed Mann–Whitney test when two groups were compared. P-values less than 0.05 were considered significant.

Although Dock2^−/− mice exhibited diminished numbers of transitional B cells and mature follicular B cells in the spleen (Figures S2B,C in Supplementary Material), LN B cells from C57BL/6 (designated Dock2^+/+) mice and Dock2^−/− mice showed similar IgM^lowIgD^hi mature phenotype due to the lack of transitional B cells (2) (Figure S2D in Supplementary Material). Therefore, to examine whether DOCK2 functions downstream of BCR, we prepared LN B cells and analyzed activation and phosphorylation of the signaling molecules. When LN B cells from Dock2^+/+ mice were stimulated with anti-IgM F(ab′)₂ antibody, the GTP-bound, activated Rac1 and Rac2 were readily detected at 0.5 min after stimulation (Figure 1A). However, BCR-mediated activation of Rac1 and Rac2 were reduced in Dock2^−/− B cells to 4.7 and 20.9% of the wild-type (WT) levels, respectively (Figure 1A). These results indicate that DOCK2 is a major Rac GEF acting downstream of BCR. On the other hand, BCR-mediated calcium influx occurred normally even in Dock2^−/− B cells (Figure 1B). In addition, we found that DOCK2 deficiency did not affect phosphorylations of other signaling molecules such as Erk, Syk, Akt, BLNK, CD19, PLCγ2, and Vav (Figures 1C,D).

Having found that DOCK2 acts downstream of BCR, we next examined whether DOCK2 deficiency affects BCR-mediated B cell functions in vitro. Although Dock2^+/+ B cells proliferated vigorously when stimulated with anti-IgM F(ab′)₂ antibody in the presence or absence of IL-4/IL-5, BCR-mediated B-cell proliferation was impaired in the absence of DOCK2 (Figure 2A). When Dock2^+/+ B cells were stimulated with anti-IgM F(ab′)₂ antibody plus IL-4 and IL-5 for 4 days, they efficiently differentiated into CD138⁺ PCs (Figure 2B). However, in the case of Dock2^−/− B cells, CD138⁺ PCs were hardly detected under the same culture condition (Figure 2B). Consistent with this, the expression of Prdm1, which encodes the transcription factor Blimp-1 important for PC differentiation (38), was readily detected in Dock2^+/+ B cells, but not Dock2^−/− B cells (Figure 2C). Importantly, BCR-mediated PC differentiation was impaired when Dock2^+/+ B cells were treated with CPYPP (Figure 2D), a small-molecule inhibitor of DOCK2 that binds to the DOCK2 DHR-2 domain and inhibits its Rac GEF activity (33). On the other hand, DOCK2 deficiency did not affect B-cell proliferation and PC differentiation in response to CD40 ligation or LPS stimulation (Figures 2A,B). Thus, DOCK2 selectively regulates BCR-mediated B cell proliferation and PC differentiation via Rac activation.

Although a previous study has indicated that B cell adhesion and IS formation are impaired in Rac2-deficient B cells (18), the physiological function of Rac1 and Rac2 activation in this process is not completely understood. To address this issue, we crossed Dock2^−/− mice with HyHEL10 mice that express a defined anti-HEL BCR and are capable of normal Ig class-switch recombination and somatic hypermutation. Irrespective of DOCK2 expression, LN B cells from HyHEL10 mice comparably bound to HEL (Figure 3A). When PKH26-labeled LN B cells from Dock2^+/− HyHEL10 mice were incubated with BHK-ICAM-HEL cells, they rapidly spread over the target membrane, where small clusters of GFP-fusion HEL were formed by 3 min within the area of interaction (Figure 3B). However, in the case of Dock2^−/− HyHEL10 B cells, a spreading response was impaired with a significant reduction of the number of BCR microclusters at the site of the contact (Figures 3B–D). Similarly, LFA-1 accumulation was reduced in the case of Dock2^−/− HyHEL10 B cells (Figures 3E,F). These results indicate that BCR-mediated IS formation critically depends on DOCK2.

To examine the role of DOCK2 in PC differentiation in vivo, we prepared LN CD45.1⁺ B cells from Dock2^+/− and Dock2^−/− HyHEL10 mice and adoptively transferred them into C57BL/6 mice (CD45.2) with HEL-conjugated SRBCs (Figure 4A). As DOCK2 deficiency reduces B cell homing to the secondary lymphoid organs (26, 39), we injected Dock2^−/− B cells twice as much as Dock2^+/− B cells to compensate the number of B cells in the lymphoid follicle (Figure S3 in Supplementary Material). In both cases, the frequencies of GL7⁺CD38⁻ B cells and IgG1⁺ B cells to the total CD45.1⁺ B cells were comparable between Dock2^+/− and Dock2^−/− B cells at day 5 and day 6 after transfer (Figures 4B,C), indicating that DOCK2 deficiency does not affect differentiation of antigen-engaged B cells to GC B cells and class-switch recombination. However, while Dock2^+/− GC B cells proliferated well from day 4 to day 5, such expansion was impaired in the case of Dock2^−/− GC B cells (Figure 4D). This was further supported by analyzing BrdU incorporation (Figure 4E). More importantly, we found that B cells from Dock2^−/− HyHEL10 mice failed to differentiate efficiently to CD138⁺ PCs (Figures 4B–D). These results indicate that DOCK2 is required for expansion of GC B cells and differentiation into PCs during TD antibody response.

To examine the B cell intrinsic role of DOCK2 under more physiological condition, we developed conditional KO mice lacking DOCK2 in a B cell-specific manner (CD19-Cre^+/−Dock2^lox/lox mice). Western blot analyses revealed that DOCK2 expression was selectively deleted in B-lineage cells in these mice (Figure S4 in Supplementary Material). We first compared B cell development between CD19-Cre^+/−Dock2^lox/lox and CD19-Cre^−/−Dock2^lox/lox mice. Although the amounts of pre/pro B cells and immature B cells in the BM were unchanged between them (Figure 5A), the number of mature recirculating B cells was reduced to 44% of the control level (Figure 5A). Similarly, CD19-Cre^+/−Dock2^lox/lox mice had diminished numbers of transitional B cells and mature follicular B cells in the spleen (Figures 5B,C), as seen in Dock2^−/− mice (Figures S2B,C in Supplementary Material). On the other hand, no phenotypic difference was found when LN B cells from CD19-Cre^+/−Dock2^lox/lox and CD19-Cre^−/−Dock2^lox/lox mice were stained for IgM and IgD, or CD21 and HSA (Figure 5D). These phenotypes were similar to those of Dock2^−/− mice (Figure S2D in Supplementary Material). Consistent with the FACS data, immunohistochemical analyses of the spleen revealed that the relative size and number of B cell follicles was significantly reduced in CD19-Cre^+/−Dock2^lox/lox mice, compared with CD19-Cre^–/– Dock2^lox/lox mice (Figure 5E). However, the organization of T cells and macrophages in the white pulp was not altered in CD19-Cre^+/−Dock2^lox/lox mice (Figure 5E).

Under basal condition, serum levels of IgG1 and IgG2b were significantly reduced in CD19-Cre^+/−Dock2^lox/lox mice, compared with CD19-Cre^−/−Dock2^lox/lox control mice (Figure 6A). We then compared TD antibody response between CD19-Cre^+/−Dock2^lox/lox and CD19-Cre^−/−Dock2^lox/lox mice. When CD19-Cre^−/−Dock2^lox/lox mice were injected intraperitoneally with OVA in CFA, antigen-specific IgG1 and IgG2b antibodies were readily detected at 14 days after immunization (Figure 6B). However, OVA-specific IgG antibody production was severely impaired in CD19-Cre^+/−Dock2^lox/lox mice (Figure 6B). These results demonstrate a critical role of DOCK2 in IgG antibody responses in vivo.