The Rac Activator DOCK2 Mediates Plasma Cell Differentiation and IgG Antibody Production

A hallmark of humoral immune responses is the production of antibodies. This process involves a complex cascade of molecular and cellular interactions, including recognition of specific antigen by the B cell receptor (BCR), which triggers activation of B cells and differentiation into plasma cells (PCs). Although activation of the small GTPase Rac has been implicated in BCR-mediated antigen recognition, its precise role in humoral immunity and the upstream regulator remain elusive. DOCK2 is a Rac-specific guanine nucleotide exchange factor predominantly expressed in hematopoietic cells. We found that BCR-mediated Rac activation was almost completely lost in DOCK2-deficient B cells, resulting in defects in B cell spreading over the target cell-membrane and sustained growth of BCR microclusters at the interface. When wild-type B cells were stimulated in vitro with anti-IgM F(ab′)2 antibody in the presence of IL-4 and IL-5, they differentiated efficiently into PCs. However, BCR-mediated PC differentiation was severely impaired in the case of DOCK2-deficient B cells. Similar results were obtained in vivo when DOCK2-deficient B cells expressing a defined BCR specificity were adoptively transferred into mice and challenged with the cognate antigen. In addition, by generating the conditional knockout mice, we found that DOCK2 expression in B-cell lineage is required to mount antigen-specific IgG antibody. These results highlight important role of the DOCK2–Rac axis in PC differentiation and IgG antibody responses.

A hallmark of humoral immune responses is the production of antibodies. This process involves a complex cascade of molecular and cellular interactions, including recognition of specific antigen by the B cell receptor (BCR), which triggers activation of B cells and differentiation into plasma cells (PCs). Although activation of the small GTPase Rac has been implicated in BCR-mediated antigen recognition, its precise role in humoral immunity and the upstream regulator remain elusive. DOCK2 is a Rac-specific guanine nucleotide exchange factor predominantly expressed in hematopoietic cells. We found that BCR-mediated Rac activation was almost completely lost in DOCK2deficient B cells, resulting in defects in B cell spreading over the target cell-membrane and sustained growth of BCR microclusters at the interface. When wild-type B cells were stimulated in vitro with anti-IgM F(ab′)2 antibody in the presence of IL-4 and IL-5, they differentiated efficiently into PCs. However, BCR-mediated PC differentiation was severely impaired in the case of DOCK2-deficient B cells. Similar results were obtained in vivo when DOCK2-deficient B cells expressing a defined BCR specificity were adoptively transferred into mice and challenged with the cognate antigen. In addition, by generating the conditional knockout mice, we found that DOCK2 expression in B-cell lineage is required to mount antigen-specific IgG antibody. These results highlight important role of the DOCK2-Rac axis in PC differentiation and IgG antibody responses.
Keywords: rac activation, DOcK2, B cell receptor, immunological synapse, plasma cell, antibody production inTrODUcTiOn B cells play an important role in protective immunity through production of antibodies that bind to and eliminate foreign antigens. During development in the bone marrow (BM), precursor B cells undergo rearrangements of the gene encoding the B cell receptor (BCR) and differentiate into immature B cells, which migrate to the spleen to complete their development via T1 and T2 transitional stages (1,2). Mature follicular B cells then enter secondary lymphoid tissues such as the lymph nodes (LNs) in search for cognate antigens. Specific recognition of antigen by the BCR triggers intracellular signaling cascades, leading to activation of mature B cells and differentiation into plasma cells (PCs) (3,4). During T cell-dependent (TD) humoral immune responses, PCs are initially produced in transient extrafollicular proliferative foci, but are subsequently derived from B cells participating in the follicular germinal center (GC) reactions (5)(6)(7). Accumulating evidence indicates that low-affinity antigens fail to induce PC differentiation (8)(9)(10). However, its underlying mechanism and cellular response are poorly understood.
Although soluble antigens can activate B cells, membranebound antigens are more effective in promoting B cell activation and are likely to constitute the dominant form of antigens responsible for B cell stimulation in vivo (11). When a mature B cell recognizes antigens tethered on the surface of a target cell such as the follicular dendritic cell (FDC), a microcluster of BCR and its cognate antigen forms and grows at the site of the contact (4), which is surrounded by adhesion molecules, leukocyte function-associated antigen-1 (LFA-1), and intercellular adhesion molecule-1 (ICAM-1) on the surface of B cells and FDCs, respectively. This structure is known as immunological synapse (IS), and its formation involves membrane polarization and cytoskeletal reorganization (4). Previous studies have indicated that the affinity of the BCR for antigen affects the extent of antigen accumulation at the contact site (12,13). Additionally, it is well established that intracellular signaling molecules also polarize to the IS, following a precise relative topology (4). Therefore, IS formation may be an important factor that determines the fate of antigen-specific B cells during humoral immune responses.
Rac is a member of Rho family GTPases that function as molecular switches by cycling between GDP-bound inactive and GTP-bound active states (14,15). Rac exists in the cytosol in the GDP-bound form and is recruited to membranes, where its GDP is exchanged for GTP by the action of one or more guanine nucleotide exchange factors (GEFs) (14,15). Once activated, Rac binds to multiple effector molecules and regulates various cellular functions including remodeling of the actin cytoskeleton. Rac is composed of three isoforms, Rac1, Rac2, and Rac3. Rac1 is ubiquitously expressed and Rac3 is highly expressed in the brain, whereas Rac2 expression is restricted largely to hematopoietic cells (15). So far, the role of Rac in B cells has been extensively analyzed using conventional Rac2 knockout (KO; Rac2 -/-) mice and/or conditional KO mice lacking Rac1 expression in B cell lineage (16)(17)(18). These results have shown that Rac2 is more important than Rac1 in B cell development, B cell adhesion, and IS formation. However, the effect of loss of Rac activation on antibody production remains unknown, because genetic deletion of both Rac1 and Rac2 in B cell lineage leads to virtually complete absence of mature B cells (17).
However, the B cell-intrinsic role of DOCK2 in antibody production remains unknown. In this study, we found that BCRmediated Rac activation and IS formation critically depend on DOCK2. By analyzing three different models, we demonstrate here that DOCK2 expression in B-lineage cells is required for PC differentiation and antigen-specific IgG production.

Plasmids and Transfection
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′-ATGAGGTCTTTGCTAATC

T TGGTGCT T TGCT TCCTGCCCCTGGCTGCTCTGG G G A A A G T C T T T G G A C G A T G T G A G -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.

calcium Flux assays
Lymph node B cells (1 × 10 6 ) 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.
homing assays 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 × 10 7 ) into C57BL/6 mice, the ratio of transferred B cells in the white pulp was compared at 48 h later.

statistical analyses
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.  (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).

DOcK2 regulates Bcr-Mediated B cell Proliferation and Pc Differentiation In Vitro
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.

DOcK2 regulates Bcr-Mediated is Formation
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 classswitch 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.

DOcK2 is required for expansion of gc B cells and Differentiation into Pcs in adoptive Transfer Model
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 antigenengaged 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.

Development and characterization of conditional KO Mice lacking DOcK2 in a B cell-specific Manner
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).

DiscUssiOn
DOCK2 regulates B cell migration and adhesion by acting downstream of chemokine receptors (26,39), yet, its role in BCR signaling is poorly understood. Here, we have shown that activations of Rac1 and Rac2 following BCR stimulation were markedly reduced in the absence of DOCK2. Our results thus identify DOCK2 as a key Rac GEF acting downstream of BCR. So far, the DH-type GEFs Vav proteins (Vav1, Vav2, and Vav3) have been considered to regulate B cell functions as the Rac GEFs (40,41). Although tyrosine phosphorylation of Vav augments its Rac GEF activity (42), BCR-mediated Vav phosphorylation was unchanged between WT and Dock2 −/− B cells. In addition, DOCK2 deficiency did not affect BCR-mediated calcium influx, which is defective in B cells from Vav1 −/− Vav2 −/− double KO mice (40,41). The precise relationship between DOCK2 and Vav proteins in BCR signaling is currently unknown. However, recent studies have shown that Vav proteins play important roles in T cells and NK cells independently of the Rac GEF activities (43,44). In light of this, it seems likely that Vav proteins act as adaptor molecules and regulate B cell functions via calcium mobilization.
Although Rac activation has been implicated in BCRmediated IS formation (4,18), its physiological relevance and the upstream signaling cascade are not completely understood. We found that antigen-driven B cell spreading and sustained growth of BCR microclusters were impaired in Dock2 −/− primary B cells. As these cellular responses are abrogated by actin polymerization inhibitors (4,45), Rac activation-induced remodeling of the actin cytoskeleton is likely to be involved. In addition, a recent study using chicken DT40 B cells revealed that the growth of BCR microclusters critically depends on PIP3, a lipid product of phosphatidylinositol 3-kinases (PI3Ks) (46). Indeed, DOCK2 binds to PIP3 through its DHR-1 domain (21,23). Therefore, it is highly conceivable that PI3K activity is required to recruit DOCK2 to the synaptic membrane and activate Rac locally for IS formation, as seen in other lymphocytes (35,46,47). While DOCK2 deficiency leads to defective IS formation, it did not affect phosphorylation of major signaling molecules downstream of BCR stimulated with a soluble cross-linking antibody. The role of DOCK2 in signal transduction might be more critical in the in vivo situations where antigen concentrations are often low and the signaling induced by antigen presented on the membrane with adhesion molecules becomes more important.
In this study, we have also shown that in the absence of DOCK2, BCR-mediated PC differentiation was severely impaired in vitro and in vivo. Similar results were obtained when WT B cells were treated with CPYPP, which binds to the DOCK2 DHR-2 domain and inhibits its Rac GEF activity (33). These results indicate that DOCK2 regulates BCR-mediated PC differentiation through Rac activation. How DOCK2-Rac signaling axis regulates PC differentiation remains to be determined. However, accumulating evidence indicates that low affinity antigens fail to induce PC differentiation (8-10). As B cell spreading and growth of BCR microclusters act to increase the number of signalosomes within the membrane (4), their defects in Dock2 −/− B cells may lead to the failure to amplify signaling above the threshold required for PC differentiation. Alternatively, in light of the fact that Rac has direct roles in the regulation of gene transcription (48,49), activated Rac may be involved in the expression of Prdm1 or its related genes during PC differentiation. Also, it may be possible that DOCK2-Rac axis regulates the expression of other molecules required for survival, growth, or differentiation during PC differentiation, because it has been reported that DOCK2 deficiency affects helper T cell differentiation by modulating cytokine receptor expression (50).
Finally, we have shown that CD19-Cre +/− Dock2 lox/lox conditional KO mice fail to mount antigen-specific IgG antibody upon immunization of OVA. This result is in marked contrast to a recent study showing that after treatment with tamoxifen to delete Rac1 in Rac2 -/-B cells, Mb1-Cre-ERT2 Rac1 lox/lox Rac2 -/mice exhibited increased IgG1 and IgG2b antibody to a TD antigen (51). As DOCK2 was deleted early during B cell development in CD19-Cre +/− Dock2 lox/lox mice, B cell trafficking is also impaired in this model. On the other hand, there is a time lag between the last tamoxifen treatment and antibody measurement in Mb1-Cre-ERT2 Rac1 lox/lox Rac2 −/− mice. These differences may affect the outcome of antibody production to The number of each subset of B cells was compared between CD19-Cre −/− Dock2 lox/lox and CD19-Cre +/− Dock2 lox/lox mice. Data are indicated as the mean ± SD of five mice. **p < 0.01 (two-tailed Mann-Whitney test). (B) FACS profiles for expression of IgM and IgD in the B220 + splenic B cells. The number of each subset of B cells was compared between CD19-Cre −/− Dock2 lox/lox and CD19-Cre +/− Dock2 lox/lox mice. Data are indicated as the mean ± SD of 5 mice. **p < 0.01 (two-tailed unpaired Student's t-test). (c) FACS profiles for expression of CD21 and heat stable antigen (HSA) in the B220 + splenic B cells. The number of each subset of B cells (T1, T2, and follicular B cells) was compared between CD19-Cre -/-Dock2 lox/lox and CD19-Cre +/− Dock2 lox/lox mice. Data are indicated as the mean ± SD of five mice. *p < 0.05; **p < 0.01 (two-tailed Mann-Whitney test). (D) FACS profiles for expression of IgM and IgD or CD21 and HSA in the B220 + peripheral LN (PLN) B cells. The number of each subset of B cells was compared between CD19-Cre −/− Dock2 lox/lox and CD19-Cre +/− Dock2 lox/lox mice. Data are indicated as the mean ± SD of five mice. *p < 0.05; **p < 0.01 (two-tailed Mann-Whitney test). (e) Immunohistochemical analyses of the spleen sections from CD19-Cre −/− Dock2 lox/lox and CD19-Cre +/− Dock2 lox/lox mice.

acKnOWleDgMenTs
We thank Jason G. Cyster for his kind permission to use HyHEL10 mice in this study. We also thank to Ayumi Inayoshi, Arisa Aosaka, and Linh Thi Hoai Nguyen for technical assistance. FUnDing