Vav Proteins Regulate the Plasma Cell Program and Secretory Ig Production

Mice

The generation of Vav^null mice has been previously described (16). Vav^null, control mice, and all breedings were maintained in the specific pathogen-free facility of Washington University School of Medicine in accordance with institutional policies for animal care and usage.

Flow cytometry

Single-cell suspensions were prepared from the spleens of 6- to 8-wk-old mice and stained with Abs according to standard protocols. Ab conjugates against the following markers were used: rat IgG2a (isotype control), CD21/CD35, CD23, B220, IgM, IgD, CD86, CD43, CD19, AA4.1, syndecan-1, CD180 (BD Biosciences), and TLR4 (eBioscience). Biotinylated Abs were detected with streptavidin-CyChrome (BD Biosciences). MZ B cells were isolated by cell sorting using anti-CD21/CD35 and anti-CD23. For intracellular staining of IgM, cells were surface stained with Abs against IgM and syndecan-1, fixed with 2% paraformaldehyde (Polysciences), and permeabilized in 0.1% saponin. Samples analyzed for intracellular Ig were again stained with anti-IgM. The change in mean fluorescence intensity (ΔMFI) is defined as MFI of intracellularly stained samples minus MFI from surface stained samples. Labeling with 1 μM CFSE (Molecular Probes) was performed on sorted MZ B cells. Labeled cells were washed in complete medium and cultured with the indicated stimuli for 72 h.

Cell culture, proliferation assays, and retroviral transductions

Unless otherwise indicated, MZ B cells were plated at 2–3 × 10⁶/ml in complete DMEM-10% FBS, stimulated with either LPS (0.5 μg/ml; Sigma-Aldrich) or 0.6 μg/ml nuclease-resistant phosphorothioate backbone oligodeoxynucleotide (CpG S-ODN) 1826: 5′-TCCATGACGTTCCTGACGTT-3′ (Integrated DNA Technologies) and cultured at 37°C in a 5% CO₂-humidified incubator. Proliferation assays using [³H]thymidine were performed as previously described (16). To analyze Vav1 rescue of PC development, retrovirus-containing supernatants were collected 48 h after transfection of the Phi-NX ecotropic packaging cell line with retrovector plasmids, centrifuged for 1 h at 10,000 × g with LPS-activated MZ B cells from wild-type (WT) or Vav^null mice, and analyzed at 72 h postactivation.

Semiquantitative and real-time PCR

MZ B cells cultured for 72 h with LPS or CpG were harvested, suspended in TRIzol (Invitrogen Life Technologies), and RNA was extracted according to the manufacturer’s instructions. Extracted RNA was treated with 10 U/ml RNase-free DNase for 20 min at 37°C. cDNA was generated using AMV reverse-transcriptase (Promega) according to manufacturer’s instructions. Semiquantitative PCR was performed on 4-fold dilutions of cDNAs. Blimp-1, Pax5, and XBP-1 expression was determined by quantitative RT-PCR (MX300P cycler; Stratagene) and normalized to GAPDH. Primers and PCR conditions are available upon request.

ELISA

Supernatant was collected from purified MZ B cells 72 h after stimulation with indicated mitogen. Serum was collected by bleeding 6- to 8-wk-old age-matched nonimmunized mice maintained in the same room of the specific pathogen-free facility. ELISA plates (Maxisorp; Nunc) coated with goat anti-mouse (H+L chain; Southern Biotechnology Associates) were blocked with 2% FBS/PBS. Samples were detected with HRP-conjugated isotype-specific Abs (Southern Biotechnology Associates) and read on Model 550 microplate reader (Bio-Rad).

Vav^null mice lack serum Ig

Humoral immunity is mediated by Ig-secreting PC following T cell-dependent and T cell-independent reactions. However, even in the absence of specific antigenic challenge, serum of normal mice contains large amounts of Igs, natural Abs thought to be produced by MZ and B1 B cells (17). Previous studies demonstrated that mice lacking Vav1 and Vav2, or all three Vav proteins (Vav^null), show defective humoral responses to immunizations with T cell-independent and T cell-dependent Ags (16, 18, 19). However, serum Ig levels in naive Vav^null mice have not been reported, whereas mice lacking Vav1, Vav2, or both show relatively normal levels of serum Igs (18, 19, 20, 21). Surprisingly, our analyses revealed a virtual lack of all serum Igs, including IgM, in Vav^null mice (Fig. 1⇓). This finding was surprising considering that B cell developmental defects in Vav^null mice are partial and, similar to Vav1/Vav2-deficient, these mice develop substantial numbers of B cells, including MZ B cells. Although Vav^null mice exhibit severe defects in the generation of B1 B cells, Vav1 single knockouts, which also show defects in B1 B cell generation, have normal levels of serum Igs (21). This finding suggests that the defective generation of B1 B cells may contribute to, but is not sufficient to explain, the striking absence of serum Igs in Vav^null mice (16, 18, 19, 21). Thus, the severe paucity of serum Ig in Vav^null mice (Fig. 1⇓) implicates Vav in the regulation of Ig production beyond a requirement in B cell development.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Analyses of all serum Igs, including IgM, in Vav^null mice. An absence of serum Igs in Vav^null mice is shown. Serum Igs were analyzed by ELISA as described in Materials and Methods. Each data point represents an individual animal. The horizontal bars represent mean concentrations. p < 0.05 for total Ig, IgG1, and IgG2A; p < 0.0001 for IgM, IgG2B, and IgG3.

MZ B cells from Vav^null mice are phenotypically mature and proliferate in response to TLR ligands

To examine whether Vav is specifically required in PC differentiation and Ig production, we focused on MZ B cells as the development of these cells appears to be largely independent of Vav (16). Indeed, analyses of WT and Vav^null splenocytes showed that populations of MZ B cells (CD21/CD35^high CD23^low/neg) were present in both types of mice, with only modest decreases in numbers and percentages in Vav^null mice (16) (Fig. 2⇓, A and B). Moreover, analyses of the expression of a panel of surface markers showed that MZ B cells from Vav^null mice were of similar size as MZ B cells from WT mice, expressed equivalent levels of IgM, IgD, B220, CD19, and CD86, but did not express AA4.1 and CD43, markers of immature B cells (Fig. 2⇓A). These data indicate that the MZ B cells in Vav^null mice are mature and phenotypically indistinguishable from WT.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Vav^null MZ B cells are mature and proliferate to TLR stimulation. A, Flow cytometry analyses of splenocytes from WT and Vav^null mice were performed. Histograms of CD21/CD35^high CD23^low cells are shown and represent Vav^null (thick line) and WT (shaded) mice. One representative of two experiments is shown. B, Total numbers and percentage of spleen MZ B cells are presented as the mean ± SD from four experiments, WT (n = 4) and Vav^null (n = 8) mice. C, The percentage of blasting MZ B cells stimulated with for 72 h is shown (n = 4 experiments) in each panel. D, As in C, the number of blasting MZ B cells stimulated with LPS or CpG with [³H]thymidine incorporation. Shown are raw cpm values, representative of two experiments with similar results. E, CFSE dilution as indicated. Histograms represent WT (shaded histogram) and Vav^null (thick line histogram) MZ B cell divisions. Representative data are shown for WT (n = 2) and Vav^null (n = 4) mice.

To determine whether proliferative responses of Vav^null MZ B cells to TLR ligands were altered, FACS-sorted MZ B cells from either WT or Vav^null mice were stimulated with either the TLR4 ligand LPS or the TLR9 ligand CpG, and their proliferation was measured by multiple assays. Given that proliferation is thought to be a mechanistic requirement for secretory Ig production (1), we reasoned that potential defects in TLR signaling and defective proliferation of Vav^null MZ B cells could underscore any defects in Ig production. Surprisingly, LPS or CpG-stimulated Vav^null MZ B cells proliferated similarly to WT, as determined by three different assays. First, the percentages of blasting cells, as defined by forward and side scatter, were similar between WT and Vav^null cells (Fig. 2⇑C). Second, [³H]thymidine incorporation by WT and Vav^null MZ B cells, in response to various doses of LPS or CpG, gave indistinguishable results (Fig. 2⇑D). Third, analyses of proliferation by CFSE dye-dilution indicated that both WT and Vav^null MZ B cells underwent similar numbers of cell divisions upon TLR4 or TLR9 stimulation (Fig. 2⇑E). Thus, we conclude that LPS- or CpG-induced proliferative responses of Vav^null MZ B cells are similar to responses in WT MZ B cells as assessed by several independent criteria.

In this context, a recent report indicated a requirement for Vav in B cell proliferative responses to LPS (22). In agreement with this report, we found that proliferation of total splenic (B220⁺) Vav^null B cells to LPS was decreased as compared with WT (Fig. 3⇓). Because the majority of B220⁺ cells in Vav^null mice are immature (16), these results are consistent with the notion that immature B cells proliferate less vigorously in response to TLR4 stimulation than mature B cells (23). However, we cannot rule out the possibility that the requirement, or lack thereof, for Vav in LPS-induced proliferative responses distinguishes MZ B cells from other B cell subsets, a precedence of which has been reported (24). Nevertheless, our results clearly indicate that in murine MZ B cells, Vav proteins are superfluous in mitogenic responses triggered by TLR4 or TLR9.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Impaired proliferation of splenic Vav^null B cells. Purified total splenic B cells from WT and Vav^null mice were stimulated with increasing concentrations of LPS and analyzed for [³H]thymidine incorporation. Shown are raw cpm values, representative of two experiments with similar results.

MZ B cells from Vav^null mice are severely impaired in PC differentiation

Having determined that MZ B cells from Vav^null mice are competent to proliferate in response to LPS or CpG (Fig. 2⇑, C–E), we next analyzed the ability of these cells to differentiate into PC and produce secretory Ig. To this end, FACS-sorted MZ B cells from WT and Vav^null mice were stimulated with LPS or CpG and then analyzed for the expression of syndecan-1 (CD138), a PC marker induced during PC differentiation but absent on the surface of naive B cells (25). As expected, WT MZ B cells cultured in the presence of LPS or CpG up-regulated the expression of syndecan-1 (Fig. 4⇓A). Strikingly, MZ B cells from Vav^null mice showed reduced percentages of syndecan-1-expressing cells under the same culture conditions (Fig. 4⇓A). To directly assess the ability of MZ B cells to produce secretory Ig, we analyzed WT and Vav^null culture supernatants for soluble IgM and IgG3. In contrast to the abundant IgM and IgG3 detected in the supernatant from WT MZ B cells, ∼10-fold less IgM and 4-fold less IgG3 were detected in the supernatant from LPS-stimulated Vav^null MZ B cells (Fig. 4⇓B). Similarly, CpG-stimulated Vav^null MZ B cells showed a significant reduction of supernatant IgM as compared with WT cells. Minimal IgM was detected in wells of MZ B cells cultured in the absence of LPS or CpG, nor did these cells express syndecan-1 (data not shown). These data indicate that Vav proteins are critical for LPS or CpG-induced syndecan-1 expression and secretory Ig production by MZ B cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Impaired syndecan-1 induction and Ig secretion by Vav^null B cells. A, Analyses of syndecan-1 expression on WT or Vav^null MZ B cells stimulated with LPS or CpG for 72 h, as indicated. The mean percentage of syndecan-1-positive cells (n = 6 mice) was WT = 41.8 ± 2.5 and Vav^null = 12.8 ± 3.7 for LPS treatment, and WT = 31.9 ± 2.3 and Vav^null = 11.2 ± 3.1 for CpG treatment. B, IgM secretion by WT or Vav^null MZ B cells, as indicated, was analyzed by ELISA (n = 4 experiments). ∗∗, p < 0.0001; ∗, p < 0.05). C, Intracellular (thick line histogram) and surface (shaded histogram) IgM staining on cells as in A and B, as indicated (n = 4 experiments). D, Syndecan-1 expression in retrovirally transfected cells, as indicated. One representative experiment of three performed is shown for WT (n = 3) and Vav^null (n = 6) mice.

PC differentiation results in the concomitant decrease of surface IgM and the increase in secreted IgM, presumably due to differential splicing of μH chain transcripts encoding membrane-bound or soluble IgM, respectively (1). Although our data (Fig. 4⇑, A and B) suggested a requirement for Vav in soluble Ig production, we reasoned that it is possible that Vav may instead be selectively required for the secretion of soluble Igs. To rule out this possibility, we analyzed surface vs intracellular expression of IgM in WT and Vav^null MZ B cells, comparing the change in MFI. In this assay, WT MZ B cells cultured in the presence of LPS showed reduced expression of surface Ig, whereas the cytoplasmic Ig staining revealed a dramatic up-regulation of total Ig levels (Fig. 4⇑C, top left). Strikingly, although Vav^null MZ B cells expressed higher levels of surface IgM than WT, these cells contained little cytoplasmic Ig protein, suggesting that the defect in soluble Ig production is due to an impairment in PC differentiation rather than a selective defect in Ig secretion (Fig. 4⇑C). Of note, we consistently detected small populations of syndecan-1-positive cells in Vav^null MZ B cell cultures (Fig. 4⇑A), suggesting that the in vitro PC differentiation defect in Vav^null MZ B cells is not absolute, even though Vav^null mice produce virtually no detectable Ig in their serum. In this regard, Vav^null syndecan-1-positive cells contained intracellular IgM and expressed lower levels of surface IgM (Fig. 4⇑C). Taken together, these data indicate that under conditions of in vitro stimulation with TLR ligands, Vav^null MZ B cells exhibit a profound, albeit leaky, defect in PC differentiation and implicate Vav as a critical factor for B cell conversion into Ig-secreting cells.

To ascertain that the PC differentiation defect of Vav^null MZ B cells was due to a lack of Vav protein expression and not due to developmental defects preceding PC differentiation, we tested whether the reintroduction of recombinant Vav could rescue PC differentiation in these cells. To this end, we infected LPS-stimulated WT and Vav^null MZ B cells with retrovirions encoding a GFP-appended Vav1 or control (GFP only) and analyzed for the induction of syndecan-1. Although a majority of WT cells expressing either GFP alone or GFP-Vav1 expressed syndecan-1, only Vav^null cells transduced with Vav1 expressed syndecan-1 at levels similar to WT (Fig. 4⇑D). Vav^null cells infected with the control vector showed no increase in syndecan-1 expression (Fig. 4⇑D). Thus, although we were unable to obtain sufficient numbers of transduced cells to measure soluble Ig production in these experiments, these data indicate that Vav^null MZ B cells are competent to differentiate into PC upon reintroduction of Vav1 protein, albeit it is likely that all three Vav proteins contribute to achieve optimal B cell function (16, 18, 19).

Vav is an upstream regulator of the Blimp-1-mediated PC developmental program

Having established that Vav proteins are critical for PC differentiation, but not for proliferative expansion of PC progenitors, we hypothesized that Vav may control a specific mechanistic step during the induction of the “plasmacytic program.” In this regard, PC development is dependent on the up-regulation of the transcriptional repressor Blimp-1. Although factors that control its own expression remain unknown at present, Blimp-1 induces the expression of several critical PC-specific genes, including XBP-1, J chain, and secretory Ig μ (4). To determine whether Vav is required for the induction of Blimp-1 or Blimp-1-dependent genes, FACS-sorted WT and Vav^null MZ B cells were cultured in the presence of LPS or CpG and then analyzed by RT-PCR and quantitative RT-PCR for their ability to induce PC-specific genes. Strikingly, these analyses revealed that, when compared with WT, Blimp-1 induction was severely decreased in Vav^null MZ B cells cultured in the presence of LPS or CpG (Fig. 5⇓A), with a greater than 4-fold decrease as assayed by quantitative RT-PCR (Fig. 5⇓B). Furthermore, consistent with Blimp-1 defects in these cells, expression of XBP-1 and J chain were also impaired in Vav^null MZ B cells (Fig. 5⇓A). Because Blimp-1 is required for the differential splicing of H chain transcripts from membrane-bound IgM (μ_m) to secretory IgM (μ_s) (4), we tested for the presence of these transcripts. We found the levels of membrane-bound transcripts were similar, or even slightly higher, in Vav^null cells, whereas transcripts of secretory IgM were lower (Fig. 5⇓A), in agreement with the levels of membrane vs secretory μ Ig H chain protein (Fig. 4⇑C). These data indicate that the Vav family is essential for the inducible expression of Blimp-1 and Blimp-1-regulated effectors. Although BCL-6 has been previously implicated in the negative regulation of Blimp-1 expression (1), we found that the levels of BCL-6 transcripts were similar between WT and Vav^null MZ B cells under both stimulated and unstimulated conditions, suggesting that Vav does not regulate the expression of BCL-6 (Fig. 5⇓A). However, we observed a modest but reproducible increase in expression of Pax-5 in LPS-stimulated Vav^null B cells (Fig. 5⇓, A and B), consistent with a requirement for Blimp-1 in Pax-5 down-regulation (26). We also note that as syndecan-1 expression is thought to be controlled by Blimp-1 (27), the decrease in syndecan-1 observed on Vav^null MZ B cells (Fig. 4⇑A) may be due to a decrease in Blimp-1.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Impaired induction of Blimp-1 and the Blimp-1-mediated transcriptional program in Vav^null MZ B cells. A, Semiquantitative RT-PCR analyses of MZ B cell RNA with 4-fold serial dilutions of cDNA, as indicated. One representative experiment for three performed is shown. B, Quantitative RT-PCR analysis of LPS induction of genes. Data shown are mean values (n = 3 experiments) normalized to GAPDH.

In contrast to Blimp-1 and Blimp-1-regulated genes, the induction of IRF-4 transcripts was indistinguishable between WT and Vav^null MZ B cells. Because IRF-4 transcripts are increased in B cells following LPS stimulation and B cells from IRF-4-deficient mice exhibit profound defects in proliferation, IRF-4 is thought to act early in PC development (1, 2). Thus, induction of IRF-4 in Vav^null MZ B cells is consistent with the ability of these cells to proliferate and suggests that either Vav acts downstream of IRF-4 or that the regulation of IRF-4 is independent of Blimp-1. Together, these data indicate that Vav is an essential upstream mediator of Blimp-1 expression and PC differentiation. Moreover, defective induction of Blimp-1 and Blimp-1-dependent effectors (Fig. 5⇑, A and B), but normal proliferation of Vav^null MZ B cells (Fig. 2⇑, C–E), indicate a dichotomy in pathways emanating from TLRs distinguished by the dependence on Vav.