SEARCH

SEARCH BY CITATION

Keywords:

  • Adjuvants;
  • B cells;
  • Cellular proliferation;
  • CpG ODN

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

Mechanisms that regulate naïve B cell proliferation and function are incompletely defined. In this study, we test the hypothesis that naïve B cell expansion, survival and ability to present antigen to T lymphocytes can be directly modulated by Toll-like receptor (TLR) agonists. In the absence of B cell receptor stimulation, CpG oligonucleotide, a TLR9 agonist, was particularly efficient in inducing naïve B cell proliferation and survival. Although the expanded naïve B cells did not mature into CD27+ or IgG+ memory B cells, these cells did differentiate into IgM-secreting cells with increased surface expression of HLA-DR, CD40 and CD80. This was associated with an increased potential for these B cells to activate allogeneic T cells. We propose that the activation and expansion of naïve B cells induced by TLR9 agonists could enhance the potential of these cells to interact with cognate antigens and facilitate cell-mediated immune responses.

Abbreviations:
CpG ODN:

CpG oligonucleotide

pDC:

plasmacytoid dendritic cell

PGN:

peptidoglycan

ssPolyU:

single-stranded polyuridine

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

Unmethylated CpG oligodeoxynucleotides (CpG ODN) enhance both innate and adaptive immunity and, consequently, have potential utilities for immunotherapeutic and vaccine adjuvant applications. These molecules mediate their effects by interacting with Toll-like receptor (TLR) 9 that is expressed on professional antigen-presenting cells (APC) 1. In humans, the professional APC that express TLR9 include plasmacytoid dendritic cells (pDC) and B lymphocytes. Activation of TLR9 on pDC may lead to maturation of these cells as well as production of type I interferon 1. Activation of memory B cells by TLR9 can lead to cellular expansion and immunoglobulin production, and this may play an important role in memory B cell homeostasis and sustained antibody production 2. The consequences of TLR9 activation in naïve B cells are less well characterized and somewhat controversial.

Previous studies have provided evidence that human adult naïve B cells fail to proliferate after exposure to CpG ODN, but these same cells are able to expand when CpG ODN exposure is coupled with B cell receptor stimulation 2. These data are compatible with the understanding that B cell receptor signals along with secondary signals are required for induction of naïve B cell division, and this serves as a checkpoint for regulating the expansion of naïve B cells in response only to cognate antigen 25. In contrast, other studies provide evidence that naïve B cells might proliferate after treatment with CpG ODN alone; however, these additional investigations relied on methodologies that did not utilize single-cell analyses to exclude the potential involvement of contaminating memory B cells. Moreover, some of these additional studies 6 relied on naïve B cells from cord blood that may be functionally different from adult B cells and include a significant fraction of transitional cells 5. In light of this controversy, we sought to examine the effects of CpG ODN on naïve B cell expansion using cells obtained from healthy adults while relying on more rigorous analytical methods.

The consequences of naïve B cell activation by CpG ODN may have additional implications besides influencing cellular proliferation. In particular, B cell activation could result in enhanced antigen presentation function stemming from increased surface expression of costimulatory or MHC molecules. To explore this possibility, we investigated the expression of costimulatory and MHC molecules on the surface of naïve B cells after incubation with CpG ODN 2006 and evaluated the capacity of these cells to induce allogeneic T cell responses. Our results provide evidence that CpG ODN can activate naïve B cells directly, resulting in cellular expansion as well as an increased potential of these cells to present antigens to T cells. These results have important implications for a role of TLR9 agonists in the enhancement of humoral and cell-mediated immunity through activation of naïve B cells.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

CpG ODN 2006 is a potent stimulator of naïve B cell proliferation

To explore the possible role of TLR ligation in naïve B cell homeostasis, we tested selected TLR agonists for their ability to promote naïve B cell expansion in vitro. Peripheral blood mononuclear cells were incubated with TLR agonists and the proliferation of naïve B cells (CD19+, CD27) was assessed by monitoring dilution of carboxyfluoroscein succinimidyl ester (CFSE). The TLR targeted by these agonists included TLR2 (peptidoglycan; PGN), TLR3 (polyinositolic polycytidylic acid; polyIC), TLR4 (lipopolysaccharide; LPS), TLR5 (flagellin), TLR8 (single-stranded polyuridine; ssPolyU) and TLR9 (CpG ODN 2006). Concentrations ranging from 0.1 to 50 µg/mL were tested and optimal concentrations of each reagent were used to generate the results depicted in Fig. 1A. In these experiments, CpG ODN 2006 induced dramatic proliferation of naïve B cells, polyIC induced intermediate proliferation responses, and the other TLR agonists induced weak proliferation responses that were nevertheless still distinguishable from the expansion seen in medium alone (Fig. 1A). These results indicated that CpG ODN alone provides sufficient stimulus to induce naïve B cell proliferation within human PBMC preparations, and suggest that TLR9 may be particularly effective relative to other TLR agonists in mediating this effect. In the more detailed studies described below, we focused on the mechanisms of the more potent TLR9-mediated naïve B cell expansion.

thumbnail image

Figure 1. Naïve B cells (CD27CD19+) proliferate in response to CpG ODN. PBMC were labeled with CFSE and cultured with or without different TLR ligands: 2 µg/mL PGN, 20 ng/mL LPS, 50 µg/mL polyIC, 0.1 µg/mL flagellin, 10 µg/mL ssPolyU and 6 µg/mL CpG ODN 2006 for 7 days. Naïve B cells were gated as CD27CD19+ and their proliferation is reflected by dilution of CFSE dye (CFSE low). These results are representative of four experiments performed using cells obtained from different donors. (B) Purified naïve B cells were obtained by negative magnetic bead selection, labeled with CFSE and incubated with or without CpG ODN 2006, CpG ODN 2395, CpG ODN 2216 or non-CpG ODN 2137 for 5 days. The numbers represent the percentages of proliferating cells (low CFSE staining). Data are representative of experiments using three different donors. (C) Cumulative data from 15 experiments, showing the median and interquartile ranges of the percentage of CFSE-low cells (y-axis) among purified CD19+CD27 naïve B cells that had been incubated with or without CpG ODN 2006 for 5 days. The p-value was determined with the Wilcoxon Signed Rank Test.

Download figure to PowerPoint

Purified naïve B cells proliferate in direct response to CpG ODN

To ascertain if the effect of CpG ODN on naïve B cell proliferation was mediated directly on the B cells or through the intervening effects on other TLR9-expressing cells within the PBMC preparation, we purified naïve B cells by negative selection, resulting in populations that were >97% CD19+ and CD27. These purified naïve B cell populations were incubated with CpG ODN 2216 (type A), CpG ODN 2006 (type B), CpG ODN 2395 (type C), or a control non-CpG ODN 2137. Naïve B cell proliferation was induced by each of the ODN tested, with type B being the most effective, followed in order by type C, A and non-CpG ODN (Fig. 1B). The modest induction of B cell activation by non-CpG ODN is consistent with results from other studies, which demonstrated low levels of B cell activation by non-CpG ODN through a TLR9-dependent mechanism 7. Based on these observations, we focused our additional experiments on type B CpG ODN and we found that although there was a wide range of responsiveness among different subjects, the induction of naïve B cell proliferation by this ODN was highly reproducible (Fig. 1C).

Importantly, naïve B cells from three samples achieved a purity ⩾99.6% by magnetic bead selection and they proliferated in response to CpG ODN stimulation (median % CFSE-low cells = 21). Two additional samples were purified to ⩾99.7% by a two-step process involving magnetic bead selection followed by flow sorting of CD20+CD27IgD+CD10 cells. Cells from these samples also proliferated in response to CpG ODN (11 and 18.9% CFSE-low cells) but did not divide in medium alone (2 and 2.2% CFSE-low cells).

Besides generating highly purified samples, we performed additional studies to ascertain if the proliferation responses were somehow dependent on other cell types, for example, transitional B cells or pDC. To rule out the involvement of transitional B cells, which are recent emigrants from the bone marrow that express a variety of phenotypic markers (CD24, CD38, CD5, CD10, CD21 and CD23) distinct from naïve B cells 5, 8, we purified CD19+CD27 B cell preparations and additionally depleted this population of transitional cells by removing CD10+ cells (depletion purity >99%; Fig. 2A). This CD10-depleted naïve B cell population was capable of expansion in vitro in response to CpG ODN 2006 (Fig. 2B). A similar result was obtained in experiments using Ki-67 induction as a measure of cell cycle entry (data not shown), confirming that the B cell population responding to TLR9 stimulation did not require the presence of phenotypically defined transitional B cells.

thumbnail image

Figure 2. Naïve B cells proliferate in response to CpG ODN in the absence of transitional CD10+ B cells. (A) Purified naïve B cells were depleted of CD10+ cells by magnetic bead depletion. Histograms show CD10 expression by purified naïve B cells before and after depletion of CD10+ cells. (B) CD10-depleted naïve B cells were stimulated with CpG ODN 2006 for 5 days and CFSE dye dilution was determined by flow cytometric analysis. These results are representative of four experiments using cells obtained from different donors.

Download figure to PowerPoint

Because pDC can be activated by CpG ODN to mediate bystander activation of other cell types such as monocytes 9 and can enhance proliferation of naïve B cells in response to BCR stimulation 10 even when present in very low frequency, we added purified pDC to cultures containing naïve B cells. The addition of pDC to purified naïve B cells at a ratio of 1 : 10, 1 : 20 or 1 : 50 did not enhance the proliferation response of naïve B cells to CpG ODN 2006 (Fig. 3A and data not shown). Thus, purified naïve B cells can be induced directly to proliferate in response to CpG ODN, independently of other cell types, and this expansion is not enhanced by the presence of pDC.

thumbnail image

Figure 3. Purified naïve B cells proliferate independently of pDC and do not mature into memory or isotype-switched memory B cells in response to CpG ODN 2006. (A) Purified naïve B cells were labeled with CFSE and incubated in the presence or absence of purified, autologous pDC (pDC/B cells = 1 : 50). Cells were treated with CpG ODN 2006 for 5 days. CFSE dye dilution is indicated on the x-axis and CD27 expression or IgD expression is shown on the y-axis. Data are representative of experiments using four different donors. (B) Purified naïve B cells were cultured with or without CpG ODN 2006 for 8 days. Cell culture supernatants were harvested and kept frozen until analyzed by ELISA for determination of IgM and IgG concentrations. (C) PBMC from two donors were depleted of CD27+ cells by magnetic bead depletion. Cells were incubated in medium alone or with CpG ODN 2006 or IL-4 (25 ng/mL). Quadrants were set based on the expression of IgM and CD80 among cells incubated in medium alone. Note that these cells were essentially all IgM+ based on isotype staining (not shown).

Download figure to PowerPoint

Naïve B cells do not mature into memory cells in response to CpG ODN

To ascertain if CpG ODN-induced naïve B cell expansion was accompanied by evidence of B cell maturation into memory B cells or plasma cells, we examined B cell surface expression of CD27, CD38, CD20, IgD, IgM and IgG as well as intracellular expression of IgM and IgG after stimulation with CpG ODN for 5–14 days. Purified naïve B cells did not increase surface expression of CD27 after culture for 5 days with CpG ODN (Fig. 3A and data not shown). Likewise >98% of purified naïve B cells expressed surface IgD after 5 days of incubation with CpG ODN (Fig. 3A and data not shown). The CD27IgD+ cell surface phenotype was maintained even among naïve B cells that had undergone cellular division (Fig. 3A). Consistent with previous studies 6 of cord blood B cells, CpG ODN increased IgM production that could be detected in cell culture supernatants (Fig. 3B). We utilized intracellular flow cytometry to investigate the induction of IgM expression on a single-cell basis. PBMC from two donors were depleted of CD27+ cells and incubated with CpG ODN, IL-4 or medium alone for 3 days. The expression of CD80 and total (cell surface and intracellular) IgM was examined on CD20+CD27 cells. Among cells incubated with CpG ODN, a subset of naïve B cells expressed total IgM above the background level observed in cells incubated in medium alone (Fig. 3C). The cells with increased total IgM expression primarily corresponded to cells that had also been induced to express CD80. Similar results were obtained by examining cells on day 7 (not shown). These results suggest that a subset of activated naïve B cells increases production of IgM in response to CpG ODN. Interestingly, IL-4 did not induce CD80 expression but did increase total IgM expression in the majority of naïve B cells (Fig. 3C), highlighting the distinguishable consequences of naïve B cell activation with these two stimuli.

Unlike IgM production, there was little evidence of IgG production in either supernatant (Fig. 3B) or within the activated B cells (median percentage of intracellular IgG+ cells = 0.4 and 0.5% for cells incubated in medium alone and cells incubated with CpG ODN, respectively; n = 6). The latter was determined by flow cytometric analysis 7–8 days after stimulation. Finally, we found little evidence for cellular differentiation into phenotypically defined, CD38+CD20, plasma cells after culture of naïve B cells with CpG ODN for 8–14 days (data not shown). These results indicate that naïve B cells are activated to proliferate and to produce IgM by CpG ODN but are not induced to undergo isotype switching or to acquire a memory or plasma cell surface phenotype. The absence of memory cell markers among naïve B cells that expanded in response to CpG ODN provides further evidence that expansion is not the consequence of contaminating memory cells in these cultures.

CpG ODN 2006 and IL-4 rescue naïve B cells from apoptosis in vitro

To characterize naïve B cell survival after stimulation with CpG ODN, we examined annexin V staining among purified naïve B cells that were incubated in the absence or presence of CpG ODN 2006 or, for comparison, with a cytokine that is known to enhance B cell viability, IL-4 11. Our results showed that CpG ODN or IL-4 enhanced naïve B cell survival, as reflected by fewer cells binding annexin V after 1, 3 and 5 days of incubation (Fig. 4 and data not shown). The anti-apoptotic effects of CpG ODN or IL-4 diminished with prolonged culture (data not shown), resulting in almost complete loss of cell viability by 14–21 days of incubation.

thumbnail image

Figure 4. CpG ODN 2006 and IL-4 each can rescue naïve B cells from apoptosis. Purified naïve B cells were cultured with CpG ODN 2006 (6 µg/mL) or IL-4 (25 ng/mL). (A) Annexin V staining was measured at day 3. (B) The median percentage of annexin+ naïve B cells was compared among naïve B cells cultured with medium, CpG ODN or IL-4. The * represents an outlier. p = 0.04, Wilcoxon Signed Rank Test, n = 5.

Download figure to PowerPoint

CpG ODN induces CD40, CD80 and HLA-DR expression on naïve B cells

We hypothesized that the increased survival and expansion of naïve B cells after activation by CpG ODN may enhance the potential of these cells to acquire antigen and interact with T helper cells. To assess the effects of CpG ODN on the antigen-presenting function of naïve B cells, we first evaluated the effects of CpG ODN on costimulatory molecule (CD40 and CD80) and HLA class II surface expression. Purified naïve B cells were cultured with or without CpG ODN 2006 or IL-4 for 48 h and then examined for expression of CD80, CD40 and MHC class II. Incubation of naïve B cells with CpG ODN resulted in increased frequencies of CD80+ cells and increased surface density of HLA class II and CD40 molecules (Fig. 5). IL-4 had little effect on CD80 expression and mediated less induction of HLA class II and CD40 expression than did CpG ODN. Since IL-4 increases the viability of B cells as efficiently as CpG ODN (Fig. 4), it is unlikely that the effects of CpG ODN are simply related to selective survival of CD80+HLA DRbrightCD40bright cells during the 48-h incubation.

thumbnail image

Figure 5. CpG ODN 2006 enhances MHC class II, CD80 and CD40 expression among naïve B cells. (A) Representative flow histograms are shown. Filled histograms represent phenotypic expression in naïve B cells cultured for 2 days without stimuli (med), solid lines represent phenotypic changes in naïve B cells cultured with CpG ODN 2006 or IL-4 as indicated. (B) Median and interquartile ranges of the percentages of CD80+ cells, CD40 MFI and MHC class II MFI are shown. **p <0.001, *p <0.05. Wilcoxon Signed Rank Test, n = 8.

Download figure to PowerPoint

CpG ODN enhances the ability of naïve B cells to activate allogeneic T cells

To evaluate the ability of CpG ODN-treated naïve B cells to induce T cell responses, naïve B cells were incubated with CpG ODN, IL-4 or medium alone for 48 h. These cells were irradiated, counted and mixed with allogeneic, CFSE-labeled T cells such that equal numbers of live B cells were added to each culture. The dose of irradiation used in these experiments prevented survival and expansion of these B cells, even among cells pretreated with CpG ODN or IL-4 (data not shown). Therefore, irradiation of the B cells ensured that the ability of these cells to induce allogeneic T cell responses was a consequence of their antigen-presenting capability and not simply a reflection of increased antigen burden due to prolonged survival or expansion of antigen-expressing cells. Incubation of naïve B cells with CpG ODN resulted in a striking increase in the ability of these cells to induce allogeneic T cell proliferation responses. Both CD4+ and CD8+ T cells expanded dramatically following stimulation with CpG ODN-treated naïve B cells, whereas relatively modest T cell expansion was achieved when alloantigens were presented by untreated or IL-4-treated naïve B cells (Fig. 6).

thumbnail image

Figure 6. Induction of allogeneic CD4+ and CD8+ T cell responses by CpG ODN-treated naïve B cells. (A) Individual histograms show CD4 and CD8 T cell proliferation responses to stimulation by naïve B cells activated with CpG ODN 2006 or IL-4 (naïve B cells: T cells = 1 : 10). These results are representative of eight experiments using cells obtained from different donors. (B) Mean percentage of CD4 and CD8 T cells that proliferated in response to B cell stimulation shown as a function of the ratio of naïve stimulator B cells to responder T cells, n = 5. Error bars represent standard error of the means (SEM).

Download figure to PowerPoint

We next examined the relationships among CD40, CD80 and HLA-DR expression on naïve B cells after CpG ODN exposure and the subsequent induction of allogeneic T cell proliferation (% CFSE-low cells) by these cells. We found a direct correlation between the expression of CD80 on naïve B cells and the induction of allogeneic CD4+ T cell proliferation responses (Spearman's correlation test, R2 = 0.37, p = 0.04), suggesting that the induction of costimulatory molecules on naïve B cells by CpG ODN exposure enhances the potential of these cells to activate antigen-reactive T lymphocytes.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

Previous studies have demonstrated that memory B cells can be induced to proliferate by TLR9 agonists independently of B cell receptor activation 2. This phenomenon may promote memory B cell homeostasis, accounting for the low levels of memory B cell proliferation, survival and persistent antibody production in the absence of antigen exposure 2. In contrast to recognized effects on memory B cells, the ability of TLR9 agonists to induce the proliferation of naïve B cells is controversial 2, 5, 10, 12, 13. Recent studies suggest that CpG ODN plus cytokines such as IL-2, IL-10 or IL-15 can result in naïve B cell expansion 12. Proliferation responses under these conditions appear to be dependent on cell-to-cell contact and consequently require an adequate density of naïve B cells in culture (1 × 105 cells/200 µL). Our experiments provide evidence that even in the absence of exogenous cytokines, CpG ODN provides sufficient signals for induction of naïve B cell proliferation at a cell density of 2 × 105 cells/200 µL. Moreover, we found that the induction of naïve B cell proliferation by CpG ODN was independent of pDC since the levels of contamination of pDC in our cell preparations was essentially undetectable and the addition of pDC to naïve B cell cultures provided no enhancement of naïve B cell proliferation responses.

Previous studies also suggested that adult naïve B cells can proliferate in direct response to CpG ODN; however, there remains uncertainty over these results due to the absence of phenotypic confirmation of the expanding cell type 13, 14. Our studies relied on single-cell analyses and highly purified cell populations to demonstrate that naïve B cells are readily induced to proliferate in response to the TLR9 agonists.

Also related to our findings, cord blood B cells are able to proliferate in response to CpG ODN alone 6, 14. Although cord blood B cells are enriched for naïve cells, a substantial fraction of these cells have a transitional B cell phenotype, making it unclear which cells are actually expanding in response to CpG ODN 5. Moreover, there may be intrinsic differences in neonatal B cells that are not reflective of adult naïve B cell populations. Our data clearly indicate that unmanipulated adult naïve B lymphocytes can be induced to expand by TLR9 ligation either within a heterogeneous PBMC preparation or when purified to near homogeneity.

Contrasting with our own observations, other groups have found that CpG ODN alone was insufficient for induction of naïve B cell proliferation 2, 15. The contrasting findings between our results and these previous reports may reflect the differences in technical approaches, including cell culture density and methods of B cell isolation. Whereas previous studies 2, 15 relied on positive selection of CD19+ cells to isolate B cell populations, we used negative selection techniques, avoiding any effects of CD19 ligation on these cells.

Cellular proliferation often can be accompanied by cellular differentiation. Therefore, it was important to ascertain if TLR-induced naïve B cell proliferation resulted in differentiation of these cells. Although the naïve B cells did not mature into CD27+ or IgG+ memory B cells after expansion in response to CpG ODN, these cells did undergo some extent of differentiation since they produced IgM and increased surface expression of MHC and costimulatory molecules. Thus, although CpG ODN can facilitate the differentiation of naïve B cells into plasma cells when accompanied by additional activation signals such as those provided by BCR stimulation, CD40/CD40L interactions or combinations of cytokines 10, 16, CpG ODN alone do not appear to mediate class switching or the generation of memory B cells or plasma cells.

Naïve B cells are thought to be short-lived cells, replaced readily by continuous production of new cells in the bone marrow 17. What then is the relevance of naïve B cell expansion in response to TLR9 stimulation? One possibility is that microbial products could have a role in naïve B cell homeostatic division. We think this is not likely since exposure to microbial products may not be required for maintaining the composition of the peripheral B cell pool, as this compartment is similar in mice raised under conventional and germ-free breeding conditions 18. Also, as indicated in rodent models, the rate of new B cell production in the bone marrow is sufficient to fully populate the peripheral B cell pool in a matter of days 17, making prolonged survival and expansion of existing naïve B cells unnecessary and potentially problematic for homeostasis. Moreover, recent studies using deuterium-labeled glucose have indicated that the in vivo turnover of CD27 B cells is relatively low when compared to that of memory B cells 19 and that naïve B cells defined as CD27 and expressing the ABC-B1 transporter rarely express the nuclear antigen Ki-67 that identifies cells in cycle 5. It is possible that the failure to detect evidence of naïve B cell proliferation may be due to the fact that the analyses were performed using peripheral blood. Conceivably, naïve B cells in gut-associated lymphoid tissues could be exposed to microbial products and induced to proliferate in response. Thus far, however, there is little evidence that naïve B cells are undergoing extensive division under normal homeostatic conditions. Therefore, the importance of peripheral homeostatic expansion of naïve B cells either induced by TLR ligation or by other mechanisms in maintaining naïve B cell survival might be questioned.

An alternative hypothesis for the physiological importance of TLR-mediated naïve B cell proliferation stems from its potential role during acute microbial challenge. We propose that TLR agonists may enhance transiently the survival and proliferation of peripheral naïve B cells during substantial microbial challenges, potentially giving these cells a better opportunity to encounter antigen during this period. If so, we reasoned that these B cells also may play a more substantial role in presenting antigens to T lymphocytes. Therefore, we examined the effects of CpG ODN on naïve B cell costimulatory molecule expression and antigen-presenting function.

We found that CpG ODN markedly enhanced expression of CD80, CD40 and HLA-DR on naïve B lymphocytes and resulted in a greater ability of these cells to activate allogeneic CD4 and CD8+ T cells. The poor immunogenic potential of resting naïve B cells in our assays is consistent with a tolerogenic role for these cells under normal physiologic conditions 20. Activation of these cells by TLR9 21 or other agonists 22, 23 promotes the immunogenicity of antigens presented by these cells and consequently may play an important role in the adjuvant effects of CpG ODN in vivo. An additional speculation stemming from this observation is that prolonged exposure to high concentrations of TLR9 agonists could predispose individuals to adverse autoimmune reactivity due to enhanced presentation of self antigens by activated naïve B cells. Further study of the role of TLR agonists in normal and abnormal naïve B cell biology will help to explore these possibilities.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

Reagents

Endotoxin-free CpG ODN 2006 (TCgTCgTTTTgTCgTTTTgTCgTT) was provided by Coley Pharmaceutical Group Inc. (Wellesley, MA). Titration experiments using a range of concentrations from 1–18 µg/mL indicated that 6 µg/mL provided an optimal effect.

Recombinant human IL-2 and IL-4 were obtained from R&D Systems (Minneapolis, MN). PGN from Bacillus subtilis, LPS from E. coli, polyIC, flagellin from B. subtilis, and single-stranded polyU/LyoVec complexes (ssPolyU) were purchased from InvivoGen (San Diego, CA).

Cell isolation

For preparation of highly purified, unmanipulated naïve B cells, the naïve B cell isolation kit II (Miltenyi Biotec, Bergisch Gladbach, Germany) was used. The negatively selected naïve B cells consisted of 97–99.8% CD19+CD27 cells and the viability was >95%; pDC contamination was measured with each experiment and always comprised fewer than 0.05% of purified naïve B cells.

Some studies involved further purification of naïve B cells by depletion of CD10+ transitional B cells. Negatively selected CD19+CD27 naïve B cells were incubated with anti-CD10-phycoerythrin (PE)-conjugated antibody (BD Pharmingen) followed by a secondary incubation with anti-PE microbeads (Miltenyi Biotec). Cells were sorted on an autoMACS cell sorter (Miltenyi Biotec) to purities >99%.

pDC were isolated from PBMC by using BDCA-4 microbeads (Miltenyi Biotec). Purified pDC were positively selected and these isolated pDC consisted of >80% lineage-negative, CD123+HLADR+ cells as determined by flow cytometry.

B lymphocyte proliferation

To evaluate cell division, whole PBMC or purified naïve B cells were labeled with CFSE. PBMC were incubated in 0.25 µM CFSE at 37°C for 10 min and washed with PBS supplemented with 10% FCS. CFSE-labeled PBMC (2 × 106/mL) were cultured in 24-well plates (BD labware, Franklin Lakes, NJ) with or without CpG ODN (6 µg/mL) for 7 days. CFSE-labeled purified naïve B cells (5 × 105 cells/well, 500 µL per well) were cultured in 48-well plates (BD labware) with or without CpG ODN for 5 days. Proliferation was estimated by measuring the percentage of cells that had diluted tracking dye.

Annexin V binding

Cell apoptosis was measured by the binding of annexin V-PE (BD Pharmigen). Purified naïve B cells were plated in 48-well plates (5 × 105 cells/well) and cultured in the absence or presence of CpG ODN or, for comparison, cells were cultured with IL-4 (25 ng/mL) for 1, 3 and 5 days. After the indicated times, cells were harvested, washed and stained with annexin V-PE according to the manufacturer's instructions. Samples were acquired on a FACSCalibur flow cytometer (Becton Dickenson, San Jose, CA) and analyzed with Cell Quest software (Becton Dickinson).

Allogeneic proliferation response

To evaluate allogeneic T cell responses, purified naïve B cells were cultured with or without CpG ODN 2006 or IL-4 (25 ng/mL) for 2 days. These cells were then irradiated (2000 rad), counted and co-cultured with allogeneic CFSE-labeled T cells. The T cells that were used as responder cells in these assays were isolated from one unrelated healthy donor by magnetic bead negative selection (Pan T cell isolation kit II; Miltenyi Biotec). These cells, which were >98% CD3+, were labeled with CFSE and cryopreserved at –80°C until thawed for allogeneic response assays. Thawed T cells were consistently >90% viable. The B cell/T cell ratios that were tested included 1 : 2, 1 : 20 and 1 : 50 with a total of 2 × 105 total cells/well (200 µL per well) in 96-well round-bottom plates (BD labware, Franklin Lakes, NJ). Proliferation was measured as the percentage of CD4+ or CD8+ T cells that diluted tracking dye. pDC contamination was measured by flow cytometric analyses and was consistently below 0.05% within the purified naïve B cells and pan T cells.

Flow cytometry

Cells were washed and stained with the following monoclonal antibodies: anti-IgD-biotin and secondary streptavidin-peridinin chlorophyll protein (PerCP), anti-CD19-allophycocyanin (APC), anti-CD27-APC, anti-CD20-PerCP, anti-IgM-APC, anti-IgG-PE, anti-CD38-APC, anti-CD4-APC, anti-CD8-PerCP, anti-Lin-FITC, anti-HLA-DR-PerCP, anti-CD123-PE, anti-BDCA2-PE, anti-CD123-biotin, streptavidin-APC and appropriate isotype control mAb (BD PharMingen, San Jose, CA). Flow cytometric analyses were performed using a FACSCalibur flow cytometer.

Immunoglobulin production and measurement

Purified naïve B cells were stimulated with or without CpG ODN 2006 in complete medium at 2 × 105 cells/well (200 µL per well) in 96-well round-bottom plates. Immunoglobulins in cell culture supernatants were measured after 8 days by IgM and IgG ELISA (Bethyl Laboratories, Montgomery, TX).

Statistical methods

Categorical variables were compared by the Wilcoxon Rank Test or one-way ANOVA. The Spearman correlation coefficient was used to determine the correlation between two continuous variables. Statistical significance was defined as p <0.05.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and methods
  7. Acknowledgements

We thank Arthur M. Krieg for helpful advice and the Coley Pharmaceutical Group for providing the CpG ODN. We also appreciate the assistance of Robert Asaad in the collection of blood samples. This work was supported in part by NIH grants: AI55793 (C.V.H., M.M.L. and S.F.S.), AI34343 (C.V.H.), AI38858 (M.M.L. and S.F.S.), and the Center for AIDS Research at Case Western Reserve University, University Hospitals of Cleveland: AI36219 (M.M.L. and S.F.S.).

  • 1

    WILEY-VCH

  • 2

    WILEY-VCH

  • 3

    WILEY-VCH

  • 4

    WILEY-VCH

  • 5

    WILEY-VCH

  • 6

    WILEY-VCH