Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 183 (12), 7661-71

B7-1/2 (CD80/CD86) Direct Signaling to B Cells Enhances IgG Secretion


B7-1/2 (CD80/CD86) Direct Signaling to B Cells Enhances IgG Secretion

Friederike C Rau et al. J Immunol.


B cell responses are regulated by Ag recognition, costimulatory signals provided by interaction with helper T cells, and by innate signals. We recently provided evidence for a link between the effects of innate and costimulatory signals on B cells during influenza virus infection, by demonstrating that most B cells in the regional lymph nodes of the respiratory tract enhance surface expression of the costimulator B7-2 (CD86) within 24-48 h following infection via a type I IFNR-dependent mechanisms, a finding we are confirming here. While the role of B7-1/2 for helper T cell activation is well documented, its role in direct B cell regulation is poorly understood. Here, our in vivo studies with mixed bone marrow irradiation chimeric mice, lacking B7-1/2 only on B cells, demonstrated that B7-1/2 expression is crucial for induction of maximal local, but to a lesser extent systemic, IgG Ab responses following influenza virus infection. In contrast to mice that completely lack B7-1/2 expression, loss of B7-1/2 on B cells alone did not significantly affect germinal center formation or the extent of CD4(+) T cell activation and IFN-gamma secretion. Instead, our in vitro studies identify a dramatic effect of B7-2 engagement on IgG, but not IgM secretion by already class-switched B cells. Concomitantly, B7-2 engagement induced expression of X-box binding protein 1 (XBP-1) and spliced XBP1, evidence for increased protein synthesis by these cells. Taken together, these results identify direct signaling through B7-1/2 as a potent regulator of IgG secretion by previously activated B cells.


Figure 1
Figure 1. Type I IFN and BCR-mediated stimuli act synergistically to induce B-7-1/2
(A) Multicolor flow cytometric analysis for expression of the indicated surface molecules was conducted on B cells from mediastinal lymph nodes and spleen cells of 2-day influenza virus A/PR8-infected BALB/c mice (open histogram, solid line) or gene-targeted B7-1/2−/− mice (dashed line, open histogram). For lymph nodes, a sample was stained with all markers, except the one displayed (solid grey) to serve as a background “fluorescence minus one” control. Overlaid histograms are from live B cells, identified by their FSC/SSC, expression of CD19, exclusion of propidium iodide and lack of staining for CD3, 4, 8, F4/80 and GR-1 (not shown). (B) Similar flow cytometric analysis conducted on MACS-purified splenic B cells of non-infected BALB/c (top), IFNR−/− (middle) and B7-1/2−/− mice (bottom panels) following culture for 16h in the absence (solid light grey) or presence of the following stimuli: (Fab)2 goat-anti-mouse IgM at 20μg/ml (solid line, open histogram); 200U IFN-β (dashed line, open histogram); and anti-IgM plus IFN-β (solid dark grey).
Figure 2
Figure 2. B7-1/2 expression required for maximal virus-specific antibody responses to influenza virus infection
(A) Frequencies of virus-specific Ig-secreting cells were determined by ELISPOT on day 12 after influenza virus infection. Shown are mean counts ± SD from 2-fold titrated duplicate cultures of MedLN and triplicate cultures of spleen cells from pooled (n=6) BALB/c (B7+) mice (filled bars) and B7-1/2 / mice (B7, open bars), respectively. Virus-specific IgG1 and IgA secreting spots were not detected (n.d.) in spleen of B7 mice. Results are a representative from two independent experiments. (B) Virus-specific Ig serum response was measured in BALB/c (B7+) (filled symbols) and B7-1/2/ mice (B7, open symbols) by ELISA at indicated time points after influenza virus infection. Data represent mean ± SD concentrations (μg/ml) of virus-specific IgG from 6 individual mice per group. *p<0.05; **p< 0.01; ***p<0.001.
Figure 3
Figure 3. B7-1/2 expression by B cells affects virus-specific antibody responses
(A) Virus-specific antibody production was assessed in mice that lacked B7-1/2 expression only on B cells (open bars) and compared to controls with B7-expressing wildtype B cells (filled bars). Mice were mixed bone marrow irradiation chimeras generated by reconstituting lethally irradiated wildtype (WT) BALB/c mice with a mix (75%/25%) of bone marrow from B cell deficient mice and either B7-1/2 / or WT BALB/c mice. MedLN and spleens from mice (n = 6/group) were pooled and analyzed by ELISPOT 10 days after infection with influenza A/PR8. Shown are mean frequencies of antibody-secreting foci ± SD calculated from two-fold titrated duplicate cultures of MedLN and triplicate cultures of spleen cells. The data are a representative of two independent experiments yielding comparable results. *p<0.05, **p< 0.01, ***p<0.001. Reduction in IgM production by B cell B71/2/ mice in MedLN in (A) was only observed in one of two experiments. (B) Virus-specific serum total IgG, IgM as well as IgG1 and IgG2a levels in chimeras with wildtype (filled symbols) and B7-1/2−/− B cells at indicated times after infection with influenza A/PR8 as measured by ELISA. Each symbol represents data from one animal. The horizontal lines represent the geometric mean for each group. *p < 0.05.
Figure 4
Figure 4. Germinal center formation is not dependent on B7-1/2 expression by B cells
(A) Shown are representative contour plots (5%) with outliers from FACS analyses of MedLN from wildtype (wt) and B7-1/2/ mice (upper panels) and of chimeras lacking B cell-expression of B7-1/2 and their controls (lower panels) at days 10 (top) and 12 (bottom) after influenza infection, respectively. Germinal center B cells (GC) were identified as CD24hi CD38lo B cells as shown, following gating on live B cells (CD3, CD4, CD8, F4/80- propidium iodide negative and CD19+ B220high). Numbers indicate frequencies among CD19+ B220hi B cells. Shown are representative samples from the respective groups, consisting of 4–8 mice each. (B) Summary of the results from the FACS analysis for individual mice analyzed as shown in (A). The horizontal lines represent the geometric mean for each group. ***p<0.001; n.s., not significant.
Figure 5
Figure 5. Lack of correlation between B7-2 expression levels on B cells and germinal center B cell frequencies in MedLN
(A) Shown are frequencies of B7-2 expressing peripheral blood B cells of individual mixed bone marrow irradiation chimeras following overnight in vitro stimulation of PBMC with anti-IgM (Fab)2 (open boxes). B7-1/2−/− (open circles) and BALB/c mice (filled circles) were used as negative and positive controls, respectively. Chimeras were generated by reconstituting lethally irradiated BALB/c mice (650rd whole body irradiation) with a mix (75%/25%) of bone marrow from B cell-deficient mice and B7-1/2−/− mice. In the chimeras, B7-2 expressing B cells are presumably radio-resistant and host-derived. (B) To determine whether a correlation exists between expression of B7-2 on B cells and germinal center formation, two groups of chimeras (n= 4) differing in the frequencies of peripheral B7-2 B cells were infected for 10 days and frequencies of germinal center B cells were determined by FACS. Chimeras were stratified in groups according to their frequencies of B cells expressing B7-2 (4–8 % open triangles; 30–32%, filled triangles) Note the lack of correlation between frequencies of B cells expressing B7-2 and germinal center B cell frequencies. (C) Levels of B7-2 expression (MFI) on CD19 + MedLN B cells (left panel) and on MedLN germinal center B cells (right panel) of individual mice were correlated with germinal center B cell frequencies. No significant correlation was observed (p > 0.05). (D) Chimeras generated with B cells from either B7+ or B7− mice as in A, but using 800rd whole body irradiation were analyzed for expression of B7-2 after in vitro stimulation of PBMC as in A (n = 4 per group, left panel). Data are mean values ± SD. Mice were infected for 11 days with influenza A/PR8 and frequencies germinal center B cells in regional lymph nodes assessed by flow cytometry (middle panel). ELISA on sera taken from mice at indicated time points were analyzed for virus-specific IgG levels. Shown are mean concentrations (μg/ml) ± SD.
Figure 6
Figure 6. Reduced CD4 T cell activation in B7-1/2−/− mice is due to expression of B7-1/2 on cells other than B cells
Shown are representative FACS contour plots (5%) with outliers to demonstrate the gating strategies used to determine the frequencies of (A) total and (B) activated (CD11a hi/CD44hi) CD4+ T cells in MedLN. Results are from chimeras lacking B7-1/2 on B cells only and their controls (far right panels) infected for 10 days with influenza virus, and BALB/c (B7+) and B7-1/2−/− mice (B7−, middle panels), infected for 12 days with influenza A/PR8 virus. Each symbol represents the result from an individual mouse. The horizontal lines represent the geometric mean for each group. Data represent results from one of at least three independent experiments performed. *** p<0.001. (C) Shown is the gating strategy used to determine frequencies of IFN-γ secreting CD4+ T cells in MedLN 7 days following influenza virus infection. Frequencies of IFN-γ producers of live T cells (CD3+, CD4+ or CD8+, CD11b, CD19 and excluding propidium iodide) were determined by intracytoplasmic staining following overnight re-stimulation with plate-bound anti-CD3 (clone 145-2C11) in the presence of monensin. Results from individual mice lacking B7-1/2 only on B cells (B cell B7; n = 4) and controls (B cell B7+; n = 4) are summarized in the right panels. The horizontal lines represent the geometric mean for each group. **p<0.01. Results are from one of two experiments that yielded similar results.
Figure 7
Figure 7. B7-2 direct stimulation does not affect B cell proliferation
Shown are results from a proliferation assay of MACS-enriched splenic B cells stimulated with/or without a combination of the following: IFN-b (200U/ml), anti-B7-2 (10 μg/ml unless otherwise noted), anti-IgM(Fab)2 (20 μg/ml). At indicated times after culture onset cells were analyzed by MTT assay for relative numbers of cells in each well (measured as adsorbance at 590–650 reference wavelengths). Stimulation of B cells with IFN-b for 16h followed by washing of the cells and then stimulation with anti-B7-2 gave similar results (data not shown). Results are from one of three experiments done that gave similar results.
Figure 8
Figure 8. B7-2 stimulation induces IgG production by class-switched B cells
(A) MACS-enriched splenic B cells were cultured in duplicate for indicated times in the presence of CD40L (0.I mg/ml), IFN-g (15ng/ml) and IL-5 (2ng/ml) with/without anti-B7-2 (10 μg/ml). Supernatants were harvested and IgG2a protein levels measured by ELISA. Horizontal lines indicate mean of 2 cultures. IgG2a levels in cultures without anti-B7-2 were below the threshold of detection (0.0165 ng/ml) at all time points analyzed. (B) Cells from similar cultures as in (A), set-up in triplicate were harvested at indicated timepoints after culture onset and RNA was extracted for qRT-PCR analysis of germline IgG2a sterile transcripts and AID expression. No significant differences in gene expression were noted between cultures in the presence/absence of anti-B7-2 at any time point. Similar results were obtained when germline IgG1 transcripts were measured in cultures containing IL-4 instead of IFN-g (data not shown). (C) Total splenic B cells and splenic B cells in which naïve IgD+ cells were removed by MACS-depletion were cultured in quadruplicate in the presence of IL-4 and IL-5 with/without anti-B7-2. IgG1 and IgG2a protein levels in supernatants were measured by ELISA 24h after culture onset. (D) Shown are relative expression levels of XBP-1 and sXBP-1 from qRT-PCR analysis of RNA extracted from B cells cultured for 24h as outlined in (A). Results are representative of 2–5 experiments conducted for each analysis.

Similar articles

See all similar articles

Cited by 40 PubMed Central articles

See all "Cited by" articles

Publication types

MeSH terms