IgE antibodies, administered to mice together with their specific antigen, enhance antibody and CD4+ T cell responses to this antigen. The effect is dependent on the low affinity receptor for IgE, CD23, and the receptor must be expressed on B cells. In vitro, IgE-antigen complexes are endocytosed via CD23 on B cells, which subsequently present the antigen to CD4+ T cells. This mechanism has been suggested to explain also IgE-mediated enhancement of immune responses in vivo. We recently found that CD23+ B cells capture IgE-antigen complexes in peripheral blood and rapidly transport them to B cell follicles in the spleen. This provides an alternative explanation for the requirement for CD23+ B cells. The aim of the present study was to determine whether B-cell mediated antigen presentation of IgE-antigen complexes explains the enhancing effect of IgE on immune responses in vivo. The ability of spleen cells, taken from mice 1–4 h after immunization with IgE-antigen, to present antigen to specific CD4+ T cells was analyzed. Antigen presentation was intact when spleens were depleted of CD19+ cells (i.e., primarily B cells) but was severely impaired after depletion of CD11c+ cells (i.e., primarily dendritic cells). In agreement with this, the ability of IgE to enhance proliferation of CD4+ T cells was abolished in CD11c-DTR mice conditionally depleted of CD11c+ cells. Finally, the lack of IgE-mediated enhancemen of CD4+ T cell responses in CD23-/- mice could be rescued by transfer of MHC-II-compatible as well as by MHC-II-incompatible CD23+ B cells. These findings argue against the idea that IgE-mediated enhancement of specific CD4+ T cell responses in vivo is caused by increased antigen presentation by B cells. A model where CD23+ B cells act as antigen transporting cells, delivering antigen to CD11c+ cells for presentation to T cells is consistent with available experimental data.
Citation: Henningsson F, Ding Z, Dahlin JS, Linkevicius M, Carlsson F, et al. (2011) IgE-Mediated Enhancement of CD4+ T Cell Responses in Mice Requires Antigen Presentation by CD11c+ Cells and Not by B Cells. PLoS ONE 6(7): e21760. doi:10.1371/journal.pone.0021760
Editor: Dennis W. Metzger, Albany Medical College, United States of America
Received: February 23, 2011; Accepted: June 6, 2011; Published: July 6, 2011
Copyright: © 2011 Henningsson et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by grants from the Swedish Research Council (www.vr.se); King Gustaf V:s 80 Years Foundation (ams.orbelon.com/gv80/); Ellen, Walter and Lennart Hesselman's Foundation; Hans von Kantzow's Foundation; Ollie and Elof Ericsson's Foundation (oestiftelse.se/); Linné Foundation; Åke Wiberg Foundation (www.ake-wiberg.com/); and Magnus Bergvall Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Apart from initiating various defense mechanisms via Fc-receptor binding or complement activation, antibodies have the capacity to regulate the immune responses against the antigen they are specific for . This phenomenon is called antibody feedback regulation and has been known for over a century , although the molecular mechanisms behind are still not completely understood . The most well known type of antibody feedback regulation is the ability of IgG to suppress antibody responses against large particulate antigens such as erythrocytes. This has been used successfully in the clinic since the 1960's to prevent Rhesus negative women from becoming immunized against fetal Rhesus positive erythrocytes, transferred via transplacental hemorrhage , . In other situations, antibodies can enhance antibody responses. IgG has a dual role and, depending on subclass, enhances responses to protein antigens via Fc-gamma-receptors  or complement . IgM is another well known enhancer of antibody responses, which utilizes the complement system , . An interesting feedback circuit is the one initiated by IgE antibodies. TNP (trinitrophenyl)-specific IgE administered intravenously (i.v.) to mice together with small protein antigens such as BSA (bovine serum albumin)-TNP or OVA (ovalbumin)-TNP induce a several 100-fold higher primary antibody response than does antigen administered alone , , , . The effect is most pronounced on the IgG response and formation of germinal centers and recall responses are also enhanced , , . The enhancing effect of IgE on antibody responses is completely dependent on the presence of CD23, the low affinity receptor for IgE , , . Murine CD23 exists in two isoforms, CD23a and CD23b. CD23a is constitutively expressed on B cells and follicullar dendritic cells (FDC), but not on dendritic cells , , and is the isoform involved in IgE-mediated enhancement of antibody responses . The CD23b isoform has been found on enterocytes in the intestine and recently also on lung epithelial cells , . It is well established by several independent groups that human as well as mouse B cells take up IgE-antigen complexes via CD23 and present the antigenic peptides to CD4+ T cells in vitro , , , , . In analogy with these findings, TNP-specific IgE administered with OVA-TNP to mice enhances proliferation and activation of OVA-specific CD4+ T cells in vivo , . The effect on T cells as well as on antibody responses requires that CD23 is expressed on B cells , . These findings are compatible with the idea that the enhancing effect of IgE on antibody and T cell responses in vivo is explained by B cell-mediated antigen presentation. Whether B cells are able to present antigen to naïve T cells or not has been debated, and there is experimental support both in favour of , , , ,  and against , , ,  this idea. In a previous study, using the same experimental approach as in the present one, we followed the transport of IgE-antigen complexes in vivo . IgE-antigen was found on the majority of B cells in the blood ten minutes after immunization and were detected in the splenic follicles on CD23hiCD21dim cells (follicular B cells) after 30 minutes . This observation provided an alternative explanation for the requirement of CD23+ B cells, suggesting that the enhancing effect of IgE on immune responses could be caused by concentrating the antigen to B cell follicles. The findings prompted us to ask whether CD23+ B cells are required both for transport and presentation of IgE-antigen complexes or whether they primarily act to transport the antigen. The results argue against the idea that presentation of IgE-antigen by CD23+ B cells is the explanation for IgE-mediated enhancement of CD4+ T cell responses in vivo.
Materials and Methods
All animal experiments were approved by Uppsala Animal Research Ethics Committee (protocol numbers C117/7 and C146/10). The mice were bred and maintained in the animal facilities at the National Veterinary Institute (Uppsala, Sweden). Skilled personnel under the supervision of the veterinarian in charge routinely observed the health status of the mice.
C57BL/6 and CD11c-DTR mice were obtained from Jackson Laboratories (Bar Harbor, ME), BALB/c mice from Bommice (Ry, Denmark), CD23-deficient (CD23-/-) mice were a kind gift from Dr Kishimoto  and were backcrossed to BALB/c for 10 generations . DO11.10 mice  were obtained from Dr Westerberg, Karolinska Institute, Stockholm. DO11.10 mice are BALB/c mice that carry transgenic TCR α and β chains, resulting in a TCR recognizing OVA323-339 together with MHC class II I-Ad. CD11c-DTR mice are BALB/c mice transgenic for a high affinity human diphtheria toxin receptor expressed under the cd11c promoter .
Offspring from heterozygous CD11c-DTR mated with wildtype BALB/c mice were used and DNA was extracted from the tail tip by digestion in 40 µl 1x modified Gitschier buffer (67 mM Tris-HCl (pH 8.8), 0.166 mM (NH4)2SO4, 6.5 mM MgCl2) with 1% 2-mercaptoethanol and 0.5% Triton X-100 at 95°C for 5 min. Thereafter, 0.5 mg/ml proteinase K (Qiagen, Hilden, Germany) was added and the digesting tissue was incubated at 55°C for 1 h followed by a 5 min incubation at 95°C. Residual tissue were removed by centrifugation at 16 000 x g for 2 min. This DNA was used in a PCR reaction with the following primers: DTR1, 5′-GCCACCATGAAGCTGCTGCCG-3′; DTR2, 5′-TCAGTGGGAATTAGTCATGCC-3′. Gene amplification was performed using reagents from Applied Biosystems (Carlsbad, CA) in a 25 µl PCR reaction mixture containing 2.5 µl 10x PCR buffer, 4 mM MgCl2, 0.2 mM dNTPs, 0.4 µM each of primers and 0.06 U/ml AmpliTaq DNA polymerase (4 min, 94°C; 15 s, 95°C; 1 min 63.1°C; 15 s 72°C, repeat previous steps for 35 cycles; 5 min 72°C). PCR products was electrophoresed on a 1.5 % agarose gel in 1x Tris-acetate-ethylenediaminetetraacetic acid (TAE) buffer at 100 V. The transgenic diphtheria receptor (DTR) gives a 625 bp band. For depletion of CD11c+ cells in vivo, CD11c-DTR mice were injected i.p. with 100 ng diphtheria toxin (Sigma-Aldrich) in 200 µl PBS. Mice were matched for age and sex within each experiment and were at least 6 weeks old.
OVA (ovalbumin) and TNP (picrylsulfonic acid/hydrate) were obtained from Sigma-Aldrich (St Louis, MO). TNP was coupled to OVA in 0.28 M cacodylate buffer, pH 6.9 as described . The coupling reaction was performed at room temperature for 90 min and stopped by addition of an excess of glycyl-glycine (1 mg/ml, Merck, Darmstadt, Germany). Free TNP was removed by dialyzing against PBS. The number of TNP residues/OVA was determined  and batches with a TNP to OVA ratio between 1 and 2 were used with similar results.
Monoclonal murine IgE-anti-TNP was derived from the B cell hybridoma IGELb4  and purified and stored as described . For flow cytometry we used PE-labeled anti-CD4 mAbs (GK1.5), APC-labeled anti-CD11c (HL3), PE-labeled MHC-II I-Ad (AMS-32.1), FITC-labeled anti-CD19 (1D3) and PE-labeled anti-B220 (RA3-6B2), all from BD Biosciences (San Jose, CA). The DO11.10 transgenic T cells were detected with FITC-labeled KJ1-26 mAb (Caltag Laboratories, Burlingame, CA) specific for this particular TCR heterodimer . For confocal microscopy, FITC-labled anti-B220 (RA3-6B2), biotinylated KJ1-26, and APC-labeled streptavidin were used (all from eBioscience, San Diego, CA).
Mice were immunized with purified IGELb4 (IgE-anti-TNP)  and OVA-TNP in the tail veins in a total volume of 200 µl PBS. IgE and antigen were pre-mixed in a test tube at room temperature and injected within 1 h. Amounts of IgE and antigen are detailed in the figure legends.
Single cell suspensions from spleens were treated with ACK lysing buffer (0.15 M NH4CL, 1.0 M KHCO3, 0.1 mM Na2EDTA, pH 7.3) for 3 min to remove erythrocytes, washed in PBS and resuspended in FACS buffer (PBS with 2% fetal bovine serum). 5×105 cells were stained at 4°C for 30 min in 100 µl FACS buffer with predetermined optimal amounts of labeled antibodies. Cells with the forward- and side-scatter properties of lymphocytes (for analysis of DO11.10 T cells) or all live cells (for analysis of CD11c+ cells) were collected on a FACScan cytometer or a LSR II cytometer (both from BD Biosciences) and analyzed using the FLOWJo Software. 2−3×105 events were analysed. OVA-specific DO11.10 T cells were identified as CD4+KJ126+ and dendritic cells as CD11c+MHC-IIhi.
Isolation of cells
The MACS system (Miltenyi Biotech, Bergisch Gladbach, Germany) was used to obtain cell suspensions depleted of, or enriched for, defined populations. Single cell suspensions from spleens were treated with ACK as described above, resuspended in MACS buffer (PBS containing 0.5% bovine serum albumin and 2 mM EDTA) and labeled with anti-CD4-conjugated magnetic beads and CD4+ cells were collected using LS columns according to the manufacturer's instructions. The isolated cell population was >96% CD4+ as determined by flow cytometry. Spleen cells were depleted of CD11c+ cells by mixing with anti-CD11c-conjugated magnetic beads followed by passage over an LD column (which is specialized for depletion of cells): more than 94% of CD11c+MHC-IIhi cells were depleted (CD11c− population in Figure 1A). To enrich for CD11c+ cells (Figure 1B) spleen cells were mixed with anti-CD11c-conjugated magnetic beads and passed over an LS column (which is specialized for enrichment of cells). Cells eluted from the column were >68% CD11c+. To deplete for B cells (Figure 1A), spleen cells were mixed with anti-CD19-conjugated magnetic beads followed by passage over an LD column. The flow through cells were >98% CD19−. To enrich for B cells, the fact that all leukocytes except mature and immature resting B cells express CD43 was used: spleen cells were mixed with anti-CD43-conjugated magnetic beads, passed over an LD column and the CD43− flow through cells (>98% CD19+) were used as B cells in transfer experiments.
Figure 1. CD11c+ cells present IgE-antigen to CD4+ T cells ex vivo.
A & B) BALB/c or CD23-/- mice were immunized with 100 µg OVA-TNP alone or together with 250 µg of IgE anti-TNP. After 4 h, spleens were removed, fractionated, and tested for ability to activate CD4+ OVA-specific DO11.10 T cells ex vivo. C) CD11c-DTR mice were treated with 100 ng diphtheria toxin or left untreated. After 24 hours, the mice were immunized as in A and 4 hours later spleen cells were removed and tested for ability to activate DO11.10 T cells ex vivo. A) is representative of two identical and 4 similar experiments, B) of three similar experiments and C) of 5 identical and 2 similar experiments.doi:10.1371/journal.pone.0021760.g001
2.3−5.0×106 CD4+ cells from DO11.10 spleens (positively selected as described above) and when applicable 17×106 CD43− B cells from BALB/c or C57BL/6 (negatively selected as described above) were adoptively transferred i.v. in 200 µl PBS.
3H-thymidin incorporation assay
BALB/c or CD23-/- mice were immunized i.v. with 100 µg of OVA-TNP in PBS with or without 250 µg of IgE anti-TNP. Controls were left unimmunized. After 4 hours, spleens from BALB/c mice were removed and spleen cells were divided into three fractions: one was left unfractionated, one was depleted of CD19+ cells and one was depleted of CD11c+ cells whereas CD23-/- spleen cells were left unfractionated (Figure 1A). In Figure 1B, CD11c+ cells were enriched on LS columns (see above). Unfractionated spleen cells were used as APC at a concentration of 6×105 cells/well. When fractionated APC populations were used (Figures 1A & B), they were adjusted to the same volume as before fractionation to contain the same numbers of the respective APC as did unfractionated spleen cells. In Figure 1B, we also used 2x, 5x and 10x the number of APC as in the unfractionated populations. APC were added in 100 µl f-DMEM (National Veterinary Institute, Uppsala, Sweden) supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin, 2 mM L-glutamine, 50 µM β-mercaptoethanol, and 5% heat-inactivated fetal bovine serum (all from Sigma-Aldrich) to 96 well tissue culture plates (Sarstedt, Nümbrecht, Germany) and irradiated (10 Gy). 1×105 CD4+ DO11.10 responder cells in 100 µl f-DMEM were added to each well. After 48 h the cells were pulsed with 1 µCi 3H-thymidine (American Radiolabeled Chemicals, St Louis, MO) for 18–24 h. Incorporation of 3H-thymidine was measured in a beta counter (Wallac 1450 MicroBeta Trilux, Perkin Elmer, Waltham, MA). Mean and SEM were calculated from five replicate wells.
Tissue sections were prepared as described previously . Briefly, spleens were frozen and embedded in O.C.T. compound (Sakura Finetek, Alphen aan den Rijn, The Netherlands) and 7 µm sections were cut using a cryostat. Air-dried sections were collected onto glass microscope slides (Mentzel-Gläser, Braunschweig, Germany) and stored frozen at −70°C. The slides were fixed in ice cold 50% acetone for 30 s and then in 100% acetone for 5 min. Rehydration of the sections was performed in PBS for 15 min at room temperature. Sections were then blocked with 5% horse serum (Sigma-Aldrich) in PBS for 1 h. After pouring off blocking solution, 2 µg/ml FITC-labeled anti-B220 and 1 µg/ml biotinylated KJ1-26, diluted in 100 µl blocking solution, was added to each section and left for 1 h at room temperature. After washing twice in PBS for 5 min, 100 µl of 2 µg/ml APC-labeled streptavidin was added, sections were stained for 1 h and washed twice in PBS. Slides were mounted in Fluoromount G (Southern Biotech, Birmingham, AL) and immunofluorescence was detected by a LSM 700 confocal microscope from Carl Zeiss (Thornwood, NY). Photos of every T cell zone on four non-consecutive sections from each individual spleen were taken by Zen 2009 software (Carl Zeiss). Numbers of KJ1-26+ T cells in each T cell zone were automatically quantified using ImageJ software (NIH, Bethesda, MD).
Statistical differences between groups were determined by Student's t-test. Values for p>0.05 (not significant, ns), p<0.05 (*), p<0.01 (**) or p<0.001 (***) are indicated in the figures.
CD11c+ cells present IgE-antigen to CD4+ T cells ex vivo
To investigate whether B cells (CD19+) or CD11c+ cells present IgE-antigen complexes in vivo, mice were immunized with OVA-TNP alone or together with IgE-anti-TNP. Four hours later their spleens were removed, cell suspensions prepared, irradiated and used as APC (antigen presenting cells) ex vivo. Spleen cells removed after 1, 2, 3, and 4 h could induce efficient antigen presentation, with the best effect seen with APC removed after 4 h (data not shown). Spleen cells removed after 13 h had lost their capacity to present antigen ex vivo (data not shown). No additional antigen or immune complexes were added to the cell cultures, ascertaining that only OVA-peptides acquired in vivo were presented on MHC-II on the various APC populations. As responder cells, CD4+ T cells from DO11.10 mice were used. These T cells have a transgenic TCR recognizing OVA323–339 together with MHC class II I-Ad . Unfractionated spleen cells from mice immunized with IgE-antigen complexes induced much more efficient T cell proliferation than spleen cells from mice given OVA-TNP alone whereas spleen cells from CD23-/- mice could not induce T cell proliferation (Figure 1A). Interestingly, spleen cells depleted of B cells (CD19−) were equally efficient as unfractionated spleen cells, whereas spleen cells depleted of CD11c+ cells (CD11c−) completely lost their antigen presenting capacity, although B cells were present in this population (Figure 1A). In the reciprocal approach, enriching spleen cells for CD11c+ cells, we found that CD11c+ cells retained the antigen presenting capacity of unfractionated spleen cells (Figure 1B). Upon increasing the number of CD11+ cells up to 10-fold of the number present amongst the unfractionated spleen cells, also the T cell proliferation increased dramatically (Figure 1B). As an alternative approach, CD11c-DTR mice, expressing the high affinity receptor for diphtheria toxin under the CD11c promoter, were used as a source of APC in the ex vivo antigen presenting assay. CD11c-DTR mice can be conditionally depleted of cells expressing high amounts of CD11c by treatment with diphtheria toxin  (Figure 2). Toxin-treated and untreated CD11c-DTR mice were immunized with OVA-TNP alone or in complex with IgE anti-TNP. Four hours later their spleens were removed, spleen cells prepared, irradiated and used as APC in the T cell proliferation assay as described above. When total spleen cells from untreated CD11c-DTR mice immunized with IgE-antigen complexes were used as APC, they induced efficient T cell proliferation (Figure 1C). In contrast, APC derived from the spleens of toxin-treated CD11c-DTR mice were unable to induce T cell proliferation (Figure 1C). Taken together, these observations suggest that CD11c+ cells, but not B cells, in mice immunized with IgE-OVA complexes present antigen efficiently enough to induce proliferation of specific CD4+ T cells ex vivo.
Figure 2. Diphtheria toxin depletes CD11c-DTR mice of CD11c+ cells.
CD11c-DTR mice were treated with 100 ng diphtheria toxin or left untreated. Twenty-four hours later total spleen cells were analyzed in flow cytometry and dendritic cells were defined as CD11c+, MHC-IIhi cells. The numbers show % of live cells in a representative experiment with 97% depletion of dendritic cells.doi:10.1371/journal.pone.0021760.g002
IgE-mediated enhancement of T cell proliferation in vivo depends on CD11c+ cells
The conclusion from Figure 1 was deduced from an ex vivo antigen presentation assay. Next, we analyzed the ability of IgE-antigen complexes to induce T-cell proliferation directly in vivo in mice lacking CD11c+ cells. CD11c-DTR mice, and in some cases littermate controls, were treated with diphtheria toxin leading to depletion of 97% of CD11c+ spleen cells in the CD11c-DTR mice (Figure 2). Mice were then adoptively transferred with DO11.10 T cells and immunized. Three days later the number of splenic OVA-specific CD4+ T cells were analyzed, either in flow cytometry (Figure 3A, B) or in confocal microscopy (Figure 3C, D). IgE enhanced T cell proliferation in wildtype littermates equally well whether the mice had been treated with diphtheria toxin or not, showing that the toxin itself did not suppress T cell responses (Figure 3B, bars 1–4). In untreated CD11c-DTR mice, IgE also enhanced T cell proliferation (Figure 3B, bar 5 and 6). In sharp contrast, the ability of IgE to enhance T cell responses was completely abolished in CD11c-DTR mice that had been depleted of CD11c+ cells by toxin treatment (Figure 3B, bar 7 and 8). Analysis by an alternative approach, counting specific T cells in spleen sections, gave similar results (Figure 3 C,D). These data, strengthen the ex vivo data (Figure 1), and show that IgE-mediated enhancement of CD4+ T cell responses in vivo requires the presence of CD11c+ cells.
Figure 3. IgE-mediated enhancement of T cell proliferation in vivo is dependent on CD11c+ cells.
CD11c-DTR mice, and in B) also wild type littermates, were treated with 100 ng diphtheria toxin or left untreated. The same day, the mice were transferred i.v. with 2.3−3.0×106 CD4+ DO11.10 spleen cells. Twenty-four hours after diphtheria toxin treatment, mice were immunized i.v. with 20 µg OVA-TNP with or without 50 µg IgE anti-TNP. Three days later, spleens were analyzed for CD4+ KJ1-26+ T cells by flow cytometry (A, B) or for KJ1-26+ cells in confocal microscopy (C, D). A) Representative dot-plot of flow cytometry data; values represent % CD4+KJ1-26+ cells of gated lymphocytes. B) Mean values ± SEM of % CD4+KJ1-26+ of CD4+ cells analyzed in flow cytometry (3 mice/group) C) Mean values ± SEM of KJ1-26+ cells in non-consecutive sections of T cell zones analyzed by confocal microscopy (3 mice/group; 3–12 T cell zones/mouse). D) Visualization of representative T cell zones from each group: KJ1-26+ cells (red) and B220+ cells (blue). B) is representative of 4 experiments and C) of 1 experiment.doi:10.1371/journal.pone.0021760.g003
MHC-II-incompatible B cells rescue IgE-mediated enhancement of T cell-proliferation in CD23-/- mice
IgE cannot enhance CD4+ T cell or antibody responses in CD23-/- mice, but transfer of syngeneic wildtype B cells can rescue these functions . In addition, bone marrow derived cells (B cells), but not stromal cells (FDC), from wildtype mice rescued the ability of IgE to enhance antibody responses in CD23-/- mice . Both observations demonstrate the requirement for CD23+ B cells in this system. T cell responses are MHC restricted. Therefore, should the effector function of CD23+ B cells be to present IgE-antigen to DO11.10 T cells, only MHC-II compatible B cells would rescue CD23-/- mice. However, should the effector function merely be to transport IgE-antigen to the spleen, both MHC compatible and incompatible wildtype B cells would rescue the response. To investigate this, CD23-/- mice on a BALB/c background (MHC-IId) were reconstituted with B cells from either C57BL/6 (MHC-IIb) or BALB/c (MHC-IId). Coming from wildtype mice, these B cells were CD23+. All animals had been transferred with DO11.10 T cells on the previous day and were immunized with OVA-TNP alone or in complex with IgE on the day of the B cell transfer. OVA-specific T cell proliferation was assessed by flow cytometry of spleen cell suspensions (data not shown) or confocal microscopy of spleen sections (Figure 4). As expected, T cells in BALB/c mice (bar 1 and 2) responded well to IgE-antigen whereas T cells in CD23-/- mice (bar 3 and 4) did not respond. Also as expected , T cell proliferation in CD23-/- mice was rescued by transfer of MHC-II compatible B cells (bar 5 and 6). Interestingly, MHC-II incompatible B cells (bar 7 and 8) were equally efficient in rescuing T cell proliferation as were MHC-II compatible B cells. This observation strongly suggests that the role of CD23+ B cells in this system is to transport rather than to present IgE-antigen complexes. There was a possibility that B cells from MHC-II incompatible mice could induce alloreactive responses by the OVA-specific T cells. However, neither BALB/c nor C57BL/6 B cells, transferred to CD23-/- mice which were left unimmunized, were able to induce proliferation of DO11.10 T cells (Figure 4, bar 9 and 10).
Figure 4. MHC-II incompatible B cells rescue IgE-mediated enhancement of T cell proliferation in CD23-/- mice.
Mice were transferred with 5×106 CD4+ DO11.10 T cells. The next day, indicated mice were transfused with B cells (spleen cells depleted of CD43+ cells on a MACS column; 98% CD19+) and subsequently immunized with 20 µg OVA-TNP, with or without 50 µg IgE anti-TNP. Three days later spleens were removed and the number of KJ1-26+ T cells in non-consecutive sections of T cell zones were analyzed by confocal microscopy (2–4 mice/group; 4–16 T cell zones/mouse; mean values ± SEM). Representative of two identical and one similar experiment.doi:10.1371/journal.pone.0021760.g004
The aim of the present study was to find out whether IgE-mediated enhancement of CD4+ T cell responses in vivo is caused by increased uptake and presentation of IgE-antigen by CD23+ B cells or whether it is more likely that these cells merely transport IgE-antigen to B cell follicles, where other cells take over and present the antigen to T cells. In summary, we were unable to find evidence for presentation of IgE-antigen complexes by CD23+ B cells in vivo. This is based on several experimental approaches. Using ex vivo antigen presentation assays we found (i) that CD19-, but not CD11c− spleen cells, taken from wildtype mice 4 h after immunization with IgE-antigen, could induce proliferation of specific CD4+ T cells (Figure 1A), (ii) that positively selected CD11c+ spleen cells could induce proliferation of specific CD4+ T cells (Figure 1B), and (iii) that spleen cells taken from diphtheria toxin-treated CD11c-DTR mice (where CD11c+ cells are conditionally depleted) 4 h after immunization with IgE-antigen, lost their ability to induce proliferation of specific CD4+ T cells (Figure 1C). In a second approach we demonstrated that IgE-antigen did not induce T cell proliferation in vivo in diphtheria toxin-treated CD11c-DTR mice (Figure 3). Finally, both MHC-II compatible and MHC-II incompatible CD23+ B cells could rescue the specific T cell response to IgE-antigen complexes in CD23-/- mice (Figure 4). As mentioned in the introduction, we have previously shown that IgE-complexed antigen can be found on peripheral B cells 10 minutes after immunizaiton and in the splenic B cell follicles (on the surface of follicular B cells) 30 minutes after immunization . This finding demonstrated a new way for antigen to enter the follicles, namely attached via IgE to CD23+ recirculating B cells. It also provided a new possible explanation to why CD23+ B cells are required for IgE-mediated enhancement of antibody and CD4 T cell responses. Supported by in vitro findings this effect had been believed to be caused by B-cell mediated antigen presentation to T cells following uptake of IgE-antigen via CD23 , , , , . The current data are suggestive of a scenario where CD23+ B cells transport IgE-antigen to the spleen where the antigen is captured and presented to CD4+ T cells by CD11c+ cells. Although we do not directly demonstrate the delivery of antigen from B cells to CD11c+ cells, we find that this is the explanation that best fits available data. In the CD11c-DTR mice, a gene that encodes the human diphtheria toxin receptor fused to green fluorescent protein (GFP) is inserted under the control of the CD11c promoter . The promoter was initially considered specific for dendritic cells, but closer investigation has revealed activity also in alveolar macrophages, splenic marginal zone and metallophilic macrophages, a subset of activated T cells, NK cells and plasmablasts (reviewed in ). Nevertheless, we find it likely that the predominant cells presenting IgE-antigen to CD4+ T cells are dendritic cells. We cannot exclude involvement of splenic macrophages but the other CD11c+ cells are not antigen presenting cells. Dendritic cells are efficient APC and are able to trigger naïve T cells, a capacity not always found amongst B cells , , . This has indeed been one of the arguments against the idea that B-cell mediated antigen presentation explains IgE-mediated enhancement. Mouse dendritic cells have not been found to express CD23. The observation that pure B cells can rescue the response to IgE-antigen in CD23-/- mice , (Figure 4) corroborates this and shows that although CD11c+ cells are required, they do not act via CD23.
We do not yet know how IgE-antigen is transferred from B cells to CD11c+ cells. Intercellular exchange of proteins can occur in several different ways, e.g. via exosomes or trogocytosis . Since dendritic cells are present at the boundary between T and B cell zones and also in B cell follicles , there should be ample opportunity for delivery of the IgE-antigen complexes from B cells to dendritic cells. The simplest explanation is that the presence of IgE-antigen on the B cell surface triggers neighboring dendritic cells to take up antigen via pinocytosis or endocytosis. Alternatively, exosomes containing CD23 and IgE-antigen bound to it, secreted by B cells after endocytosis of surface bound CD23 , may be very palatable to nearby dendritic cells. Antibody-antigen-complement complexes are transported on marginal zone B cells, expressing high levels of complement receptor 1 and 2, from the marginal zone to B cell follicles where they are delivered to FDC , . Apart from B cells, FDC are the only cells in B cell follicles that express CD23. However, FDC do not seem to be involved in IgE-mediated enhancement, shown by the fact that B cells alone rescue the response in CD23-/- mice  (Figure 4) and that bone marrow chimeras with CD23- FDC respond well to IgE-antigen . Therefore, a similar role for FDC in capturing IgE-antigen from follicular B cells as they have in capturing antigen-antibody-complement from marginal zone B cells, is unlikely. Instead our data suggest an alternative route for antigen handling where IgE-antigen complexes are transported by peripheral CD23+ B cells to B cell follicles where they are delivered to dendritic cells. Although we are not aware of any previous examples of antigen delivery from B cells to dendritic cells, the reverse has been observed , .
Several laboratories, including our own, have demonstrated that B cells via CD23 can take up and present IgE-antigen to T cells in vitro , , , , . Nevertheless, the data presented here strongly argue against that B-cell-mediated antigen presentation explains why IgE is such a potent enhancer of T cell responses in vivo. The ex vivo assay (Figure 1) differs from previous in vitro studies in that the APC are only able to interact with the antigen while they are still in the animal - no further antigen is added to the cultures. Allowing cells and IgE-antigen to incubate in vitro may facilitate for CD23+ B cells to take up the complexes, as compared to the in vivo situation where the effective antigen concentration is smaller and the B cells are mobile. In addition, antigen transportation is not required in vitro and the ability of B cells to endocytose IgE-antigen via CD23 therefore may become predominant. This is an additional example of discrepant results obtained from in vivo and in vitro studies of the antibody feedback regulation system. IgG-mediated suppression of erythrocyte responses was independent of FcγRIIB in vivo but not in vitro ,  and complement-tagged antigen could be taken up via complement receptors 1 and 2 on B cells and presented to T cells in vitro , ,  but not in vivo .
Apart from its role in IgE-mediated enhancement of immune responses, CD23 is a negative regulator of IgE production. This is evidenced by increased IgE levels in CD23 knockout mice  and in mouse strains where CD23 is mutated ,  as well as by reduced IgE production in transgenic mice overexpressing CD23 . Interestingly, IgE-mediated enhancement works well in mice overexpressing CD23, showing that the negative and positive regulatory effects of CD23 are not mutually exclusive . This is also what would be expected if the negative effect is caused by changes in B cell signaling whereas the positive effect is caused by transportation only.
Another interesting function of CD23 is to mediate transepithelial transport of IgE-antigen across intestinal barriers and across respiratory epithelium , , . This effect is dependent on the CD23b isoform, whereas IgE-mediated enhancement of in vivo antibody responses is dependent on CD23a . Nevertheless, it is an interesting parallel that IgE-antigen complexes can be transported bound to CD23 both on enterocytes , , airway epithelial cells  and B cells .
CD23 is known as the low affinity receptor for IgE, but does in fact have a dual affinity ranging from 2−7×106 M−1 to 2−7×107 M−1 . The higher affinity is associated with the oligomeric, membrane bound form of the receptor  and up to 80% of peripheral B cells have recently been shown to be preloaded with IgE bound to CD23 , . Presumably this IgE is sampled from the total IgE in the individual, thus reflecting the antigen experience of the host, and would enable rapid capture and transport of many different antigens by CD23+ B cells. One argument against a significant role of IgE in immune regulation has been that the serum levels are exceedingly low. However, the observations that B cells can be preloaded with IgE ,  suggest that they may have acquired high enough concentrations of relevant IgE to mediate e.g. antigen transport regardless of a low serum concentration at the time of antigen entry. CD23 is a C-type lectin, a family of receptors known to act as pattern recognition receptors recognizing carbohydrate structures on pathogens . Therefore, CD23 most likely can bind directly to certain carbohydrates expressed on pathogens. This may be an additional, IgE-independent, way for CD23+ B cells to transport antigen.
In summary we have presented data supporting the idea that peripheral CD23+ B cells capture IgE-antigen and transport it to splenic B cell follicles where antigen is delivered to dendritic cells which subsequently present it to T cells. The data argue against a role for CD23+ B cells in presenting IgE-antigen to T cells in vivo.
Conceived and designed the experiments: FH K-OG JH BH. Performed the experiments: FH ZD JSD ML FC. Analyzed the data: FH ZD JSD ML JH BH. Wrote the paper: FH ZD BH.
- 1. Heyman B (2000) Regulation of antibody responses via antibodies, complement, and Fc receptors. Annu Rev Immunol 18: 709–737.
- 2. von Behring E, Wernicke E (1892) Über Immunisierung und Heilung von Versuchstieren bei der Diphterie. Z Hyg Infektionskrankheit 12: 10–44.
- 3. Clarke CA, Donohoe WTA, Woodrow JC, Finn R, Krevans JR, et al. (1963) Further experimental studies on the prevention of Rh haemolytic disease. Br Med Journal 1: 979–984.
- 4. Bowman JM (1988) The prevention of Rh immunization. Transfus Med Rev 2: 129–150.
- 5. Wernersson S, Karlsson M, Dahlström J, Mattsson R, Verbeek JS, et al. (1999) IgG-mediated enhancement of Ab responses is low in FcRγ chain deficient mice and increased in FcγRII deficient mice. J Immunol 163: 618–622.
- 6. Diaz de Ståhl T, Dahlström J, Carroll MC, Heyman B (2003) A role for complement in feedback-enhancement of antibody responses by IgG3. J Exp Med 197: 1183–1190.
- 7. Heyman B, Pilström L, Shulman MJ (1988) Complement activation is required for IgM-mediated enhancement of the antibody response. J Exp Med 167: 1999–2004.
- 8. Youd ME, Ferguson AR, Corley RB (2002) Synergistic roles of IgM and complement in antigen trapping and follicular localization. Eur J Immunol 32: 2328–2337.
- 9. Heyman B, Liu T, Gustavsson S (1993) In vivo enhancement of the specific antibody response via the low affinity receptor for IgE. Eur J Immunol 23: 1739–1742.
- 10. Fujiwara H, Kikutani H, Suematsu S, Naka T, Yoshida K, et al. (1994) The absence of IgE antibody-mediated augmentation of immune responses in CD23-deficient mice. Proc Natl Acad Sci USA 91: 6835–6839.
- 11. Gustavsson S, Wernersson S, Heyman B (2000) Restoration of the antibody response to IgE/antigen complexes in CD23-deficient mice by CD23+ spleen or bone marrow cells. J Immunol 164: 3990–3995.
- 12. Hjulström S, Landin A, Jansson L, Holmdahl R, Heyman B (1995) No role of interleukin-4 in CD23/IgE-mediated enhancement of the murine antibody response in vivo. Eur J Immunol 25: 1469–1472.
- 13. Gustavsson S, Hjulström S, Liu T, Heyman B (1994) CD23/IgE-mediated regulation of the specific antibody response in vivo. J Immunol 152: 4793–4800.
- 14. Westman S, Gustavsson S, Heyman B (1997) Early expansion of secondary B cells after primary immunization with antigen complexed with IgE. Scand J Immunol 46: 10–15.
- 15. Hjelm F, Karlsson MCI, Heyman B (2008) A novel B-cell mediated transport of IgE-immune complexes to the follicle of the spleen. J Immunol 180: 6604–6610.
- 16. Getahun A, Hjelm F, Heyman B (2005) IgE enhances antibody and T cell responses in vivo via CD23+ B Cells. J Immunol 175: 1473–1482.
- 17. Rao M, Lee WT, Conrad DH (1987) Characterization of a monoclonal antibody directed against the murine B lymphocyte receptor for IgE. J Immunology 138: 1845–1851.
- 18. Maeda K, Burton GF, Padgett DA, Conrad DH, Huff TF, et al. (1992) Murine follicular dendritic cells and low affinity Fc receptors for IgE (FcεRII). J Immunol 148: 2340–2347.
- 19. Yu LCH, Montagnac G, Yang P-C, Conrad DH, Benmerah A, et al. (2003) Intestinal epithelial CD23 mediates enhanced antigen transport in allergy: evidence for a novel splice form. Am J Physiol Gastrointes Liver Physiol 285: G223–G234.
- 20. Palaniyandi S, Tomei E, Li Z, Conrad DH, Zhu X (2011) CD23-Dependent Transcytosis of IgE and Immune Complex across the Polarized Human Respiratory Epithelial Cells. J Immunol 186(6): 3484–96.
- 21. Kehry MR, Yamashita LC (1989) Low-affinity IgE receptor (CD23) function on mouse B cells: role in IgE-dependent antigen focusing. Proc Natl Acad Sci USA 86: 7556–7560.
- 22. Pirron U, Schlunck T, Prinz JC, Rieber EP (1990) IgE-dependent antigen focusing by human B lymphocytes is mediated by low-affinity receptor for IgE. Eur J Immunol 20: 1547–1551.
- 23. Santamaria LF, Bheekha R, van Reijsen FC, Perez Soler MT, Suter M, et al. (1993) Antigen focusing by specific monomeric immunoglobulin E bound to CD23 on Epstein-Barr Virus-Transformed B cells. Human Immunology 37: 23–30.
- 24. van der Heijden FL, van Neerven RJJ, van Katwijk M, Bos JD, Kapsenberg ML (1993) Serum-IgE-facilitated allergen presentation in atopic disease. J Immunol 150: 1–8.
- 25. Carlsson F, Hjelm F, Conrad DH, Heyman B (2007) IgE enhances specific antibody and T cell responses in mice overexpressing CD23. Scand J Immunol 66: 261–270.
- 26. Constant S, Schweitzer N, West J, Ranney P, Bottomly K (1995) B lymphocytes can be competent antigen-presenting cells for priming CD4+ T cells to protein antigens in vivo. J Immunol 155: 3734–3741.
- 27. Milich DR, Chen M, Schodel F, Peterson DL, Jones JE, et al. (1997) Role of B cells in antigen presentation of the hepatitis B core. Proc Natl Acad Sci U S A 94: 14648–14653.
- 28. Constant SL (1999) B lymphocytes as antigen-presenting cells for CD4+ T cell priming in vivo. J Immunol 162: 5695–5703.
- 29. Attanavanich K, Kearney JF (2004) Marginal zone, but not follicular B cells, are potent activators of naive CD4 T cells. J Immunol 172: 803–811.
- 30. Crawford A, Macleod M, Schumacher T, Corlett L, Gray D (2006) Primary T cell expansion and differentiation in vivo requires antigen presentation by B cells. J Immunol 176: 3498–3506.
- 31. Fuchs FJ, Matzinger P (1992) B cells turn off virgin but not memory T cells. Science 258: 1156–1159.
- 32. Eynon EE, Parker DC (1992) Small B cells as antigen-presenting cells in the induction of tolerance to soluble protein antigens. J Exp Med 175: 131–138.
- 33. Ronchese F, Hausmann B (1993) B lymphocytes in vivo fail to prime naive T cells but can stimulate antigen-experienced T lymphocytes. J Exp Med 177: 679–690.
- 34. Jenkins MK, Khoruts A, Ingulli E, Mueller DL, McSorley SJ, et al. (2001) In vivo activation of antigen-specific CD4 T cells. Annu Rev Immunol 19: 23–45.
- 35. Murphy KM, Heimberger AB, Loh DY (1990) Induction by antigen of intrathymic apoptosis of CD4+CD8+TCRlo thymocytes in vivo. Science 250: 1720–1723.
- 36. Jung S, Unutmaz D, Wong P, Sano G, De los Santos K, et al. (2002) In vivo depletion of CD11c(+) dendritic cells abrogates priming of CD8(+) T cells by exogenous cell-associated antigens. Immunity 17: 211–220.
- 37. Good AH, Wofsy L, Henry C, Kimura J (1980) Preparation of hapten-modified protein antigens. In: Mishell BB, Shiigi SM, editors. Selected Methods in Cellular Immunology. San Francisco, CA: W.H. Freeman and Co. pp. 343–350.
- 38. Rudolph AK, Burrows PD, Wabl MR (1981) Thirteen hybridomas secreting hapten-specific immunoglobulin E from mice with Iga or Igb heavy chain haplotype. Eur J Immunol 11: 527–529.
- 39. Haskins K, Kubo R, White J, Pigeon M, Kappler J, et al. (1983) The major histocompatibility complex-restricted antigen receptor on T cells. I. Isolation with a monoclonal antibody. J Exp Med 157: 1149–1169.
- 40. Bar-On L, Jung S (2010) Defining in vivo dendritic cell functions using CD11c-DTR transgenic mice. Methods Mol Biol 595: 429–442.
- 41. Rechavi O, Goldstein I, Kloog Y (2009) Intercellular exchange of proteins: the immune cell habit of sharing. FEBS Lett 583: 1792–1799.
- 42. Lindquist RL, Shakhar G, Dudziak D, Wardemann H, Eisenreich T, et al. (2004) Visualizing dendritic cell networks in vivo. Nat Immunol 5: 1243–1250.
- 43. Mathews JA, Gibb DR, Chen BH, Scherle P, Conrad DH (2010) CD23 Sheddase A disintegrin and metalloproteinase 10 (ADAM10) is also required for CD23 sorting into B cell-derived exosomes. J Biol Chem 285: 37531–37541.
- 44. Ferguson AR, Youd ME, Corley RB (2004) Marginal zone B cells transport and deposit IgM-containing immune complexes onto follicular dendritic cells. Int Immunol 16: 1411–1422.
- 45. Cinamon G, Zachariah MA, Lam OM, Foss FW , Cyster JG (2008) Follicular shuttling of marginal zone B cells facilitates antigen transport. Nat Immunol 9: 54–62.
- 46. Huang NN, Han SB, Hwang IY, Kehrl JH (2005) B cells productively engage soluble antigen-pulsed dendritic cells: visualization of live-cell dynamics of B cell-dendritic cell interactions. J Immunol 175: 7125–7134.
- 47. Bergtold A, Desai DD, Gavhane A, Clynes R (2005) Cell surface recycling of internalized antigen permits dendritic cell priming of B cells. Immunity 23: 503–514.
- 48. Karlsson MCI, Wernersson S, Diaz de Ståhl T, Gustavsson S, Heyman B (1999) Efficient IgG-mediated suppression of primary antibody responses in Fc-gamma receptor-deficient mice. Proc Natl Acad Sci U S A 96: 2244–2249.
- 49. Karlsson MCI, Getahun A, Heyman B (2001) FcγRIIB in IgG-mediated suppression of antibody responses: different impact in vivo and in vitro. J Immunol 167: 5558–5564.
- 50. Thornton BP, Vetvicka V, Ross GD (1994) Natural antibody and complement-mediated antigen processing and presentation by B-lymphocytes. J Immunol 152: 1727–1737.
- 51. Boackle SA, Holers VM, Karp DR (1997) CD21 augments antigen presentation in immune individuals. Eur J Immunol 27: 122–129.
- 52. Cherukuri A, Cheng PC, Pierce SK (2001) The role of the CD19/CD21 complex in B cell processing and presentation of complement-tagged antigens. J Immunol 167: 163–172.
- 53. Carlsson F, Getahun A, Rutemark C, Heyman B (2009) Impaired antibody responses but normal proliferation of specific CD4+ T cells in mice lacking complement receptors 1 and 2. Scand J Immunol 70: 77–84.
- 54. Yu P, Kosco-Vilbois M, Richards M, Köhler G, Lamers MC (1994) Negative feedback regulation of IgE synthesis by murine CD23. Nature 369: 753–756.
- 55. Lewis G, Rapsomanakiki T, Bouriez T, Crockford T, Ferry H, et al. (2004) Hyper IgE in New Zealand black mice due to a dominant-negative CD23 mutation. Immunogenetics 56: 564–571.
- 56. Ford JW, Sturgill JL, Conrad DH (2009) 129/SvJ mice have mutated CD23 and hyper IgE. Cell Immunol 254: 124–134.
- 57. Payet ME, Woodward EC, Conrad DH (1999) Humoral response suppression observed with CD23 transgenics. J Immunol 163: 217–223.
- 58. Yang P-C, Berin MC, Yu LCH, Conrad D, Perdue MH (2000) Enhanced intestinal transepithelial antigen transport in allergic rats is mediated by IgE and CD23 (FcεRII). J Clin Invest 106: 879–886.
- 59. Tu Y, Salim S, Bourgeois J, Di Leo V, Irvine EJ, et al. (2005) CD23-mediated IgE transport across human intestinal epithelium: inhibition by blocking sites of translation or binding. Gastroenterology 129: 928–940.
- 60. Dierks SE, Bartlett WC, Edmeades RL, Gould HJ, Rao M, et al. (1993) The oligomeric nature of the murine FceRII/CD23. Implications for function. J Immunol 150: 2372–2382.
- 61. Mwinzi PN, Ganley-Leal L, Black CL, Secor WE, Karanja DM, et al. (2009) Circulating CD23+ B cell subset correlates with the development of resistance to Schistosoma mansoni reinfection in occupationally exposed adults who have undergone multiple treatments. J Infect Dis 199: 272–279.
- 62. Griffith QK, Liang Y, Onguru DO, Mwinzi PN, Ganley-Leal LM (2011) CD23-bound IgE augments and dominates recall responses through human naive B cells. J Immunol 186: 1060–1067.
- 63. McGreal EP, Miller JL, Gordon S (2005) Ligand recognition by antigen-presenting cell C-type lectin receptors. Current Opinion in Immunology 17: 18–24.