Uppsala University
Ovalbumin-anti-CD23 conjugates enhance
ovalbumin-specific immune responses in a CD23-dependent manner
Final reportspecific immune responses in a CD23-dependent manner
Final report
Version: 1.0
Details:
Name: Lenny van Mechelen
lp.vanmechelen@student.avans.nl
School details: Avans University of Applied Sciences
School of life science and environmental technologies Lovensdijkstraat 61
4818 AJ Breda The Netherlands
Education: Biology and Medical Laboratory Sciences Teacher supervisor: Ans Arets
jmc.arets@avans.nl
Date of internship: September 2012 – April 2013 Company details: Uppsala University
Department of Medical Biochemistry and Microbiology Biomedical Centre, Building 8, Level 2
Husargatan 3, Box 582 751 23 Uppsala Sweden
Company supervisor: Frida Henningson Johnson, PhD
Preface
This report represents the graduation project of my Bachelor of Applied Science in Biology and Medical Laboratory Sciences at Avans University of Applied Sciences in Breda, the Netherlands. The project was performed at the Department of Medical Biochemistry and Microbiology, unit of Experimental Immunology at Uppsala University in Sweden under the supervision of Frida Henningson Johnson, PhD.
I would like to thank Frida Henningson Johnson, PhD for the opportunity to perform my graduation project in her research group and for the excellent guidance in the project, including the practical and theoretical support, the critical view in the project and the guidance in writing this report. In addition I want to thank my co-supervisor Prof. Birgitta Heyman for her depth knowledge in theoretical and practical immunology always willing to share with me and teach to me. Finally I want to thank all the members in the group for their great practical support and guidance in the lab.
Abstract
A humoral immune response against blood-borne protein antigens is initiated in the white pulp of the spleen and requires activation of both B cells and helper T cells. Before a humoral immune response can be initiated, the antigen has to be transported to splenic follicles, because antigens have not access to the follicles by themselves. It has been shown that CD23+ B cells in vivo can transport IgE-antigen complexes into the follicles. CD23 is the low affinity receptor for IgE and is primarily expressed on B cells and follicular dendritic cells in mice. When mice are immunized with IgE-antigen complexes, an enhanced immune response can be seen. In the current study the transport function of CD23 on B cells was used to investigate whether ovalbumin conjugated to anti CD23 antibodies can facilitate antigen transport into the B cell follicles and enhance an antigen-specific immune response. Using CD23 as a transporting molecule, the ovalbumin conjugates can be efficiently transported to the follicles of the spleen.
First, ovalbumin was conjugated to anti-CD23 antibodies and to an isotype control. Subsequently, BALB/c and CD23 deficient mice were immunized with the ovalbumin conjugates and ovalbumin-specific IgG was measured at different time points. To analyze ovalbumin-ovalbumin-specific T cell proliferation, BALB/c mice were adoptively transferred with DO11.10 cells, carrying a transgenic T cell receptor for a specific ovalbumin peptide, and immunized with ovalbumin conjugates. The number of transgenic CD4+ T cells was determined by flow cytometry of splenic cells. To analyze germinal center formation, spleens were isolated from conjugate immunized BALB/c mice and flow cytometry was performed to determine the number germinal center B cells. Also spleen sections were immunofluorescent stained and the percentage of germinal center containing follicles was determined by counting the germinal centers and follicles.
The primary ovalbumin-specific IgG antibody response in this study was significantly enhanced in mice immunized with ovalbumin-anti-CD23 conjugates compared to the mice immunized with the isotype conjugates or ovalbumin alone and compared to the immunized CD23 deficient mice. The proliferation of ovalbumin-specific CD4+ T cells was not enhanced in mice immunized with ovalbumin-anti-CD23 conjugates in comparison with the control groups. The ovalbumin-specific T cell response was even significantly lower than the mice immunized with isotype conjugates. The percentage of germinal center B cells in the mice immunized with ovalbumin-anti-CD23 conjugates was significantly higher than the percentage of the ovalbumin alone group. Also the percentage of germinal center containing follicles was higher in mice immunized with ovalbumin-anti-CD23 conjugates compared to the isotype control group and ovalbumin alone group. The increased germinal center formation in the mice immunized with ovalbumin-anti-CD23 conjugates is in correlation with the enhanced ovalbumin-specific antibody response.
In conclusion, ovalbumin-anti-CD23 conjugates immunized in mice enhance ovalbumin-specific immune responses in a CD23-dependent manner.
Table of contents
1
INTRODUCTION ... 5
2
THEORETICAL BACKGROUND ... 6
2.1 Humoral immune response ... 6
2.1.1 Spleen structure ... 6
2.1.2 T cell-dependent antibody response ... 7
2.1.3 Antigen transport to follicles ... 8
2.2 CD23 ... 9
2.2.1 Structure ... 9
2.2.2 Functions ... 10
3
MATERIAL AND METHODS... 12
3.1 Mice ... 12
3.2 Antibody purification... 12
3.3 Conjugation of OVA and B3B4/C0H2 antibody ... 13
3.4 Purification of conjugates ... 13
3.5 OVA-specific IgG by ELISA ... 14
3.6 OVA-specific T cell proliferation ... 14
3.7 Germinal center formation ... 15
3.8 Statistical analysis... 15
4
RESULTS ... 16
4.1 B3B4 specificity... 16
4.2 Conjugation of OVA and B3B4/C0H2 antibody ... 16
4.3 OVA-specific antibody response ... 17
4.4 OVA-specific T cell proliferation ... 19
4.5 Germinal center formation ... 19
5
DISCUSSION AND CONCLUSIONS ... 21
6
RECOMMENDATIONS ... 23
1 Introduction
To initiate a humoral immune response, antigens or antigens in complex with antibodies must be transported into the follicles of peripheral lymphoid organs, because they do not have access to the follicles by themselves. Different cells can transport these antigens into the splenic follicles. It has been shown that CD23+ B cells in vivo can transport IgE-antigen complexes and deliver the antigen to CD11c+ cells, which capture and present the antigen to CD4+ cells. CD23 is the low affinity receptor for IgE and is primarily found on B cells and follicular dendritic cells (FDCs) in mice. CD23 has an important role in enhancing immune responses of IgE-antigen complexes and in their transport. CD23 and not IgE, is the necessary molecule to enhance these immune responses. This was shown by Squire, C.M. et al, 1994, where ovalbumin (OVA) was covalently coupled to anti-CD23 (B3B4) antibodies. The same level of immune enhancing effect was seen as when OVA in complex with IgE antibodies was used.
In this study the aim was to investigate whether antigen-B3B4 conjugates can facilitate antigen transport into the B cell follicles and enhance an antigen-specific immune. Using CD23 as a transporting molecule, the antigen-B3B4 conjugates can be transported to the follicles of the spleen. By using B3B4 antibodies instead of IgE antibodies, unwanted effects of degranulation of mast cells will be avoided. OVA is used as antigen in this study. First, it was investigated if the same antibody enhancement could be seen when OVA is coupled to the anti-CD23 antibody B3B4, as described by Squire, C.M. et al, 1994. When this method was successfully set up, there was also investigated whether the OVA-specific T cell proliferation could be enhanced by OVA-B3B4 conjugates and whether germinal center formation was increased.
In the next chapter the theoretical background about the events in a humoral immune response and CD23 and its structure and functions will be described. The material and methods are described in chapter 3. Subsequently in chapter 4 the results are described and chapter 5 will present the discussion and conclusions. Finally chapter 6 will represent recommendations in relation to this project.
2 Theoretical background
2.1 Humoral immune response
A humoral immune response is the immune response mediated by antibodies, produced by a subset of B lymphocytes called plasma cells. This type of response can be stimulated by both protein and non-protein antigens. Antigens in tissues end up in the lymph nodes and blood-borne antigens end up in the spleen, where an adaptive immune response is initiated [1]. Before describing the processes involved in initiating humoral immune responses, the structure of the spleen will be discussed. In this report the focus will be on the humoral immune response of blood-borne antigens.
2.1.1 Spleen structure
The spleen is a peripheral lymphoid organ whose main functions are to remove older erythrocytes, blood-borne microorganisms and cellular debris from the circulation. Also initiation of adaptive immune responses takes place in the spleen.
The spleen can be divided into the red pulp, the blood-filled region, and the lymphocyte-rich white pulp. The many arterial vessels in the spleen contribute to efficient blood-filtering. Blood enters the spleen through the afferent splenic artery, which divides in smaller branches in the spleen (Fig. 1a). Some of these smaller arteries end up in the red pulp, where the blood is filtered from microbes, damaged cells, antibody-coated cells and microbes. Other branched arteries, called central arterioles, end up in the marginal sinus, an area in the marginal zone. The marginal zone is the region between the red and the white pulp and serves to transit cells from the blood to the white pulp. Cells found in the marginal zone include marginal zone macrophages (outer layer), marginal zone metallophilic macrophages (inner layer), marginal zone B cells and dendritic cells (DCs) (in between the layers). The white pulp is the lymphocyte-rich region in the spleen where the adaptive immune response is initiated. It consists of two main compartments, the T cell zone and the B cell zone (B cell follicle) (Fig. 1b) [1, 2].
Figure 1: Schematic overview of the spleen (a). Structure of the white pulp in the spleen in mice and human (b) [2]. In the B cell follicles, germinal centers arise after antigen exposure. Germinal centers are structures in the follicle where affinity maturation, isotype switching, generation of B cells and long-lived plasma cell differentiation takes place. It consists of a dark zone and a light zone, where the dark zone contains proliferating B cells and the light zone contains mostly FDCs, non-proliferating B cells, naive B cells and follicular helper T cells (TFH cells) [3]. Most of the events during a humoral immune
response against blood-borne protein antigens occur in the white pulp of the spleen. These events will be discussed in the next paragraph.
2.1.2 T cell-dependent antibody response
T cell-dependent responses occur in the presence of protein antigens and require activation by the protein antigen of both helper T cells and B cells. When antigen presenting cells (APCs), mostly DCs, encounter an antigen, they can capture, internalize and process the antigen to subsequently present antigenic peptides via class II major histocompatibility complex (MHC) molecules to naive CD4+ helper T cells (Fig. 2). This process occurs in the T cell zones of peripheral lymphoid organs. The helper T cells are then initially activated and proliferate and migrate toward the follicles.
B cells are being activated by the antigen in the follicles. The protein antigen binds to the B cell receptor (membrane bound IgM or IgD) and the antigen will be internalized, processed and presented via class II MHC molecules on the B cells. Subsequently, the B cells migrate toward the T cell zones. At the boundary of the T cell zone and the follicle, the activated B cells present the antigen to activated helper T cells. The cells then interact and some of the activated helper T cells
differentiate into TFH cells which migrate into the follicles. Also some of the activated B cells migrate
back to the follicles and proliferate (in the dark zone of germinal centers) [1].
Figure 2: Sequence of events in humoral immune responses to T cell-dependent protein antigens [1].
In the germinal center, germinal center B cells undergo isotype switching leading to the production of IgG, IgE or IgA. First, in the dark zone, germinal center B cells will proliferate and somatic mutation takes place in the variable region of the immunoglobulin genes which leads to different affinities for the antigen. The B cells migrate to the light zone of the germinal center where FDCs present the antigen to these B cells. The antibodies on the B cells with the highest affinity bind to the antigen and survive. Antibodies on the B cells with low affinity for the antigen undergo apoptosis. This process gets help from TFH cells by promoting the survival of high affinity B cells. These B cells then undergo
isotype switching by recombination of the variable heavy chain and constant region genes and differentiate into antibody-secreting plasma cells or memory cells. The antibody-secreting plasma cells migrate to the bone marrow and the memory cells remain partly in the lymphoid organ and others migrate out of the germinal centers to recirculate between the lymphoid organs and the blood. Because memory B cells are generated, a second exposure to the same antigen will cause a more rapid and larger immune response then the first exposure. This is because memory B cells do not have to undergo the whole process again of affinity maturation and isotype switching [1, 3].
2.1.3 Antigen transport to follicles
Before a humoral response to blood-borne antigens can be initiated, antigens or antigen-antibody complexes must be transported to the follicles in the spleen, because they have not access to the follicles by themselves. Different cells can transport these antigens into the splenic follicles, like macrophages, follicular or plasmacytoid dendritic cells or marginal zone B cells [1].
When antigens in complex with IgM and complement arrive in the marginal zone, they bind to complement receptor 1/2 on marginal zone B cells, which transport the antigen into the follicles to FDCs [4, 5]. Small soluble blood-borne antigens and molecules can be distributed through the splenic white pulp by a stromal network in the spleen called the splenic conduit system. This distribution was shown to be size-dependent. Small molecules were shown in the conduit system in the white pulp whereas bigger molecules were not seen in the conduit system, but only in the red pulp and in some
blood vessels [6]. In addition, it has been shown that IgE-antigen complexes can be transported to splenic follicles by CD23+ B cells [7], which will be discussed more in detail later.
In this study, the focus will be on CD23 and its function as a transporting molecule on B cells. More detailed information about CD23 will be discussed in the next chapter.
2.2 CD23
There are two different Fc receptors that can bind to IgE. One of them is the high affinity receptor, FcεRI, expressed on mast cells and basophils. The binding of IgE to this receptor plays a role in allergic responses. The low affinity receptor, FcεRII or CD23, is the second receptor for IgE. In both humans and mice, CD23 exists in two forms, CD23a and CD23b. CD23a is primarily expressed on B cells in humans [8], but it has also been seen on human intestinal epithelial cells [9]. In mice, this receptor has been found on primarily FDCs and B cells [10].
CD23b is expressed on several cells in humans, including B cells, some T cells, eosinophils and lung and intestinal epithelial cells [8, 11, 12]. In mice the CD23b form is found on enterocytes in the intestine [13].
2.2.1 Structure
In contrast to most other immunoglobulin receptors, CD23 belongs to the C-type lectin family [14]. The receptors belonging in this family are calcium-dependent and bind carbohydrates [1]. CD23 exists as a membrane-bound form and a soluble form. The membrane-bound form of CD23 consists of an extracellular trimeric α-helical coiled-coil stalk ending in three C-type lectin domain heads and a short intracellular N-terminal (figure 3). The stalk region is cleavable, resulting in soluble CD23, which have cytokine-like activities. The intracellular N-terminal has two splice variants, resulting in CD23a or CD23b. In the head domain are binding sites for IgE, CD21 (complement receptor 2) and for some integrins, including CD11b and CD11c [14].
2.2.2 Functions
Studies have indicated that CD23 can function as a negative regulator for IgE. In transgenic mice, overexpressing CD23, production of IgE was reduced [15]. In contrast, increased levels of IgE were found in CD23 deficient (CD23-/-) mice [16].
Besides as a function in IgE regulation, CD23 has a role in enhancing antibody responses. When OVA-2,4,6-trinitrophenyl (TNP), was administered in mice together with TNP-specific IgE, a >100 fold higher antibody response was seen than when OVA was administered alone [17]. In addition, antigen-specific IgE-mediated enhancement was impaired in CD23-/- mice [18], however the enhancement was restored when CD23+ B cells were transferred to CD23-/- mice [19]. Also an enhanced OVA-specific T cell response was seen and both the enhanced antibody response and T cell response was shown to be CD23-dependent [20].
As mentioned earlier, CD23 plays a role in antigen transport. In vivo, CD23+ B cells can transport IgE-antigen complexes into the follicles of the spleen. OVA-TNP administered together with TNP-specific IgE in mice, was found on B cells 5 minutes after immunization and on follicular B cells in splenic follicles after 30 min [7]. In vitro it has been shown that CD23+ B cells are able to capture IgE-antigen complexes via CD23, endocytose and present the antigen peptides to CD4+ T cells [21, 22].
First, it was thought that in vivo CD23+ B cells not only transport the IgE-antigen complexes into the follicles, but also endocytose and present the antigen like they can in vitro. However, a study has shown that CD23+ B cells in vivo only transport the IgE-antigen complex and deliver the antigen to CD11c+ cells, which capture and present the antigen to CD4+ cells [23].
CD23 does not only play a role in antigen transport to the follicles of the spleen, but also to other locations. For instance, CD23 mediates transepithelial antigen transport via IgE in intestinal epithelial cells of mice [13, 24], rats [25] and humans [9, 12]. It is also demonstrated that CD23 on human respiratory epithelial cells can transport IgE [11]. Furthermore, it is thought that IgE is selectively transported by probably CD23, to mammary secretions of sheep [26].
The above mentioned ways of transportation all involves CD23 and IgE. CD23 has a role in enhancing antigen specific immune responses by enhanced antigen transport into the follicle. CD23, and not IgE, is the necessary molecule to enhance these immune responses. This was shown by a study, where OVA was covalently coupled to B3B4 antibodies. The same level of immune enhancing effect was seen as when OVA in complex with IgE antibodies was used. Immunizing mice with OVA-B3B4 conjugates resulted in a significant IgG1 response and a detectable IgE response. However, when administering OVA alone, no detectable response was seen [27].
In this study we used the transport function of CD23 on B cells to investigate whether antigen-B3B4 conjugates can facilitate antigen transport into the B cell follicles and enhance an antigen-specific immune response. Using CD23 as a transporting molecule, the antigen-B3B4 complex can be efficiently transported to the follicles of the spleen (Fig.4).
Figure 4: OVA in complex with IgE bound to CD23 on a B cell (A). OVA conjugated to anti-CD23 bound to CD23 on a B cell (B).
3 Material and methods
3.1 Mice
The mice used in this project were wild type BALB/c mice, CD23-/- mice and DO11.10 mice, all with a BALB/c background. DO11.10 mice are transgenic BALB/c mice having T cell receptors specific for OVA peptides, presented by MHC class II molecules. The receptor is expressed in 80-90% of the T cells in the animal. All mice used were 9-23 weeks old females and the age was matched within every experiment.
3.2 Antibody purification
Cell cultureHybridomas, producing anti-CD23 rat monoclonal antibodies (B3B4) (IgG2a isotype) and non-specific isotype control antibodies (C0H2) were cultured in Dulbecco's Modified Eagle's Medium (Sigma-Aldrich) supplemented with 0.01 M HEPES, 100 U/ml penicillin, 100 μg/ml streptomycin, 0.05 mM β-mercaptoethanol, 1 mM sodium pyruvate, 2 mM L-glutamin and 5% fetal calf serum (FCS) (all from Sigma-Aldrich).
Affinity chromatography
The antibodies were purified from supernatant from the cultured hybridomas using a protein G Sepharose column. First the column was equilibrated with binding buffer (20 mM sodium phosphate with 0.02% NaN3, pH 7). Filtered (0.45 μm membrane filter, Millipore) supernatant with 0.02% NaN3
was run over the column and the bound antibodies were eluted with elution buffer (0.1 M glycine-HCl with 0.02% NaN3, pH 2.5-3.5) in 2 ml fractions. Antibody containing fractions were determined by
absorbance at 280 nm and were pooled. The pooled antibodies were concentrated and washed with phosphate buffered saline (PBS) using a 50000 MW Amicon Ultra Centrifugal Filter (Millipore). Finally, the purified antibodies were sterile filtered with a sterile syringe filter (0.2 μm Cellulose Acetate Membrane, VWR International).
ELISA
To confirm the specificity of B3B4 antibodies, an enzyme-linked immunosorbent assay (ELISA) was performed. 96-well EIA/RIA plates (Costar) were coated with the purified B3B4 (20, 40 and 100 μg/ml) in PBS with 0.02% NaN3 at 4° C overnight. As a positive control, B3B4 (40 μg/ml), with a
known high specificity, was coated. Plates were blocked with PBS with 5% dry milk and 0.02% NaN3
for 3 hours at room temperature. Subsequently, recombinant CD23 in dilution buffer (PBS with 0.25% dry milk, 0.05 % Tween 20 and 0.02% NaN3) was added in different concentrations. The plates
were incubated at 4° C overnight. As primary antibody, rabbit polyclonal anti-CD23 IgG (Santa Cruz Biotechnology) was added to the plates and incubated 3 hours at room temperature. Subsequently, alkaline phosphatase (AP)-goat anti-rabbit IgG (Invitrogen) was added as secondary antibody and again the plates were incubated for 3 hours at room temperature. Finally phosphatase substrate (Sigma Aldrich) (1 mg substrate/ml 1M diethanolamine buffer with 0.5 mM MgCl2, pH 9.8) was added
(Molecular Devices) was used. Between all steps, the plates were washed with PBS and/or with PBS with 0.05 % Tween 20. All incubation steps were performed in a humid chamber except after adding the substrate.
3.3 Conjugation of OVA and B3B4/C0H2 antibody
ConjugationTo conjugate OVA to B3B4 or C0H2 antibodies, OVA was thiolated using Traut’s reagent (Thermo Scientific) according the manufactures instructions to introduce sulfhydryl groups to OVA. B3B4 or C0H2 antibodies were modified using Sulfo-MBS (m-maleimidobenzoyl-N-hydoxysuccinimide ester) (Thermo Scientific) according the manufactures instructions. Sulfo-MBS adds maleimide groups to the antibodies which can react with the sulfhydryl groups on OVA resulting in a stable thioether bond. Shortly, OVA was treated 1.5 hours at room temperature with a 15 fold molar excess of Traut’s reagent. B3B4 and C0H2 antibodies were treated 1 hour at room temperature with a 50 fold molar excess of Sulfo-MBS. Excess of reagent was separated from the modified OVA and B3B4 and C0H2 antibodies by using Zeba Spin Desalting Columns (Thermo Scientific). Finally the modified OVA and B3B4 or C0H2 were combined with a molar ration of 7:1 or 4:1 and incubated for 2 hours at room temperature.
SDS-PAGE
To analyze the conjugates, a non-reducing SDS-PAGE was performed according to the manufacturer’s instructions (Life Technologies). Protein samples were prepared with Bolt SDS sample buffer (4x) (Life Technologies) and separated by electrophoresis on a 4-12% Bis-Tris Plus gel (Life Technologies) in Bolt MES SDS running buffer (1x) (Life Technologies). SeeBlu Plus2 Pre-stained standard (Invitrogen) was used as a protein size marker.
The proteins on the gel were visualized by staining with 0.25% Coomassie Brilliant Blue R-250 (Sigma Aldrich) in 40% methanol and 10% acetic acid overnight. Destaining was performed using a 40% methanol and 10% acetic acid solution.
3.4 Purification of conjugates
Two batches of conjugates were obtained. One batch was purified by size-exclusion chromatography and the other batch by centrifugal filtering.
Size-exclusion chromatography
To purify the conjugates from free OVA, one batch of conjugates was purified by size-exclusion chromatography (LCC-500 and all other equipment from Farmacia) with a Sepharose Cl6b column. Before start, the column was equilibrated with degassed PBS with 0.04% NaN3. Subsequently, the
protein sample was loaded via the loop and elution was performed with degassed PBS with 0.04% NaN3 in 9 ml fractions using a fraction collector Frac-100 (Farmacia). For all steps, a flow rate of 0.3
ml/min was used. The elution profile was generated by measuring absorbance at 280 nm. To analyze the protein containing fractions, SDS-PAGE was performed as described before.
Centrifugal filtering
The second batch of conjugates was purified using 100.000 MW Amicon Ultra Centrifugal Filters (Millipore). The OVA (45 kDa) will be in the flow-through, whereas the conjugates (> 195 kDa) will not pass the filter. The centrifugal steps were performed 6 times with PBS at 4000 rpm for 15 minutes in a Heraeus Megafuge 40R Centrifuge (Thermo Scientific).
Again, purified conjugates were analyzed by SDS-PAGE as described before.
3.5 OVA-specific IgG by ELISA
BALB/c mice and CD23-/- mice were immunized intravenously (i.v.) in the tail veins with purified OVA-B3B4, OVA-C0H2 or OVA alone in PBS (for concentrations see figure legends). Blood samples were taken from the tail on day 7, 14 and 21 after immunizations. After 3 weeks of rest, mice were immunized again with OVA alone in PBS. Again blood samples were taken at different time points after immunization.
Serum samples were measured for OVA-specific IgG by ELISA. The ELISA was mainly performed under the same conditions as described before. Shortly, plates were coated with OVA (50 μg/ml) (Sigma Aldrich) in PBS with 0.02% NaN3 at 4° C overnight. Then a blocking step was performed and sera in
different dilutions were added. Known amounts of IgG anti-OVA were used as standards. Finally, AP-anti-IgG sheep anti-mouse (Jackson Immuno Research) was added as enzyme-linked antibody and the subsequent steps are similar as described before.
3.6 OVA-specific T cell proliferation
Adoptive transferDO11.10 mice were sacrificed by cervical dislocation and spleens were isolated and smashed to make single cell suspensions. To remove erythrocytes, the single cell suspensions were treated with Ammonium-Chloride-Potassium (ACK) lysing buffer (0.15 M NH4CL, 1 M KHCO3, 0.1 mM Na2EDTA, pH
7.3) for 3 minutes and washed and resuspended in PBS. During these steps, the cells were kept on ice. The lymphocytes with CD4+ OVA-specific T cells were adoptively transferred i.v. in the tail veins of BALB/c mice in 200 μl PBS.
The next day BALB/c mice were immunized i.v. in the tail veins with purified OVA-B3B4, OVA-C0H2 or OVA alone in 200 μl PBS (for concentrations see figure legends).
Flow cytometry
3 days after immunizations, the BALB/c mice were sacrificed by cervical dislocation and spleens were isolated. Single cell suspensions were treated with ACK as described before, washed with PBS and resuspended in FACS buffer (PBS with 2% FCS). Cells were stained in 100 μl FACS buffer for 30 minutes at 4° C in the dark with PE-labeled anti-CD4 (GK1.5) (e Bioscience) to select CD4+ T cells and FITC-labeled anti-DO11.10 T cell receptor (KJ1-26) (e Bioscience) to select transgenic OVA-specific T cells. After staining, the cells were washed twice with FACS buffer. Lymphocytes were selected using the forward and side scatter with a FACScan (BD Biosciences) and analyzed using FlowJo software. OVA-specific DO11.10 T cells were identified as CD4+KJ126+.
3.7 Germinal center formation
Flow cytometryBALB/c mice were immunized i.v. in the tail veins with purified OVA-B3B4, OVA-C0H2 or OVA alone in 200 μl PBS (for concentrations see figure legends). 10 days after immunizations, mice were sacrificed by cervical dislocation and spleens were isolated, and half the spleen was stored in PBS for analysis with flow cytometry and the other half was frozen in Tissue-Tek O.C.T. Compound (Sakura Finetek) for immunofluorescence staining (see next paragraph).
The half spleens were prepared for flow cytometry as described before. For half a spleen, only 2 ml of ACK was used instead of 3 ml. Cells were stained in 100 μl FACS buffer with CD16/CD32 Fc-block (2.4G2) (BD Biosciences) for 30 minutes at 4° C with PaBl-labeled anti-B220 (RA3-6B2) (BD Biosciences) to select B cells and biotinylated PNA (self-made) and Alexa 647-labeled GL7 (BD Biosciences) to select germinal center B cells and incubated en washed as described before. Because biotinylated PNA was used, second staining step was performed with PE-labeled Streptavidin (SA) (BD Biosciences). For compensating, beads (CompBeads, BD Biosciences) were stained for 15 minutes in the dark at room temperature with PE-labeled anti-B220 (RA3-6B2) (e Bioscience), PaBl-labeled anti-B220 (RA3-6B2) (BD Biosciences) or Alexa 647-labeled anti-CD34 (RAM34) (BD Biosciences). After staining, the beads were washed with FACS buffer. Lymphocytes were selected using the forward and side scatter with a LRS II (BD Biosciences) and analyzed using FlowJo software. Germinal center B cells were identified as B220+GL7+PNA+.
Immunofluorescence staining
8 μm sections were cut of the half spleens using a Cryostar NX70 (Thermo Scientific) at a temperature of -20° C and collected on microscope slides (Menzel-Gläser, Menzel GmbH). Slides were fixed with 4% paraformaldehyde and rehydrated in PBS twice for 5 minutes at room temperature. Subsequently, the sections were blocked with 5% horse serum (Sigma-Aldrich) in PBS for 30 minutes at room temperature. Sections were stained for 1 hour at room temperature in the dark in 100 μl 5% horse serum with PaBl-labeled anti-B220 (RA3-6B2) (BD Biosciences) to visualize B cells, FITC-labeled MOMA (AbD SeroTec) for metallophilic macrophages and biotinylated PNA (self-made) for germinal center B cells (all with a final concentration of 2 μg/ml). The slides were washed twice with PBS for 5 minutes. To detect the biotinylated PNA, the staining step was repeated with PE-labeled SA (BD Biosciences). The slides were washed again twice with PBS for 5 minutes. Finally, the slides were mounted in Fluoromount G (Southern Biotech) and air-dried in the dark. Fluorescence was detected with a LSM700 confocal microscope (Carl Zeiss) and the sections were analyzed using Zen 2009 software. The percentage of germinal center containing follicles was determined by counting the number of follicles and germinal centers in the spleen sections.
3.8 Statistical analysis
Statistical differences between the experimental and control groups were determined by Student’s t-test. The values p>0.05 (ns, not significant), p<0.05 (*), p<0.01 (**) or p<0.001 (***) are indicated in the figures.
4 Results
4.1 B3B4 specificity
In this study, the aim was to investigate whether OVA conjugated to B3B4 antibodies can facilitate antigen transport via CD23 to the B cell follicles and enhance an antigen-specific immune response. Therefore, B3B4 and isotype control antibodies were purified from hybridoma supernatant and subsequently conjugated to OVA. Before conjugation the specificity of B3B4 for CD23 was tested by ELISA. The purified B3B4 showed specificity for recombinant CD23 comparable with the control B3B4 antibody (Fig. 5). The absorbance of the 20 μg/ml and 40 μg/ml B3B4 coating is almost similar (around 2.5) when a concentration of 1 μg/ml recombinant CD23 is used. The 100 μg/ml B3B4 coating shows an absorbance around 3.0 by that concentration.
0.0 0.2 0.4 0.6 0.8 1.0 0 1 2 3 4 control B3B4 20 µg/ml B3B4 40 µg/ml B3B4 100 µg/ml Concentration recombinant CD23 (µg/ml)
A
b
s
o
rb
a
n
c
e
(
4
0
5
n
m
)
Fig 5: B3B4 specificity for recombinant CD23. The specificity of B3B4 for recombinant CD23 was determined by ELISA by using different concentrations of B3B4 and recombinant CD23. As control, B3B4 (40 μg/ml), with a known high specificity, was used.
4.2 Conjugation of OVA and B3B4/C0H2 antibody
OVA was conjugated to purified B3B4 or C0H2 antibodies. First, OVA was conjugated to B3B4 in a molar ratio of 7:1 and the conjugates were purified by size-exclusion chromatography (Fig. 6A). Because there was still a lot of unreacted OVA left, the subsequent conjugations were performed in an OVA:B3B4/C0H2 molar ratio of 4:1 and an alternative method for purification (centrifugal filtering) of the conjugates was used (Fig. 6B). In all conjugations, no free antibodies could be seen, which means that the antibodies had completed the reaction with OVA molecules. Purification by size-exclusion chromatography led to the complete removal of unreacted OVA. However, by centrifugal filtering, there was a small amount of unreacted OVA left after purification of OVA-C0H2.
Figure 6: OVA-B3B4 conjugates purified by size-exclusion chromatography (A). OVA-B3B4 and OVA-C0H2 conjugates purified by centrifugal filtering (B). The proteins were separated on a 4-12% Bis-Tris Plus gel under non-reducing conditions and visualized by Coomassie staining. 190 kDa IgE was included as a size marker.
4.3 OVA-specific antibody response
To investigate whether the OVA-B3B4 conjugates can enhance OVA-specific antibody responses, BALB/c and CD23-/- mice were immunized with size-exclusion purified OVA-B3B4 conjugates or OVA alone in PBS. Blood samples were taken on day 7, 14 and 21 after immunization. The mice were rested for 3 weeks and a secondary immunization (boost) was performed with 50 μg OVA in PBS per mouse. Again, blood samples were taken after 7 and 32 days. OVA-specific IgG in serum was detected by ELISA (Fig. 7). On day 7 and 14 after primary immunization, OVA-specific IgG levels in BALB/c mice immunized with OVA-B3B4 conjugates were twice as high as the control groups and on day 21 they were 1.5 times higher. The OVA-specific IgG response is extremely high and more equalized between the groups on day 7 and 32 after secondary immunization.
Day 7 BA LB/c OVA -B3B 4 BA LB/c OVA CD23 -/- O VA -B3B 4 CD 23-/- OVA 0 50 100 150 O V A -s p e c if ic I g G ( u g /m l) Day 14 BA LB/c OVA -B3B 4 BA LB/c OVA CD23 -/- O VA -B3B 4 CD 23-/- OVA 0 50 100 150 O V A -s p e c if ic I g G ( u g /m l) Day 21 BA LB/c OVA -B3B 4 BA LB/c OVA CD23 -/- O VA -B3B 4 CD 23-/- OVA 0 50 100 150 O V A -s p e c if ic I g G ( u g /m l) Day 7 boost BA LB/c OVA -B3B 4 BA LB/c OVA CD 23-/- OVA -B3B 4 CD 23-/- OVA 0 5000 10000 15000 O V A -s p e c if ic I g G ( u g /m l) Day 32 boost BA LB/c OVA -B3B 4 BA LB/c OVA CD 23-/- OVA -B3B 4 CD23 -/- O VA 0 500 1000 1500 2000 2500 O V A -s p e c if ic I g G ( u g /m l)
Figure 7: OVA-specific IgG in conjugate immunized mice. BALB/c and CD23-/- mice were i.v. immunized with 0.2 ml size-exclusion purified OVA-B3B4 (OD: 0.80) (one mouse only received 0.1 ml) or 100 μg OVA in PBS. The primary OVA-specific IgG levels were measured on day 7, 14 and 21 after immunization. Subsequently, after 3 weeks of rest, the mice had a secondary immunization (boost) with 50 μg OVA in PBS. Secondary OVA-specific IgG levels were measured on day 7 and 32 after secondary immunization. Graphs show mean ± the standard error of the mean.
Because the amount of OVA molecules immunized in the form of OVA-B3B4 is unknown, the OVA and OVA-B3B4 groups are scientifically not comparable. Therefore, the experiment was repeated with OVA-C0H2 conjugates as isotype control to compare between OVA-B3B4 and OVA-C0H2. This time, only BALB/c mice were immunized with B3B4, C0H2 or OVA alone in PBS. OVA-specific IgG was detected on day 7, 14 and 21 after immunization (Fig. 8). One of the mice immunized with OVA-B3B4 conjugates had similar levels as the control groups and the OVA-specific IgG levels were more varying in this group than the control groups. However, the mice immunized with OVA-B3B4 conjugates showed a significantly higher OVA-specific IgG response than the control groups.
Day 7 OV A-B 3B4 OV A-C 0H2 OVA 0 20 40 60 80 ** ** O V A -s p e c if ic I g G ( u g /m l) Day 14 OVA -B3B 4 OVA -C0H 2 OVA 0 20 40 60 80 ** ** O V A -s p e c if ic I g G ( u g /m l) Day 21 OV A-B 3B4 OV A-C 0H2 OVA 0 20 40 60 * ** O V A -s p e c if ic I g G ( u g /m l)
Figure 8: OVA-specific IgG in conjugate immunized mice. BALB/c mice were i.v. immunized with 0.2 ml centrifugal filtering purified OVA-B3B4 (OD: 0.76), OVA-C0H2 (OD: 0.70) or 100 μg OVA in PBS. The primary OVA-specific IgG levels were
4.4 OVA-specific T cell proliferation
Next, it was analyzed whether OVA-B3B4 conjugates could induce a proliferation in OVA-specific T cells. BALB/c mice were adoptively transferred with DO11.10 splenic cells and after 1 day, i.v. immunized with OVA-B3B4, OVA-C0H2 or OVA alone in PBS. 3 days after immunization, the number of OVA-specific CD4+ T cells was analyzed by flow cytometry. This experiment was performed twice and in both experiments there were significantly less OVA-specific CD4+ T cells in the mice immunized with OVA-B3B4 conjugates than in the mice immunized with OVA-C0H2 (Fig. 9), although an increase was expected.
OVA-specific T cell proliferation
OVA -B3B 4 OVA -COH2 OVA 0 2 4 6 8 * * % O V A -s p e c if ic CD4 + T c e lls o f to ta l CD4 + T c e lls
OVA-specific T cell proliferation
OVA -B3B 4 OVA -COH2 OVA 0 1 2 3 4 5 * ns % O V A -s p e c if ic CD4 + T c e lls o f to ta l CD4 + T c e lls A B
Figure 9: OVA-specific T cell proliferation in conjugate immunized mice. A and B represent two different experiments. BALB/mice were adoptively transferred with DO11.10 splenic cells and after 1 day, i.v. immunized with 0.2 ml centrifugal filtering purified OVA-B3B4 (OD: 0.76), OVA-C0H2 (OD: 0.70) or 100 μg OVA in PBS. 3 days after immunization, the number of OVA-specific CD4+ T cells was analyzed by flow cytometry. Graphs show mean ± the standard error of the mean.
4.5 Germinal center formation
To analyze germinal center formation in splenic follicles from mice immunized with OVA-B3B4 conjugates, BALB/c mice were immunized with OVA-B3B4, OVA-C0H2 or OVA alone in PBS. After 10 days, the spleens were analyzed by flow cytometry to determine the number of germinal center B cells of total B cells and by confocal microscopy to determine the percentage of germinal center containing follicles.
The number of germinal center B cells in the mice immunized with OVA-B3B4 conjugates was higher than the control groups (Fig. 10A). Even though the number was significantly higher than the OVA control group, it was not significantly higher than the OVA-C0H2 control group.
The percentage of germinal center containing follicles was significantly higher than the C0H2 control group (Fig. 10B). The percentage of germinal center containing follicles was determined by counting the number of follicles and germinal centers in the spleen sections (Fig. 10C).
Germinal center B cells OV A-B 3B4 OV A-C0 H2 OVA 0.0 0.5 1.0 1.5 ns * % g e rm in a l c e n te r B c e lls o f to ta l B c e lls Germinal centers OV A-B 3B4 OV A-C0H2 OV A 0 20 40 60 * ns % g e rm in a l c e n te r c o n ta in in g f o lli c le s A B
Figure 10: Germinal center formation in conjugate immunized mice. BALB/c mice were i.v. immunized with 0.2 ml centrifugal filtering purified OVA-B3B4 (OD: 0.76), OVA-C0H2 (OD: 0.70) or 100 μg OVA in PBS. Levels of germinal center B cells were determined by flow cytometry (A) and the amount of germinal center containing follicles was determined by confocal microscopy (B). Example of B220 (blue) MOMA (green) and PNA (red) stained spleen section with follicles and germinal centers (C). Graphs show mean ± the standard error of the mean.
1 mm germinal center follicle
C
germinal center5 Discussion and conclusions
Previous studies have demonstrated that IgE-antigen complexes enhance antigen-specific immune responses [17, 18, 19] and that CD23 on B cells transport these complexes to the follicle of the spleen [7, 23].
The aim of this study was to investigate whether antigens conjugated to B3B4 antibodies can facilitate antigen transport into the B cell follicles and enhance an antigen-specific immune response. As antigen, OVA was used and covalently coupled to B3B4 and isotype control C0H2 antibodies. First, the conjugates were purified by size-exclusion chromatography. Because this purification step took 3 days, we changed to another method that took only one afternoon, namely centrifugal filtering. As a non-specific control in this study OVA-C0H2 was used to directly compare with OVA-B3B4. Because it is unknown how many OVA molecules are coupled to B3B4 or C0H2 antibodies and thus immunized, OVA alone is not the best suitable control. However OVA alone is included because a minimum enhanced antibody response was expected, as shown before [17, 28].
Only one publication can be found where OVA-specific antibody responses were determined after immunization in mice with OVA-B3B4 conjugates [27]. There they showed an enhanced antigen-specific immune response in the mice immunized with OVA-B3B4 conjugates compared to the control groups. However they only used wild type BALB/c mice, not CD23-/- mice.
The primary OVA-specific IgG antibody response in the current study was significantly enhanced in mice i.v. immunized with OVA-B3B4 conjugates compared to the mice immunized with OVA-C0H2 or OVA alone and compared to the immunized CD23-/- mice. This shows that the enhancement is CD23 dependent and that our method works. However the secondary antibody response did not show an enhancement compared to the control groups. This is in contrast with the data published by Squire, C.M. et al, 1994. They show that also the secondary response was enhanced after secondary subcutaneously immunization with OVA or OVA-B3B4 conjugates [27]. However they measured only the OVA-specific IgG1 antibody response, whereas in our study, total IgG levels were determined and secondary immunizations were done with OVA alone, not B3B4 conjugates. The total OVA-specific IgG levels after secondary immunization were extremely high, which could be the reason that they reached a certain level, too high to be reasonable values to compare.
The proliferation of OVA-specific CD4+ T cells was not enhanced in mice immunized with OVA-B3B4 conjugates in comparison with the control groups. The OVA-specific T cell response was even significantly lower than the mice immunized with OVA-C0H2. This was not expected because the OVA-specific antibody response is enhanced. To get this antibody response, B cells need help from OVA-specific CD4+ T cells. Earlier it was shown in our group that the OVA-specific T cell proliferation was significantly enhanced in mice immunized with OVA-TNP together with IgE anti-TNP, compared to mice immunized with OVA-TNP alone [23]. In that study 20 µg of OVA-TNP was immunized in mice and in our study 100 µg of OVA. This could be the reason why the mice immunized with OVA alone in our study show a percentage of 5.5 and 1.8% of OVA-specific CD4+ T cells of total CD4+ T cells and in the study of Henningson, F. et al, 2011 only 0.4%. The OVA-conjugates were prepared by introducing sulfhydryl groups to OVA and adding maleimide groups to the B3B4 and C0H2 antibodies. It could be possible that the reagents modified that particular peptide recognized by the T cell receptor in DO11.10 mice. The T cell receptors in these mice recognize the OVA323-339 peptide [29].
The transgenic T cell receptor is expressed in 80-90% of the T cells in DO11.10 mice, that’s why some peptide modifications could lead to a relatively major decrease in magnification.
A few days after initial exposure to a T cell dependent antigen, germinal centers will arise in the follicles, where affinity maturation, isotype switching, generation of B cells and long-lived plasma cell differentiation takes place. To analyze germinal center formation the number of germinal center B cells and the percentage of germinal center containing follicles was determined. 1.1% germinal center B cells of total B cells were detected in spleens from mice immunized with OVA-B3B4 conjugates over 0.8% in the isotype control group and 0.6% in the OVA alone group. The percentage of germinal center B cells in the mice immunized with OVA-B3B4 conjugates was significantly higher than the percentage of the OVA alone group. Also the percentage of germinal center containing follicles was higher in mice immunized with OVA-B3B4 conjugates (48%) compared to the isotype control group (32%) (significantly higher) and OVA alone group (38%). The increase in germinal centers and germinal center B cells in the mice immunized with OVA-B3B4 conjugates is correlating with the enhanced OVA-specific IgG response.
First, the mechanism behind IgE-mediated immune enhancement was unknown. It was thought that B cells via CD23 capture and present IgE-antigen complexes more efficiently than when antigen alone is administered. This is indicated by several in vitro studies [21, 22].
However in vivo it was shown that CD23+ B cells only transport the IgE-antigen complex to the spleen and that antigen presentation is performed by CD11+ cells (probably DCs) [23]. The enhancing immune response is CD23-dependent [20]. Thus CD23 seems to be an important molecule for antigen transport. The enhancing effect is caused by an efficient transport of antigens to the follicles, which could lead to enhanced concentration of antigens compared to when antigen alone is administered. By replacing IgE for anti-CD23 antibodies, conjugated to antigens, unwanted effects of mast cell degranulation caused by IgE can be avoided. By targeting antigens via CD23 to B cells, efficient transport can lead to enhanced immune responses what could be interesting as clinical purposes, like vaccinations. Some types of vaccines, like subunit vaccines (whole inactivated pathogens, purified protein antigens or recombinant peptides), have the disadvantage of a weakened immunogenicity [30]. To overcome this problem, adjuvants can be added to the vaccines. In this case B3B4 can act like an ‘adjuvant’ in vaccines by enhancing an immune response specific for the antigen caused by the efficient transport via CD23.
In conclusion, however the OVA-specific CD4+ T cell proliferation was not increased in the mice immunized with OVA-B3B4 conjugates, the primary OVA-specific IgG responses was significantly enhanced. Because a low OVA-specific IgG response was seen in CD23-/- mice and in the isotype control group, it can be concluded that the enhancement of the antibody response was CD23-dependent.
6 Recommendations
The secondary antibody response was not enhanced in the mice primary immunized with OVA-B3B4 conjugates compared to the control groups. As secondary immunization 50 µg of OVA per mouse was i.v. administered to the mice. This turned out to result in enormous high levels of OVA-specific IgG. This experiment could be repeated with 5 µg OVA per mouse to obtain more reliable IgG levels. At this moment, it is unknown how many OVA molecules are conjugated to one antibody, leading to an unknown concentration of OVA immunized in the mice. The conjugation could be optimized by using a lower molar excess of OVA during conjugation. By using an OVA:B3B4 ratio of 4:1, there is still some unreacted OVA left. When lowering the molar excess of OVA, all OVA can react with the B3B4. Then it can be estimated how many molecules are bound to the antibody and subsequently the concentration can be determined.
To be able to get an antibody response to protein antigens, B cells need help from specific T cells. In this study the specific T cell proliferation was not enhanced in the mice immunized with OVA-B3B4 conjugates compared to the control groups. It could be that the OVA peptides recognized by the receptor were modified during conjugation. This experiment could be repeated with a new batch of conjugates, preferably with known concentrations and a lower concentration of OVA in the OVA alone group.
In this study OVA is used as antigen conjugated to B3B4 and C0H2 antibodies. To confirm whether the enhanced immune response can also be seen when other antigens are used, all experiments could be repeated by using other antigens, for example bovine serum albumin or keyhole limpet hemocyanin.
References
[1] Abbas, A.K., Lichtman, A.H., Pillai, S. (2011). Cellular and molecular immunology. 7th ed. Philadelphia: Elsevier Inc, Saunders.
[2] Mebius, R.E., Kraal, G. (2005). Structure and function of the spleen. Nature Reviews
Immunology, 5: 606-616.
[3] Victora, G.D., Nussenzweig, M.C. (2012). Germinal centers. Annual Review of Immunology, 30: 429-457.
[4] Ferguson, A.R., Youd, M.E., Corley, R.B. (2004). Marginal zone B cells transport and deposit IgM-containing immune complexes onto follicular dendritic cells. International Immunology, 16: 1411-1422.
[5] Cinamon, G., Zachariah, M.A., Lam, O.M., Foss, F.W. Jr., Cyster, J.G. (2008). Follicular shuttling of marginal zone B cells facilitates antigen transport. Nature Immunology, 9: 54-62.
[6] Nolte, M.A., Beliën, J.A.M., Schadee-Eestermans, I., Jansen, W., Unger, W.W.J., van Rooijen, N., Kraal, G., Mebius, R.E. (2003). A Conduit System Distributes Chemokines and Small Blood-borne Molecules through the Splenic White Pulp. The Journal of Experimental Medicine, 198: 505-512.
[7] Hjelm, F., Karlsson, M.C.I., Heyman, B. (2008). A novel B cell-mediated transport of IgE-immune complexes to the follicle of the spleen. Journal of Immunology, 180: 6604-6610. [8] Yokota, A., Kikutani, H., Tanaka, T., Sato, R., Barsumian, E.L., Suemura, M., Kishimoto, T.
(1988). Two species of human Fcε receptor II (FcεRII/CD23): tissue-specific and IL-4-specific regulation of gene expression. Cell, 55: 611-618.
[9] Li, H., Nowak-Wegrzyn, A., Charlop-Powers, Z., Shreffler, W., Chehade, M., Thomas, S., Roda, G., Dahan, S., Sperber, K., Berin, M.C. (2006). Transcytosis of IgE-antigen complexes by CD23a in human intestinal epithelial cells and its role in food allergy. Gastroenterology, 131: 47-58. [10] Maeda, K., Burton, G.F., Padgett, D.A., Conrad, D.H., Huff, T.F., Masuda, A., Szakal, A.K., Tew,
J.G. (1992). Murine follicular dendritic cells (FDC) and low affinity Fc-receptors for IgE (FcεRII).
Journal of Immunology, 148: 2340-2347.
[11] Palaniyandi, S., Tomei, E., Li, Z., Conrad, D.H., Zhu, X. (2011). CD23-dependent transcytosis of IgE and immune complex across the polarized human respiratory epithelial cells. Journal of
Immunology, 186 (6): 3484-3496.
[12] Tu, Y., Salim, S., Bourgeois, J., Di Leo, V, Irvine, E.J., Marshall, J.K. Perdue, M.H. (2005). CD23-mediated IgE transport across human intestinal epithelium: Inhibition by blocking sites of
[13] Yu, L.C.H., Montagnac, G., Yang, P., Conrad, D.H., Benmerah, A., Perdue, M.H. (2003). Intestinal epithelial CD23 mediates enhanced antigen transport in allergy: evidence for a novel splice form. American Journal of Physiology - Gastrointestinal and Liver Physiology, 285: G223-G234.
[14] Gould, H.J., Sutton, B.J. (2008). IgE in allergy and asthma today. Nature, 8: 205-217.
[15] Payet, M.E., Woodward, E.C., Conrad, D.H. (1999). Humoral response suppression observed with CD23 transgenics. Journal of Immunology, 163: 217-223.
[16] Yu, P., Kosco-Vilbois, M., Richards, M., Köhler, G., Lamers, M.C. (1994). Negative feedback regulation of IgE synthesis by murine CD23. Nature, 369: 753-756.
[17] Heyman, B., Tianmin, L., Gustavsson, S. (1993). In vivo enhancement of the specific antibody response via the low-affinity receptor for IgE. European Journal of Immunology, 23: 1739-1742.
[18] Fujiwara, H., Kikutani, H., Suematsu, S., Naka, T., Yoshida, K., Yoshida, K., Tanaka, T., Suemura, M., Matsumoto, N., Kojima, S. (1994). The absence of IgE antibody-mediated augmentation of immune responses in CD23-deficient mice. Proceedings of the National
Academy of Sciences of the United States of America, 91: 6835-6869.
[19] 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. Journal
of Immunology, 164: 3990-3995.
[20] Getahun, A., Hjelm, F., Heyman, B. (2005). IgE enhances antibody and T cell responses in vivo via CD23+ B cells. Journal of Immunology,175: 1473-1482.
[21] Carlsson, F., Hjelm, F., Conrad, D.H., Heyman, B. (2007). IgE enhances specific antibody and T cell responses in mice overexpressing CD23. Scandinavian Journal of Immunology, 66: 261-270.
[22] Kehry, M.R., Yamashita, L.C.. (1989). Low-affinity IgE receptor (CD23) function on mouse B cells: role in IgE-dependent antigen focusing. Proceedings of the National Academy of
Sciences, 86 (19): 7556-7560.
[23] Henningson, F., Ding, Z., Dahlin, J.S., Linkevicius, M., Carlsson, F., Grönvik, K., Hallgren, J., Heyman, B. (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.
[24] Yu, L.C.H., Yang, P., Berin, M.C., Di Leo, V., Conrad, D.H. McKay, D.M., Satoskar, A.R., Perdue, M.H. (2001). Enhanced transepithelial antigen transport in intestine of allergic mice Is mediated by IgE/CD23 and regulated by interleukin-4. Gastroenterology, 121: 370-381.
[25] Yang, P.C., Berin, M.C., Yu, L.C., Conrad, D.H., Perdue, M.H. (2000). Enhanced intestinal transepithelial antigen transport in allergic rats is mediated by IgE and CD23 (FceRII). Journal
of Clinical Investigation, 106: 879-886.
[26] Hine, B.C., Hunt, P.W., Beasley, A.M., Windon, R.G., Glover, S.A., Colditz, I.G. (2010). Selective transport of IgE into ovine mammary secretions. Research in Veterinary Science, 89: 184-190. [27] Squire, C.M., Studer, E.J., Lees, A., Finkelman, F.D., Conrad, D.H. (1994). Antigen presentation
is enhanced by targeting antigen to the FcεRII by antigen-anti- FcεRII conjugates. Journal of
Immunology, 152: 4388-4396.
[28] Matsumoto, N., Kojima, S. (1994). The absence of IgE antibody-mediated augmentation of immune responses in CD23-deficient mice. Proceedings of the National Academy of Sciences
of the United States of America, 91: 6835-6869.
[29] Murphy, K.M., Heimberger, A.B., Loh, D.Y. (1990). Induction by antigen of intrathymic apoptosis of CD4+CD8+TCRlo thymocytes in vivo. Science, 250: 1720-1723.
