| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Microbiology and Immunology, Virginia Commonwealth University, Richmond, VA 23298; and
Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
50% decrease in IgE responses (5).
In these mice, overexpression of CD23 was driven by using the Thy-1
promoter and predominant transgene expression was thus restricted to T
cells while normal expression levels were retained on B cells. Helminth
infections with Nippostrongylus brasiliensis also caused a
modest decrease in IgE production over time while IgG1 and IgM levels
remained unaffected. Although these studies indicate that CD23 expressed on T cells has a modest effect on IgE production, they do not address its role on B cells, where CD23 is more typically expressed in high concentration. Consequently, Tg mice were developed with the CD23 transgene controlled by the class I, H-2Kb promoter combined with the Ig enhancer (6). In common with other Tg mice using this promoter/enhancer system (7, 8), expression of the transgene appears to be limited to the hemopoietic system. Studies of these Tg mice revealed that serum Ig levels were reduced and the soluble form of CD23 was elevated before immunization (6). Induction of primary and secondary IgE levels were observed to be significantly suppressed in response to Ag/alum immunizations, N. brasiliensis infections, and stimulation with goat-anti-mouse IgD. Furthermore, enzyme-linked immunospot analysis of activated B lymphocytes revealed a reduced number of IgE Ab-forming cells. Humoral Ig levels were mildly dampened after immunization, suggesting that the transgene affected production of all isotypes, potentially through interactions of CD23 with an alternate ligand, like CD21 (6).
Given the above results, we hypothesized that high levels of CD23 on B cells would suppress IgE production, and we sought to determine whether CD23 overexpression on B cells is responsible for IgE suppression in these Tg mice. In vitro culture with purified Tg B cells revealed that the synthesis machinery was intact, as IgE production was normal. Furthermore, adoptive transfer of Tg lymphocytes into normal mice further confirmed that B lymphocytes alone were not sufficient to cause suppression. Finally, lymphokine studies and studies with Leishmania major revealed no functional abnormalities in the Tg T cells. Thus, the B cell contribution to the IgE suppression observed in CD23-Tg mice was minimal and suppression appeared most likely dependent on a cell type other than B or T cells. FDC express surface CD23 (9), are irradiation resistant (10), provide iccosomal Ag for B cells (11), and are found in germinal centers where recall responses are initiated (12, 13). Therefore, we reasoned that these cells might play a role in the IgE suppression observed in these CD23-Tg mice. Compared with normal FDC, the IgE responses of B cells in vitro were suppressed by Tg FDC. Further, high levels of CD23 expression were observed in germinal centers where FDC reside. Taken together, these results support the concept that high-level expression of CD23 on lymphoid cells does not fully explain the observed suppression of IgE in Tg mice, but elevated CD23 levels associated with the FDC and B cells in the germinal center may explain the observed suppression.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
BALB/c and C57BL/6 mice were purchased from the National Cancer Institute (Frederick, MD) or The Jackson Laboratory (Bar Harbor, ME). Tg mice were constructed at the Mouse Core Facility at the University of Pennsylvania and maintained at the Medical College of Virginia campus of Virginia Commonwealth University. All mice used in experiments were between 6 and 14 wk of age.
RPMI 1640 supplemented with 10% heat-inactivated FCS (HyClone, Logan, UT), 2 mM glutamine, 1 mM sodium pyruvate, 1x nonessential amino acids, penicillin (100 U/ml), streptomycin (100 U/ml; all from Life Technologies, Rockville, MD), and 5 x 10-5 M 2-ME (Sigma, St. Louis, MO) (B cell medium). For lymph node cultures, B cell medium was supplemented with 1 mM HEPES (Life Technologies). FDC cultures used B cell medium except that 2-ME was not added.
B cell isolation
B cells were isolated from disrupted spleens by using a Percoll gradient as described previously (14). Resting B cells were primarily collected from the 66–70% interface, although the 60–66% interface, also primarily resting B cells, was also used in some experiments when larger cell numbers were needed.
B cell proliferation
Proliferation of splenic B lymphocytes was assayed in vitro by culturing increasing doses of purified resting B cells in complete B cell medium. Cells were activated with a cocktail of recombinant CD40 ligand (CD40LT, 0.1 ug/ml; Ref. 15 ; Immunex, Seattle, WA), IL-4 (50,000 U/ml; Baculovirus supernatants were a gift from Dr. Bill Paul, National Institutes of Health, Bethesda, MD), and IL-5 (5 ng/ml; R&D Systems, Minneapolis, MN) at 37°C in a 5% CO2 incubator. During the final 8 h of a 72-h culture, 1 µCi/well of [3H]thymidine (sp. act. 6.7 Ci/mmol; DuPont/NEN, Wilmington, DE) was incorporated into the DNA of actively growing and dividing cells (16). Plates were harvested with a Packard Filtermate 196 Harvester (Packard Instrument, Downers Grove, IL) onto a Unifilter 96-well plate (Packard Instrument), containing a glass-fiber filter. A total of 2.4 ml of Microscint-20 (Packard Instrument) scintillation fluid was added to the Unifilter and counted in a Top Count scintillation counter (Packard Instrument).
In vitro IgE induction
Doses of resting B cells (0–100,000 cells/well) were activated in vitro in a 96-well plate (Costar, Cambridge, MA) for 10 days with 50,000 U/ml IL-4, 5 ng/ml IL-5, 0.1 µg/ml CD40LT, and 0.1 µg/ml of the anti-leucine zipper mAb M15 at 37°C in complete B cell medium in a 5% CO2 incubator. These activation conditions previously have been shown to be optimal (17). Supernatant fluids were collected and Ig levels were determined by ELISA.
To test the ability of primed lymphocytes to make IgE in culture, draining lymph nodes were collected from the DNP-keyhole limpet hemocyanin (KLH)/alum-immunized mice 9 days after secondary injection. A single-cell suspension was made, 1 x 106 lymph node cells/ml were cultured in a 96-well plate with increasing Ag doses (0–100 ng/ml DNP-KLH) for 10 days in lymph node medium, and IgE levels were determined by ELISA.
Adoptive transfer
Adoptive transfer methods were those described previously with minor modifications (18). Donor Tg and littermate (LM) control mice were immunized with DNP-KLH/alum as described previously (6). Recipients were irradiated with 600 rad under the maximum scatter conditions by using a 137Cs source (Mark I, Model 68–0146; JL Shepherd & Associates, San Fernando, CA). This dose was sufficient to arrest the lymphocyte populations but leave the myeloid and FDC populations. Twenty-four hours after irradiation, mice were reconstituted i.p. with 2–5 x 107 donor splenocytes. Control transfers (LM to LM and Tg to Tg) were done to test the efficacy of the experiment in relation to the previous DNP-KLH/alum experiments. Immune cells from Tg animals were transferred to LM animals, and LM immune cells were transferred to Tg animals as the two experimental groups. Forty-eight hours after adoptive transfer of whole splenocytes, mice were immunized s.c. with 100 µg of DNP-KLH in 4 mg of alum, and a booster of 100 µg of DNP-KLH in 2 mg of alum were administered s.c. in the front and hind footpads 14 days later. The serum levels of total IgE and Ag-specific IgG1 were monitored throughout the primary and secondary responses by ELISA. Serum was collected from the tail vein.
L. major parasite isolation, infection, and quantitation
L. major clone V1 (MHO/IL/80/Friedlin; Ref. 19) was cultured in Dr. David Sacks’ laboratory (National Institutes of Health) in M199 medium supplemented with 20% HI-FCS (HyClone), 100 U/ml penicillin, 100 µg/ml streptomycin, 2 mM L-glutamine, 40 mM HEPES, 0.1 mM adenine (in 50 mM HEPES), 5 mg/ml hemin (in 50% triethanolamine), and 1 mg/ml 6-biotin (M199/S) at 26°C. Infective stage metacyclic promastigotes of L. major were isolated from stationary culture (5–6 day old) by negative selection with peanut agglutinin (Vector Laboratories, Burlingame, CA). Mice were infected with 105 purified metacyclics in the left hind footpad. Lesion size was measured to monitor disease progression with a metric caliper (Thomas Scientific, Swedesboro, NJ), and swelling attributable to the infection was calculated by subtracting the size of the contralateral noninfected footpad.
ELISAs
Levels of IgE, IgG1, and IgG2a in supernatants or serum were determined by ELISA as described previously (6). Briefly, a pair of rat anti-mouse IgE monoclonals, B1E3 (5 µg/ml) and biotinylated R1E4, were used as the capture and biotinylated secondary Ab, respectively. Biotin then was recognized by streptavidin coupled to alkaline phosphatase (Southern Biotechnology Associates, Birmingham, AL). Total IgG1 and IgG2a both were determined by using an unlabeled primary goat anti-mouse Ab at 5 ng/ml and detected with goat anti-mouse class-specific Ab coupled to alkaline phosphatase (all Abs are from Southern Biotechnology Associates). To detect Ag-specific IgG1 in serum from immunized mice, the ELISA plates were coated with 5 ng/ml DNP-KLH, and the rest of the ELISA protocol was completed in the same manner as total IgG1. In the case of Ag-specific IgG1, the concentrations were relative because the standard curve used an anti-IgG coating Ab instead of Ag. All isotype levels were detected by colorometric change by using p-nitrophenyl phosphate as substrate. Absorbance was measured at 450 nm with background noise correction at 605 nm. Data was analyzed by using Softmax Pro version 1.2 (Molecular Devices, Sunnyvale, CA), and all (duplicate) samples were tested at multiple dilutions.
FDC isolation
FDC were isolated from the lymph nodes (popliteal, brachial, axillary, inguinel, periaortic, and mesenteric) of CD23-Tg and LM mice as described previously (20, 21, 22). In brief, the FDC donor mice were irradiated to reduce contaminating lymphocytes in the lymph nodes and enrich for the irradiation-resistant FDC. The level of irradiation used in this study was increased from the typical 600 rad to 1000 rad to eliminate more lymphocytes and to minimize any contribution by lymphocytes contaminating the FDC preparation. Three days after irradiation, the lymph nodes were removed and cut with 26-gauge sterile needles to facilitate enzymatic digestion. The nodes then were incubated with 1 ml of 8 mg/ml collagenase D (lot no. FIA148; Boehringer Mannheim, Indianapolis, IN) and 0.5 ml of 10 mg/ml DNase I (lot no. 32H9545; Sigma), in 1 ml of complete DMEM at 37°C. After a 1-h incubation, the cells were released from the digested stroma by gentle pipetting. The medium containing the released cells were collected and the cells were directly layered onto a continuous 50% Percoll gradient and centrifuged for 20 min at 700 x g. The low-density (1.050–1.060 g/ml) FDC-enriched fraction then was removed and washed twice. Finally, adherent macrophages were removed by incubating the cells in petri dishes at 37°C for 1 h. The nonadherent cell suspension typically contained 25–50% FDC as determined by flow cytometry by using the FDC-specific mAb FDC-M1 (21). The vast majority of the contaminating cells were medium to large lymphocytes.
To confirm CD23 expression on FDCs, the cells were further purified
from the enriched preparations by using rat mAb FDC-M1 in conjunction
with anti-rat-coated magnetic beads (Dynal, Great Neck, NY). These
positively selected FDC have been found to be 90% pure by using
this purification procedure (22). The enriched FDC
preparation used for cell culture experiments contain between 20 and
40% FDC, 10 and 20% B cells, and the balance of the cells are T cells
with a small percentage of macrophages and other as yet unidentified
cells. In the present study, the magnetic bead separation system
removed 30% of the cells, leaving most contaminating T and B cells in
the remaining 70% of cells. Messenger RNA was isolated from purified
FDCs by using the TRIzol protocol (Life Technologies) exactly as
described by the manufacturer. As a control, mRNA was also isolated
from purified murine B cells by using the same protocol. Real-time PCR
to analyze the mRNA samples was performed with the Taqman One-Step
RT-PCR Mastermix Reagent Kit, in which first-strand cDNA synthesis and
amplification are performed in a single reaction. Analyses were
performed in a model 7700 ABI Prism Sequence Detector (Applied
Biosystems, Foster City, CA). RNA samples were normalized by using a
probe for murine 18S rRNA (Applied Biosystems) as an endogenous
standard. A standard curve was prepared by using a dilution series of
RNA from CD23-transfected Chinese hamster ovary (CHO) cells
(23) probed with the CD23 primer and probe set (see
below). Each mRNA sample was analyzed in triplicate at two different
mRNA concentrations. The CD23 primer and probe set consisted of two
amplificaton primers, 5'-CCGGCCAGTGGAACGA-3' and
5'-GGGTGGGCCTTGTTGGA-3', and the TaqMan probe
5'-FAM-CCTTCTGCCGCAGCTACTTGGATGC-TAMRA-3', targeted to the lectin
domain. No template and no reverse transcriptase controls yielded no
detectable product, demonstrating a lack of contamination of the
samples with genomic DNA or PCR product.
FDC cell cultures
Enriched FDC preparations (1 x 105 cells) were placed in 96-well tissue culture plates (catalogue no. 3595; Costar) containing 200 µl of RPMI 1640 culture medium plus 10% FCS. FDC were incubated for 6 days at 37°C in a 5% CO2 incubator. The collagenase and DNase treatment used in the FDC isolation procedure rapidly cleaved CD23 from the surfaces of both FDC and B cells. Incubation of FDC for a minimum of 4 days was necessary to restore easily detectable levels of CD23 to the cell surfaces (our unpublished observations). B cell preparations at 500 cells/well then were added to FDC cultures. At the same time, CD40LT (0.1µg/ml), IL-4 (50,000 U/ml), IL-5 (5 ng/µl), and M15 (0.1 µg/ml) also were added. Cultures then were incubated for another 14 days, with medium collected and replaced at day 7 and 14. The levels of IgG and IgE produced between days 7–14 was determined by using an ELISA as described above.
Immunocytochemistry
The purpose of immunocytochemistry was to determine the extent of CD23 labeling in normal and CD23-Tg mice by using FITC anti-mouse CD23 (catalog no. 01234D, lot Mo26738; BD PharMingen, San Diego, CA). To amplify the FITC signal, anti-FITC-HRP incubation was followed by binding of tyramide-FITC. These reagents are available in the Tyramide Signal Amplification kit (catalog no. NEL701A; NEN Life Science Products, Boston, MA) and were used as recommended by the manufacturer (see also Ref. 24). After an initial overnight incubation of lymph node cryostat sections, the signal was amplified (see above), and normal and Tg lymphnode sections were compared for FITC fluorescence intensity of follicular areas (FDCs and germinal center B cells). The results were recorded by using a digital fluorescecence microscope.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The first objective was to determine whether the suppressed IgE
responses in Tg mice can be explained by overexpression of CD23 on B
cells. Cho et al. (23) reported that interaction of B
cells with CD23-transfected CHO cells resulted in decreased B cell
proliferation and IgE production in vitro. Thus, proliferation
decreased with increased expression of membrane CD23 (23).
However, when resting splenic B lymphocytes from Tg animals were
activated with CD40LT and cytokines, B cell proliferation was unaltered
when compared with LM controls (Fig. 1
A). Thus, in this Tg system,
increased CD23 levels did not affect the growth/proliferative ability
of B lymphocytes.
|
B. In two of the six experiments, the
IgE production in Tg B cells was lower than that in LM at some B cell
concentrations, but the reverse was not seen. In short, no clear trend
was seen for differences in in vitro IgE production between Tg or
normal B cells. Similar variability was seen when CD23-Tg mice that
used the Thy-1 promoter were analyzed in a this manner
(23). Finally, mice were immunized in footpads with
DNP-KLH, and 9 days after secondary immunization, draining popliteal
lymph nodes were removed. Lymph node cells (1 x
106 cells/ml) were placed in culture for 10 days
with increasing doses of DNP-KLH. Suppression of the IgE response by
primed Tg lymphocytes was not observed (Fig. 1C). This
contrasts with in vivo results where the same immunization regimen
results in a profound suppression of serum IgE (6). Th1/Th2 cytokine ratios were not altered by transgene expression
We reasoned that an alteration in the Th1/Th2 profile could play a
role in regulating IgE responses in vivo. The Tg mice have been bred
for at least seven generations on a BALB/c background and a natural Th2
profile was expected. To determine whether Th2 cytokine profiles were
altered, Tg and LM were infected with L. major. Being an
intracellular pathogen, a Th1-type response is necessary to
clear/control the disease. Therefore, if Tg mice on the BALB/c
background had shifted their cytokine profiles to Th1, they would
control parasite replication. To ensure that results were not affected
by the site of Tg insertion, we examined mice from 2A Tg founder lines
M21 and M11. In each experiment, we compared the Tg animals with their
LM controls as well as the two classical models, BALB/c and C57BL/6. As
is seen in Fig. 2, the footpad swelling
in Tg mice was similar to that in the BALB controls, indicating a Th2
profile.
|
and IL-4 production was
seen, arguing against an alteration in the Th1/Th2 profile (data not
shown). This is further supported by analyzing IgE and IgG2a levels in
the infected mice 6 wk after infection (Fig. 2B); both IgE
and IgG2a levels were strongly (p < 0.05 in
both situations) suppressed. Because IgE is elevated by a Th2 and IgG2a
elevated by a Th1 response, respectively, this supports the concept
that the humoral response is suppressed in the Tg animals. Adoptive transfer
To determine whether nonlymphoid cells play a role in suppressing
IgE responses in Tg mice, an "adoptive transfer" study was
undertaken. The lymphoid cells of the recipient are destroyed by
irradiation while nonlymphoid cells that are less susceptible to
irradiation injury persist. Normal responses were seen in Tg-to-Tg (low
IgE and IgG1) and LM-to-LM (high IgE and IgG1) control transfers (Fig. 3). If lymphocyte CD23 overexpression was
responsible for the suppression, Tg donor lymphocytes transferred into
the LM recipient would be expected to cause suppression of IgE, and the
normal response would be expected in a LM donor to Tg recipient
transfer. However, the opposite results were observed. When Tg
lymphocytes were transferred into an irradiated non-Tg recipient, high
IgE (Fig. 3A) and IgG1 (Fig. 3B) levels were
observed, and when non-Tg lymphocytes were transferred into a Tg
animal, suppression was restored, suggesting that nonlymphoid Tg
tissues are important in the CD23-mediated suppression of IgE.
|
The adoptive transfer studies indicated that the cells responsible
for inhibiting the IgE response persist in the irradiated Tg host.
Given that FDC are irradiation resistant (12), express
surface CD23 (9), and present iccosomal Ag to B cells
(12, 13), we reasoned that Tg FDC might play a role in
regulating the IgE response in the Tg mice. To begin testing this
possibility, we incubated FDC from Tg and LM animals with wild-type B
cells stimulated with CD40LT, IL-4, and IL-5 and compared the effects
on IgE and IgG production. Addition of Tg FDC supported a lower IgE
response (60% lower) but had no significant effect on IgG
production (Fig. 4). In three experiments
of this type, the average IgG response in cultures with Tg FDC was
99.8% of the LM control, whereas the IgE response was only 31% of the
LM control (p < 0.01). Stimulated B cells
alone made IgE in the day 0 to day 7 period, but in the absence of FDC,
B cells made very little IgE in the day 7 to day 14 period.
|
The above studies prompted a shift in focus toward analysis of
CD23 on FDCs. FDCs normally express CD23, but unfortunately the
collagenase used in the FDC isolation procedure rapidly cleaves CD23
from the surfaces of B cells and FDC (our unpublished observation).
Consequently, it was difficult to determine whether FDCs from Tg
animals overexpress CD23 by FACS analysis. Immunohistochemistry of
lymph nodes from immunized M21 and LM animals revealed that germinal
centers rich in B cells and FDC did express much higher levels of CD23
than did controls (Fig. 5). In contrast,
at least in the M21 Tg line, expression of CD23 on T cells was not
evident. These results are consistent with CD23 levels observed on B
and T cells by FACS analysis (6). The labeling indicated a
high level of CD23 expression in the area of FDC networks, but it was
difficult to be certain that FDC were responsible for the high levels
of CD23 because germinal center B cells also express CD23.
|
90% pure, the overexpression of CD23 on the
Tg FDCs was confirmed. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This concept of dual roles of CD23 is interesting, especially in light of the other primary known role for CD23, namely enhancement of Ag presentation. In the late 1980s, Kehry et al. (32) in the mouse and Pirron et al. (33) in the human sytem demonstrated that CD23 on B cells could considerably augment the Ag presentation capacity of B lymphocytes. Indeed, the efficiency of Ag presentation via CD23 approached the efficiency seen when targeting B cell surface Ig. Squire et al. (34) further demonstrated that covalently linking Ag to anti-CD23 would augment the humoral response and proposed that this property of CD23 could be a useful vaccine strategy. The Heyman laboratory has performed the most extensive study of this property of CD23. This laboratory noted that ingestion of IgE-Ag complexes by B cells resulted in an enhanced IgG response to the ingested Ag (35). Indeed, the loss of this CD23 Ag focusing property was the only abnormality reported in the CD23-/- mice produced by Fujiwara et al. (3).
Clearly, the Ag presentation capacity of CD23 points in exactly the opposite direction from the suppression seen in the paper describing the phenotype of the new CD23-Tg animals (6). A potential explanation for this apparent paradox comes from adoptive transfer studies. The Heyman’s laboratory (36) adoptively transfered CD23+ lymphocytes into CD23-/- mice to demonstrate that CD23 expression on B lymphocytes is necessary for the IgE/Ag/CD23-mediated enhancement of the IgG response. This enhancement was not observed when CD23-/- lymphocytes were transferred into irradiated normal hosts. Thus, the only difference between the two transfers was the presence or absence of CD23 on lymphocytes.
In the studies reported here, the adoptive transfer experiments pointed
to the importance of the host environment (Fig. 3). Namely, transfer of
Tg lymphocytes into irradiated normal recipients resulted in
restoration of the IgE (and IgG1) response, whereas transfer of normal
lymphocytes into Tg hosts resulted in suppression. Taken in context,
these studies indicate that B cell expression of CD23 is not a
sufficient cell with respect to regulation of the IgE response. CD23
expression on nonlymphoid tissue clearly is quite important for the
regulatory effects observed.
Another potential mechanism to help explain the suppressive results
seen is that the Th1/Th2 ratio has been changed by Tg expression.
L. major infection was used as a classic model in which Th1
responses are necessary for clearance of the intracellular pathogen
(37). The two phenotypes most fully characterized with
respect to this pathogen are BALB/c and C57BL/6 mice because they
represent the two extremes of the response (37). Footpad
swelling, parasite burden, and cytokine responses were assessed to
determine whether a change occurred in the Tg animals. In separate
studies (data not shown) cytokine responses of
TCROVA (38)-CD23 double-Tg mice were
measured. In both models, IL-4/IFN- cytokine profiles were not
altered, arguing against this playing a role in the inhibition of serum
IgE observed.
Obviously, because both membrane and soluble CD23 levels are enhanced in the Tg mice, the observed phenotype could be attributable to either membrane or soluble CD23 (or both). Because soluble CD23-Tg mice had no observable phenotype (5), membrane CD23 is favored as being responsible. The current model of CD23 predicts that three monomers interact with each other to form a functional trimer on the cell surface. This model is based on the noted heptad repeat pattern found in the stalk region of the molecule (39) and chemical cross-linking studies (40). CD23 is cleaved by an as yet unidentified metalloprotease (41) and the cleaved monomeric product interacts with only a single low affinity with IgE (42). Recently, Kelly et al. (17) found that a soluble CD23 oligomer with high affinity/avidity for IgE could be produced by attachment of a modified leucine zipper to the amino terminus of the stalk region of CD23. It will be interesting to see whether Tg animals producing this soluble oligomer have a phenotype similar to the membrane produced by Tg mice, and such studies are in progress.
The adoptive transfer studies, combined with high levels of CD23
expression in the germinal center, suggests expression on FDC is
playing an important role in the phenotype. This concept was supported
by the demonstration that IgE synthesis was significantly reduced by
culturing B cells in the presence of Tg FDCs (Fig. 5). Also of note are
data indicating that mouse FDC-CD23 expression is highest after
immunization with CFA (9). CFA is known to give a poor IgE
response, and this study speculates that FDC expression of CD23 may
partly be responsible for the low IgE responses when CFA is used
(9). In addition, the phenotype of the CD23-Tg mice has
similarities with that seen for CR2-/- animals
(43). In vitro responses of CR2-/-
B cells are normal, but in vivo T-dependent humoral responses are
severely depressed (43). However, adoptive transfer
studies in the CR2-/- animals indicated that B
cell CR2 expression is the relevant entity (43).
In combination, data pointing to the importance of the FDC/B cell interactions suggest that conditions that give elevated CD23 expression on FDCs (e.g., CFA or a CD23 Tg), give a negative signal(s) to B cells and thus inhibit differentiation into IgE Ab-producing cells. Although CD23 has a number of known ligands, CD21 and sIgE are the most likely to be involved in the humoral response regulation described in this report. IgE in the form of sIgE is likely present on B cells committed to production of IgE, and CD21 is present on all B cells. The inhibitory mechanism is unknown, but interaction of the CD23 on the FDC with CD21 on the B cell could help explain the observation that multiple isotypes tended to be suppressed. The fact that the suppression is clearly greatest with respect to IgE suggests the possibility that the interaction of FDC-CD23 with the sIgE may somehow deliver a stronger signal that turns off the IgE response specifically. In this regard, the Cho et al. (23) studies are supportive of the FDC model. Namely, in vitro culture of B cells with CHO cells transfected with high levels of CD23 resulted in suppression, with the strongest effects observed with IgE. Because similar results are reported here for B cells cultured with Tg FDCs, potentially the CD23+ CHO cells in vitro were acting in the same manner as the CD23+ FDCs in vivo. Although additional experiments are clearly required, these results are quite supportive of the concept that a high level of FDC-CD23 may not support the production of IgE. An obvious additional experiment would be to inhibit the CD23-mediated suppression with anti-CD23. However, as noted in the study that used CD23+-CHO cells (23), the currently available Abs were ineffective in this regard. Development of Ab reagents against the stalk region, which would have the potential to inhibit CD23 oligomerization, may have more potential to inhibit this activity and this will be tested as these reagents become available.
Because CD23 also is induced on B lymphocytes as well, one could ask why do B cells also not show the same activity, namely suppressed Ig production. By analogy with the culture with CD23+-CHO cells (23), the key well could be that CD23 must be elevated at the initiation of the response. B cell CD23 is not maximally elevated until day 3 after activation. Even with Tg B cells, the level may not be sufficient at initiation of the culture. An obvious question is why did CD23 evolve to have what seems to be essentially opposite functions: one that suppresses IgE and one that enhances the general humoral response? Although this remains unclear, there are several observations that may be useful. First, IgE is inhibited when CD23 is expressed on FDCs, and induction of CD23 in a normal animal occurs under conditions where a cellular response, e.g., CFA, is strongly induced or when CD23 is inappropriately elevated on this cell, as in the transgene situation. Under Th2 conditions, which give a strong IgE response and high levels of B cell CD23, FDC-CD23 is low and it is quite probable that the augmentation response is acting through enhancement of Ag processing. This hypothesis certainly is testable by developing mice that have transgene expression only on B cells or only on FDCs and these lines are open for future experimentation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892.
3 M.J.P. and S.L.T.H. contributed equally to this work and should be considered co-first authors.
4 Current address: Research Institute for Genetic Engineering and Biotechnology, 41470 Gebze-Kocaeli, Turkey.
5 Address correspondence and reprint requests to Dr. Daniel H. Conrad, Virginia Commonwealth University, Department of Microbiology and Immunology, Sanger Hall, 1101 East Marshal Street, Box 980678, MCV Station, Richmond, VA 23298.
6 Abbreviations used in this paper: FDC, follicular dendritic cells; Tg, transgenic; CD40LT, recombinant CD40 ligand; KLH, keyhole limpet hemocyanin; LM, littermate; CHO, Chinese hamster ovary.
Received for publication June 15, 2000. Accepted for publication February 5, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
8.1 chain transgenic mice. EMBO J. 8:719.[Medline]
RII). J. Immunol. 148:2340.[Abstract]
RI. J. Immunol. 161:6696. receptor II (CD23) function on mouse B cells: role in IgE dependent antigen focusing. Proc. Natl. Acad. Sci. USA 86:7556.RII by antigen-anti-FcRII conjugates. J. Immunol. 152:4388.[Abstract]
-Helical coiled-coil stalks in the low-affinity receptor for IgE (FcRII/CD23) and related C-type lectins. Proc. Natl. Acad. Sci. USA 89:753.RII/CD23: implications for function. J. Immunol. 150:2372.[Abstract]
RII) release from cell membranes is mediated by a membrane-bound metalloprotease. Biochem. J. 333:573.
RII: sites of cleavage and requirements for dual affinity interaction with IgE. J. Immunol. 154:4240.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  G. Pongratz, J. W. McAlees, D. H. Conrad, R. S. Erbe, K. M. Haas, and V. M. Sanders The Level of IgE Produced by a B Cell Is Regulated by Norepinephrine in a p38 MAPK- and CD23-Dependent Manner. J. Immunol., September 1, 2006; 177(5): 2926 - 2938. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sukumar, D. H. Conrad, A. K. Szakal, and J. G. Tew Differential T Cell-Mediated Regulation of CD23 (Fc{epsilon}RII) in B Cells and Follicular Dendritic Cells. J. Immunol., April 15, 2006; 176(8): 4811 - 4817. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B.-H. Chen, M. A. Kilmon, C. Ma, T. H. Caven, Y. Chan-Li, A. E. Shelburne, R. M. Tombes, E. Roush, and D. H. Conrad Temperature Effect on IgE Binding to CD23 Versus Fc{epsilon}RI J. Immunol., February 15, 2003; 170(4): 1839 - 1845. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Rabah, S. Grant, C. Ma, and D. H. Conrad Bryostatin-1 Specifically Inhibits In Vitro IgE Synthesis J. Immunol., November 1, 2001; 167(9): 4910 - 4918. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Kilmon, R. Ghirlando, M.-P. Strub, R. L. Beavil, H. J. Gould, and D. H. Conrad Regulation of IgE Production Requires Oligomerization of CD23 J. Immunol., September 15, 2001; 167(6): 3139 - 3145. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) were stimulated with CD40LT
and cytokines for 48 h in a 96-well tissue culture plate.
A, The proliferative ability of the lymphocytes was
measured by an 8-h [3H]thymidine pulse at the end of the
culture period. Cells then were harvested onto a 96-well filter plate
and counted in a Top Count scintillation counter. B,
Resting splenic B cells were stimulated in culture as above except that
culture time was 10 days. The B cell dosage is indicated on the
x-axis. Supernatants were harvested and tested by ELISA
for total IgE. C, Mice were immunized as described with
DNP-KLH/alum, and 9 days after booster, primed lymphocytes from
draining nodes (106 cells/ml) were incubated for 10 days
with increasing concentrations of DNP-KLH. Supernatants were harvested
and tested for total IgE by ELISA. IgE produced by Tg lymphocytes was
not different from controls. Error bars are ±SE.
) mice, and Tg lymphocytes were transferred into Tg
(
) and LM (
) mice. Mice were then immunized with the standard
protocol of DNP-KLH/alum immunization and tested for total IgE
(A) and Ag-specific IgG1 (B) throughout
the primary and secondary responses. Error bars are ±SE. This is a
representative experiment of a total of five indicating the same
relationships (three times for the M21 line and two times for the M11
line).
