Download A new multivalent B cell activation model

iim$$$6246
International Immunology, Vol. 9, No. 2, pp. 239–248
© 1997 Oxford University Press
A new multivalent B cell activation model—
anti-IgD bound to FcγRI: properties and
comparison with CD40L-mediated activation
Sung-weon Cho and Daniel H. Conrad
Department of Microbiology and Immunology, Virginia Commonwealth University, Richmond, VA 23298,
USA
Keywords: germinal centers, Ig synthesis, IgE synthesis, IgG1, IgM, IL-4, IL-5, mouse B cells
Abstract
CHO cells permanently transfected with mouse FcγRI α chain were prepared and used as a model
to polyclonally activate murine B cells. The transfected CHO cells were treated with mitomycin C
and placed into culture with varying quantities of anti-IgD. Using this model, murine splenic B cells
(from BALB/c or C57Bl/6) were activated by mouse IgG2a–anti-IgD (10.4.22 or AF3.33) in a manner
that is analogous to the activation of B cells seen with highly polyvalent anti-IgD (Hδa/1) prepared
by chemical cross-linking to dextran. Efficient B cell activation was seen with nanogram quantities
of anti-IgD. In the presence of IL-4 and IL-5, IgG1 production levels were equivalent to or better
than seen when stimulation was with Hδa/1–dextran; however, IgE induction was not seen in either
situation. The Ig production capacity was compared to that seen when B cells were activated with
CD40L, using either CD40L-transfected CHO or a soluble CD40L construct. In the presence of IL-4
and IL-5, once a critical threshold of B cells was present, IgE and to a lesser extent IgG1
production was inversely proportional to B cell number when CD40L was the activating agent. In
contrast, with FcγRI–anti-IgD, IgM and IgG1 production was directly proportional to B cell number,
while IgE production was never seen. Finally, when B cells were co-activated with immobilized antiIgD and CD40L simultaneously, the IgE production from B cells induced by CD40L was strongly
inhibited, while IgG1 and IgM production were not affected. Since B cell co-activation via sIg and
CD40L would be a common scenario in secondary follicles, this inhibition of IgE production may
be one of the reasons why serum IgE levels are much below IgG in normal immune situations.
Introduction
For some time high level sIg cross-linking has been realized
to be an excellent B cell activation signal via the sIg pathway.
The studies were done initially with a repeat epitope antigen
or anti-sIg and greatly extended by the use of a highly
multivalent anti-sIg, prepared by covalently linking the antisIg to dextran (1). The latter studies demonstrated that highlevel B cell activation was evident with anti-sIg doses that
were at least 1000-fold lower than required using anti-sIg
alone (1). This system has been exploited extensively as a
model for B cell activation analogous to T-independent type 2
(TI-2) antigens (reviewed in 2). Analysis of cytokine involvement has revealed that such a system can be used to model
isotype switching to IgG subclasses (3,4) or IgA (5), but
curiously, not IgE (4). With regard to IgE, the key role of the
CD40L–CD40 interaction in in vivo production of IgE is now
well recognized in both human and murine systems (reviewed
in 4,6). A number of laboratories have shown IgE production
using either recombinant CD40L-transfected cells or antiCD40, providing that IL-4 is also supplied. Thus, the CD40L–
CD40 interaction, in conjunction with IL-4 production, is
sufficient to give an IgE response, while other cytokines,
especially IL-5 in mouse, will further enhance IgE production (7,8).
In an ongoing humoral response antigen–antibody complexes are a feature that plays a recognized role in both
pathophysiology as well as the regulation of the immune
response. A natural consequence of these complexes is
binding to Fc receptors and such binding, especially in
germinal centers, would be expected to result in extensive
cross-linking of B cell sIg, given that multiple antigen–antibody
complexes are bound to the surface of FcR-positive cells.
Indeed fibroblasts transfected with FcγRII have been used to
Correspondence to: D. H. Conrad
Transmitting editor: Z. Ovary
Received 20 August 1996, accepted 25 October 1996
240 FcγRI–anti-IgD B cell activation
model B cell activation and study the effect of other B cell
surface components on this activation (9,10). In this study we
have used FcγRI-transfected cells to model murine B cell
activation, using IgG2a–anti-IgD mAb. This Fc receptor has
an appreciable affinity [53108 M–1 (11,12)] for monomeric
IgG2a and thus, in contrast to the FcγRII/III, will bind at
significant levels to both monomeric as well as complexed
IgG2a. The rationale for using this receptor to model Fcdependent activation relates to this higher affinity which
should decrease monomeric IgG–anti-sIg and give enhanced
sensitivity to the anti-sIg signal. We have found this model to
be an effective system for inducing B cell activation. The
model required the presence of both IL-4 for significant B
cell proliferation, and both IL-4 and IL-5 for significant Ig
production. Using these conditions significant levels of isotype
switching to IgG1, but not IgE, is seen. In secondary follicle
germinal centers, B cells would be exposed to both Fcpresented antigen–antibody complexes as well as CD40L.
Thus, we have used this model to also explore the effect of
having both stimuli present in the same culture well. Using
both CD40L-transfected cells as well as a soluble CD40L
trimer construct (CD40LT), we observed a strong inhibition of
IgE production, when compared to CD40L stimuli alone. IgM
and IgG1 production was not significantly affected. We and
others have observed that switching to IgE and IgE production
by CD40L stimuli is quite efficient; indeed, levels of IgE and
IgG1 are similar under optimal (low B cell input) conditions.
The sensitivity of the IgE switching/production mechanism in
the presence of Fc-presented complexes may be one of the
mechanisms explaining why IgE is a minor component of
serum Ig.
were further purified by flow cytometric sorting using an
EPICS-750 sorter after staining the B cells with FITC–B220, a
mAb directed against a B cell-specific form of CD45. Sorted
B cells were .99% pure, as judged by re-analysis on a
FACScan using phycoerythrin–anti-IgM.
Preparation of transfected cells
The cDNA for mouse CD40L was prepared by RT-PCR using
primers corresponding to positions 1–13 (sense) and 850–
867 (antisense) using the published sequence (19) and mRNA
from the CDC-35 T cell line (20) that had been activated by
overnight culture with anti-CD3 (2C11), using the Superscript
RT-PCR kit from Gibco (Gaithersburg, MD). Isolation of mRNA
was as described (21) using the GITC protocol. The 0.8 kb
Methods
Antibodies, cytokines, cell lines and B cell preparation
10.4.22 (13), Hδa/1 and AF3.33 (14) (a gift of Dr Fred
Finkelman) were anti-IgD mAb directed against IgDa or IgDb
(AF.3.33) allotypes respectively; the relevant ascites were
produced in nude mice and purified by ion exchange
chromatography. Hδa/1–dextran was a generous gift from Dr
Fred Finkelman. Anti-CD40L, termed MR-1, (15) was initially
a gift of Dr R. Noelle and phycoerythrin- or biotin-conjugated
MR-1 was purchased from PharMingen (San Diego, CA).
Recombinant mouse IL-4 was a generous gift from Dr William
Paul (NIH, Bethesda, MD); mouse IL-5 was purchased from
R&D Systems (Minneapolis, MN). CHO-K1 or CHO/DHFR–
(16) were obtained from the ATCC (Rockville, MD) and were
maintained in supplemented DMEM or IMDM medium with
nucleosides respectively. A cell suspension of mouse splenic
cells was first depleted of adherent cells by culturing for 30
min at 37°C in B cell media (RPMI 1640 containing 10% heat
inactivated FCS, 10 mM HEPES, 10 mM Na pyruvate, 5310–5
M 2-mercaptoethanol, and 100 U/ml penicillin and streptomycin). Subsequently, B cells were purified by negative
selection using anti-CD8, anti-CD4 and anti-Thy-1.1 as
described previously (17), and resting B cells were obtained
by using discontinuous Percoll gradients (18); cells at the 66–
70% interface were considered to be resting B cells and were
used in the assays described. When indicated these cells
Fig. 1. Anti-IgD bound to FcγRI (RG1 cells) is an effective B cell
activation agent. (A) BALB/c B cells (13105/well) were placed in
culture with optimal IL-4 and IL-5 levels plus 6000 mitomycin C-treated
RG1 (d), CHO control cells (u) or media alone (m) along with the
indicated amount of anti-IgD (10.4.22). For comparison, the indicated
levels of Hδa/1–dextran was also used as the activation agent (s).
Proliferation, measured by [3H]thymidine pulse at day 3, is shown
with Hδa/1–dextran activation on the right axis and other samples on
the left axis. (B) Activation as a function of B cell input per well;
symbols, proliferation and time scale are the same as in (A). Values
shown are the average of triplicate wells 6 1 SD of mean.
FcγRI–anti-IgD B cell activation 241
Fig. 2. Cytokine requirements for anti-IgD/FcγRI-mediated activation. BALB/c B cells (13105/well) were placed in culture with 1 µg/ml 10.4.22
and mitomycin C-treated RG1 cells as in Fig. 1. In (A) and (B) increasing doses of IL-4 are added in the presence of 5 ng/ml IL-5; (C) and
(D) have 10,000 U/ml of IL-4 with the indicated dose of IL-5. Proliferation (d) was determined as in Fig. 1 in (A) and (C), while IgM (j), IgG1
(s) or IgE (m) production was determined by ELISA after 8 days of culture is shown in (B) and (D). ELISA values shown are mean 6 1 SE.
RT-PCR product was blunt-ended with Klenow, phosphorylated with T4 kinase and, after agarose gel purification, ligated
into EcoRV cut pCDNA1 vector (Invitrogen, San Diego, CA).
Confirmation of correct CD40L sequence was obtained by
partial sequencing of the insert. Permanent transfectants were
obtained using CHO/DHFR– by co-transfection of CD40LpCDNA1 and pSV2-DHFR (5:1 ratio) using the calcium phosphate procedure (22). Positive clones were identified by
FACS using phycoerythrin–MR-1 (PharMingen) and CD40L
expression was further amplified using methotrexate. The
positive line used herein is termed 40L-2 and is maintained
in MEM-α medium containing 4 µM methotrexate. The preparation of the soluble CD40LT chimera is described elsewhere (23). The activity of the soluble CD40LT was further
enhanced by using an antibody, termed M15, against the
leucine zipper portion of the chimera; M15 and soluble
CD40LT were a generous gift of William Fanslow (Immunex,
Seattle, WA). In preliminary experiments, the optimal dose of
soluble CD40LT and M15 for B cell activation was determined
to be 0.1 µg/ml for both agents. CHO expressing FcgRI were
prepared and maintained in a manner similar to 40L-2, again
using co-transfection with pSV2-DHFR. A plasmid (mFcγRIpEE6) having the FcγRI cDNA (24) was a gift from Mark
Hogarth (Heidelberg, Australia) and the FcγRI-expressing
CHO is termed mRG1–1. Analysis of FcγRI expression is via
FACS using mouse IgG2a and F(ab9)2–anti-mouse IgG.
Proliferation assays
Support cells (40L-2, mRG1-1 or CHO control) were collected
by EDTA treatment and 1–33106 cells were incubated with
25 µg/ml mitomycin C for 30 min at 37°C in a foil wrapped tube.
Support cells were then washed four times by centrifugation,
resuspended in B cell media and the indicated cell number
added to a 96-well flat-bottomed culture plate (Costar,
Cambridge, MA). B cells, cytokines and anti-IgD were added
as indicated with a final volume of 200 µl/well. On day 3 after
culture initiation, the wells were pulsed for 8 h with 1 µCi/well
[3H]thymidine (6.2 Ci/mmol; NEN, Boston, MA). Cells were
then harvested using a PHD harvester (Cambridge Technology) and counted in a model 1220 LKB liquid scintillation
counter. All samples were run in triplicate wells and data is
expressed as c.p.m. 6 1 SD.
Ig production analysis
IgM, IgG1 and IgE production was determined in respective
supernatants after 8 days of culture. All Ig analyses were
performed by ELISA essentially as described previously (25).
Alkaline phosphatase-labeled or unlabeled polyclonal antiIgM and IgG1 were obtained from Southern Biotechnology
(Birmingham, AL). IgE assays utilized a pair of rat anti-mouse
IgE mAb, R1E4 and B1E3, again as described previously (26).
All ELISA determinations were performed with a quadruplicate
series of dilutions and read at dual wavelength (405/650 nm)
242 FcγRI–anti-IgD B cell activation
Fig. 3. IgG1 and IgM production is proportional to proliferation, while IgE is not seen, when activation is via FcγRI/RG1 (A) or anti-IgD–dextran
(B). Increasing doses/well of C57Bl/6 B cells (A) or BALB/c (B) B cells were activated with 1 µg/ml AF3.33 plus RG1 cells or 1 µg/ml Hδa/1–
dextran respectively. Media contained excess IL-4 and IL-5. Proliferation (d) is measured on day 3 and IgG1 (j), IgM (m) or IgE (r) production
was assayed after 8 days of culture. ELISA values shown are mean 6 1 SE.
with a Vmax ELISA reader (Molecular Devices, Sunnyvale
CA). Standard curves were run with the corresponding purified
myeloma/hybridoma proteins and four-parameter analysis
with Molecular Devices software. Dilution values that fell in
the linear portion of the curve were used for analysis.The
values shown represent duplicate samples determined at
multiple dilutions 6 1 SE.
Results
Efficient B cell activation with nanogram amounts of anti-IgD
bound to FcγRI
In order to determine whether FcγRI-bound anti-IgD would
efficiently cause B cell activation, CHO cells transfected with
the FcγRI were prepared, treated with mitomycin C and
allowed to bind to the wells of a culture plate. Subsequently,
purified mouse B cells were added along with an excess of
IL-4 and IL-5. The indicated amount of anti-IgDa or anti-IgDb
was then added and after 48 h, the cells were pulsed
with [3H]thymidine and harvested. When FcγRI1 cells were
present, nanogram levels of anti-IgD caused significant B cell
proliferation (Fig. 1). In contrast, the anti-IgD mAb are not
stimulatory at these concentrations, either by themselves or
when used with control untransfected CHO cells. Also
shown in Fig. 1 is a dose response for stimulation by anti-IgD
(Hδa/1)–dextran; the latter was initially noted to be a potent B
cell stimulator by Brunswick et al. (1). Note that the sensitivity
of the two systems is quite similar, although the level of
proliferation (day 3) is higher for Hδa/1–dextran. In Fig. 1(B),
the relationship of proliferation to B cell input is shown at a
constant anti-IgD level and, again, significant proliferation is
seen only when FcγRI is present.
Cytokine requirements for proliferation and Ig production
The requirements for the cytokines IL-4 and IL-5 was next
examined, and the results are shown in Fig. 2. At a constant
FcγRI–anti-IgD B cell activation 243
Fig. 4. Cytokine requirements for CD40L activation. 40L-2 cells (6000/well) were placed in culture with BALB/c B cells (5000/well) and the
indicated level of IL-4 added in the presence (d) or absence (n) of 5 µg/ml IL-5. Proliferation (6 1 SD) was measured on day 3 and ELISA
measurements (6 1 SE) were after 8 days of culture.
B cell level, increasing doses of IL-4 (at a constant IL-5 level)
resulted in both increasing proliferation and in Ig production.
Increasing doses of IL-5 at a constant IL-4 level had no effect
on proliferation; in other experiments (data not shown), IL-4
alone was shown to induce proliferation. However, a clear
dose response to IL-5 was seen for IgG1 and IgM production.
Note that in spite of the production of microgram quantities
of IgG1, indicating that IL-4-induced isotype switching to
IgG1 was evident, IgE production was not seen. Supernatants
were also assayed for production of IgG2a, IgG2b and IgG3,
and only low levels of these isotypes are produced under
these conditions (data not shown). Figure 3(A) shows the Ig
and proliferation response as a function of B cell concentration, using AF3.33 and C57Bl/6 B cells. For comparison
purposes, the response of BALB/c B cells to Hδa/1–dextran
is also shown (Fig. 3B). The only difference between the
response using C57Bl/6 and BALB/c B cells was a higher
IgM response with the C57Bl/6 system (Fig. 2 and data not
shown). Note that both proliferation and Ig production (IgG1
and IgM) are directly proportional to B cell input with the RG1
and the Hδa/1–dextran activation systems.
In order to compare the B cell activation parameters of
proliferation and Ig production of the polyvalent Ig activation
model used above with CD40L, both CD40L-transfected CHO
cells and a soluble CD40LT construct were used. Figure 4
shows the cytokine requirements when using CD40L-CHO
(40L-2) and, as can be seen, the main cytokine requirement
is IL-4. The only effect of IL-5 is an increase in IgM production.
When the activating agent was CD40LT, again enhanced IgM
production was seen and, to a variable degree, enhanced
IgG1 (data not shown). Note now that, as expected, an
excellent IgE response directly proportional to IL-4 levels
was seen with 40L-2 activation. The most striking difference
between the CD40L and anti-IgD activation systems was seen
when the variable was B cell input at a constant level of
cytokines and CD40L. With CD40L-mediated activation (either
the soluble construct or CD40L-transfected CHO), only IgM
production is proportional to B cell number. In contrast, IgE,
and to a lesser extent IgG1, production is most efficient at
low B cell inputs. Indeed with the CD40L activation system
the maximum IgE production was seen with as few as 500 or
5000 B cells/well with CD40L-CHO and CD40LT respectively.
In the experiment shown the lowest number of cells added
was 500 cells/well; however, we have noted in additional
experiments that the peak IgE response is between 300 and
1000 cells/well with the 40L-2 cell activation system. Note
that production of IgE equals or even slightly exceeds IgG1
under these optimal conditions of high IL-4 levels and low B
244 FcγRI–anti-IgD B cell activation
6 shows the results of an experiment in which both the
polyvalent anti-sIg system and CD40L-transfected cells were
placed into the same culture well and increasing numbers of
B cells were added along with media containing optimal
levels of IL-4 and IL-5. While proliferation was proportional to
B cell number, as seen with the CD40L activation system
alone, IgG1 and IgE production decreased with increasing B
cell numbers. In addition, the presence of the polyvalent
activation model induced a further inhibition of IgE production.
Note that little IgE is produced at any B cell concentration
when both activation systems are present. While the primary
effect was on IgE, decreases in IgG1 were also seen in some
experiments (data not shown). Consistent differences in either
IgM or in proliferation were not seen. A similar inhibition of
primarily IgE production was seen when co-activation was
performed using CD40LT instead of the 40L-2 cells (data
not shown).
This decrease was further confirmed by holding all reagents
constant except for anti-IgD and, as seen in Fig. 7, a dosedependent decrease in IgE production was again seen.
In the same experiment, proliferation, and IgG1 and IgM
production was also measured; as in the results shown in
Fig. 6, the most dramatic effect was on IgE production. Note
also that these experiments were performed with sort purified
B cells (.99% pure), arguing against that the effects seen
were due to any action of contaminating cells such as NK or
other non-B, non-T cells. In addition, the experiments shown
in Fig. 6 were with B cells from BALB/c while Fig. 7 used
C57Bl/6 B cells, further demonstrating that the effects seen
were not strain related. Note that at higher concentrations the
AF3.33 exhibits some inhibitory effect on IgE production, but
the effect is greatly accentuated by binding to the FcγRI.
Fig. 5. IgM production is proportional to proliferation while IgE and
IgG1 production is inversely proportional to B cell input using CD40LT
or 40L-2 activation. BALB/c B cells were activated with soluble
CD40LT and M15 (A) or 40L-2 cells (B) in the presence of excess
IL-4 and IL-5; the indicated number of B cells/well was added.
Proliferation (d) was measured on day 3 (6 1 SD) and IgM (m), IgG1
(u) or IgE (r) levels were determined after 8 days of culture by
ELISA (6 1 SE).
cell number. Indeed, in repeat experiments the maximum IgE
levels were seen at B cell inputs of 500–10,000 cells/well for
the 40L-2 and soluble CD40LT respectively, and always
decreased in a manner as shown in Fig. 5 at higher B cell
inputs. IgG1 production was more variable in appearance—
in some experiments a quite similar decrease to that seen
with IgE was seen while in others the IgG1 production
simply leveled off or slightly decreased. In contrast, with
the polyvalent Ig production model (Fig 3), IgG1 and IgM
production is proportional to proliferation and IgE production
is not detected.
Simultaneous activation by both CD40L and polyvalent
anti-sIg
In a secondary follicle, the presence of antigen–antibody
complexes and activated T cells would give a scenario where
both anti-sIg activation and CD40L activation are occurring
essentially simultaneously. Thus, we have explored the use
of this model to examine the effect of this co-activation. Figure
Discussion
Since its discovery as the antigen receptor for B lymphocytes,
sIg has been proposed for two relatively global roles. Firstly,
sIg serves to focus antigen on the correct B cell, resulting in
internalization, antigen processing/presentation and subsequent involvement of the resulting MHC class II peptide in
T cell activation. Secondly, cross-linking of sIg under defined
conditions itself results in B cell activation. Early studies with
purified anti-sIg demonstrated that co-cross-linking of the
FcγRII and sIg inhibited this activation (27), and more recent
studies have indicated that the mechanism for this inhibition
involves de-phosphorylating the B cell receptor (28,29). However, in an ongoing humoral immune response IL-4 would be
expected to largely prevent this inhibitory action (30). Thus,
the presence of antigen–antibody complexes could be anticipated to stimulate B cells by cross-linking sIg receptors, and
the actual humoral immune response seen would result from
stimulation of B cells via both CD40L and sIg. Note that the
reversal of this inhibition by IL-4 (30) also means that the
increased B cell activation seen with RG1 cells and anti-IgD
would not be expected to be a result of having the FcγRI
simply preventing the engagement of FcγRII on the B cell
surface. In further support of this concept, addition of the
FcγRII blocking mAb 2.4G2 (31) did not alter the B cell
activation capacity of either 10.4.22 or AF3.33 anti-IgD mAb
at the concentrations used in this study (data not shown).
FcγRI–anti-IgD B cell activation 245
Fig. 6. Co-activation by anti-IgD/RG1 and CD40L results in strong inhibition of IgE production. BALB/c B cells were activated with 6000 40L-2
cells/well plus either 6000 CHO control (u) or RG1 plus 1 µg/ml anti-IgD (10.4.22) (d). All CHO cells were mitomycin-C treated. Proliferation
(day 3) and Ig production (day 8) is shown as a function of B cell input/well. All wells contained excess IL-4 and IL-5.
To study this dual stimulation, we wished to first set up
a murine model where antigen–antibody complexes would
efficiently stimulate B cells. FcγRI-transfected CHO cells were
chosen since the higher affinity of the FcγRI (11,32) would
mean less monomeric IgG2a would be present in the culture
well. The results of Figs 1 and 2 clearly demonstrate that this
model was successful for B cell activation, and in the presence
of appropriate cytokines both proliferation and Ig synthesis
are seen. The model is most analogous to the polyvalent antiIgD–dextran system first described by Brunswick et al. (1),
and the anti-IgD–dextran model is directly compared in Figs
1 and 3. B cell activation is seen at lower input levels of antisIg and mAb that are essentially non-mitogenic give good
activation under polyvalent conditions, as has been observed
with the dextran model as well. Isotype switching can be
induced as is evidenced by IgG1 production when IL-4 is
present; however, analogous to the anti-IgD–dextran model
(4), IgE production was not seen under any of the conditions tested.
The generally accepted manner for induction of IgE in both
human and rodent models is now CD40L stimulation of B
cells in the presence of IL-4 or IL-13 (humans only) and the
secretion level of IgE, at least in mice, is further augmented
by IL-5 (for review see 33,34). An in vitro IgE response is
induced by activating mouse B cells with either recombinant
soluble or membrane bound CD40L in the presence of IL-4
and IL-5 (7,19). The critical nature of these components is
especially seen in that mice rendered genetically deficient in
either CD40L (35,36) or CD40 (37) or IL-4 (38) do not exhibit
a detectable IgE response. We prepared CD40L-CHO and
obtained a soluble CD40L construct (CD40LT) for analysis.
Figure 4 demonstrates that the primary cytokine required for
IgE and IgG1 is IL-4; this differs somewhat from earlier studies
where a requirement for IL-5 was also reported (7,8). However,
a clear enhancement of IgM production was seen with
increased IL-5 and when the soluble CD40LT construct was
used, IL-5 resulted in an increase in both IgM and IgG1. As
Fig. 5 illustrates, the most striking characteristic of the CD40L
activation is that the level of IgE produced is inversely
proportional to the B cell concentration, once a critical number
of cells are present. Thus, with CD40L-transfected cells or
soluble recombinant CD40L maximal IgE production is seen
at an input of 500–5000 cells/well respectively. This inverse
dependence on B cell levels for increased IgE production in
vitro was first observed using an IL-4 and lipopolysaccharide
activation scenario (39), although the results are certainly
more dramatic when the activating agent is CD40L. IgG1
levels also initially level off and then decrease with increasing
246 FcγRI–anti-IgD B cell activation
Fig. 7. Inhibition of IgE production occurs with sort purified B cells and is dose dependent on anti-IgD concentrations. Sort purified B cells
from C57Bl/6 mice (10,000/well) were activated with 40L-2 cells plus either CHO control (d) or RG1 (s) cells as in Fig. 6. The indicated dose
of AF3.33 anti-IgD was added and proliferation (day 3) or Ig production (day 8) was then determined. All wells contained excess IL-4 and IL-5.
B cell number, while IgM production remains directly proportional to B cell number, indicating that the decrease seen is
amplified when isotype switching is required. Since the
majority of IgE switching evidently occurs in two stages (40),
the intermediate being IgG1, this may relate to the increased
effect seen for IgE. The reason for this inverse relationship is
unknown. One agent that is known to inhibit IL-4 action on B
cells is IFN-γ (39), however, this suppression is not seen when
recombinant CD40L is used, suggesting that the mechanism
of the IFN-γ effect may relate to T cell CD40L expression
(41,42) rather than being a direct effect on B cells. In any
case, the effect is still seen when highly purified (sorted) B
cells are used (Fig. 7 and data not shown), arguing against
an agent made by a contaminating cell causing the suppression. We are currently examining whether the suppression
can be seen when supernatants from dense B cell cultures
are added to cultures producing the maximal levels of IgE.
We noted that the levels of IgE and IgG1 were quite similar
in situations where IgE production was optimized. However,
in vivo, levels of IgE are orders of magnitude lower than IgG
concentrations (mg/ml versus ng/ml); suggesting that IgE
synthesis is highly susceptible to inhibition. In developing
germinal centers, where isotype switching occurs, B cells are
being stimulated by both sIg and CD40L. Indeed, recent
evidence indicates that the sIg stimulation is necessary to
prevent apoptosis of CD40L-activated B cells, potentially via
a fas-mediated mechanism (43,44). Thus, we examined the
effect on CD40L induction of IgE that simultaneous sIg
stimulation would cause. The result was a strong downregulation of IgE production; while to a certain extent other
Ig classes were also effected, the most dramatic effect was
on IgE production. This result is analogous to that seen
when B cells were activated with activated T cells or T cell
membranes in the presence of anti-IgD–dextran (45), with
respect to the inhibition of IgE production. However, in contrast
to those studies, we did not see an enhancement of IgG1
and IgM production under co-activation conditions. In a latter
study, the effect of anti-IgD–dextran on B cells activated with
a sub-optimal dose of CD40L found that Ig production,
including IgE, can be enhanced (8). Using the polyvalent
FcγRI–anti-IgD model and varying doses of CD40L, we have
not seen significant enhancement of any isotype production.
While trivial explanations such as differences in CD40L construct cannot be completely ruled out, the results suggest
that inhibition of IgE production would be the expected result
when B cells are co-activated with immune complexes and
CD40L in vivo. Thus, one of the mechanisms operating to
modulate the level of IgE production in vivo is potentially the
FcγRI–anti-IgD B cell activation 247
B cell activation via sIg that occurs in germinal centers in
a time frame that is quite similar if not identical to the
CD40L signal.
16
17
Acknowledgements
The authors acknowledge the expert technical assistance of Elaine
Studer, Michelle Kilmon and Claud Johnson during the course of
these studies. Supported by NIH grants AI18797 and AI34631.
C. S. W. was supported in part by a teaching assistantship from
Virginia Commonwealth University.
18
19
Abbreviations
RG1 cells
40L-2
CHO transfected with the mouse FcγRI cDNA
CHO transfected with mouse CD40L cDNA
References
1 Brunswick, M., Finkelman, F. D., Highet, P. F., Inman, J. K., Dintzis,
H. M. and Mond, J. J. 1988. Picogram quantities of anti-Ig
antibodies coupled to dextran induce B cell proliferation. J.
Immunol. 140:3364.
2 Mond, J. J., Lees, A. and Snapper, C. M. 1995. T cell-independent
antigens type 2. Annu. Rev. Immunol. 13:655.
3 Snapper, C. M., McIntyre, T. M., Mandler, R., Pecanha, L. M. T.,
Finkelman, F. D., Lees, A. and Mond, J. J. 1992. Induction of
IgG3 secretion by interferon gamma: a model for T cellindependent class switching in response to T cell-independent
type 2 antigens. J. Exp. Med. 175:1367.
4 Snapper, C. M., Pecanha, L. M. T., Levine, A. D. and Mond, J. J.
1991. IgE class switching is critically dependent upon the nature
of the B cell activator, in addition to the presence of IL-4. J.
Immunol. 147:1163.
5 McIntyre, T. M., Kehry, M. R. and Snapper, C. M. 1995. Novel
in vitro model for high-rate IgA class switching. J. Immunol.
154:3156.
6 Kehry, M. R. and Hodgkin, P. D. 1994. B-cell activation by helper
T-cell membranes. Crit. Rev. Immunol. 14:221.
7 Maliszewski, C. R., Grabstein, K., Fanslow, W. C., Armitage, R.,
Spriggs, M. K. and Sato, T. A. 1993. Recombinant CD40 ligand
stimulation of murine B cell growth and differentiation: cooperative
effects of cytokines. Eur. J. Immunol. 23:1044.
8 Snapper, C. M., Kehry, M. R., Castle, B. E. and Mond, J. J.
1995. Multivalent, but not divalent, antigen receptor cross-linkers
synergize with CD40 ligand for induction of Ig synthesis and class
switching in normal murine B cells: a redefinition of the TI-2 vs T
cell-dependent antigen dichotomy. J. Immunol. 154:1177.
9 Carter, R. H. and Fearon, D. T. 1992. CD19: lowering the threshold
for antigen receptor stimulation of B lymphocytes. Science
256:105.
10 Fearon, D. T. and Carter, R. H. 1995. The CD19/CR2/TAPA-1
complex of B lymphocytes: linking natural to acquired immunity.
Annu. Rev. Immunol. 13:127.
11 Unkeless, J. C. and Herman, N. E. 1975. Binding of monomeric
immunoglobulins to Fc receptors of mouse macrophages. J. Exp.
Med. 142:1520.
12 Canfield, S. M. and Morrison, S. L. 1991. The binding affinity of
human IgG for its high affinity Fc receptor is determined by
multiple amino acids in the CH2 domain and is modulated by the
hinge region. J. Exp. Med. 173:1483.
13 Oi, V. T. and Herzenberg, L. A. 1979. Localization of murine Ig1b and Ig-1a (IgG 2a) allotypic determinants detected with
monoclonal antibodies. Mol. Immunol. 16:1005.
14 Stall, A. M. and Loken, M. R.1984. Allotypic specificities of
murine IgD and IgM recognized by monoclonal antibodies. J.
Immunol. 132:787.
15 Noelle, R. J., Roy, M., Shepherd, D. M., Stamenkovic, I., Ledbetter,
J. A. and Aruffo, A. 1992. A 39-kDa protein on activated helper
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
T cells binds CD40 and transduces the signal for cognate
activation of B cells. Proc. Natl Acad. Sci. USA 89:6550.
Urlaub, G. and Chasin, L. A. 1980. Isolation of Chinese hamster
cell mutants deficient in dihydrofolate reductase activity. Proc.
Natl Acad. Sci. USA 77:4216.
Bartlett, W. C., Kelly, A. E., Johnson, C. M. and Conrad, D. H.
1995. Analysis of murine soluble FcεRII: sites of cleavage and
requirements for dual affinity interaction with IgE. J. Immunol.
154:4240.
DeFranco, A. L., Raveche, E. S., Asofsky, R. and Paul, W. E. 1982.
Frequency of B lymphocytes responsive to anti-immunoglobulin. J.
Exp. Med. 155:1523.
Armitage, R. J., Fanslow, W. C., Strockbine, L., Sato, T. A., Clifford,
K. N., Macduff, B. M., Anderson, D. M., Gimpel, S. D., DavisSmith, T., Maliszewski, C. R., Clark, E. A., Smith, C. A., Grabstein,
K. H., Cosman, D. and Spriggs, M. K. 1992. Molecular and
biological characterization of a murine ligand for CD40. Nature
357:80.
Tony, H. P. and Parker, D. C. 1985. Major histocompatibility
complex-restricted, polyclonal B cell responses resulting from
helper T cell recognition of anti-immunoglobulin presented by
small B lymphocytes. J. Exp. Med. 161:223.
Greene, J. M. and Struhl, K. 1989. Preparation and analysis of
RNA. In Ausubel, F. M. Kingston, R. E., Moore, D. D., Seidman,
J. G., Smith, J. A. and Struhl, K., eds, Current Protocols in
Molecular Biology. Greene/Wiley-Interscience, New York.
Kingston, R. E., Chen, C. A. and Okayama, H. 1993. Calcium
phosphate transfection. In Ausubel, F. M., Kingston, R. E., Moore,
D. D., Seidman, J. G., Smith, J. A. and Struhl, K. eds, Current
Protocols in Molecular Biology, p. 9.1.1. Greene/WileyInterscience, New York.
Fanslow, W. C., Srinivasan, S., Paxton, R., Gibson, M. G., Spriggs,
M. K. and Armitage, R. J. 1994. Structural characteristics of
CD40 ligand that determine biological function [Review]. Semin.
Immunol. 6:267.
Quilliam, A. L., Osman, N., McKenzie, I. F. C. and Hogarth,
P. M. 1993. Biochemical characterization of murine FcgammaRI.
Immunology 78:358.
Campbell, K. A., Lees, A., Finkelman, F. D. and Conrad, D. H. 1992.
Co-crosslinking FcεRII/CD23 with B cell surface immunoglobulin
modulates B cell activation. Eur. J. Immunol. 22:2107.
Keegan, A. D., Fratazzi, C., Shopes, B., Baird, B. and Conrad,
D. H. 1991. Characterization of new rat anti-mouse IgE
monoclonals and their use along with chimeric IgE to further define
the site that interacts with FcεRII and FcεRI. Mol. Immunol. 28:1149.
Phillips, N. E. and Parker, D. C. 1984. Cross-linking of B
lymphocyte Fc gamma receptors and membrane immunoglobulin
inhibits anti-immunoglobulin-induced blastogenesis. J. Immunol.
132:627.
D’Ambrosio, D., Hippen, K. L., Minskoff, S. A., Mellman, I., Pani,
G., Siminovitch, K. A. and Cambier, J. C. 1995. Recruitment and
activation of PTP1C in negative regulation of antigen receptor
signaling by Fc gamma RIIB1 [see comments]. Science 268:293.
Pani, G., Kozlowski, M., Cambier, J. C., Mills, G. B. and
Siminovitch, K. A. 1995. Identification of the tyrosine phosphatase
PTP1C as a B cell antigen receptor-associated protein involved
in the regulation of B cell signaling [see comments]. J. Exp.
Med. 181:2077.
O’Garra, A., Rigley, K. P., Holman, H., McLaughlin, J. B. and
Klaus, G. G. B. 1987. B cell stimulatory factor 1 reverses Fc
receptor-mediated inhibition of B lymphocyte activation. Proc.
Natl Acad. Sci. USA 84:6254.
Unkeless, J. C. 1979. Characterization of a monoclonal antibody
directed against mouse macrophage and lymphocyte Fc
receptors. J. Exp. Med. 150:580.
Sears, D. W., Osman, N., Tate, B., McKenzie, I. F. C. and Hogarth,
P. M. 1990. Molecular cloning and expression of the mouse high
affinity Fc receptor for IgG. J. Immunol. 144:371.
Noelle, R. J. 1995. The role of gp39 (CD40L) in immunity. Clin.
Immunol. Immunopathol. 76:S203.
Banchereau, J., Bazan, F., Blanchard, D., Brière, F., Galizzi, J. P.,
Van Kooten, C., Liu, Y. J., Rousset, F. and Saeland, S. 1994. The
CD40 antigen and its ligand. Annu. Rev. Immunol. 12:881.
248 FcγRI–anti-IgD B cell activation
35 Renshaw, B. R., Fanslow, W. C., III, Armitage, R. J., Campbell,
K. A., Liggitt, D., Wright, B., Davison, B. L. and Maliszewski, C. R.
1994. Humoral immune responses in CD40 ligand-deficient mice.
J. Exp. Med. 180:1889.
36 Xu, J., Foy, T. M., Laman, J. D., Elliott, E. A., Dunn, J. J.,
Waldschmidt, T. J., Elsemore, J., Noelle, R. J. and Flavell, R. A.
1994. Mice deficient for the CD40 ligand. Immunity 1:423.
37 Kawabe, T., Naka, T., Yoshida, K.,, Tanaka, T., Fujiwara, H.,
Suematsu, S., Yoshida, N., Kishimoto, T. and Kikutani, H. 1994.
The immune responses in CD40-deficient mice: impaired
immunoglobulin class switching and germinal center formation.
Immunity 1:167.
38 Kühn, R., Rajewsky, K. and Müller, W. 1991. Generation and
analysis of interleukin-4 deficient mice. Science 254:707.
39 Coffman, R. L., Ohara, J., Bond, M., Carty, J., Zlotnik, A. and
Paul, W. E. 1986. B cell stimulatory factor-1 enhances the IgE
response of lipopolysaccharide-activated B cells. J. Immunol.
136:4538.
40 Yoshida, K., Matsuoka, M., Usuda, S., Mori, A., Ishizaka, K. and
Sakano, H. 1990. Immunoglobulin switch circular DNA in the
41
42
43
44
45
mouse infected with Nippostrongylus brasiliensis: evidence for
successive class switching from µ to ε via gamma1. Proc. Natl
Acad. Sci. USA 87:7829.
Gauchat, J., Aubry, J., Mazzei, G., Life, P., Jomotte, T., Elson, G.
and Bonnefoy, J. 1993. Human CD40-ligand: molecular cloning,
cellular distribution and regulation of expression by factors
controlling IgE production. FEBS Lett. 315:259.
Roy, M., Waldschmidt, T., Aruffo, A., Ledbetter, J. A. and Noelle,
R. J. 1993. The regulation of the expression of gp39, the CD40
ligand, on normal and cloned CD41 T cells. J. Immunol. 151:2497.
Rothstein, T. L., Wang, J. K. M., Panka, D. J., Foote, L. C., Wang,
Z., Stanger, B., Cui, H., Ju, S.-T. and Marshak-Rothstein, A. 1995.
Protection against Fas-dependent Th1-mediated apoptosis by
antigen receptor engagement in B cells. Nature 374:163.
Garrone, P., Neidhardt, E. M., Garcia, E., Galibert, L., Van Kooten,
C. and Banchereau, J. 1995. Fas ligation induces apoptosis of
CD40-activated human B lymphocytes. J. Exp. Med. 182:1265.
Pecanha, L. M. T., Yamaguchi, H., Lees, A., Noelle, R. J., Mond,
J. J. and Snapper, C. M. 1993. Dextran-conjugated anti-IgD
antibodies inhibit T cell-mediated IgE production by augment the
synthesis of IgM and IgG. J. Immunol. 150:2160.