| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
RIIB on Germinal Center Cells: Implications for Selection of High-Affinity B Cells1
Kimmel Cancer Center and Department of Microbiology and Immunology, Thomas Jefferson University, Philadelphia, PA 19107
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RIIB, mediates
inhibition of B cell receptor-triggered events in primary B cells. We
investigated the expression of FcRIIB on germinal center (GC) cells
to better understand its role in memory B cell development.
Immunohistological analyses demonstrated differential regulation of
FcRIIB on GC cells. Its levels are markedly down-regulated on GC B
cells and up-regulated on follicular dendritic cells (FDC) at all times
during the GC response. Analyses of surface expression of FcRIIB by
flow cytometry and FcRIIB mRNA levels by RT-PCR analysis confirmed
that this FcR is down-regulated in GC B cells. In mice lacking
FcRIIB, the development of the secondary FDC reticulum in GCs is
substantially delayed, although the overall kinetics of the GC response
are unaltered. These findings have direct implications for models
proposed to account for the selection of high-affinity B cells in the
GC and suggest a role for FcRIIB in promoting the maturation of the
FDC reticulum. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
R) are widely distributed on
hemopoietic cells and play an important role in immune regulation by
linking Ab-mediated responses with effector functions. The murine
high-affinity FcRI and the low-affinity FcRII and FcRIII
(1) are distinguished by their structure, function,
distribution, and affinities for IgG (1, 2). These FcR
contain homologous extracellular domains; however, they vary
considerably in their transmembrane and cytoplasmic regions. FcRI
and FcRIII are associated with a common -chain, which contains
the immunoreceptor tyrosine-based activation motif in its cytoplasmic
tail. In contrast, FcRII contains in its cytoplasmic tail an
immunoreceptor tyrosine-based inhibitory motif (3).
FcRIIB is encoded by a single gene in the mouse. However,
alternative mRNA splicing results in two FcRII isoforms, FcRIIb1
and FcRIIb2 (4, 5). Murine B cells exclusively express
FcRIIb1 and lack expression of the activation FcRs
(6). Coaggregation of the B cell Ag receptor
(BCR)5 and FcRIIb1
on B cells results in the inhibition of Ag internalization, activation,
proliferation, and Ab secretion (7, 8, 9, 10). Previous in vitro
studies have also shown that FcRIIB is capable of independently
generating an apoptotic signal upon clustering
(10, 11, 12).
Germinal centers (GC) are inducible lymphoid microenvironments that
arise primarily in response to T-dependent Ags. During the GC response,
a high frequency of mutations is introduced in the variable regions of
the BCRs expressed by GC B cells. Ag present in the form of immune
complexes (IC) retained on follicular dendritic cells (FDC) within the
GC is thought to drive the selection of high-affinity B cell mutants,
ultimately resulting in the formation of the memory B cell compartment.
During this process of interaction between ICs and GC B cells, it is
believed that B cells bearing high-affinity BCR are preferentially
stimulated, resulting in the uptake of Ag and its subsequent
presentation to Th lymphocytes (13, 14, 15). This model
considers only the role of BCR but ignores the role of FcRIIB in B
cell selection upon interaction with ICs. Given the general inhibitory
role of FcRIIB, it is unclear how GC B cells, on interaction with
ICs, could be activated and internalize the Ag for presentation to
Ag-specific Th cells.
In this context, it has been suggested that the consequence of
coengagement of BCR and FcRIIB by ICs retained on FDC in GC is
dependent on the balance between concurrent activation and inhibitory
signals, leading to stimulatory, inhibitory, or apoptotic responses
(10). FcRIIB expression is also induced on FDC within
the GC, and it has been proposed that this high level expression allows
FDCs to convert ICs to a highly immunogenic form for B cell activation
(16). Interestingly, none of the studies leading to these
models has examined the expression levels of FcRIIB on GC B cells.
Recent studies have indicated that the surface expression of FcRIIB
is reduced on GC B cells in autoimmune strains of mice (17, 18). In contrast, in vitro experiments have indicated that
activated B cell blasts differ from resting B cells in that they
express significantly elevated levels of FcRIIB, as well as its
downstream signaling molecules Src homology domain 2-containing
inositol 5-phosphatase and Src homology domain 2-containing inositol
5-phosphatase-2 (19). These data highlight the
importance of elucidating the role of FcRIIB in memory B cell
development and the GC reaction.
In the present study, we examined whether expression of FcRIIB is
modulated in GC B cells during a T-dependent immune response and have
characterized the GC response in mice lacking this receptor. Our
results indicate that both FcRIIB mRNA and protein expression are
significantly down-regulated in GC B cells as compared with non-GC B
cells. In addition, whereas the kinetics and magnitude of the GC
response appear unaltered in FcRIIB-deficient mice, the development
of FDC reticula in GCs is delayed, supporting a role for this receptor
in FDC maturation. The implications of these results are discussed in
the context of models proposed to account for selection of
high-affinity B cells in the GC.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FcRIIB-/- mice on a C57BL/6
background (20) and C57BL/6 mice were purchased from
Taconic Farms (Germantown, NY). The generation and characterization of
common -chain-deficient mice have been described previously
(21). These mice, obtained from Dr. J. Ravetch
(Rockefeller University, New York, NY), had been backcrossed to C57BL/6
mice for eight generations. All mice were maintained in pathogen-free
conditions. Mice (10–12 wk old) were immunized i.p. with 3 x
108 SRBC (Lampire Biological Labs, Pipersville,
PA) in PBS or 100 µg 4-hydroxy-3-nitrophenyl acetyl-chicken
-globulin (NP13-CGG) in alum and were
sacrificed at different time intervals for analysis.
Antibodies
The following Abs were used for immunohistology or flow
cytometric analysis: PE-Texas Red-conjugated anti-B220 (clone
RA3-6B2; Caltag Laboratories, Burlingame, CA); FITC-labeled GL7;
PE-labeled anti-FcRII/FcRIII (clone 2.4G2); PE-labeled
anti-B220 (clone RA3-6B2); biotin-labeled anti-CD35 (clone
8C12) (all purchased from BD PharMingen, San Diego, CA); unconjugated
rat IgG Ab to mouse FDCs (FDC-M1 and FDC-M2, gifts from Dr. M.
Kosco-Vilbois, Serono Pharmaceutical Research Institute,
Plan-les-Quates, Switzerland); FITC-labeled MOMA-2 (Serotec, Oxford,
U.K.); biotin-labeled (Fab')2 mouse anti-rat
IgG (Jackson ImmunoResearch Laboratories, West Grove, PA);
biotin-labeled peanut agglutinin (PNA; Sigma-Aldrich, St. Louis, MO);
PE or biotin-labeled anti-IgD (clone 11-26; Southern Biotechnology
Associates, Birmingham, AL) and streptavidin (SA)-PE (Molecular Probes,
Eugene, OR). The K9.361 hybridoma was a kind gift from Dr. U.
Hammerling (Sloan Kettering Memorial Hospital, New York, NY).
Histology
Spleens from immunized mice were obtained at various times after immunization and processed for histological analysis as previously described (22). Briefly, spleens were embedded in Tissue-Tek OCT compound (Fisher Scientific, Bridgewater, NJ) by freezing in a 2-methylbutane bath cooled with liquid nitrogen. Frozen spleens were stored at -70°C until sectioned. Sections of 6 µm were cut on a cryostat microtome and mounted onto 0.01% poly-L-lysine (Sigma-Aldrich)-coated slides. The sections were air dried for 1 h, fixed in acetone for 15 min, air dried again for at least 2 h, and stored at -70°C until further analysis.
Frozen sections were rehydrated in TBS for 20 min followed by blocking in TBS, 5% BSA, 0.05% Tween 20 for another 20 min. Double immunohistology was then performed by staining with fluorescent Abs described above for 30 min. For unconjugated Abs, as a secondary step, sections were stained for 30 min with biotinylated mouse anti-rat Ab. The slides were then washed three times in TBS, 5% BSA, 0.05% Tween 20, and the biotinylated Abs were revealed by SA-PE. The slides were washed and stored in PBS in the dark. The stained sections were analyzed using a fluorescence microscope (Leitz Diaplan, Leitz, Wetzlar, West Germany), and digital images were captured using a Kodak DC290 Zoom digital camera and MDS-290 software (Eastman Kodak, Rochester, NY).
Flow cytometry and cell sorting
Cell suspensions prepared from spleens excised from mice on day
8 post-SRBC immunization were depleted of erythrocytes by ammonium
chloride Tris lysis (23). For four-color surface
staining, 2 x 106 cells resuspended in PBS
containing 2% BSA were incubated with pretitered dilutions of
GL7-FITC, biotinylated anti-IgD, anti-B220-PET-Texas Red, and
2.4G2-PE for 30 min at 4°C. SA Red 670 was used as a second-step
reagent. Cells were analyzed using a Coulter Epics XL/MCL analyzer, and
the data were analyzed by WinMDI software (The Scripps Research
Institute, La Jolla, CA). For sorting, spleen cells prepared from
C57BL/6 or Fc chain-/- mice on day 8
post-SRBC immunization were stained with GL7-FITC, anti-B220-PE and
anti-IgD-biotin followed by SA Red 670. GC and non-GC B cells were
separated on a Coulter Epics Elite flow cytometric cell sorter.
RT-PCR assay
mRNA was extracted from sorted GC and non-GC B cells using the
RNeasy Miniprep kit (Qiagen, Valencia, CA) according to the
manufacturer’s recommendation. Using 0.2 or 0.5 µg of RNA from each
sample, cDNA synthesis and FcRIIB-specific PCR amplification was
performed in a single step using the Superscript one-step RT-PCR system
(Life Technologies, Gaithersburg, MD). FcRIIB cDNA was amplified
using the 5' primer 5'-AAGTCTAGGAAGGACACTGC-3' and the 3' primer
5'-ATCCTGGCCTTCTGGCTTGC-3'. As an internal control, 
-actin mRNA
was amplified using primers purchased from Clontech Laboratories (Palo
Alto, CA). RT-PCR was conducted in 50-µl volumes with one cycle
programmed to perform cDNA synthesis at 50°C for 30 min and 94°C
for 2 min, followed immediately by PCR amplification for 30 cycles at
95°C for 1 min, 55°C for 45 s, and 72°C for 1 min. A final
extension was done at 72°C for 5 min. One-fifth volume of the PCR was
electrophoresed on a 1.5% agarose gel, and the bands were visualized
under UV light.
Gel hybridization
Agarose gels were dried at 65°C for 2 h under vacuum,
denatured in 0.5 M NaOH, 1.5 M NaCl for 1 h, and neutralized for
1 h in 0.5 M Tris, 3 M NaCl. The gels were then prehybridized at
42°C in prehybridization buffer solution containing 6x 0.6 M NaCl,
0.15 M Tris (pH 8.0), 6 mM EDTA; 0.05 M Denhardt’s solution; 0.1%
SDS; and 250 µg of herring sperm DNA. Probes were prepared from PCR
products by 32P labeling using a random primed
DNA labeling kit (Roche Molecular Biochemicals, Mannheim, Germany),
denatured, and added to the prehybridization buffer, and the gels were
hybridized overnight at 42°C. The gels were washed at low stringency
in 5x SSC, 5% SDS for 1 h followed by two 30-min washes at
43°C. Subsequently, a high stringency wash in 0.1x SSC, 0.1% SDS at
60°C was performed. The gels were scanned on a PhosphorImager
(Molecular Dynamics, Sunnyvale, CA) after overnight exposure. The bands
were quantitated after subtracting the background signal using Image
Quant software (Molecular Dynamics). The integrated intensity of each
band representing individual PCR products was measured and normalized
using the -actin band intensity for each individual sample.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SRBC have been previously shown to induce robust GC responses in
an adjuvant-independent manner (24, 25). However, only
limited information was available regarding the magnitude and kinetics
of this response. Therefore, we first examined the number and sizes of
GC in the spleens of SRBC immunized C57BL/6 mice. Spleen sections
obtained from days 4–40 after immunization were stained with GL7. The
total number of GCs in each section were counted and categorized into
three different sizes by counting GL7+ cells as
previously described (26). A summary of the results is
presented in Fig. 1
. On day 4 after SRBC
immunization, the majority of GL7+ cells were
scattered in follicles. Small foci of GL7+ cells
localized within the FDC network as well as a few well-developed small
GCs were detectable (data not shown). By day 6, there was a large
expansion in the number of GC comprising all sizes. This large number
of GCs was maintained until day 12, after which the numbers declined.
Large GCs were present from day 6, reached a peak by day 10, and
thereafter declined with a corresponding increase in medium and small
GCs. Although the number of medium sized GCs started to decline after
day 12, the small GCs continued to increase, perhaps due to the
dissolution of some of the large and medium sized GCs. The numbers of
small GC gradually declined from day 16 onward, and by day 40 only a
few GCs were present, almost all of which were small.
|
Although a strong correlation exists between the patterns of PNA
and GL7 staining, earlier studies reported the presence of both
PNA+GL7+ as well as
PNA+GL7- cell populations
in spleen (27, 28). Immunohistological analysis of spleen
sections from C57BL/6 mice derived 8 days after primary SRBC
immunization was performed by staining with GL7 and PNA. As shown in
the inset in Fig. 2, most
PNA+ cells are GL7+ (and
thus appear yellow), although some areas stained positive for either
GL7 only or PNA only, suggesting heterogeneity in the expression of
their respective ligands. Further, a rim of PNA+
cells that were GL7- (Fig. 2, arrow) was also
detected surrounding the GC. Although the significance of these
differences in the staining pattern of GL7 and PNA is presently
unclear, the results show that there is a more restricted pattern of
expression of the ligands for GL7 than of the ligands for PNA in GCs.
Therefore, we used GL7 as a marker to define GC cells in our subsequent
analyses.
|
RIIB expression on GC cells
We next examined whether there was any difference in the level of
FcRIIB expression on GC and follicular B cells. In contrast to
macrophage/monocyte lineage cells, B cells exclusively express the b1
isoform of FcRII and have been reported to lack expression of other
Fc family receptors such as FcRI, FcRIII, and the b2 isoform
of FcRII (1, 3). The mAb 2.4G2 recognizes the
extracellular domains of FcRII and FcRIII. To avoid FcRIII
staining with 2.4G2, we immunized Fc
chain-/- mice with SRBC. These mice express
neither FcRI nor FcRIII due to the absence of the common
-chain, required for assembly and membrane transport of these
receptors. However, FcRIIB expression is not perturbed in these
mice. Immunohistological analysis of spleen sections from Fc
chain-/- mice that were immunized with SRBC 8
days earlier was performed using 2.4G2 in combination with GL7 and
anti-B220. Representative images are shown in Fig. 2
. As shown in
Fig. 2B, follicular mantle B cells clearly stain with 2.4G2,
and the marginal zone stains even more strongly with 2.4G2, suggesting
relatively high levels of FcRIIB expression.
GL7+ GCs revealed an interesting pattern of 2.4G2
staining. Although FcRIIB expression was undetectable on B cells
toward one side of the GC, intense 2.4G2 staining was observed
polarized to the other side of the GC, suggesting up-regulated
FcRIIB expression in this area. Similar results were obtained when
sections were stained with PNA and 2.4G2 (Fig. 2A). This
sectored pattern of FcRIIB expression was consistently observed.
Parallel sections stained with Ab to B220 and 2.4G2 were then analyzed.
As shown in Fig. 2C, B220+ marginal
zone cells exhibited relatively high levels of FcRIIB (and thus
appear orange in Fig. 2C), and the
B220+ follicular mantle cells also coexpressed
FcRIIB (and thus appear yellow in Fig. 2C). In contrast,
the B220+ areas corresponding to GC exhibited
sectoring, with undetectable levels of FcRIIB in one area (thus
appearing green; Fig. 2C, arrow) and high levels of
FcRIIB in the other area.
We then examined whether the observed differential regulation of
FcRIIB was also a feature of GC responses to the well-characterized
model Ag, NP. Day 10 spleen sections from
NP13-CGG-immunized Fc
chain-/- mice were stained with 2.4G2 and GL7
or anti-B220. These experiments also showed sectoring, with high
levels of FcRIIB in one region and low to undetectable levels in the
other region of GCs (Fig. 2, D and E). The
analysis was further extended by using an FcRIIB-specific mAb,
K9.361 (29, 30, 31), that was made available to us at a later
time during the study. K9.361 exclusively recognizes FcRIIB (Ly17.2
allotype of Ly17.1/Ly17.2 alloantigen system) in C57BL/6 mice and does
not bind to FcRIII (31). Immunohistological analysis of
day 8 spleen sections from SRBC-immunized C57BL/6 mice with K9.361 and
GL7 or anti-B220 revealed exactly the same sectored pattern of GC
staining observed with 2.4G2 (Fig. 2, F and G).
These data collectively indicate that the expression of FcRIIB is
differentially modulated in these two areas of GCs.
FcRIIB expression is markedly reduced on GC B cells
and up-regulated on FDC
We previously demonstrated that FcRIIB is up-regulated on FDC
within the GC (16). In the experiments described above
(Fig. 2), we could not evaluate whether the GC B cells present in the
FDC-rich regions of the GC also expressed FcRIIB. To
investigate this issue, chimeric mice were created by lethally
irradiating FcRIIB-/- mice and
reconstituting them with bone marrow from
FcRIIB+/+ C57BL/6 mice. Because FDCs are
resistant to radiation (32), they are derived from the
host in these chimeras and therefore cannot express FcRIIB. The
donor cells, including B cells, however, can express the FcRIIB
receptor. Day 9 spleen sections from NP-CGG-immunized chimeric mice
were examined for expression of FcRIIB in GCs. In the absence of
FcRIIB expression on FDC, any staining of GC cells with the 2.4G2 Ab
should result from FcR expression on B cells. As shown in Fig. 3, although FDCs were present in the GC,
there was a lower level of FDC-M1 staining in general (Fig. 3A) and as expected, 2.4G2 staining on FDC was absent. B
cells in these GCs, including those present in the FDC reticulum;
exhibited markedly reduced levels of 2.4G2 staining compared with
mantle zone B cells (Fig. 3B).
|
RIIB at all
stages of the GC response
We further investigated whether reduced expression of FcRIIB on
GC B cells is characteristic of all stages of the GC response.
FcRIIB expression was evaluated on SRBC-immunized C57BL/6 mice from
day 4 to day 30 and day 4 to day 16 in Fc
chain-/- mice. Spleen sections were stained
with GL7 in combination with 2.4G2, FDC-M1, or FDC-M2. Representative
images are shown in Fig. 4. The results
of this analysis revealed "sectoring" of GC due to up-regulated
FcRIIB levels in the region rich in FDCs and low to undetectable
levels of FcRIIB expression on GC B cells where the FDC network was
absent. These observations were made at all time points of the GC
response starting from day 4 when GC clusters first appeared (Fig. 4A). Furthermore, this was consistent for all GC sizes.
|
chain-/- mice, we observed an
accelerated kinetics of the GC response. In contrast to C57BL/6 mice,
there were many large GCs by day 4, and by day 16 most of the GCs were
small and the numbers were substantially reduced (data not shown).
These changes in GC kinetics might be due to reduced clearance and
increased deposition of immune complexes in Fc
chain-/- mice as we previously reported
(22). FDC organization and development, however, appeared
normal in these mice (Fig. 4B). Consistent with the findings
in C57BL/6 mice (Fig. 4A), the GCs in Fc
chain-/- mice exhibited sectoring, with
increased levels of FcRIIB on FDC and low to undetectable levels on
GC B cells in FDC-deficient areas at all time points (Fig. 4B).
Flow cytometric and RT-PCR analysis of FcRIIB expression in GC B
cells
As an alternative assay for FcRIIB expression on splenic GC B
cells, we used flow cytometry. The mAbs 2.4G2 and K9.361 were used to
determine the levels of FcRIIB on B cells from Fc
chain-/- and C57BL/6 mice, respectively.
Four-color immunofluorescence analysis was performed, and surface
expression of FcRIIB was analyzed on B220+
gated GL7+IgD- (GC) and
GL7-IgD+ (non-GC) B cells.
As shown in Fig. 5, compared with the
GL7-IgD+ non-GC B cell
population, there was a substantial decrease of FcRIIB expression on
GL7+IgD- GC B cells. The
extent of surface FcRIIB down-regulation on GC B cells detected with
2.4G2 (Fig. 5A) and K9.361 (Fig. 5B) was similar,
thus confirming our immunohistology observations (Fig. 2). It is
possible that due to the differences in the sensitivities of flow
cytometry and fluorescence microscopy techniques, low level expression
of FcRIIB on GC B cells was not detectable in immunohistological
analyses. However, the unimodal distribution of FcRIIB staining
indicates that all GC B cells exhibit this low expression level with no
evidence for the presence of any minor population of GC B cells that
express FcRIIB at levels similar to
GL7-IgD+ B cells.
|
RIIB, RNA prepared from sorted
GL7+IgD- GC cells from
immunized C57BL/6 mice was analyzed using RT-PCR to determine the
levels of FcRIIB encoding mRNA. These levels were compared with
those in the
GL7-IgD+-naive B cell
population. FcRIIB mRNA from both cell populations was amplified
using gene-specific primers (see Materials and Methods).
-Actin mRNA served as an internal control to ensure that equal
quantities of cDNA were amplified in the reactions. To test the
linearity of PCR amplification and the extent of down-regulation, PCR
products removed after different numbers of PCR cycles were subjected
to agarose gel electrophoresis. The gel was hybridized with an
FcRIIB-specific probe, and the relative intensity of each band
representing individual PCR products was measured on a PhosphorImager
and expressed as units of integrated band intensity (see
Materials and Methods). The results of this analysis showed
nearly a 6-fold reduction of FcRIIB mRNA in GC B cells compared with
non-GC IgD+ B cells (Fig. 6).
|
RIIB-deficient mice
The differential regulation of FcRIIB on GC B cells and FDC
prompted us to examine the GC response in mice lacking this receptor.
Histological analysis of spleen sections from day 6 to day 22 post-SRBC
immunization was performed on FcRIIB-/- and
C57BL/6 mice with GL7, FDC-M1, and FDC-M2. In C57BL/6 mice, the GC FDC
reticulum was seen as fine processes extending throughout one side of
the GC (see Fig. 4). This was true at all time points after
immunization. In contrast, as shown in Fig. 7, this reticular pattern of staining was
very weak in FcRIIB-deficient GCs at early and intermediate times in
the GC response. In these GCs, strong FDC-M1 and FDC-M2
staining exhibited a largely punctate pattern until day 16. Thereafter,
FDC staining was similar to that seen in C57BL/6 mice. The primary FDC
reticula in FcRIIB-/- mice, however,
appeared normal at all stages of the GC response as judged by staining
with the anti-CD35 Ab-8C12. Interestingly, even the punctate
pattern of FDC staining observed during the early GC response of
FcRIIB-/- mice was largely restricted to one
side of the GC. To investigate the cellular origin of this punctate
staining, parallel sections from the experiment illustrated in Fig. 7
were stained with GL7, FDC-M1, and the MOMA-2 mAb. MOMA-2 has been
previously reported to detect monocytes and tingible body macrophages
(33). Representative images from this analysis are shown
in Fig. 8. At early time points in
FcRIIB-/- mice (day 8 is illustrated in Fig. 8), MOMA-2 and FDC-M1 staining were largely colocalized to the punctate
bodies, and these were rare in C57BL/6 GCs. At later times (day 20 is
illustrated), whereas the reticular pattern of FDC-M1 staining in
FcRIIB-/- GCs appeared similar to C57BL/6
GCs, Moma-2/FDC-M1 double-positive cells were still clearly visible but
remained rare in C57BL/6 GCs. Interestingly, we also observed MOMA-2
positive, FDC-M1 negative cells in the GCs of
FcRIIB-/- mice at these later time
points.
|
|
RIIB-/- mice correlated with a difference
in the time course of the GC response, the kinetics of this
response were examined in FcRIIB-/- mice
from day 4 to day 22 after SRBC immunization. As shown in Fig. 9, there was no apparent difference in
the kinetics of appearance or disappearance of total or different sized
GL7+ GCs in FcRIIB-/-
mice as compared with C57BL/6 mice (Fig. 1). These results suggest that
while the development of the "mature" FDC reticulum is delayed, the
general kinetics and magnitude of the GC B cell response remain
unchanged due to absence of FcRIIB.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RIIB is also
regulated in GC B cells. The levels of this receptor on GC B cells
remain low at all stages of the response. Further, our data reveal that
FDC "maturation" is perturbed in the absence of FcRIIB,
suggesting that this receptor may play an important role in activation
or secondary development of FDCs. These observations have important
implications for models proposed to account for the selection of
high-affinity B cells in the GC.
Early models (13, 14, 35) explaining the mechanisms
involved in memory B cell development proposed that B cells with
somatically mutated Ag receptors are selected based on their capacity
to be activated by Ag held as ICs on FDC. These models, however, did
not consider the role of FcRIIB. Later models that implicated a role
for FcRIIB in the selection of GC B cells were primarily based on
data obtained from in vitro studies of primary B cells. Ravetch and
Lanier (36) proposed that negative or positive selection
of GC B cells depends on whether FcRIIB alone is engaged or
coengaged with the BCR, respectively. Interaction of ICs on FDC with B
cells through FcRIIB alone results in apoptosis, whereas
coengagement with the BCR results in survival of B cells. Because our
results demonstrate that FcRIIB is substantially down-regulated on
GC B cells, negative selection due to an apoptotic signal generated
through FcRIIB seems unlikely.
A model proposed by Tarlinton and Smith (37) suggested
that early during the GC response, centrocytes usually initiate
differentiation to Ab-forming cells (AFC). Ab produced by these AFC
competes with centrocytes for the Ag present in the form of ICs on FDC
in GC. Late in the primary response, when the Ab titer or affinity
reaches a maximum, AFC differentiation diminishes, and a switch to
memory B cell production takes place. This switch requires coengagement
of the BCR and FcRIIB. This model implies that FcRIIB is not
involved in the differentiation to AFC early in the GC response but is
a requirement for memory formation later in the GC response. However,
our data on the time course expression of FcRIIB on GC B cells (Fig. 4) do not support this model. However, we did not examine whether
FcRIIB expression is reduced on GC B cells present in FDC-rich
regions at all times during the GC response. Further studies will be
required to determine whether a small subpopulation of GC B cells
expresses normal levels of FcRIIB at a specific time in the GC
response.
In contrast to those described above, Tew et al. (38)
proposed a model that takes into account the significance of FcRIIB
on FDC in selection of high-affinity GC B cells. According to this
model, the high density of FcRIIB on FDC allows them to bind most of
the Fc regions in the ICs on their surface. This minimizes the
interaction of these Fc regions with FcRIIB on GC B cells, thus
preventing the inhibition of B cell activity. The studies that led to
the proposal of this model, however, did not examine the expression of
FcRIIB on GC B cells. Based on our findings, it appears that the
primary mechanism by which GC B cells escape the inhibitory FcRIIB
signal is by down-regulating the expression of this receptor. Our flow
cytometric and RT-PCR analyses (Figs. 5 and 6), however, did reveal low
level expression of FcRIIB on GC B cells. Flow cytometric analysis
performed on day 4 and later time points (day 16 and day 22) showed
similar results (S. P. Rao and T. Manser, unpublished
observations). Therefore, it is possible that both Fc blocking by
FcRIIB on FDC and the down-regulation of expression of FcRIIB on
B cells synergize to allow B cell stimulation by ICs. In any case, the
functional relevance of such low FcRIIB levels is not clear and will
be the subject of future investigations.
Recent studies have suggested a role for FcRIIB down-regulation in
the development of autoimmunity. One study reported a 10-fold
down-regulation on GC B cells in autoimmune-prone New Zealand Black
mice and a 4-fold reduction on GC B cells in New Zealand White mice
(17). In another study by the same group
(18), FcRIIB expression on the GC B cells from
nonautoimmune mice, including C57BL/6, was up-regulated. The reasons
for the discrepancy of our results and the results of this previous
study are presently unclear. Factors that may account for this
discrepancy are use of keyhole limpet hemocyanin as an immunogen and
examining the expression of FcRIIB on GC B cells after secondary
challenge in their study. More likely, the differences are due to the
criteria used for defining GC B cells. Although in our study FcRIIB
expression was analyzed on GC B cells that were defined as
GL7+IgD-, in the previous
study all PNAhigh cells were considered GC B
cells. It was recently demonstrated that no single marker can
unambiguously distinguish GC B cells from other subsets, and flow
cytometric analyses showed that PNAhigh cells
include IgD+ cells (27). Therefore,
analyses of surface expression of FcRIIB on GC B cells defined on
the basis of PNA binding alone may be misleading. These data emphasize
the need to more thoroughly investigate the phenotypic criteria used in
defining GC B cells.
It was also previously reported that FcRIIB deficiency on a C57BL/6
background results in the development of severe autoimmune
glomerulonephritis in old mice (39). These findings formed
the basis for the suggestion (36) that absence or abnormal
down-regulation of FcRIIB is a mechanism that allows autoreactive
IgG autoantibodies to be produced, predisposing to autoimmune disease.
However, because our data show that down-regulation of FcRIIB on GC
B cells is characteristic of mice that are not autoimmune prone, the
mechanistic role for FcRIIB in the development of autoimmune
conditions needs further evaluation. For example, there is no formal
evidence demonstrating that autoreactive B cells in the GC, which in
normal situations are believed to be negatively selected (13, 14), can be recruited into the memory pool due to
down-regulation or absence of FcRIIB.
Our data also provide new insights into the role of FcRIIB in the
"maturation" of FDC. We have reported that FDC-specific Abs,
especially FDC-M1, recognize Ags that are induced on FDC in secondary
but not primary follicles (16), suggesting that the
expression of these Ags is associated with the maturation of FDC. Our
data indicate that the gross GC response in
FcRIIB-/- mice is comparable in kinetics and
magnitude with that of normal mice. However, the time course
histological analysis with FDC-specific Abs revealed a significant
delay in the normally strong reticular expression pattern of the FDC-M1
and FDC-M2 Ags, indicating a defect in the timing of maturation of FDC
in the GC. Moreover, during the early phases of the GC response in
FcRIIB-/- mice, strong FDC-M1 staining
colocalizes in a punctate pattern with MOMA-2 staining. Further studies
are required to determine the nature of the cells that give rise to
this staining pattern. They may be a subset of tingible body
macrophages that express the FDC-M1 Ag, but it is also interesting to
speculate that they are FDC precursors (40) that have yet
to develop dendritic morphology.
B cell recall responses are thought to develop in GCs in response to
ICs trapped on FDC. In this context, a recent study (16)
investigated the significance of FcRIIB on the accessory activity of
FDC. In contrast to normal FDCs, those from FcRIIB-deficient mice
were incapable of augmenting IC-mediated B cell recall responses in
vitro. Given our findings, it is possible that this lack of accessory
activity of FDCs may be due not only to absence of FcRIIB but to a
defect in FDC maturation as well. More generally, our data suggest that
the function of FcRIIB on FDC is not merely to trap ICs but also to
promote FDC maturation and expression of hitherto unknown molecules
that may be involved in the regulation of GC B cell selection, memory
development, and recall responses. Based on the correlation between the
appearance of the FDC-M1 and FDC-M2 Ags and FcRIIB up-regulation on
FDC during immune responses, it is tempting to speculate that the
pathways involved in the regulation of expression of these molecules
are one and the same. Clearly, more studies are required to elucidate
the role of FcRIIB and mechanisms regulating the expression of this
receptor on GC B cells and FDCs.
Another observation made in our study was the sectoring of GCs into two
clearly discernible compartments when stained with Abs specific for FDC
or FcRIIB. This sectoring was observed at all times during the GC
response and raises the possibility that these areas in murine GC may
correspond to the light zone composed of centrocytes (area in which the
FDC network is organized) and the dark zone composed of centroblasts
(area where the FDC network is absent) reported in earlier studies on
human tonsillar GCs (41, 42). If so, the delay in FDC
maturation in FcRIIB deficient mice may perturb the selection,
affinity maturation and memory B cell development of centrocytes in the
light zone. However, in the mouse, formal evidence that GC B cells in
these areas actually represent centroblasts and centrocytes is
currently lacking.
|
Footnotes |
|---|
2 Current address: Department of Immunology and Inflammation, Biogen, Cambridge, MA 02142.
3 Current address: Merck & Co., 126 East Lincoln Avenue, Rahway, NJ 07065.
4 Address correspondence and reprint requests to Dr. Tim Manser, Kimmel Cancer Institute, Bluemle Life Sciences Building, 708, 233 South 10th Street, Philadelphia, PA 19107. E-mail address: manser{at}lac.jci.tju.edu
5 Abbreviations used in this paper: BCR, B cell Ag receptor; GC, germinal center; FDC, follicular dendritic cell; IC, immune complex; NP, 4-hydroxy-3-nitrophenyl acetyl; CGG, chicken -globulin; PNA, peanut agglutinin; AFC, Ab-forming cell; SA, streptavidin.
Received for publication February 22, 2002. Accepted for publication June 5, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
receptors and membrane immunoglobulin inhibits anti-immunoglobulin-induced blastogenesis. J. Immunol. 132:627.[Abstract]
receptor-mediated inhibition of mouse B cell activation. J. Immunol. 134:2835.[Abstract]
RIIb. Curr. Opin. Immunol. 10:306.[Medline]
RIIB-induced B cell apoptosis. Immunity 10:753.[Medline]
receptor IIB on follicular dendritic cells regulates the B cell recall response. J. Immunol. 164:6268.RIIB1 in germinal center B cells associated with hyper-IgG and IgG autoantibodies in murine systemic lupus erythematosus. Int. Immunol. 11:1685.RIIB signaling in resting and activated B cells. J. Immunol. 167:204.RII-deficient mice. Nature 379:346.[Medline]
chain deletion results in pleiotrophic effector cell defects. Cell 76:519.[Medline]
-chain does not alter maturation of the B cell response. J. Immunol. 159:2116.RII is a negative regulator of FcRIII in IgG immune complex-triggered inflammation but not in autoantibody-induced hemolysis. Eur. J. Immunol. 30:481.[Medline]
)RIIB-deficient mice results from strain-specific epistasis. Immunity 13:277.[Medline]
This article has been cited by other articles:
|
|
 |
|
  F. Coffey, X. Liu, and T. Manser Primary Development and Participation in a Foreign Antigen-Driven Immune Response of a Chromatin-Reactive B Cell Clonotype Are Not Influenced by TLR9 or Other MyD88-Dependent TLRs J. Immunol., November 15, 2007; 179(10): 6663 - 6672. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Al-Alwan, Q. Du, S. Hou, B. Nashed, Y. Fan, X. Yang, and A. J. Marshall Follicular Dendritic Cell Secreted Protein (FDC-SP) Regulates Germinal Center and Antibody Responses J. Immunol., June 15, 2007; 178(12): 7859 - 7867. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. S. M. Rahman and T. Manser Failed Up-Regulation of the Inhibitory IgG Fc Receptor Fc{gamma}RIIB on Germinal Center B Cells in Autoimmune-Prone Mice Is Not Associated with Deletion Polymorphisms in the Promoter Region of the Fc{gamma}RIIB Gene J. Immunol., August 1, 2005; 175(3): 1440 - 1449. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Aydar, S. Sukumar, A. K. Szakal, and J. G. Tew The Influence of Immune Complex-Bearing Follicular Dendritic Cells on the IgM Response, Ig Class Switching, and Production of High Affinity IgG J. Immunol., May 1, 2005; 174(9): 5358 - 5366. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Su, J. Wu, J. C. Edberg, X. Li, P. Ferguson, G. S. Cooper, C. D. Langefeld, and R. P. Kimberly A Promoter Haplotype of the Immunoreceptor Tyrosine-Based Inhibitory Motif-Bearing Fc{gamma}RIIb Alters Receptor Expression and Associates with Autoimmunity. I. Regulatory FCGR2B Polymorphisms and Their Association with Systemic Lupus Erythematosus J. Immunol., June 1, 2004; 172(11): 7186 - 7191. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Su, X. Li, J. C. Edberg, J. Wu, P. Ferguson, and R. P. Kimberly A Promoter Haplotype of the Immunoreceptor Tyrosine-Based Inhibitory Motif-Bearing Fc{gamma}RIIb Alters Receptor Expression and Associates with Autoimmunity. II. Differential Binding of GATA4 and Yin-Yang1 Transcription Factors and Correlated Receptor Expression and Function J. Immunol., June 1, 2004; 172(11): 7192 - 7199. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Manser Textbook Germinal Centers? J. Immunol., March 15, 2004; 172(6): 3369 - 3375. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Aydar, P. Balogh, J. G. Tew, and A. K. Szakal Altered Regulation of Fc{gamma}RII on Aged Follicular Dendritic Cells Correlates with Immunoreceptor Tyrosine-Based Inhibition Motif Signaling in B Cells and Reduced Germinal Center Formation J. Immunol., December 1, 2003; 171(11): 5975 - 5987. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. SM. Rahman, S. P. Rao, S. L. Kalled, and T. Manser Normal Induction but Attenuated Progression of Germinal Center Responses in BAFF and BAFF-R Signaling-Deficient Mice J. Exp. Med., October 20, 2003; 198(8): 1157 - 1169. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
6-fold lower in GC B cells compared with
non-GC B cells. C57BL/6 mice were immunized with SRBC. On day 8
postimmunization, mRNA was extracted from sorted GC and non-GC B cells
followed by cDNA synthesis and Fc
, GC B cells; 
, non-GC B
cells.
