| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,,,
*
Laboratory of Immunology and Vascular Biology, Department of Pathology, Stanford University, and
Departments of Medicine, Microbiology, and Immunology, Stanford University School of Medicine, Stanford, CA 94305; and
Veterans Affairs Palo Alto Health Care System, Palo Alto, CA 94304
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
4
7, helps target lymphocytes to Peyer’s
patches (PP) and intestinal lamina propria (ILP). We have previously
shown that protective immunity to rotavirus (RV), an intestinal
pathogen, resides in memory B cells expressing
47. In this study, using a novel FACS
assay, we have directly studied the phenotype of B cells that express
surface RV-specific Ig during the in vivo RV immune response. During
primary infection, RV-specific B cells first appear as large
IgD-B220low47-
and 47+ cells (presumptive
extrafollicular, Ab-secreting B cells), and then as large and small
IgD-B220high47-
cells (presumptive germinal center B cells). The appearance of B cells
with the phenotype of large
IgD-B220low47+
cells in PP and most notably in mesenteric lymph nodes coincides with
the emergence of RV-specific Ab-secreting cells (ASC) in the ILP. Thus,
these B lymphocytes are good candidates for the migratory population
giving rise to the RV-specific ASC in the ILP. RV-specific long-term
memory B cells preferentially accumulate in PP and express
47. Nine months after infection most
RV-specific IgA ASC are found in PP and ILP and at lower frequency in
bone marrow and spleen. This study is the first to follow changes in
tissue-specific homing receptor expression during Ag-specific B cell
development in response to a natural host, tissue-specific pathogen.
These results show that 47 is tightly
regulated during the Ag-specific B cell response to RV and is expressed
concurrently with the specific migration of memory and effector B cells
to intestinal tissues. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
47 on the lymphocytes
and the mucosal addressin cell adhesion molecule-1 has been established
as key in directing the migration of lymphocytes to Peyer’s patches
(PP)4 (3, 4) and intestinal lamina propria (ILP) (2, 4, 5, 6).
The integrin 47 is
expressed by B and T cells responsible for immunity to intestinal Ags
(7, 8, 9, 10). Our group has evaluated the importance of
47+
lymphocytes in defense against rotavirus (RV), an infectious model of
intestinal immunity. Adoptive transfer experiments using FACS-sorted
memory B cells from RV-immune animals revealed that cells capable of
mediating RV clearance of chronically infected T and B
cell-deficient mice (RAG-2-/-) preferentially
express
47+
(11). Mice receiving
47+
memory B cells
(B220highIgD-), but not
those receiving
47-
memory or naive
(B220highIgD+) B cells,
produced RV-specific IgA in the stool, cleared the virus, and were
protected against reinfection (11). In addition, recent
passive transfer experiments using lymphocytes from
7-/- mice as a source of
donors cells showed that expression of
47 is essential for
B, but not T, cell migration to the intestine and is required for the
transfer of RV immunity (12, 13). These studies
demonstrated the importance of
47+
B cells in the gut immune response but did not investigate the
developmental stage, frequency of RV-specific cells expressing
47, or kinetics of
appearance and tissue localization of the emergent of
47+
cells during the primary immune response.
Therefore, the present study was undertaken to evaluate the time course
of the appearance, B cell subset compartmentalization, and
47 expression level
of Ag-specific cells during the primary humoral response to RV
infection, and to relate these observations to the physical
distribution and dissemination of RV-specific Ab-secreting cells (ASC).
To accomplish this we used a FACS assay capable of identifying B cells
that express RV-specific surface Ig in combination with an RV-specific
ELISPOT assay. Our findings define critical developmental transitions
of B cells during the primary anti-RV response and demonstrate that
47 expression in
Ag-specific B cells is up-regulated during the intestinal response to
an authentic intestinal pathogen at a time correlating with seeding of
the ILP with ASC. Moreover, we show that ASC persist in intestinal
sites up 9 mo postinfection, and that long-term RV-specific memory B
cells are concentrated in PP and express
47, a characteristic
that may facilitate their recirculation through the PP and intestines,
thereby enhancing the efficiency of the anamnestic response.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
C57BL/6 mice were obtained from Charles River Breeding Laboratories (Hollister, CA) and were bred in the Palo Alto Veterans Administration breeding facility. Six- to 8-wk-old mice were inoculated by gastric gavage with 105 shedding dose 50 (SD50)/mouse of EC RV after receiving 100 µl of 1.33% sodium bicarbonate to neutralize stomach acid. The EC RV was prepared as an intestinal homogenate as previously described, and its SD50 titer was determined by oral inoculation of mice by serial 10-fold dilution (14). Mice were sacrificed at different time points after the primary inoculation of EC for analysis of RV-specific B cells by ELISPOT and FACS. Viral Ag in fecal samples was measured by ELISA (14). Mice analyzed 9 mo after primary infection were challenged orally 2 mo after primary infection with 105 SD50/mouse of murine EC RV. These mice were thus the same age and infected in the same manner as mice used in our previous study (11). As previously shown, we could not detect a boosting effect in RV-specific Ab titers in these mice after rechallenge (data not shown) (15). For this reason the mice that received RV twice are considered to have only been infected with RV once and are shown simultaneously in the graphs with the mice that only received RV once.
Isolation of lymphocytes
At the desired time points after infection mice were sacrificed by cervical dislocation. Mesenteric lymph nodes (MLN), peripheral lymph nodes (PLN), PP, and splenic lymphocytes were harvested as previously described (16). ILP lymphocytes were isolated from collagenase- or dispase-digested intestine from which epithelium and PP had been removed (by EDTA treatment and dissection, respectively). ILP lymphocytes were then purified by density gradient centrifugation (16). Bone marrow (BM) cells were obtained from excised femurs by trimming the ends of the bone with a razor blade and flushing the interior with DMEM with 10% FCS and 2.5% 1 M HEPES buffer using a 22-gauge needle and syringe (11, 16).
B cell ELISPOT for ASC
ELISPOT assays for detection of ASC were performed as previously
described (11). Briefly, 96-well Millipore filtration
plates with Immobilon-P membranes (Millipore, Bedford, MA) were coated
overnight at 4°C with VP2 and VP6 virus-like particles (VLP) made
from baculovirus recombinants expressing heterologous bovine RF RV
(17). Plates were blocked for 30 min at 37°C with RPMI
1640, 10% FCS, 2 mM L-glutamine, 1 mM sodium pyruvate, 20
mM HEPES, 100 U/ml penicillin, and 100 µg/ml streptomycin. Cells were
added to plates at 10-fold dilutions in RPMI 1640 and cultured
overnight at 37°C in 5% CO2. Plates were
washed and incubated with HRP-conjugated goat anti-mouse IgG,
anti-IgA, or anti-IgM (Kirkegaard & Perry, Gaithersburg, MD).
The spots formed by individual ASC were visualized by adding the
precipitating peroxidase substrate, 3-amino-9-ethylcarbazole. Cells
were also added to wells coated overnight at 4°C with 1 µg/ml goat
anti-mouse IgA, IgG, or IgM (Kirkegaard & Perry) to detect total
ASC and to wells without Ag to determine background ASC. Background ASC
were detected only with anti-IgM. In 
90% of the samples tested,
the background was 0. When background IgM-ASC exceeded 50% of
RV-specific IgM-ASC (<5% of samples), data were discarded. In the
remaining RV-specific IgM ASC samples, background was 
30% and was
subtracted from RV-specific ASC.
FACS assay
FACS analysis of Ag-specific B cells was based on the techniques
described by McHeyzer-Williams et al. (18). Small
lymphocytes were discriminated from large immunoblasts by light scatter
profile. B cells that express RV-specific surface Ig were identified
based on the binding of biotinylated VLP. VLP were cesium-purified
twice; purity was checked by PAGE with Coomassie staining and
commercially biotinylated (Chromoprobe, Mountain View, CA). Binding of
biotinylated VLP to B cells that express surface Ig was detected with
the use of streptavidin-PerCP (BD Biosciences, San Jose, CA). Abs were
purchased from BD PharMingen (La Jolla, CA) unless otherwise noted.
Cells were stained with a mixture of FITC-conjugated Abs (anti-IgD,
clone 11-26; anti-CD3, clone 145-2C11; anti-GR1, clone RB6-8C5;
anti-MAC-1, clone M1/70) to identify activated (IgD-negative) B
cells. B cell populations were further discriminated by staining with
anti-B220-allophycocyanin (clone RA3-6B2),
anti-CD138/Syndecan-1-PE (clone 281-2), and
47-PE or
allophycocyanin (clone DATK-32, prepared in our laboratories and
allophycocyanin-conjugated by Chromoprobe). Two million cells were
stained, permitting acquisition of 1 x
106 cells for FACS analysis.
Statistics
Statistical analysis was performed using StatView software (SAS Institute, Cary, NC). ELISPOT data were analyzed using a Kruskal-Wallis test. If p < 0.05, pairwise comparisons between the values from uninfected mice and values at each day after infection were performed using a Mann-Whitney test. To compensate for the performance of multiple comparisons, p < 0.01 was considered significant when performing the Mann-Whitney tests. For FACS data, pairwise comparisons between the values from uninfected mice and values on each of the days after infection were performed with Fisher’s least significance test (p < 0.05 was considered significant).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To determine the distribution of RV-specific ASC (RV-ASC), adult
C57BL/6 mice were orally infected with EC RV and sacrificed at
different time points after infection, and RV-ASC in PP, MLN, spleen,
ILP, and BM were identified by ELISPOT assay. At 4 days after infection
IgM-RV-ASC were detected in PP and MLN (Fig. 1
, upper panels). By day 7,
IgM-RV-ASC were also detected in spleen and ILP (5 x
105 total cells; Figs. 1 and 2, upper panels), and by day
28 IgM-RV-ASC were not detected in these four organs. In BM, IgM-RV-ASC
were only detected 28 days after infection (Fig. 2, upper
panels). At 7 days, IgG and IgA-RV-ASC appeared in all organs
studied, but with a clear predominance of IgA-ASC in the ILP (Figs. 1 and 2, upper panels). In PP, IgA-RV-ASC were predominant
over IgG and continued to rise in number, peaking at 9 mo, while the
highest number of IgG-ASC was detected on day 28. In MLN, similar
numbers of IgG and IgA-RV-ASC were detected, peaking on days 14 and 7,
respectively, and were almost completely absent at 9 mo. Thus, PP, and
not SP or MLN, are the secondary lymphoid organs where the vast
majority of long-term (9 mo) RV-specific IgA-ASC reside. In ILP (as in
PP) IgA-RV-ASC were clearly predominant over IgG-RV-ASC, both peaking
on day 28 and declining almost 10-fold by 9 mo after infection.
However, despite this decline from the peak, the frequency of
IgA-RV-ASC in the ILP was higher than in all other tissues tested,
including PP, and represents 6.2% of total IgA ASC present in the ILP.
In spleen and BM similar numbers of IgG and IgA-RV-ASC were
found on day 28, increasing slightly or remaining at stable levels at 9
mo, respectively. Thus, most long-term (9 mo) RV-specific
IgA-ASC were localized to PP and ILP, while most IgG ASC were found in
spleen and BM. These ELISPOT data correspond with measurements of
RV-specific Abs in serum and stool IgA measured by ELISA (data not
shown) (14).
|
|
and 2, lower panels). In these organs,
IgG-RV-ASC represented a maximum of 10% of total IgG-ASC by day 28
after infection and <2% by 9 mo. IgA-RV-ASC gradually rose to
represent 25% of the total IgA-ASC in PP by 9 mo. In contrast, in MLN
they represented a maximum of 20% of the total IgA-ASC on day 28, but
<0.3% by 9 mo. The extraordinarily high frequency of RV-specific ASC
so long after infection may result from retention of Ag in the PP
(19). In the ILP, IgA-RV-ASC represented a maximum of 12%
of total IgA-ASC on day 28 and were reduced to 6% by 9 mo. In the
spleen, IgA-RV-ASC represented a maximum of 26% of total IgA-ASC on
day 14, and this percentage diminished gradually to 11% 9 mo after
infection. In contrast, IgG-RV-ASC never represented >3% of the total
splenic IgG-ASC 28 days after infection. Similarly, IgA-RV-ASC
gradually increased to 20% of the total IgA-ASC in BM 9 mo after
infection, while IgG ASC were never >4%. Thus, in spleen and BM the
ratio of IgA-RV-ASC/total IgA-ASC was higher than the ratio of
IgG-RV-ASC/total IgG ASC. Phenotype of B cells from the primary anti-RV humoral immune response
To study the phenotype of B cells that comprise the primary immune
response against RV we used a novel flow cytometric assay based on the
binding of labeled RV-Ag (in the form of VLP) to B cells that express
RV-specific surface Ig. Our method is illustrated in Fig. 3, which shows examples of the three
types of RV-specific B cells identified in this study. Using a similar
FACS strategy, the first hapten-specific B cells detected during a
primary immune response are extrafollicular cells and were found to be
large IgD- cells with low levels of B220
expression, secreting class-switched, nonsomatically hypermutated Ab,
and to arise before the formation of germinal centers (GC) (18, 20).
|
, A and B, upper panels, and Fig. 3C, middle panels). Subsequently, RV-specific GC
B cells were detected (Fig. 3, A–C, upper
panels); these cells were shown to be IgD-
and express high levels of B220 (18, 20). ASC derived from
GC are IgD- and down-regulate B220 expression to
levels similar to those produced by extrafollicular ASC, which arise
first (18); therefore IgA ASC observed in MLN, spleen, and
LP later than day 7 may include an increasing number of these
GC-derived ASC. These data are presented in detail in Fig. 4, A and B, and
below. RV-specific memory B cells appear last and are characterized as
small, IgD-B220high cells
(Fig. 3, A–C, lower panels). These cells are
present up to 9 mo after infection, and the majority of the
cells are
47+.
These data are presented in detail in Fig. 4C and are
described in the last section of Results.
|
47+
vs
47-
staining were set using an isotype-matched, irrelevant,
allophycocyanin-conjugated Ab (rat IgG2a-allophycocyanin; BD
PharMingen), RV-VLP binding vs nonbinding quadrants were determined
using second-stage control staining (cells stained with
streptavidin-PerCP, omitting biotinylated RV-VLP) and by comparison
with identical B cell subsets obtained from uninfected littermate
mice.
RV-specific extrafollicular cells are first detected in PP by day 4
after infection and predominate in MLN from days 7–28 as large
47+ cells
RV-specific (VLP binding) B cells were first detected in PP 4 days
after infection and had the phenotype of
47-
extrafollicular cells
(IgD-B220low; see Fig. 4A, upper panel). As mentioned above, this
phenotype has been used to identify the first Ag-specific ASC
(class-switched, non-GC B cells) appearing following immunization
(18, 20). The frequency of these cells peaked in both PP
and MLN on day 7 (see Fig. 3, gate R4, for representative FACS data).
Although a small increase in virus-specific cells with this
phenotype was also observed on day 7 in the spleen, these
differences were not statistically significantly from background levels
observed in uninfected mice. In PP, on days 7 and 14 after infection,
20 and 60% of these cells expressed
47, respectively
(Fig. 4A). Cells of the extrafollicular phenotype were
detected in MLN by day 7, and by days 14 and 28 >50% of them
expressed 47 (Fig. 4A, middle panel). Cells with this phenotype were
no longer detectable in PP 28 days after infection or in MLN at 4 mo.
The appearance of extrafollicular
47+
B cells, especially in MLN, is coincident with the appearance of large
numbers of ILP IgA-RV-ASC (Figs. 2 and 4A).
RV-specific GC cells appear in PP, MLN, and spleen as both
large and small
IgD-B200high47-
B cells
On day 7 after infection both large and small RV-specific B cells
with the phenotype of GC cells
(IgD-B220high47-)
were first detected in MLN (Fig. 4, B and C).
Large cells with this phenotype peaked in PP, MLN, and spleen by day 14
and returned to background levels in these organs by 4 mo (Fig. 4B). At 14 days, total small RV-specific
IgD-B220high cells peaked
in MLN on day 14 (90%
47-),
returning to uninfected levels by day 28 (Fig. 4C). This
contrasted sharply with the kinetics of the appearance these cells in
PP, where they were present at high frequency on day 14 after infection
and remained at high levels up to 9 mo after infection (see the
description of the kinetics of memory B cell frequencies below).
Syndecan-1+ cells are found in extrafollicular B cell compartment
The overwhelming majority (>95%) of Ag-specific
Syndecan-1+ B cells have been shown to be ASC
(defined by Ag-specific ELISPOT assays), and Syndecan-1 is thus a
reliable and selective marker for ASC (18, 20). To
determine whether RV-specific B cells express Syndecan-1, we
substituted an anti-Syndecan-1 Ab for the
anti-47 Ab used
in the experiments illustrated in Fig. 3. As shown in Fig. 5, RV-specific
Syndecan-1+ cells were detected in the large
IgD-B220low subset in PP,
MLN, and spleen on days 7 and 14 after infection. RV-specific
Syndecan-1+ cells were not detected in the large
or small IgD-B220high (GC
or memory phenotype) subsets (data not shown). Unlike the total and
47+B220low
subsets (Fig. 4A), the
RV+Syndecan-1+ cells were
no longer detectable 28 days after infection in any of the lymphoid
organs tested (Fig. 5). In addition, the percentage of the total
RV-specific B220low population expressing
Syndecan-1 was lower than the percentage expressing
47 (compare Figs. 4A and 5). These results suggested that the
Syndecan-1+ and
47+
subpopulations of
IgD-B220low cells could be
independent. To determine whether the Syndecan-1+
and
47+
subpopulations overlap, we stained MLN cells from mice 7 days after
infection using our standard FACS protocol using both
anti-Syndecan-1 and
anti-47 mAbs
(omitting B220 staining). As shown in Fig. 6, 50% of RV-specific
Syndecan-1+ cells coexpress
47. Thus, many MLN
RV-specific ASC, whether identified as B220low or
Syndecan-1+, express
47 and therefore have
the potential to home to the ILP.
|
|
Syndecan-1+, RV-specific B cells were only
found in the extrafollicular subset
(IgD-B220low phenotype),
suggesting that this subset includes the RV-specific ASC detected in
our ELISPOT experiments. To directly test this hypothesis, we sorted
selected subpopulations of B cells from the MLN of mice 7 days after RV
infection and determined by ELISPOT their capacity to secrete
RV-specific Ig (Table I). As expected,
RV-specific ASC were found predominantly in the large
IgD-B220low subpopulation,
and within this B cell subset the percentages of RV-specific
cells detected by ELISPOT and FACS were comparable (1.97 and 1.80%,
respectively). However, in the large
IgD-B220high (GC
phenotype) B cell subset few RV-specific ASC were detected by ELISPOT
(0.03%), while large numbers of RV-specific B cells were detected by
FACS (2.83%). Taken together, these data show that the ELISPOT and
FACS techniques detect similar numbers of RV-specific cells in the B
cell subset likely to contain ASC (large,
IgD-B220low
extrafollicular B cells) and provide overlapping information on the
RV-specific cells detected (i.e., Ab isotype by ELSPOT and
47 expression and
developmental phenotype by FACS). Additionally, the FACS technique is
capable of detecting RV-specific cells that do not secrete Ab, such as
GC and memory B cells.
|
47+
cells
We have previously shown that passively transferred purified
47+
splenic memory B cells (IgD-), but not
47-
cells taken from donor mice 9 mo after infection with murine RV,
mediated viral clearance in chronically infected
RAG-2-/- mice (11). To directly
demonstrate the presence of
47+
memory B cells we stained splenic, PP, and MLN cells from mice that had
been infected orally with RV as illustrated in Fig. 3, gate R5. As
illustrated in the summary of these results (Fig. 4C,
upper panel), RV-specific memory B cells (small,
IgD-B220high) were
detected in PP and, as expected from our previous findings, >60% of
these cells expressed
47 9 mo after
infection. It should be noted that the frequency of these RV-specific
B220high B cells is approximately twice the
frequency of RV-specific IgA-ASC detected by ELISPOT (Fig. 1). While
our FACS sorting data show that in MLN 7 days post-RV infection,
ELISPOT detected IgA ASC are restricted to large,
B220low/- cells, we cannot exclude the
possibility that the long-term RV-specific memory B cells detected by
FACS and ASC are overlapping populations. At this time in MLN we
observed a statistically significant increase in the frequency of
47+,
RV-specific memory B cells, although the number of total RV-specific
memory B cells was not significantly different from that in uninfected
mice (Fig. 4C, middle panel).
Analysis of mice 4 mo after infection revealed small, but
significant, numbers of
47+
RV-specific memory B cells
(IgD-B220high) in the
spleen (Fig. 4C, lower panel). At this time
significant numbers of these cells were not present in MLN. Four months
after infection, of the highest frequency of RV-specific memory B cells
was found in PP, one-third of them
47+
(Fig. 4C, upper panel).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
47+)
B cells following oral RV infection (Fig. 4A). The
appearance of this cell population in PP and MLN coincides with the
appearance of RV-specific IgA-ASC in the ILP (Fig. 2), suggesting that
these cells are the migratory population that gives rise to RV-specific
ASC in the ILP during the anti-RV immune response. In addition, we
have provided evidence that most long-term RV-specific memory B cells
(small IgD-B220high cells)
express the intestinal homing receptor
47 and that these
memory B cells accumulate preferentially in PP (Fig. 4C).
The results reported in Figs. 1, 2, and 4 are summarized schematically
in Fig. 7.
|
There is less evidence of selective intestinal migration of small,
resting memory B cells than for immunoblast migration. In the murine
model our previous passive transfer experiments demonstrated that
47+B220highIgD-
splenic B cells (memory phenotype) conferred immunity to RV, while
47-
cells did not (11). Our findings that long-term (9 mo
after infection) RV-specific memory B cells are preferentially
localized to PP and that two-thirds of them are
47+
(Fig. 4C) reinforces this hypothesis. This conclusion is
also supported by data included here (Fig. 4C) that 4 mo
after infection RV-specific memory B cells were found in the spleen,
and a significant proportion of them were
47+.
In contrast to the data from 9 mo after infection, at 4 mo after
infection a smaller fraction (approximately one-third) of the small
IgD-B220high cells in PP
were
47+
(Fig. 4C). As evidenced by the peak of these small
B220high cells on day 14 after infection (when
almost all of them were
47-
(Fig. 4C) some of these cells could be GC, noncirculating
cells.
We are unaware of studies analyzing and comparing the distribution of
Ag-specific memory and effector B cells using a physiological natural
pathogen as a model. The model most comparable to ours used vesicular
stomatitis virus (VSV), a virus that does not replicate extraneuronally
in mice, and a functional passive transfer method to evaluate the
presence of memory B cells (30). These authors found that
ASC are induced at the site where Ag is present (spleen and peripheral
LN after i.v. and s.c. inoculation, respectively) and in the BM, while
memory B cells then recirculate throughout the lymphoid system
independently of Ag localization. Twenty-one days after oral submucosal
immunization ASC were found to be restricted to the submandibular
draining lymph nodes (LN), but not nondraining LN. In contrast, memory
B cells were found at similar numbers in both draining and non-draining
LN. These findings suggest that the compartmentalization of the mucosal
immune response does not apply to memory B cells and would be at odds
with our findings that indicate that mucosally primed memory
RV-specific B cells predominantly accumulate in PP. These data would
also be at odds with our findings that after mucosal immunization
RV-specific B cells were not detected in PLN 7 days and 9 mo after
infection (data not shown). The differential localization of mucosally
induced memory B cells that we have observed (Fig. 4C) is
more apparent at time points past the 21 days after infection studied
in the VSV model, suggesting a time-dependent accumulation of the
intestinal memory B cells in PP, but not in spleen or MLN. Another
important parameter that could explain the differences between the
models is RV’s mucosal route of sensitization (Ag presentation in PP)
vs the submucosal route used for VSV (Ag presentation in
submandibular LN).
Detailed FACS surface phenotype analysis and correlation with ASC
function of Ag-specific murine B cells has been conducted with hapten
adjuvant models using a FACS strategy similar to ours (18, 20). In these studies Ag-specific B cells were identified as
being IgM-IgG1+ cells and
not IgD- cells as we found. The phenotype and
correlation of phenotype with Ab secretion of the principal subsets
analyzed seem to be the same: GC cells are
B220highPNA+, while
extrafollicular ASC cells are
B220-/lowSyndecan-1+, and
resting memory small lymphocytes are
B220highIgD-. An important
difference between the results obtained using the hapten model and our
results is the kinetics of the appearance of extrafollicular
Syndecan-1+ B cells with Ag-specific surface Ig.
After hapten immunization the frequency of these cells peaks on day 7
and almost completely disappears by day 10 (18); however,
we found that at 14 days the frequency of RV-specific
Syndecan-1+ is still 50% the peak levels on day
7 (Fig. 5). Similarly, in the VSV model high numbers of VSV-specific
GCs associated with persisting Ag were present for >1 mo after
immunization, considerably longer than has been observed for
anti-hapten responses (19). Viral replication in the
case of RV and viral Ag organization of both RV and VSV may
account for the differences between the hapten and viral models, and
these results highlight the utility of viruses as probes in
understanding the physiology of the immune system. Smith et al.
(20) went on to investigate the rapid disappearance of
extrafollicular ASC during a primary immune response and analyzed the
kinetics of extrafollicular Ag specific B cell induction after i.p.
immunization in relation to the appearance of ASC in BM. They concluded
that the rapid decline of extrafollicular ASC was not probably due to
migration to the BM, because none was detected in this organ at time
points when extrafollicular cells were disappearing from the spleen.
They therefore proposed that the decline in extrafollicular splenic ASC
was due to cellular apoptosis in situ, which was demonstrated
immunohistologically by nick end labeling. In our model disappearance
of extrafollicular ASC does correlate with appearance of ASC in ILP
and, to a lesser degree, in spleen and BM (Figs. 1 and 2). In addition,
we have detected the intestinal homing receptor,
47, on at least 50%
of cells with the extrafollicular phenotype, both RV-specific cells and
the total subpopulation (Fig. 4A). Our results suggest that
in a physiological immune response to an intestine-specific pathogen,
the decline in extrafollicular ASC seen during the primary immune
response could be due at least in part to migration to tertiary immune
organs and the ILP in particular. Although Smith et al.
(20) did not study the presence of ASC in ILP it is not
probable that the decline in spleen extrafollicular cells in their
model could be explained by migration to ILP, because i.p. immunization
is not very efficient in inducing an intestinal mucosal immune response
(11). Inherent differences between the secondary organs
where the immune response originates (PP in our study vs spleen) and
probably virus vs a nonreplicating Ag could lead to different fates of
extrafollicular ASC.
The results reported here support our previous findings that homing of
RV-specific B cells to ILP is dependent on
47 (11, 13). In previous studies parenteral immunization with rhesus RV
did not induce intestinal RV-specific ASC (11). In
preliminary experiments we found that 7 days after s.c. inoculation
with murine RV, RV-specific extrafollicular
(IgD-B220low) B cells in
draining PLNs do not express
47. RV-specific cells
of this phenotype in MLN following natural intestinal RV infection are
largely
47+,
suggesting that up-regulation of
47 expression is
a consequence of Ag-specific activation within the intestinal
compartment. Other authors, using murine virus as a parenteral
immunogen, have succeeded in identifying RV-specific mucosal ASC
(31). However, in subsequent passive transfer experiments
these authors found that B cells from PLN of parenterally
immunized animals migrated to PP where they presented viral Ag,
initiating an intestinal response, but did not produce virus-specific
Abs (32). Collectively, our data suggest that contrary to
what we have recently shown for RV-specific CD8+
lymphocytes induced after parenteral immunization (12),
Ag-specific B cells seem to be highly dependent on
47 for migration to
the intestine.
In conclusion, our findings define critical developmental transitions
of B cells during the primary anti-RV immune response and
demonstrate that 47
expression in RV-specific B cells in intestinal secondary lymphoid
tissue is up-regulated at a time correlating to appearance of
RV-specific ASC in the ILP ASC. Moreover, we show that long-term
RV-specific memory B cells are concentrated in PP and also express
47, which may serve
to target their recirculation to intestinal tissues.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 K.R.Y. and M.A.F. contributed equally to this work, and E.C.B. and H.B.G. contributed equally as senior authors.
3 Address correspondence and reprint requests to Dr. Kenneth R. Youngman, Veterans Affairs Palo Alto Health Care System, Building 101, Room C4-131, Mail Code 154-B, 3801 Miranda Avenue, Palo Alto, CA 94304. E-mail address: kry{at}stanford.edu or Dr. Manuel A. Franco at the current address: Instituto de Genetica Humana, Pontificia Universidad Javeriana, Bogota, Colombia. E-mail address: mafranco{at}javeriana.edu.co
4 Abbreviations used in this paper: PP, Peyer’s patch; ASC, Ab-secreting cell; BM, bone marrow; GC, germinal center; ILP, intestinal lamina propria; LN, lymph node; MLN, mesenteric LN; PLN, peripheral LN; RV, rotavirus; RV-ASC, RV-specific ASC; SD50, shedding dose 50; VLP, virus-like particle; VSV, vesicular stomatitis virus.
Received for publication June 22, 2001. Accepted for publication December 18, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
47 and LFA-1 in lymphocyte homing to Peyer’s patch-HEV in situ: the multistep model confirmed and refined. Immunity 3:99.[Medline]
4-integrins in lymphocyte homing to mucosal tissues in vivo. J. Immunol. 52:3282.
7 integrins in formation of the gut-associated lymphoid tissue. Nature 382:366.[Medline]
47. J. Immunol. 161:4227.47 independent pathway for CD8+ T cell mediated intestinal immunity to rotavirus. J. Clin. Invest. 106:1541.[Medline]
47 integrin expression but does not require IgA antibody production. J. Immunol. 166:1894.This article has been cited by other articles:
|
|
 |
|
  H. E. Mei, T. Yoshida, W. Sime, F. Hiepe, K. Thiele, R. A. Manz, A. Radbruch, and T. Dorner Blood-borne human plasma cells in steady state are derived from mucosal immune responses Blood, March 12, 2009; 113(11): 2461 - 2469. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Chachu, D. W. Strong, A. D. LoBue, C. E. Wobus, R. S. Baric, and H. W. Virgin IV Antibody Is Critical for the Clearance of Murine Norovirus Infection J. Virol., July 1, 2008; 82(13): 6610 - 6617. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. T. Pals, D. J. J. de Gorter, and M. Spaargaren Lymphoma dissemination: the other face of lymphocyte homing Blood, November 1, 2007; 110(9): 3102 - 3111. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Di Martino, C. Basset, A. Ogier, A. Charpilienne, D. Poncet, and E. Kohli Distribution and phenotype of rotavirus-specific B cells induced during the antigen-driven primary response to 2/6 virus-like particles administered by the intrarectal and the intranasal routes J. Leukoc. Biol., October 1, 2007; 82(4): 821 - 828. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Graham, J. A. O'Donnell, Y. Tan, N. Sanders, E. M. Carrington, J. Allison, and B. S. Coulson Rotavirus Infection of Infant and Young Adult Nonobese Diabetic Mice Involves Extraintestinal Spread and Delays Diabetes Onset J. Virol., June 15, 2007; 81(12): 6446 - 6458. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Feng, M. C. Jaimes, N. H. Lazarus, D. Monak, C. Zhang, E. C. Butcher, and H. B. Greenberg Redundant Role of Chemokines CCL25/TECK and CCL28/MEC in IgA+ Plasmablast Recruitment to the Intestinal Lamina Propria After Rotavirus Infection J. Immunol., May 15, 2006; 176(10): 5749 - 5759. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. L. Graham, F. E. Fleming, P. Halasz, M. J. Hewish, H. S. Nagesha, I. H. Holmes, Y. Takada, and B. S. Coulson Rotaviruses interact with {alpha}4{beta}7 and {alpha}4{beta}1 integrins by binding the same integrin domains as natural ligands J. Gen. Virol., December 1, 2005; 86(12): 3397 - 3408. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Valosky, H. Hishiki, T. E. Zaoutis, and S. E. Coffin Induction of Mucosal B-Cell Memory by Intranasal Immunization of Mice with Respiratory Syncytial Virus Clin. Vaccine Immunol., January 1, 2005; 12(1): 171 - 179. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Parez, A. Garbarg-Chenon, C. Fourgeux, F. Le Deist, A. Servant-Delmas, A. Charpilienne, J. Cohen, and I. Schwartz-Cornil The VP6 Protein of Rotavirus Interacts with a Large Fraction of Human Naive B Cells via Surface Immunoglobulins J. Virol., November 15, 2004; 78(22): 12489 - 12496. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Jaimes, O. L. Rojas, E. J. Kunkel, N. H. Lazarus, D. Soler, E. C. Butcher, D. Bass, J. Angel, M. A. Franco, and H. B. Greenberg Maturation and Trafficking Markers on Rotavirus-Specific B Cells during Acute Infection and Convalescence in Children J. Virol., October 15, 2004; 78(20): 10967 - 10976. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hieshima, Y. Kawasaki, H. Hanamoto, T. Nakayama, D. Nagakubo, A. Kanamaru, and O. Yoshie CC Chemokine Ligands 25 and 28 Play Essential Roles in Intestinal Extravasation of IgA Antibody-Secreting Cells J. Immunol., September 15, 2004; 173(6): 3668 - 3675. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. R. Vuillemenot, J. F. Rodriguez, and G. W. Hoyle Lymphoid Tissue and Emphysema in the Lungs of Transgenic Mice Inducibly Expressing Tumor Necrosis Factor-{alpha} Am. J. Respir. Cell Mol. Biol., April 1, 2004; 30(4): 438 - 448. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. H. Lazarus, E. J. Kunkel, B. Johnston, E. Wilson, K. R. Youngman, and E. C. Butcher A Common Mucosal Chemokine (Mucosae-Associated Epithelial Chemokine/CCL28) Selectively Attracts IgA Plasmablasts J. Immunol., April 1, 2003; 170(7): 3799 - 3805. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
4). *, Data points statistically significant different
from day 0 (p < 0.05).
