| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Centre d’Immunologie et de Biologie Parasitaire, Unité Institut National de la Santé et de la Recherche Medicalé, Unité 547, Institut Pasteur, 59019 Lille Cedex, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI expressed by human eosinophils is involved in
IgE-mediated cytotoxicity reactions toward the parasite
Schistosoma mansoni in vitro. However, because receptor
expression is low on these cells, its functional role is still
controversial. In this study, we have measured surface and
intracellular expression of FcRI by blood eosinophils from
hypereosinophilic patients and normal donors. The number of unoccupied
receptors corresponded to 
4,500 Ab binding sites per cell, whereas
50,000 Ab binding sites per cell were detected intracellularly.
Eosinophils from patients displayed significantly more unoccupied
receptors than cells from normal donors. This number correlated to both
serum IgE concentrations and to membrane-bound IgE. The lack of FcRI
expression by mouse eosinophils has hampered further studies. To
overcome this fact and experimentally confirm our findings on human
eosinophils, we engineered IL-5 x hFcRI
double-transgenic
mice, whose bone marrow, blood, spleen, and peritoneal eosinophils
expressed FcRI levels similar to levels of human eosinophils, after
4 days culture with IgE in the presence of IL-5. Both human and mouse
eosinophils were able to secrete IL-10 upon FcRI engagement. Thus,
comparative analysis of cells from patients and from a relevant animal
model allowed us to clearly demonstrate that FcRI-mediated
eosinophil activation leads to IL-10 secretion. Through FcRI
expression, these cells are able to contribute to both the regulation
of the immune response and to its effector
mechanisms. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI, has been considered for a long
time to be expressed only on mast cells and basophils and to be
responsible for triggering immediate hypersensitivity reactions
(1). More recently, FcRI expression was demonstrated on
other human cell types, including eosinophils (2),
platelets (3), epidermal Langerhans cells (4, 5), dendritic cells (6) and monocytes/macrophages
(7), allowing the receptor to play a role in
anti-parasitic effector function in vitro (2, 3) or in
Ag presentation (6, 8). Despite the low amounts of FcRI
expressed by human eosinophils, this cell population is able to mediate
Ab-driven cellular cytotoxicity (ADCC) reactions toward
Schistosoma mansoni larvae in vitro, and to release
eosinophil peroxydase, a pharmacologically active mediator, upon
cross-linking of FcRI with anti-FcRI-chain mAb
(2).
By contrast, in mouse, FcRI expression is restricted to mast cells
and basophils (9, 10), suggesting a species polymorphism
linked to the cellular distribution of FcRI. The lack of FcRI
(and CD23) expression on wild-type mouse eosinophils (9)
provided the beginning of an explanation to the long lasting debate
about the role of IgE and eosinophils in mouse immunity to
schistosomiasis.
Furthermore, two recent papers (11, 12) reopened the
controversy about the role of FcRI expressed by human eosinophils.
On one hand, Seminario et al. (11) were unable to detect
FcRI at the surface of eosinophils, while demonstrating high amounts
of FcRI inside the cell and released in medium, as a soluble
receptor. On the other hand, Kita et al. (12) detected low
levels of surface expression but failed to measure any biological
effect (degranulation, superoxide anion production, or leukotriene C4
release) upon receptor activation by IgE and anti-IgE. To reassess
the levels of FcRI expression, we took advantage of a recent method
for quantification of surface or intracellular binding sites by flow
cytometry. We determined the number of unoccupied FcRI molecules
expressed by purified eosinophils from a large series of patients with
eosinophilia, as well as from normal donors.
We also intended to confirm our results on an experimental model that
was more relevant to the human situation than WT mice. Therefore,
transgenic
(Tg)4
mice expressing human FcRI (hFcRI) under the control of its
own promoter elements were first produced (13). These
animals expressed a "humanized" receptor with a cellular
distribution similar to humans, including eosinophils (after infection
by S. mansoni). Because naive mice have a very low number of
eosinophils compared with humans and rats, hFcRI Tg mice were
crossed with IL-5 Tg mice exhibiting massive eosinophilia in different
organs (14). Eosinophils from these hFcRI x
IL-5 double-Tg animals expressed a low number of surface FcRI and
contained high amounts of intracellular hFcRI. Nevertheless,
receptor activation was sufficient to trigger IgE-dependent adherence
of eosinophils to S. mansoni larvae and, as for human
eosinophils, a significant IL-10 release.
Taken together, our results demonstrate that low levels of unoccupied
FcRI at the surface of eosinophils endow these cells with both
effector and regulatory function in immune response.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A total of 24 different hypereosinophilic patients and 6 normal
donors were selected for this study, after informed consent.
Hypereosinophilia was associated with skin diseases, hypereosinophilic
syndromes, allergy, and hematological disorders. The characteristics of
eosinophil donors, eosinophil preparations and serum IgE levels are
summarized in Table I
.
|
Animals were bred and housed in a specific pathogen-free
facility. IL-5 x hFcRI Tg mice were obtained by crossing
IL-5 Tg animals (14), in which IL-5 expression in T cells
is driven by the human CD2 promoter, with hFcRI Tg animals
(13), where the transgene is expressed under the control
of its own promoter elements. Six- to 12-wk-old
F2 IL-5 x hFcRI Tg animals and their
IL-5 Tg littermate controls were used for the experiments. Expression
of the IL-5 transgene was assessed by monitoring blood eosinophilia,
whereas presence of the hFcRI transgene was analyzed by Southern
blot as previously described (13).
S. mansoni cycle and infections
A Guadeloupean strain of S. mansoni was maintained using Biomphalaria glabrata snail as the invertebrate intermediate host and, in the mice, as vertebrate definitive host. For in vivo experiments, animals were infected percutaneously with 50 cercariae after shaving the abdominal skin, and were sacrificed 46 days later for cell collection. For in vitro experiments schistosomula were collected in MEM after the application of cercariae to isolated pieces of Swiss mouse abdominal skin for 3 h.
Reagents
Anti-human CD16- and CD3-coated magnetic beads, anti-mouse
CD45R (B220)-, CD8 (Ly-2)-, and CD90 (Thy1.2)-coated magnetic beads
and the MACS system were purchased from Miltenyi Biotec (Bergisch
Gladbach, Germany). Percoll was obtained from Pharmacia (Uppsala,
Sweden). RPMI 1640 and MEM, glutamine, penicillin, streptomycin,
G418, and FCS were obtained from Life Technologies (Paisley, U.K.).
BSA, paraformaldehyde, and saponin were obtained from Sigma (St. Louis,
MO). The blocking anti-FcRI (15.1, a mouse IgG1
(mIgG1)) mAb (5) and rabbit polyclonal
anti-FcRI antiserum (997) (15) were a kind gift
from Dr. J.-P. Kinet (Harvard Medical School, Boston, MA). Both 15-1
and mIgG1 isotype control (Diaclone, Besançon, France) were
FITC-labeled in our laboratory. Recombinant human IL-5 (rhIL-5) was
obtained from Diaclone. Mouse anti-FcRI mAb CRA-1
(IgG2b nonblocking) and CRA-2 (IgG1 blocking) (16) were
kind gift from Dr. C. Ra (Juntendo University, Tokyo, Japan), and 22E7
(IgG1 nonblocking) (17) was provided by Dr. A. Tsicopoulos
(Institut Pasteur de Lille, Lille, France). Mouse anti-rat
FcRI mAb (3A92) was kindly provided by Dr. T. Flemming (Beth
Israel Deaconess Medical Center, Boston, MA). Human myeloma IgE (hIgE)
was purchased from Bernett Laboratories (Laguna Niguel, CA). Hybridoma
supernatant containing chimeric hIgE (cIgE) molecules (composed of the
Fc portion of hIgE and Fab portion of
anti-4-hydroxy-3-nitrophenacetyl mIgE) was prepared in our
laboratory (18). Biotinylated rat anti-mouse 
1 and
2 L chain, rat anti-mouse Fc
RII/RIII (2.4G2)
(19) and mouse anti-human IgE were purchased from
PharMingen (San Diego, CA). PE-conjugated donkey anti-mouse IgG (H
+ L) F(ab')2 and FITC-conjugated goat
anti-mouse IgG (Fc specific) were obtained from Jackson
ImmunoResearch Laboratories (West Grove, PA). PE-conjugated
streptavidin (SA-PE) was obtained from Molecular Probes (Eugene, OR).
PE-conjugated rat anti-mIL-10 was purchased from Caltag
Laboratories (Burlingame, CA). PE-conjugated rat IgG1 was obtained from
Immunotech (Coulter, Miami, FL). The anti-mIgG
F(ab')2 and FITC-labeled anti-hIgE were
obtained from Sigma. Quantum Simply Cellular quantification kit and
Qifikit were purchased from Dako (Glostrup, Denmark).
Cell purification
Human eosinophils were isolated from the venous blood of patients, using immunomagnetic beads and the MACS system, as previously described with minor modifications (20). Diluted whole blood (1:1) was layered onto a Percoll gradient (density = 1.082 g/l) and centrifuged at 1800 rpm for 20 min. The granulocyte pellet, containing mainly neutrophils and eosinophils, was harvested and depleted of erythrocytes by hypotonic saline lysis. Briefly, the granulocyte pellet was incubated for 30 min at 4°C with anti-CD16- and anti-CD3-coated immunomagnetic beads to remove neutrophils and contaminating lymphocytes, respectively. Purified eosinophils were obtained by passage of the cells through the field of a permanent magnet. After isolation, eosinophil preparations were cytocentrifuged and cytospins were stained with May Grünwald Giemsa (RAL 555, Rieux, France). The purity of eosinophil preparations was usually above 97%.
Mouse peripheral blood cells were obtained by retro-orbital puncture.
Peritoneal cells were obtained by flushing the peritoneal cavity with
10 ml of ice-cold PBS. Splenocytes were obtained by gentle dissociation
of the spleen in ice-cold PBS. Bone marrow cells were isolated from
femur and tibia of mice by flushing the bone marrow cavities with
ice-cold PBS. Aggregates were removed from cell suspensions by
filtration on a nylon filter and erythrocytes were lysed using
hypotonic saline. After washing, the cells were resuspended in PBS for
the experiments. For activation experiments, splenic eosinophils were
purified using a MACS (21). Nonfractionated cell
suspension (1 x 108 cells/ml) was incubated
for 15 min with CD90 (Thy1.2), CD45R (B220), and CD8 (Ly-2) magnetic
beads. Purified eosinophils were obtained by passage of the cells
through the field of a permanent magnet. After isolation, eosinophil
preparations were cytocentrifuged and the cytospins were stained with
May Grünwald Giemsa (RAL 555). Purity of splenic eosinophil
preparations was ranging between 90 and 99%.
Cell culture
Culture medium consisted of RPMI 1640 supplemented with 10%
heat-inactivated FCS, 2 mM L-glutamine, 100 IU/ml
penicillin, and 100 µg/ml streptomycin (complete medium). Purified
human eosinophils, purified mouse splenic eosinophils, and
unfractionated mouse cells were cultured for 4 days in complete medium
with the addition of 2.5 ng/ml rhIL-5 with 0–10 µg/ml of cIgE.
Unfractionated mouse cells were used for flow cytometry analyses for
surface and intracellular expression of FcRI. Rat basophilic
leukemia cells were kept in complete medium. Chinese hamster ovary
(CHO) cells stably transfected with the three chains of the
human FcRI (13) were kept cultured in complete medium
containing 1 mg/ml G418.
In vivo receptor up-regulation
IL-5 x hFcRI Tg mice were injected i.v. 4 times at
24 h interval with 100 µg cIgE as previously reported for mouse
IgE (22). Animals were sacrificed 24 h after the last
injection. Peritoneal and splenic cells were obtained as described
above and analyzed for FcRI expression. Blood samples were taken to
determine IgE serum concentration at the time of sacrifice.
Flow cytometric analysis of surface FcRI
Freshly purified human eosinophils were resuspended at 4 x
106/ml in PBS-1% BSA. Aliquots of 50 µl were
incubated with FITC-conjugated anti-FcRI (15.1),
FITC-conjugated isotype-matched Ab at a final concentration of 2.5
µg/ml for 1 h at 4°C in round bottom 96-well plates. Staining
specificity was controlled by preincubating the cells with hIgE for 15
min on ice before the addition of the FITC-15.1 mAb. After two washes
in PBS, cells were resuspended in PBS-0.5% BSA before analysis.
Membrane-bound IgE was detected using FITC-conjugated anti-hIgE
(1:200). For cells cultured for 4 days in the presence of cIgE,
staining was performed using PE-conjugated anti-mouse IgG (H + L)
F(ab')2 (1:200) after additional saturation
with cIgE.
Mouse cells were resuspended at 2 x 106/ml
in PBS containing 0.1% BSA and 0.05% sodium azide. One
hundred-microliter aliquots were used per sample. Unless otherwise
specified, all incubation steps were performed on ice for 30 min.
Surface expression of hFcRI was analyzed after saturation of
FcRII/RIII receptors with 150 µg/ml 2.4G2. Except for the
determination of occupied receptors following injection of cIgE, cells
were first incubated with cIgE, then after washing, with a biotinylated
anti-mouse 1 and 2 L chain (1:100) followed by SA-PE (1:200).
Anti-mouse 1 and 2 L chain was omitted in control samples. For
quantification, cIgE was detected using PE-conjugated anti-mouse
IgG (H + L) F(ab')2 (1:200). For eosinophils from
S. mansoni-infected mice, murine IgE already bound to
hFcRI at cell surface was detected with biotinylated anti-mouse
IgE followed by SA-PE (1:200). Biotin-conjugated anti-mouse IgE was
omitted for control samples. Eosinophils were identified on the basis
of their forward and side scatters. Ten thousand events were usually
acquired per sample. Thresholds were set on control stainings (included
for every sample at every time point).
Samples were analyzed on a FACSCalibur using CellQuest software (Becton Dickinson, Mountain View, CA).
Flow cytometric analysis of intracellular FcRI
Human eosinophils were fixed with 2% paraformaldehyde in PBS for 10 min. After washing in PBS, cells were resuspended at 4 x 106/ml in PBS containing 1% BSA and 0.1% saponin (permeabilization buffer) for 10 min at room temperature. The samples were then incubated for 30 min with 15.1 mAb or isotype-matched Ab at a final concentration of 25 µg/ml (saturating concentration) in permeabilization buffer. After washing with permeabilization buffer, cells were incubated for 10 min with 5 µl normal goat serum to block nonspecific binding, then FITC-conjugated anti-mouse IgG F(ab')2 was added for 20 min. Samples were washed twice in permeabilization buffer, once in PBS, and were resuspended in PBS-0.5% BSA for analysis.
For mouse eosinophils, permeabilization was performed as described for
human eosinophils. Detection of intracellular FcRI was performed,
after saturation of surface FcRI with 150 µg/ml cIgE and 150
µg/ml 2.4G2, using anti-hFcRI (15-1) mAb (10 µg/ml) and
FITC-conjugated anti-mouse IgG (Fc specific) (1:100).
Anti-hFcRI was replaced by an isotype-matched Ab in control
samples.
Measurement of membrane and intracellular unoccupied FcRI
For human eosinophils, the number of unoccupied receptors at the
cell surface and intracellularly were determined with the Quantum
Simply Cellular quantification kit (direct staining) and the
Quifikit (indirect staining), respectively, according to the
manufacturer’s instructions. For murine eosinophils, both surface and
intracellular unoccupied receptors were determined using the Quifikit.
Quantum kit is based on goat anti-mouse IgG coated-microbeads with
different Ab binding capacities (ABC). FITC-conjugated 15-1 and
FITC-conjugated isotype-matched Ab were incubated with the beads. A
calibration curve was obtained for each Ab by plotting the median
fluorescence intensity (MFI) values against the ABC reported for the
beads. For each experimental sample, ABC values were deducted by
interpolation of the MFI on the calibration curve. The number of 15-1
specific binding sites, thus grossly reflecting the number of FcRI
molecules, was calculated by subtracting the ABC value for the isotype
control Ab from the ABC value for 15-1. Based on a similar principle,
Quifikit beads are coated with different amounts of mouse
anti-human CD5 Ab. Beads were incubated with the following relevant
secondary Abs: PE-conjugated anti-mouse-IgG (H + L)
F(ab')2 (surface expression on mouse eosinophils,
RBL, and CHO transfected with hFcRI
), FITC-conjugated
anti-mouse IgG (Fc specific) (intracellular expression and
surface expression following cIgE injection for mouse eosinophils) or
FITC-conjugated anti-mouse IgG F(ab')2
(intracellular expression in human eosinophils). Calibration curves
were obtained as for the Quantum kit. For each experimental sample, ABC
values were obtained for the incubation of the secondary Abs following
incubation with the relevant primary Ab or with its isotype-matched
control. The specific ABC was calculated by subtraction of the ABC
value for the isotype control from the ABC value for the primary
Ab.
Detection of mouse IL-10 by intracellular flow cytometry
After fixation and permeabilization (as described above), mouse eosinophils were incubated first with 5 µl normal rat serum followed by the addition of 5 µg/ml PE-conjugated rat anti-mouse IL-10 or PE-conjugated rat IgG1. After 30 min, cells were washed twice in permeabilization buffer, once in PBS and were resuspended in PBS-0.5% BSA for analysis.
Eosinophil activation
Highly purified human eosinophils (2 x 106/ml in a 24-well plate) were incubated first with cIgE for 1 h at 37°C followed by the addition of anti-hIgE at 10 µg/ml. Alternatively cells were stimulated with 10 µg/ml 15.1 mAb, followed by the addition of 10 µg/ml anti-mIgG F(ab')2. Supernatants were collected after 18 h and analyzed for cytokine release.
Purified murine splenic eosinophils (2 x
106cells/ml in a 24-well plate) were incubated
for 4 days with 5 µg/ml cIgE or with 5 µg/ml control ascites, in
the presence of 2.5 ng/ml rhIL-5. FcRI activation was achieved by
the addition of 10 µg/ml anti-human IgE to the culture.
Supernatants were collected after 18 h and analyzed for cytokine
release.
Cytokine quantification
IL-10 was assayed in eosinophil supernatants using specific Elisa kit (Diaclone, and R&D Systems, Minneapolis, MN, for human and mouse, respectively) according to the manufacturer’s instructions. The lower detection limit was 5 pg/ml for hIL-10 and <4 pg/ml for mIL-10.
Statistical analyses
Statistical significance was determined using Student’s
t test for unpaired groups with a 95% confidence level.
Correlation between IgE levels in the serum from the patients and
FcRI surface expression was established using Spearman’s rank
coefficient. Analyses were performed using Statview software.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI expression by human eosinophils
FcRI expression on freshly purified human eosinophils was first
analyzed by flow cytometry using an anti-hFcRI Ab. This
highly specific monoclonal Ab, which interacts with the IgE binding
site and thus only detects unoccupied receptors, has been successfully
used to detect hFcRI on human eosinophils (2, 12),
epidermal Langerhans cells (4), monocytes
(7), and platelets (3). A fluorescein-labeled
conjugate (FITC-15.1 mAb) was used. Specific binding of this
anti-receptor Ab was detected (Fig. 1A). This binding was significantly inhibited upon incubation
of cells with saturating amounts of hIgE before the staining (Fig. 1A). Detection of hFcRI on human eosinophils was
further confirmed using two other mAbs, CRA-1 (nonblocking) and CRA-2
(blocking) (16), and a rabbit polyclonal antiserum (997)
(15) (data not shown). Because the various Abs provided us
with similar results and because specific binding of 15-1 could easily
be assessed by preincubation with IgE, this later was used for the rest
of the experiments. The number of unoccupied receptors present at the
cell surface was then determined by quantitative flow cytometry on
eosinophils from 22 hypereosinophilic patients and 4 healthy donors. In
a preliminary experiment, we validated our quantification method using
two cell lines expressing FcRI. Using a mouse anti-rat
FcRI (3A92), we determined that rat basophilic leukemia cells
expressed 2.2 x 105 Ab binding sites at
their surface, a number in agreement with a previously published study
(23) (Fig. 1B). Likewise quantification of Ab
binding sites on CHO cells stably transfected with the three chains of
the human FcRI with 15-1, 22E7 (a nonblocking anti-hFcRI),
and CRA-1 was giving nearly identical results (1.5 x
105 Ab binding sites) (Fig. 1B).
Eosinophils expressed an average of 4548 ± 1090 Ab binding sites
at their surface. Although surface expression of FcRI was detected
in every sample analyzed, expression levels were heterogenous. Sorting
according to the pathology revealed that the number of unoccupied
receptors was significantly higher on eosinophils from patients with
hematological disorders and skin diseases (1.6- and 1.2-fold,
respectively) when compared with normal donors (Fig. 1C). By
contrast, FcRI expression on eosinophils from patients with allergy
or hypereosinophilic syndromes was similar to that found on cells from
normal donors (Fig. 1C). Nevertheless, as previously
reported by others (12), a correlation (r
= 0.625) was found between IgE levels in the serum from patients and
FcRI surface expression (n = 20) (Fig. 1D). If a correlation between serum IgE levels and
membrane-bound IgE (to both FcRI and FcRII/CD23) was fully
anticipated (Fig. 1E), we also found a highly significant
correlation (r = 0.912) between receptor-bound IgE and
unoccupied FcRI (Fig. 1F). This thus reflects that the
number of unoccupied receptors (available for further IgE binding) does
increase with IgE serum concentrations.
|
RI at their surface, we investigated whether
culture in the presence of IgE was able to increase receptor
expression, as previously reported for mast cells (24) and
basophils (25). After 4 days of culture (with 2.5 ng/ml
IL-5) in the absence of IgE, FcRI was expressed at higher levels
than on freshly purified cells from normal donors (Fig. 2, thin-line histogram). Furthermore, a dose-dependent increase of
FcRI expressed at the surface of eosinophils was observed when cells
were cultured in the presence of cIgE (Fig. 2, inset). Thus,
it demonstrates that the correlation between IgE concentrations and
FcRI expression also holds true in vitro, at least within a certain
range. In contrast with eosinophils from normal donors, such a
dose-dependent increase was not found for eosinophils from patients
that displayed only a moderate increase at the lowest dose of IgE
before reaching a plateau. Additionally, eosinophils from patients
displayed slightly increased receptor expression after 4 days of
culture in the absence of cIgE (data not shown).
|
RI were then compared on
the same samples (n = 5). An average of 46,202 ±
8,974 Ab binding sites were detected in the cytoplasm, whereas
4,260 ± 310 Ab binding sites were detected at the cell surface
(Fig. 3). Thus far, the pool of cytoplasmic hFcRI molecules exceeds the
number of receptors at the cell surface of human eosinophils.
|
RI on eosinophils from IL-5 x hFcRI
double-Tg mice
Because eosinophils from patients displayed heterogeneous FcRI
expression, we sought to obtain a relevant animal model, which would
allow us to study FcRI expression and function on eosinophils in
reproducible conditions and without the inconvenience of material
availability. Therefore, we crossed hypereosinophilic IL-5 Tg mice with
hFcRI Tg animals expressing a humanized FcRI with the same
cellular distribution as humans. IL-5 x hFcRI Tg mice
displayed massive eosinophilia in several organs: bone marrow
(50.3 ± 6.3% eosinophils), peripheral blood (51.7 ± 11.5%
eosinophils), spleen (42.2 ± 7.6% eosinophils), and peritoneal
cavity (50.8 ± 11.2% eosinophils).
As for human eosinophils, expression of hFcRI on eosinophils from
these animals was investigated. Expectedly, flow cytometric analysis
allowed us to detect low expression levels (822 Ab binding sites) for
humanized FcRI (hFcRI) at the surface of freshly isolated splenic
(Fig. 4A) and peritoneal (data not shown) eosinophils from IL-5
x hFcRI Tg animals, whereas no receptor expression was detected
on eosinophils from IL-5 Tg mice (data not shown).
|
RI expression on murine mast cells (22) and basophils
(25) in vivo, we investigated whether such a phenomenon
was also occurring on eosinophils. IL-5 x hFcRI Tg mice
were injected 4 times with 100 µg cIgE and sacrificed 24 h after
the last injection. The number of occupied Ab binding sites and, upon
saturation with cIgE, total Ab binding sites present at the surface of
freshly isolated splenic and peritoneal eosinophils was determined by
flow cytometry. We found that splenic and peritoneal eosinophils had
1950 and 2248 occupied Ab binding sites, respectively, and 5013 and
9094 total Ab binding sites, respectively. A typical histogram is
represented for splenic eosinophils on Fig. 4
B. Taking this
number into account, we deducted that, following the injection of 400
µg of cIgE, 3000 and 6850 unoccupied Ab binding sites,
respectively, were present at the surface of splenic and peritoneal
eosinophils. Using a blocking () or a nonblocking
(22E7) anti-hFcRI Ab to directly determine the number of
unoccupied Ab binding sites, as performed on human cells, we obtained
similar results (3205 with 15-1 and 3019 with 22E7) for splenic
eosinophils and lower values (4660 and 4075) for peritoneal
eosinophils. Taken together, these data show that IgE-mediated FcRI
up-regulation does occur at the surface of eosinophils from IL-5
x hFcRI Tg mice in vivo, whether the unoccupied or total number
of receptors are considered.
As we found for eosinophils from normal donors, expression of hFcRI
after culture with IgE showed a similar dose-dependent increase when
compared with eosinophils cultured in the absence of IgE (Fig. 4C and inset). However, in contrast with human
eosinophils, the mere culture in the absence of IgE did not increase
receptor expression (Fig. 4C, thin-line histogram).
Number of Ab binding sites at the surface and of unoccupied
intracellular hFcRI molecules were then determined for bone
marrow, blood, and splenic and peritoneal eosinophils. Resting
eosinophils isolated from the 4 organs expressed <1000 Ab binding
sites at their surface. After 4 days culture with 5 µg/ml cIgE, the
number of Ab binding sites ranged from 4410 to 7804 at the cell
surface, according to the origin of eosinophils, whereas culture in the
absence of IgE barely affected the number of receptors (Fig. 5A). Thus, in the presence of IgE, eosinophils from IL-5
x hFcRI Tg mouse express similar levels of surface FcRI as
human eosinophils.
|
RI molecules in
eosinophils freshly isolated from different organs was at least 10
times higher than surface expression by the corresponding cells (Fig. 5, B compared to A). Upon culture in the presence
of IL-5 but without IgE, splenic and peritoneal eosinophils had
increased numbers of unoccupied intracellular hFcRI (Fig. 5B). Except for bone marrow eosinophils, the number of
unoccupied intracellular hFcRI further increased upon culture
with both cIgE and IL-5 (Fig. 5B). Peritoneal eosinophils
had significantly more unoccupied intracellular FcRI than blood
(p < 0.02) or splenic
(p < 0.01) eosinophils (Fig. 5B).
As for human blood eosinophils, bone marrow, blood, and splenic and
peritoneal eosinophils, IL-5 x hFcRI Tg mice harbor, under
these experimental conditions, a large intracellular pool of unoccupied
hFcRI and express low number of surface FcRI.
Expression of FcRI by S. mansoni-infected mouse
eosinophils
Elevated IgE levels are a hallmark of helminthic infections in
both human and rodents (26). Furthermore, human
(2) and rat (27) eosinophils (but not WT
mouse eosinophils) have been shown to participate to FcRI-dependent
ADCC toward S. mansoni larvae in vitro. To investigate
whether up-regulation of eosinophil-expressed FcRI by IgE was also
taking place in vivo, during the course of schistosomiasis, we
determined the number of Ab binding sites on hFcRI x IL-5
Tg mouse eosinophils after 46 days infection with S.
mansoni, at a time when IgE levels begin to increase. Eosinophils
isolated from the different organs from infected animals displayed a
significant increase in FcRI surface expression when compared with
cells from noninfected mice (Fig. 6). The maximum increment was observed for peritoneal and splenic
eosinophils. Interestingly, this increased membrane expression of
FcRI was associated to decrease in the intracellular pool of
unoccupied hFcRI in bone marrow, blood, and peritoneal
eosinophils (significant only for peritoneal eosinophils). These
experiments show that infection by S. mansoni is able to
up-regulate surface expression of FcRI on eosinophils in vivo,
increased IgE levels are likely to play an important part in this
phenomemon.
|
RI
Having established that both human and Tg mouse eosinophils
expressed comparable numbers of FcRI, we investigated the
contribution of FcRI to both effector and regulatory function of
eosinophils. Because IgA immune complexes have been recently shown to
induce the release of IL-10 by human eosinophils (20), it
was worth investigating whether IgE and FcRI would also mediate
IL-10 release by human or Tg mouse eosinophils. Upon receptor
engagement with cIgE and secondary Ab, human eosinophils were able to
release 62 ± 41 pg/ml IL-10 (above control), whereas receptor
cross-linking with anti-FcRI mAb (15.1 mAb) and anti-mIgG
released up to 200 pg/ml IL-10 (Fig. 7A). We then verified that, in our experimental model, mouse
eosinophils contained IL-10 as previously demonstrated upon S.
mansoni infection (28). Intracellular staining,
revealed that eosinophils from IL-5 x hFcRI Tg mice
contained high amounts of IL-10 (Fig. 7, inset). After
culture in the presence of cIgE, receptor cross-linking with
anti-hIgE led to a release of 234 ± 126 pg/ml mIL-10 (above
control) (Fig. 7B). These results demonstrate that both
human and IL-5 x hFcRI Tg mouse eosinophils are able to
secrete an immunoregulatory cytokine upon FcRI-dependent
activation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI expression by human eosinophils and its involvement in
IgE-mediated ADCC reactions toward S. mansoni larvae had
been first reported in 1994 (2). Using quantitative and
sensitive methods, we have shown here that levels of unoccupied
receptors at the surface of human eosinophils correspond to 4500 Ab
binding sites. The use of an anti-receptor Ab interacting with the
IgE binding site without prior treatment with lactic acid (which would
have damaged cellular integrity) prevented us from detecting
IgE-occupied FcRI. Nevertheless, in agreement with a previous study,
we were able to find a correlation between receptor expression level
and serum IgE concentrations (12). The correlations found
between serum IgE levels, membrane-bound IgE, and unoccupied FcRI
reflect that, even in patients with high IgE levels, these receptors
are not saturated, probably due to IgE-mediated receptor up-regulation.
Otherwise these correlations would have been observed only for patients
with (very) low IgE levels. We were also able to demonstrate that the
number of unoccupied receptors was significantly increased in some
pathologies (hematological disorders and skin diseases) but not in
others (allergies and hypereosinophilic syndromes) when compared with
normal donors. Hematological disorders (tumors, myeloproliferative
lymphoma, or eosinophilic leukemia) included in this study are
characterized by high levels of cell proliferation; thus, it is
possible that undifferentiated, proliferating eosinophils or eosinophil
precursors would be more sensitive to receptor up-regulation. An
increased IgE receptor expression would provide eosinophils with a
means to exert anti-tumoral activity through IgE-mediated ADCC. It
has been shown on one hand that eosinophils could play a role in IL-4
anti-tumoral activity (29) and, on the other hand,
that FcRI participated in anti-tumoral activity (30, 31). Nevertheless, even in hematological disorders, surface
expression of unoccupied FcRI by eosinophils is low, when compared
with basophils, which express up to 6 x 105
receptors at their surface (32). The absence of FcR,
which has recently been shown to increase surface expression of the
receptor (33) in eosinophils might account at least in
part for theses differences between cell types.
Along the same lines, we were also able to increase receptor expression
on eosinophils from normal donors in vitro, upon a 4-day culture in the
presence of cIgE. Such a dose-dependent increase was not observed on
eosinophils from patients, which already expressed more receptors and
with high IgE concentrations in their serum. A proportionally
comparable dose-dependent increase of receptor expression on
eosinophils from patients would represent a much more important
increase in the number of additional receptors expressed over such a
short period of time. Aside from the time factor, the limited range of
IgE-driven FcRI up-regulation on eosinophils in vitro might also be
attributed in part to their lack of FcR protein,
In contrast, we demonstrated that eosinophils contain large amounts of
unoccupied intracellular FcRI. These differences in the number of
unoccupied receptors between surface and intracellular compartments
could explain why some studies failed to detect surface expression, but
were able to report on the presence of FcRI inside the cells
(11). Three factors could contribute to this excess of
intracellular unoccupied FcRI. First, high levels of proteolysis
could lead to the release of almost all the surface-expressed
FcRI as a soluble receptor. Second, an extensive storage of
receptors in eosinophil granules before their exportation to the
surface could take place. Third, an overproduction of FcRI
compared with FcR might occur. FcR is necessary for surface
expression and/or function of FcRI, FcRI (34), and
FcRI (35), all expressed by human eosinophils,
and thus might be the key factor limiting surface expression of these
three receptors, considered as a whole. Such a competition has
been demonstrated between FcRI and FcRIII on murine bone
marrow-derived mast cells (36).
To obtain a confirmation of our results in a relevant experimental
model and because WT mice do not express FcRI (13), we
crossed previously generated hFcRI Tg mice expressing the
receptor on eosinophils (under the control of hFcRI promoter)
(13) with hypereosinophilic IL-5 Tg animals
(14). These IL-5 x hFcRI Tg mice not only
provided us with an abundant and reproducible source of material, but
also allowed us to study eosinophils from different organs. Freshly
isolated cells obtained from naive animals expressed <1000 Ab binding
sites at their surface, whereas the receptor was not detectable on
cells from IL-5 Tg animals. Such a low surface expression on
eosinophils from IL-5 x hFcRI Tg animals might be due to
the virtual absence of IgE in naive mice. As previously reported for
other cell types (22, 25, 37) as well as for rat
eosinophils (27), injection of IgE or culture in the
presence of IgE led to a significant increase in surface expression,
which was then comparable to the levels observed in humans. Although
the molecular mechanisms underlying this phenomenon are only partially
understood, it is now widely admitted that IgE stabilizes receptors
anchored at the membrane, while allowing more receptors to be
synthesized and targeted to the membrane, thus concurring to the
increased expression. Nevertheless, unoccupied intracellular
hFcRI was also present in high amounts in mouse eosinophils, when
compared with surface expression.
We have shown that eosinophils obtained from various organs expressed
different levels of FcRI at their surface. Bone marrow eosinophils
expressed the lower number of receptors, followed by blood and splenic
and peritoneal eosinophils. This leads us to envision a pathway along
which eosinophils would mature and, among other parameters, increase
their expression levels of FcRI. In humans and most likely in naive
animals, eosinophils mature in the bone marrow for 3–4 days, then
migrate and stay for several hours to 1 day in the blood stream to
finally reach the organs where they remain for 2–7 days in tissues
(38). This would likely explain why peritoneal eosinophils
display the highest expression and are probably the most
differentiated. Because spleen has an open circulatory system and blood
cells freely migrate into the splenic stroma without transendothelial
migration, splenic eosinophils might have lacked the necessary stimuli
provided, during transendothelial migration, through adhesion
molecules-mediated cell-cell contacts and might thus be biologically
different from other tissue eosinophils, such as peritoneal ones
(39, 40).
During experimental infection with S. mansoni, FcRI
surface expression was greatly increased in particular on splenic and
peritoneal eosinophils from double-Tg mice. According to our proposed
model, it seems logical that the most differentiated types of
eosinophils would be found in tissues.
Expectedly, higher levels of surface expression were reached after a
46-day in vivo infection when compared with a 4-day in vivo treatment
with IgE and all the more with an in vitro IgE-induced up-regulation
experiment. In the former case, IgE is not the only factor that is
likely to promote FcRI expression over such an extended period; IL-4
is another one (24, 41, 42, 43). A similar phenomenon had been
observed on intestinal rat mast cells, where infection by
Nippostrongylus brasiliensis was more efficient at
increasing FcRI expression (measured by detection of FcR) than
the mere injection of IgE (44). A kinetic study of FcRI
expression on eosinophils from IL-5 x hFcRI Tg mice along
the course of S. mansoni infection might provide information
about the respective roles of the infection by itself, of IL-4 and of
IgE on the increase in receptor expression. Nevertheless, we have shown
already that eosinophils from S. mansoni-infected
hFcRI Tg mice displayed IgE-dependent cytotoxicity toward
S. mansoni larvae, whereas eosinophils from similarly
infected WT animals were ineffective. In contrast, we have also shown
that hFcRI Tg displayed decreased granuloma volume when compared
with WT animals (45). In these animals, expression is not
restricted to eosinophils, but also extends to APC (monocytes and
epidermal Langerhans cells); however, one can argue that FcRI might
not only affect serum IgE levels through Ag presentation, but also
eosinophil activation.
Measurements, on the same eosinophil samples, of surface and
intracellular FcRI, allowed us to get some additional insight
about the mechanisms of ligand-induced receptor up-regulation. When IgE
is present in the biological fluids (i.e., upon S.
mansoni infection or in vitro in culture medium), the amount of
surface FcRI increases by remaining for longer period of time at the
surface. IgE-receptor complexes are reinternalized more slowly, and are
protected from proteolysis and thus released in the medium at a slower
rate than unoccupied receptors. Even if the synthesis of new receptors
is stimulated at the transcriptional and/or at the transductional level
upon helminthic infection, the large intracellular pool of free
FcRI, existing in cells from naive animals is partially depleted.
This phenomenon was more strikingly observed on peritoneal eosinophils.
It thus means that, even if basal levels of surface FcRI expression
are low, they can be greatly increased in some situations (parasitic
infection, inflammatory diseases, etc.) and play a greater role than
previously inferred by some recent studies (11, 12).
We were also able to demonstrate that, upon FcRI engagement with IgE
and anti-hIgE, both human and IL-5 x hFcRI Tg mouse
eosinophils were able to release IL-10, which is abundant inside cells
from both species (20, 28). Aside from their involvement
in cytotoxic reactions, eosinophils are thus likely to act not only as
effectors but also as modulators of the immune response. IL-10, also
released by human eosinophils upon triggering with secretory IgA
(20) affects, among other cell types, the Ag presenting
capacity of macrophages and decreases type 1 cytokine production by
lymphocytes.
In conclusion, we have here demonstrated, using two parallel systems,
human and IL-5 x hFcRI Tg mice, that eosinophils express
low amounts of FcRI at their surface, while possessing a large
intracellular pool of unoccupied FcRI. Surface expression can be
up-regulated and allows eosinophils to participate to FcRI-mediated
reactions. The use of a relevant animal model, more faithfully
reproducing the human situation, should provide us with more
information about the role of this receptor in eosinophil function and
in human diseases.
|
Acknowledgments |
|---|
RI (15.1) mAb and for allowing the use of
FcRI Tg mice and to Dr. C. Ra for providing us with CRA-1
and CRA-2 Abs. |
Footnotes |
|---|
2 Current address: Department of Clinical and Laboratory Medicine, Akita University School of Medicine, Hondo Akita 010, Japan.
3 Address correspondence and reprint requests to Dr. Monique Capron, Unité Institut National de la Santé et de la Recherche Medicalé, Unité 167, Institut Pasteur, 1 rue du Prof. Calmette, BP 245, 59019 Lille Cedex, France. E-mail address: monique.capron{at}pasteur-lille.fr
4 Abbreviations used in this paper: Tg, transgenic; hFcRI, human FcRI; mIgG1, mouse IgG1; rhIL-5, recombinant human IL-5; SA-PE, PE-conjugated streptavidin; ABC, Ab binding capacities; MFI, median fluorescence intensity; hIgE, human myeloma IgE; cIgE, chimeric hIgE; ADCC, Ab-driven cellular cytotoxicity;
Received for publication June 28, 2000. Accepted for publication May 8, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RI): from physiology to pathology. Annu. Rev. Immunol. 17:931.[Medline]
RI). J. Exp. Med. 175:1285.RI. J. Exp. Med. 175:1353.RI) mediates IgE-dependent allergen presentation. J. Immunol. 154:6285.[Abstract]
RI) on monocytes of atopic individuals. J. Exp. Med. 179:745.RI as a complex composed of FcRI- and FcRI-chains and can use this receptor for IgE-mediated allergen presentation. J. Immunol. 157:607.[Abstract]
receptors on murine eosinophils: implications for the functional significance of elevated IgE and eosinophils in parasitic infections. Blood 89:3826. is a molecular amplifier of IgE- and IgG-mediated in vivo responses. Immunity 8:517.[Medline]
RI by human eosinophils. J. Immunol. 162:6893.RI chain is necessary for efficient folding in the endoplasmic reticulum. J. Biol. Chem. 270:8249.RI) in human platelets and its’ intracellular expression in human megakaryocytes. Blood 93:2543.RI). Analysis of functional domains of the subunit with monoclonal antibodies. J. Biol. Chem. 266:11245. receptor binding site to the third constant region domain of IgE. EMBO J. 10:101.[Medline]
): inhibition by immunoglobulin a complexes. J. Exp. Med. 190:487.RI expression in vitro and in vivo: evidence for a novel amplification mechanism in IgE-dependent reactions. J. Exp. Med. 185:663.-, - and -subunits of the receptor for immunoglobulin E. Mol Immunol 22:1045.[Medline]
receptor I expression and IgE-dependent release of histamine and lipid mediators from human umbilical cord blood- derived mast cells: synergistic effect of IL-4 and IgE on human mast cell Fc receptor I expression and mediator release. J. Immunol. 162:5455.RI expression in vivo. J. Immunol. 158:2517.[Abstract]
RI on rat eosinophils and macrophages. J. Immunol. 165:1266. RI bearing cells. Hum. Antibodies 8:169.[Medline]
RI- subunit. Immunity 12:515.[Medline]
chain deletion results in pleiotrophic effector cell defects. Cell 76:519.[Medline]
chain. Molecular basis for CD89/FcR chain association. J. Biol. Chem. 270:29781.RI chain results in upregulation of FcRIII-dependent mast cell degranulation and anaphylaxis: evidence of competition between FcRI and FcRIII for limiting amounts of FcR and chains. J. Clin. Invest. 99:915.[Medline]
4 integrin have distinct roles in eosinophil tethering and arrest on vascular endothelial cells under physiological flow conditions. J. Immunol. 159:3929.[Abstract]
RI -chain messenger RNA in eosinophils. J. Allergy Clin. Immunol. 96:1161.[Medline]
RI) on human mast cells by IL-4. Int. Immunol. 8:1367. receptor type I expression in recombinant human stem cell factor-dependent fetal liver-derived human mast cells. J. Immunol. 159:2911.[Abstract]
RI expression in the rat intestinal mucosa and tongue is enhanced during Nippostrongylus brasiliensis infection and can be up-regulated by in vivo administration of IgE. J. Immunol. 158:3805.[Abstract]
RI transgenic (hFcRI) mice as a model for the study of human schistosomiasis. F. D. Finkelman, and S. J. Galli, and J. F. Urban, and R. L. Coffman, eds. Keystone Symposia on Molecular and Cellular Biology: Type 2 Cytokines in Allergy and Helminth Infection 307. Incline Village, NV. (Abstr.). This article has been cited by other articles:
|
|
 |
|
  E. A. Nigro, A. T. Brini, E. Soprana, A. Ambrosi, D. Dombrowicz, A. G. Siccardi, and L. Vangelista Antitumor IgE Adjuvanticity: Key Role of Fc{epsilon}RI J. Immunol., October 1, 2009; 183(7): 4530 - 4536. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Madan, F. Demircik, S. Surianarayanan, J. L. Allen, S. Divanovic, A. Trompette, N. Yogev, Y. Gu, M. Khodoun, D. Hildeman, et al. Nonredundant Roles for B Cell-Derived IL-10 in Immune Counter-Regulation J. Immunol., August 15, 2009; 183(4): 2312 - 2320. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. N. Karagiannis, M. G. Bracher, J. Hunt, N. McCloskey, R. L. Beavil, A. J. Beavil, D. J. Fear, R. G. Thompson, N. East, F. Burke, et al. IgE-Antibody-Dependent Immunotherapy of Solid Tumors: Cytotoxic and Phagocytic Mechanisms of Eradication of Ovarian Cancer Cells J. Immunol., September 1, 2007; 179(5): 2832 - 2843. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Ganley-Leal, P. N. Mwinzi, C. B. Cetre-Sossah, J. Andove, A. W. Hightower, D. M. S. Karanja, D. G. Colley, and W. E. Secor Correlation between Eosinophils and Protection against Reinfection with Schistosoma mansoni and the Effect of Human Immunodeficiency Virus Type 1 Coinfection in Humans Infect. Immun., April 1, 2006; 74(4): 2169 - 2176. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Spik, C. Brenuchon, V. Angeli, D. Staumont, S. Fleury, M. Capron, F. Trottein, and D. Dombrowicz Activation of the Prostaglandin D2 Receptor DP2/CRTH2 Increases Allergic Inflammation in Mouse J. Immunol., March 15, 2005; 174(6): 3703 - 3708. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Decot, G. Woerly, M. Loyens, S. Loiseau, B. Quatannens, M. Capron, and D. Dombrowicz Heterogeneity of Expression of IgA Receptors by Human, Mouse, and Rat Eosinophils J. Immunol., January 15, 2005; 174(2): 628 - 635. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Woerly, K. Honda, M. Loyens, J.-P. Papin, J. Auwerx, B. Staels, M. Capron, and D. Dombrowicz Peroxisome Proliferator-activated Receptors {alpha} and {gamma} Down-regulate Allergic Inflammation and Eosinophil Activation J. Exp. Med., August 4, 2003; 198(3): 411 - 421. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Woerly, P. Lacy, A. B. Younes, N. Roger, S. Loiseau, R. Moqbel, and M. Capron Human eosinophils express and release IL-13 following CD28-dependent activation J. Leukoc. Biol., October 1, 2002; 72(4): 769 - 779. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or after 4 days culture in the
presence of 2.5 ng/ml rhIL-5 without (
) 5 µg/ml
cIgE. A, Surface expression of Fc
). Eosinophils
were stained for membrane and intracellular Fc