| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
on Human Eosinophils: II. IL-5 Down-Modulates Its Receptor Via a Proteinase-Mediated Process1
* Allergy and Immunology and
Pulmonary and Critical Care Sections of Department of Medicine, and Departments of
Biomolecular Chemistry and
Pediatrics, University of Wisconsin, Madison, WI 53792; and
¶ Department of Internal Medicine, Mayo Clinic and Mayo Foundation, Rochester, MN 55905
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
expression is attenuated on
bronchoalveolar lavage eosinophils, soluble (s)IL-5R is detectable
in BAL fluid in the absence of increased steady state levels of
sIL-5R mRNA, and BAL eosinophils become refractory to IL-5 for ex
vivo degranulation. We hypothesized that IL-5 regulates its receptor
through proteolytic release of mIL-5R, which in turn contributes to
the presence of sIL-5R. Purified human peripheral blood eosinophils
were incubated with IL-5 under various conditions and in the presence
of different pharmacological agents. A dose-dependent decrease in
mIL-5R was accompanied by an increase in sIL-5R in the
supernatant. IL-5 had no ligand-specific effect on mIL-5R or
sIL-5R mRNA levels. The matrix metalloproteinase-specific inhibitors
BB-94 and GM6001 and tissue inhibitor of metalloproteinase-3 partially
inhibited IL-5-mediated loss of mIL-5R, suggesting that sIL-5R
may be produced by proteolytic cleavage of mIL-5R. IL-5 transiently
reduced surface expression of 
-chain, but had no effect on the
expression of GM-CSFR. Pretreatment of eosinophils with a dose of
IL-5 that down-modulated mIL-5R rendered these cells unable to
degranulate in response to further IL-5 stimulation, but they were
fully responsive to GM-CSF. These findings suggest that
IL-5-activated eosinophils may lose mIL-5R and release sIL-5R in
vivo, which may limit IL-5-dependent inflammatory events in diseases
such as asthma. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Eosinophil responses to IL-5 are mediated through the IL-5R, a
heterodimer consisting of an IL-5-specific subunit and a
signal-transducing subunit (-chain
(c)),3 which is
identical with the signal transduction subunit for IL-3R and GM-CSFR
(2, 5, 9). Through alternative splicing,
IL-5R-chain transcripts can be expressed in several isoforms, one
that includes a transmembrane domain and encodes a membrane (m)-bound
receptor, and another that encodes a soluble (s) form
(10). rsIL-5R protein has been shown to bind to IL-5
(11) and inhibit IL-5-induced signal transduction,
mediator release, and survival in vitro (12). In addition,
sIL-5R can inhibit allergen-induced in vitro differentiation of
eosinophils (13). It has been suggested that sIL-5R
plays an immunoregulatory role in vivo (11, 14, 15), but,
to date, the production of sIL-5R protein by human eosinophils has
not been demonstrated.
In the accompanying study, we demonstrated that, following local airway
Ag challenge, eosinophils recruited to the airway exhibit a loss of
mIL-5R and a loss of responsiveness to IL-5, and that the
bronchoalveolar lavage (BAL) fluid contains an increased concentration
of sIL-5R in the absence of increased steady state levels of
sIL-5R mRNA in BAL eosinophils (40). Although
eosinophil activation and their migration from the circulation to the
airway lumen are associated with the presence of IL-5, our data suggest
that within the airway, eosinophil activation by IL-5 is probably
limited and may in fact be switched to an IL-5-independent
mechanism.
A number of mechanisms may contribute to the down-modulation of
mIL-5R on airway eosinophils and to the increased presence of
sIL-5R in the BAL fluid. However, the close association of these two
events led us to hypothesize that mIL-5R may be released from the
eosinophil cell surface via a proteolytic process. We have previously
demonstrated that matrix metalloproteinases (MMPs) are increased in BAL
fluid following Ag challenge (16), and they are known to
be required for eosinophil migration through basement membrane
components (17). Furthermore, MMPs, proteases, and ADAMs
(membrane-associated proteins containing a disintegrin and
metalloproteinase domain) have recently gained considerable attention
for their role as sheddases for the release of integral plasma membrane
proteins (18, 19, 20).
Because peripheral blood eosinophils activated with IL-5 share many
characteristics with BAL eosinophils isolated following airway Ag
challenge (1, 21, 22, 23, 24), we used an in vitro model system to
investigate the effect of IL-5 on expression of mIL-5R on
circulating eosinophil, to determine whether IL-5 induces the release
of sIL-5R from eosinophils, and to establish mechanisms by which
these events occur. Purified human peripheral blood eosinophils were
treated with IL-5 for various times, and cell surface expression of
mIL-5R was determined by flow cytometry, steady state levels of
sIL-5R and mIL-5R were determined by RT-PCR, and concentrations
of sIL-5R in cell culture supernatant fluids were determined by
ELISA. In addition, MMP inhibitors were used to evaluate the role of
MMPs on modulation of IL-5R expression.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Eosinophils were purified from the heparinized peripheral blood of atopic volunteer donors, with eosinophils composing 2–10% of the peripheral blood leukocytes. Briefly, peripheral blood was layered over 1.090 g/ml Percoll and centrifuged for 20 min at 700 x g. The granulocyte mixture pellet was collected, RBCs were lysed by hypotonic shock, and neutrophils were depleted by incubation with anti-CD16-conjugated immunomagnetic beads and collected by exposure to a magnetic field (AutoMac system; Miltenyi Biotec, Auburn, CA) (25). The resulting eosinophils were >98% pure and >97% viable.
Eosinophil treatment
To determine the effect of IL-5 on expression of eosinophil cell
surface mIL-5R and release of sIL-5R, freshly isolated peripheral
blood eosinophils were treated in duplicate with medium (RPMI 1640 with
5% FCS and 1% penicillin/streptomycin), IL-5 (0.1 pg/ml to 100
ng/ml), or GM-CSF (10 ng/ml) at 2 x 106/ml
in a final volume of 0.5 ml in 48-well plates, and incubated at 37°C,
5% CO2, in a humidified incubator for the times
indicated in each experiment. In one experiment, eosinophils were
incubated at both 37°C and 4°C. All cytokines were purchased from
R&D Systems (Minneapolis, MN). In additional experiments, eosinophils
were preincubated with MMP inhibitors: 0.5 mM 1–10, phenanthroline
(Sigma-Aldrich, St. Louis, MO), 50 µM BB-94 (Batimastat, a kind gift
from British Biotech Pharmaceuticals, Oxford, U.K.), 50 µM GM6001
(Galardin; Chemicon, Temecula, CA), 2 µg/ml tissue inhibitor of MMP
(TIMP)-1 (Calbiochem, San Diego, CA), 0.4 µg/ml TIMP-2 (Chemicon), 5
µg/ml TIMP-3 (Chemicon), 20 µM phosphoramidon (Calbiochem), 2 mM
MMP inhibitor-1 (Calbiochem), or protease inhibitors 500 µg/ml
-aminocaproate acid (Sigma-Aldrich) and 500 µM leupeptin
(Calbiochem). Dose-response curves were performed to determine optimal
concentrations (lowest amount of inhibitor that resulted in greatest
inhibition without inducing cell death). Eosinophils were preincubated
with the above pharmacological inhibitors for 2 h, then exposed to
IL-5 for an additional 3 h. In some experiments, eosinophils were
cultured in duplicate at 4 x 106/ml in a
final volume of 0.5 ml. Cell pellets were collected and pooled together
for IL-5R mRNA measurements. Culture supernatant fluids were
removed, and fluids from duplicate wells were pooled and stored at
-20°C until analyzed for the presence of sIL-5R. Immediately
before analysis, supernatant fluids were concentrated to 5x at 4°C
using Ultrafree-4 centrifugal filter units (Millipore, Bedford, MA)
with a molecular mass cutoff limit of 3 kDa.
Flow cytometric analysis
For analysis of cell surface receptors, purified eosinophils
from each indicated experiment were washed and resuspended in ice-cold
FACS buffer (1x PBS containing 2% BSA and 0.2% sodium azide) at a
concentration of 2 x 106/ml. Cells (1
x 105) were then incubated for 30 min on ice
with appropriate Abs. Abs included PE-conjugated mAb to IL-5R
(CD125; BD PharMingen, San Diego, CA), c (CD131; eBioscience, San
Diego, CA), and CD69 (BD Immunocytometry Systems, San Jose, CA),
FITC-conjugated anti-GM-CSFR (CD116, BD PharMingen), and PE- and
FITC- conjugated mouse IgG isotype controls (BD Immunocytometry
Systems). After incubation, cells were washed twice and resuspended in
250 µl FACS buffer. To exclude the dead/dying cells and cell debris,
propidium iodide (at a final concentration of 3 µl/ml) was added
immediately before analysis. For analysis, 10,000 events were collected
using a BD Immunocytometry Systems FACScan II, and data analyses were
performed using the CellQuest software package (BD Immunocytometry
Systems).
Detection of sIL-5R in cell culture supernatant fluids by ELISA
A sensitive two-step sandwich-type ELISA was developed to
measure sIL-5R in 5x concentrated cell culture supernatant fluids.
Half-area-well ELISA plates (catalogue 3690; Corning, Corning,
NY) were coated overnight at 4°C with a predetermined optimal
concentration of purified monoclonal anti-human IL-5R (clone
A17; BD PharMingen). Nonspecific binding sites were blocked with 10%
dialyzed newborn calf serum. Test samples were incubated overnight at
4°C on Ab-coated plates, and then sIL-5R was detected with a
biotinylated goat anti-human IL-5R polyclonal Ab (R&D Systems).
Streptavidin conjugated to a HRP polymer (POLY-HRP-40; Research
Diagnostics, Flanders, NJ) was used to increase assay sensitivity. A
one-component substrate, 3,3',5'5'-tetramethylbenzidine (Kirkegaard &
Perry Laboratories, Gaithersburg, MD), was used for color development,
and the reaction was stopped by addition of 0.18 M sulfuric acid.
OD450 was determined with a Dynatech MR500 microplate
reader, and data were analyzed with Biolink Software (Dynatech
Laboratories, Chantilly, VA). The concentration of sIL-5R in
supernatant fluids was calculated by comparison with a standard curve
generated with known amounts of human rsIL-5R (Sigma-Aldrich). The
sensitivity for sIL-5R was 
12 pg/ml.
Eosinophil-derived neurotoxin (EDN) analysis
Freshly isolated peripheral blood eosinophils were pretreated with medium, 0.01 or 10 ng/ml IL-5, for 4 h. Culture medium contained RPMI 1640/5% FCS to prevent degranulation during the preincubation phase. Cells were then washed and restimulated with medium or an activating concentration of IL-5 (1 ng/ml) or GM-CSF (1 ng/ml) for additional 4 h at 37°C. During the activation phase, medium contained HBSS supplemented with 0.03% gelatin (HBSS/gel; Sigma-Aldrich) to allow eosinophil adhesion and degranulation. The cell-free supernatant fluids were frozen at -20°C until EDN was measured by RIA (26). The sensitivity for EDN was 2 ng/ml. Total cellular EDN was determined from parallel cultures to which an equal volume of 1% Triton X-100 in 0.1 N HCl was added.
Detection of IL-5R mRNA by RT-PCR
Total RNA was extracted from 4 x 106
IL-5- or medium-treated eosinophils using a one-step phenol/chloroform
extraction reagent (Tri Reagent; Sigma-Aldrich). Total RNA was treated
with DNase (RQ1 RNase-free DNase; Promega, Madison, WI) to degrade DNA,
and cDNA was synthesized, as previously described (27).
PCR was performed by transferring 4 µl of cDNA to a 650 µl
thin-walled PCR tube along with 2.5 U platinum Taq
(Invitrogen Life Technologies, Carlsbad, CA), 5 µl 10x PCR
buffer, 0.01 µM dNTPs, 50 mM MgCl2, and 0.2
µM of primer in a final volume of 50 µl. A forward primer specific
for mIL-5R, position 1033–1056, and two different reverse primers,
sIL-5R at position 1279–1298 and mIL-5R at position 1542–1564,
were constructed using published sequences (11). Upstream
and downstream primers were separated by introns so that any genomic
DNA amplified by these procedures could be discriminated from cDNA
based on size. The predicted size of cDNA fragments was 266 bp for
sIL-5R and 527 bp for mIL-5R. PCR was conducted at 94°C for 2
min, then 24 cycles of 96°C for 30 s, 60°C for 30 s, and
72°C for 30 s. The number of PCR cycles (24) was
optimized to maintain a linear relationship between mRNA and PCR
products. Controls included in each PCR run included samples containing
reagents with no cells and samples that had not been reverse
transcribed. A DNA probe (205 bp, position 1033–1237) for both forms
of IL-5R was synthesized and labeled with HRP, and PCR products were
detected by Southern blot analysis (ECL system; Amersham,
Piscataway, NJ).
Real-time PCR was used to assay for rRNA. Real-time PCR allows for the detection of a target sequence by continuous measurement of a fluorescent dye label generated during the course of amplification. The assay was performed on the ABI PRISM 7000 Sequence Detection System from Applied Biosystems (Foster City, CA). TaqMan ribosomal RNA control reagents with a probe bound to both a VIC reporter dye and a Tamara quencher and TaqMan Universal PCR Master Mix were purchased from Applied Biosytems. Duplicate reactions were conducted in 50 µl in a 96-well format using 0.8 µl of cDNA. The thermal cycler protocol was 95°C for 10 min, then 40 cycles of 95°C for 15 s and 60°C for 1 min. A relative standard was assayed with the samples, and a standard curve was generated with a slope of -3.01 and an R value of 0.9946. Values for the standard curve and samples were based on threshold cycles (CT), set in the geometric phase of the analysis.
Statistical analysis
Statistical analysis was performed using the SigmaStat software package (Jandel Scientific Software, San Rafael, CA). Data are expressed as medians with 25 and 75 interquartiles (or means ± SEM for normally distributed data). The Wilcoxon signed rank test (or a paired t test for normally distributed data) was used to compare different time points with 0 h. Correlations were made using Spearman rank order correlation test. A p value of <0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and sIL-5R
Purified peripheral blood eosinophils were incubated for
4 h with medium alone or increasing concentrations of IL-5 (Fig. 1
A). IL-5 caused a
dose-dependent loss of mIL-5R on eosinophils; at 10 ng/ml (300 pM)
of IL-5, mIL-5R was nearly undetectable. There was also a
corresponding increase in the early activation marker
(28), CD69. To confirm that the inability to detect
mIL-5R was not due to ligand occupancy of the receptor, eosinophils
were exposed to IL-5 at 4°C. Under these conditions, neither
mIL-5R nor CD69 expression was significantly changed from baseline
(Fig. 1B). Nonspecific activation of eosinophils with PMA or
calcium ionophore induced CD69, but had no effect on mIL-5R (data
not shown). Following IL-5 exposure, the reduction in mIL-5R
expression on the cell surface (Fig. 2A) was paralleled by an
increase in the concentration of sIL-5R in the cell culture
supernatant fluids (Fig. 2B). Furthermore, the levels of
sIL-5R in the cell culture supernatant fluids inversely correlated
(Spearman’s correlation coefficient
(rs) = -0.776, p <
0.001) with eosinophil cell surface expression of mIL-5R (Fig. 2C).
|
|
To determine the ex vivo kinetics of receptor expression on
eosinophils, purified peripheral blood eosinophils were cultured in
medium alone, IL-5 (10 ng/ml), or GM-CSF (10 ng/ml) for 0, 0.5, 4, and
24 h. Eosinophil surface expression of mIL-5R, c, and
mGM-CSFR was then determined by flow cytometry. After 30 min of IL-5
exposure, mIL-5R was dramatically reduced and was nearly
undetectable at 4 and 24 h (Fig. 3A). Re-expression of
mIL-5R did not occur when eosinophils were evaluated again at 48 and
72 h (data not shown). In contrast, c expression was markedly
reduced at 30 min, began to return by 4 h, and was completely
re-expressed at 24 h (Fig. 3B). Levels of mGM-CSFR
did not change significantly following incubation of eosinophils with
IL-5 (Fig. 3C). Interestingly, there was also a 2-fold
decrease in mIL-5R at 24 h in the medium control group, which
was not observed for c or mGM-CSFR. After GM-CSF treatment, the
change in mIL-5R (Fig. 3D), c (Fig. 3E),
and mGM-CSFR (Fig. 3F) followed the same pattern as that
induced by IL-5 exposure. However, the degree of GM-CSF-induced
mIL-5R down-modulation was significantly less
(p < 0.01) than IL-5-induced mIL-5R
down-modulation at each time point.
|
mRNA expression in peripheral blood eosinophils following
IL-5 treatment
To determine whether the increased levels of sIL-5R seen in
culture supernatant fluids of IL-5-stimulated blood eosinophils were
due to greater production of mRNA, sIL-5R mRNA in purified
peripheral blood eosinophils was measured by RT-PCR, and the identity
of the PCR product was confirmed by Southern blot analysis using
sIL-5R-specific probes. Data were normalized to the housekeeping
gene, rRNA. Steady state levels of sIL-5R mRNA were significantly
decreased at 2 h after culture either in medium alone or IL-5
(Fig. 4A). There was no
ligand-specific effect of IL-5 on sIL-5R mRNA. In contrast to the
pattern of sIL-5R mRNA, concentrations of sIL-5R protein in
corresponding culture supernatant fluids increased over time (Fig. 4B), suggesting that its presence is not related to
increased steady state levels of mRNA.
|
mRNA decreased rapidly within the
first 2 h of culture either with medium or IL-5 (Fig. 4
C). Although mIL-5R mRNA levels returned to normal by
24 h of culture, protein expression remained nearly undetectable
(Fig. 4D).
Effects of MMP inhibitors on IL-5-mediated regulation of
mIL-5R and sIL-5R
To determine whether proteolytic enzymes contribute to the
modulation of IL-5R, purified blood eosinophils were preincubated
with various MMP and protease inhibitors for 2 h, then cultured
for an additional 3 h in medium alone or with IL-5. Eosinophil
expression of mIL-5R was determined by flow cytometric analysis, and
the release of sIL-5R into culture supernatant fluids was measured
by ELISA. Fig. 5 shows that the
MMP-specific inhibitors, BB-94 and GM6001, significantly reduced the
down-modulation of mIL-5R expression on IL-5-treated blood
eosinophils in a dose-dependent manner. Eosinophil viability was >95%
at inhibitor concentrations less than 100 µM. At higher
concentrations, significant cell toxicity was observed (<70%
viability at 200 µM BB-94 or GM6001). As expected, following exposure
of peripheral blood eosinophils to IL-5, the mIL-5R expression,
denoted as median channel fluorescence (MCF), was significantly
diminished compared with untreated cells (Fig. 6A). IL-5-mediated reduction
in mIL-5R expression was partially inhibited by pretreatment of
eosinophils with the hydroxamate compound BB-94 (Fig. 6A).
In contrast, IL-5-mediated reduction in c was not inhibited by
pretreatment of eosinophils with BB-94 (Fig. 6B); in fact,
further reduction was observed in the presence of IL-5 and BB-94. In
addition to inhibiting mIL-5R loss, BB-94 blocked the IL-5-mediated
increase in sIL-5R in cell culture supernatant fluids (Fig. 6C). To confirm the specific contribution of MMPs to the
IL-5-mediated reduction in mIL-5R expression, natural inhibitors of
MMPs, TIMPs 1–3 were added to the culture system. TIMP-3, but neither
TIMP-1 nor TIMP-2, significantly inhibited IL-5-mediated receptor
down-modulation (Table I). Receptor
expression was not affected by the metalloendopeptidase phosphoramidon,
MMP inhibitor-1, or the protease inhibitors -aminocaproate acid and
leupeptin (Table I). In contrast, proteasome inhibitors
N-Ac-Leu-Leu-norleucinal, MG132, and -lactone, and bafilomycin
A1 partially blocked IL-5-mediated loss of
mIL-5R (Table I). There was no additional effect when MMP and
proteasome inhibitors were both added to the cells (data not shown). In
contrast to mIL-5R, the loss of c was not prevented by
MMP-specific inhibitors (Table I). In fact, expression of c was
further down-regulated in the presence of IL-5 and either BB-94 or
GM6001. IL-5-mediated down-regulation of c was significantly
diminished by the proteasome inhibitors, N-Ac-Leu-Leu-norleucinal,
MG132, and -lactone. Interestingly, the nonspecific MMP inhibitor,
phenanthroline, also blocked c down-regulation. Addition of
pharmacologic agents in the absence of IL-5 had no effect on mIL-5R
or c expression (data not shown).
|
|
|
Eosinophil degranulation was used to assess the potential
functional significance of the reduction of mIL-5R on
IL-5-pretreated peripheral blood eosinophils. Freshly isolated
peripheral blood eosinophils were pretreated with IL-5 at
concentrations shown to decrease mIL-5R, then restimulated either
with 1 ng/ml IL-5 or GM-CSF. During the pretreatment phase, cells were
cultured in medium containing HBSS and FCS to prevent adherence and
degranulation. Under these conditions, total cellular EDN was
equivalent for each reaction condition (Fig. 7A). Cells were subsequently
restimulated in HBSS containing 0.03% gelatin to allow degranulation.
Pretreatment with medium alone or a low dose (0.01 ng/ml) of IL-5,
conditions under which mIL-5R expression is retained, resulted in
high levels of EDN release upon subsequent stimulation with IL-5. In
contrast, pretreatment with 10 ng/ml IL-5, conditions under which
mIL-5R is lost, resulted in an inability of these cells to be
further activated with IL-5 to release EDN (Fig. 7B). In
fact, after loss of mIL-5R, EDN release was not significantly
different between IL-5-stimulated eosinophils (136 ± 34 ng/ml) and
cells stimulated with medium alone (112 ± 47 ng/ml). Finally,
incubation with IL-5 at either concentration had no effect on
subsequent GM-CSF-activated EDN release. These data indicate that a
reduction in mIL-5R expression is associated with a decrease in the
ability of these cells to respond with the release of EDN to IL-5,
i.e., there was a demonstrable functional correlate for the observed
change in receptor expression.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
protein and a concomitant increase of sIL-5R in cell
culture supernatant fluid. The release of sIL-5R occurred in the
absence of increased steady state levels of sIL-5R mRNA in the
IL-5-stimulated blood eosinophils. The loss of mIL-5R protein from
the cell surface was associated with a selective loss in responsiveness
of the eosinophil to IL-5. These in vitro findings complement those in
the accompanying study, in which mIL-5R on the surface of BAL
eosinophils was decreased compared with circulating eosinophils, and
sIL-5R in BAL fluid increased 48 h after local airway Ag
challenge (40). Similar to the finding in our in vitro
model, the increased presence of sIL-5R in BAL fluid was not
associated with increased steady state levels of sIL-5R mRNA in BAL
eosinophils, and the loss of mIL-5R protein from the eosinophil
surface was associated with a lack of responsiveness to IL-5. The in
vitro data indicate that the source of sIL-5R in BAL fluid may be
eosinophils. Furthermore, we provide evidence that a MMP is required
for both the IL-5-induced reduction in mIL-5R and the release of
sIL-5R. Thus, it appears that the receptor can be shed by
proteolytic cleavage, a mechanism that has not, to date, been proposed
for the regulation of IL-5R expression.
Detection of sIL-5R in IL-5-stimulated human eosinophils is a unique
finding. Soluble receptors for cytokines may be produced by alternative
mRNA splicing (IL-7R, GM-CSFR), proteolytic cleavage (IL-1R, IL-2R,
TNFR), or both (IL-4R and IL-6R) (29). We present evidence
that proteolytic cleavage contributes, at least in part, to the
decrease in mIL-5R and the increase in sIL-5R. First, the
dramatic IL-5-mediated decrease in mIL-5R on the eosinophil surface
was inversely correlated with the concentration of sIL-5R in the
cell culture supernatant fluid. Second, there were no corresponding
ligand-specific changes in steady state levels of sIL-5R mRNA.
Third, the rapid kinetics for IL-5-mediated loss of mIL-5R and
increase in sIL-5R are consistent with a proteolytic event, rather
than transcriptional control of the IL-5R subunits (30).
Finally, both the down-modulation of mIL-5R from the cell surface
and the increased release of sIL-5R into the culture supernatant
fluid were partially inhibited by the MMP-specific inhibitor BB-94.
Because phenanthroline and the more specific MMP inhibitors BB-94 and
GM6001 prevented mIL-5R down-regulation, it is likely that a MMP is
responsible for the IL-5-induced loss of mIL-5R. TIMP-3, but neither
TIMP-1 nor TIMP-2, inhibited the action of the proteinase, suggesting
that the enzyme may be a membrane-associated disintegrin and
metalloproteinase (ADAM), rather than a soluble MMP. Receptors for
other cytokines, including IL-6 and TNF, also appear to be cleaved by
ADAMs, and the shedding of those receptors is inhibited by a similar
panel of inhibitors, as we have described in this study for IL-5R
(19, 31, 32, 33, 34, 35). Further study is required to identify the
specific MMP responsible for IL-5R cleavage.
It is of interest to note that a decreased expression of mIL-5R also
occurred when eosinophils were cultured in medium alone (Figs. 3, A and D, and 4). We speculate that this may be
due to normal membrane turnover combined with a reduced level of
transcription or to preapoptotic events in cells cultured for extended
periods in suboptimal conditions (i.e., in the absence of factors known
to enhance viability in vitro, such as IL-5 (36) and
GM-CSF (37)). In addition, because the MMP inhibitor BB-94
was able to reduce the level of sIL-5R in the culture medium to a
level significantly lower than that seen with medium alone (Fig. 6C), there may be a baseline level of MMP activity in the
cultures, even in the absence of IL-5.
Based on observation made using an IL-5-responsive erythroleukemia cell
line, Martinez-Moczygemba and Huston (38) proposed a model
for IL-5R receptor desensitization whereby, upon engagement of both
IL-5R subunits by IL-5, the signal-transducing cytoplasmic domain of
c is truncated via proteasomal degradation and the complex is
endocytosed. In our study, proteasome inhibitors prevented the loss of
both c and mIL-5R, indicating that proteasome activity and
endocytosis may account, at least in part, for the loss of surface
mIL-5R expression. Membrane-associated IL-5R may be endocytosed
along with c or released into the medium after MMP cleavage and
proteasomal degradation. Proteasome inhibitors appear to stabilize the
/c complex (38), and thus may inhibit shedding of
the -chain by strengthening its association with the extracellular
portion of c. Together, these findings suggest that both MMP and
proteasome activity are required for down-regulation of IL-5R on the
eosinophil cell surface.
Our findings that protein expression of c is transiently decreased
at 30 min, but fully re-expressed at 24 h (Fig. 3B) are
consistent with the proteasome model (38), which predicts
transient reduction in surface c expression upon IL-5 exposure due
to proteasomal degradation and endocytosis. Down-regulation of c is
enhanced by the MMP inhibitor BB-94 (Fig. 6B), but inhibited
by proteasome inhibitors (Table I), suggesting that proteasome, but not
MMP, activity is required for c down-modulation, and that MMPs may,
in fact, inhibit c endocytosis. Clearly, expression of the and
c subunits of the IL-5R is regulated by different mechanisms.
The differential regulation of the and c subunits of IL-5R
may contribute to selective desensitization of IL-5R compared with
other members of the IL-5R family. Indeed, we have shown that
eosinophils pre-exposed to 10 ng/ml IL-5 became unresponsive to
subsequent degranulation by IL-5, but remained fully responsive to
GM-CSF (Fig. 7B). This is, at least in part, contrary to the
hypothesis of heterotypic desensitization of the IL-5R family that was
proposed by Martinez-Moczygemba and Huston (38). Using an
erythroleukemia cell line, they demonstrated that pre-exposure to IL-5
for 1 h resulted in a lack of signal transduction events (tyrosine
phosphorylation of c and activation of Janus kinase 2 and STAT5) in
response to subsequent stimulation with IL-3 or GM-CSF. We have not
determined whether similar heterotypic desensitization occurs in human
eosinophils at this early time point; however, it seems likely because
c is significantly down-modulated within 30 min of IL-5 exposure. If
heterotypic desensitization does occur, it is likely to be transient,
as c is re-expressed within 4 h of IL-5 exposure. Because
IL-5R is not re-expressed, it is not surprising that the
responsiveness of cells to GM-CSF returns, but responsiveness to IL-5
does not. In this way, IL-5-activated eosinophils may become
selectively unresponsive to IL-5, thus inhibiting IL-5-dependent
inflammatory processes and, perhaps, promoting IL-5-independent
processes.
In vitro incubation of peripheral blood eosinophils with IL-5 is
clearly a simplified model for IL-5-mediated eosinophil responses in
the airway. Although IL-5-activated eosinophils and BAL eosinophils
share many characteristics (1, 21, 22, 23, 24), eosinophils
recruited to the lung after Ag challenge have been exposed to a variety
of ligands (e.g., cytokines, chemokines, matrix proteins, lipid
mediators, complement, and Igs) during the process of transmigration
and activation (1, 39). Therefore, it is unlikely that the
phenotypic changes induced as a result of this process could be exactly
mimicked by exposure to a single cytokine. Nevertheless, our findings
of reduced mIL-5R and increased sIL-5R without corresponding
changes in mRNA levels, along with reduced responsiveness to IL-5, but
not GM-CSF in blood eosinophils exposed to IL-5, are strikingly similar
to those found in BAL cells after Ag challenge (40),
suggesting that IL-5 and MMPs may play a similar role in vivo. Our
findings with blood eosinophils do, however, differ from our BAL
findings (40) in several respects: 1) blood eosinophils
showed only a transient reduction in c expression after IL-5
exposure, whereas BAL eosinophils had significantly reduced c
48 h post-Ag challenge; 2) blood eosinophils displayed no change
in GM-CSFR expression in response to IL-5, whereas BAL eosinophils
had increased GM-CSFR expression after Ag challenge; and 3) blood
eosinophils treated with IL-5 under conditions that markedly reduced
expression of mIL-5R remained fully responsive to GM-CSF for EDN
release, whereas BAL eosinophils had only a moderate response to
GM-CSF. These differences indicate that the primary distinction between
the in vivo and in vitro models lies within the GM-CSFR, and suggest
that this receptor may be affected by additional factors to which the
BAL eosinophils may have been exposed in vivo.
In conclusion, exposure of peripheral blood eosinophils to IL-5 results
in a rapid, sustained loss of mIL-5R that is, at least in part,
dependent on MMP activity. This receptor modulation renders eosinophils
less responsive to IL-5. In addition, proteolytic cleavage of the
receptor results in the release of sIL-5R, which may have
immunoregulatory effects in vivo. In the accompanying study, we
demonstrated similar findings in BAL fluid following Ag challenge, and
speculated that a comparable sequence of events occurs in vivo in
eosinophil-associated diseases such as asthma. The loss of mIL-5R on
the eosinophil surface and the release of sIL-5R may be important
mechanisms that limit IL-5-mediated eosinophil function.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Elizabeth A. (Becky) Kelly, Section of Pulmonary and Critical Care Medicine, 600 Highland Avenue, CSC K4/928, University of Wisconsin School of Medicine, Madison, WI 53792-9988. E-mail address: eak{at}medicine.wisc.edu
3 Abbreviations used in this paper: c, -chain; ADAM, a disintegrin and metalloproteinase; BAL, bronchoalveolar lavage; EDN, eosinophil-derived neurotoxin; m, membrane form; MCF, median channel fluorescence; MMP, matrix metalloproteinase; s, soluble form; TIMP, tissue inhibitor of matrix metalloproteinase.
Received for publication July 2, 2002. Accepted for publication October 1, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
subunit. Proc. Natl. Acad. Sci. USA 89:7041. chain and a chain shared with the receptor for GM-CSF. Cell 66:1175.[Medline]
47 integrins by monoclonal antibodies or ligands enhances survival of human eosinophils in vitro. J. Immunol. 163:6217. receptors on the surface of human colon carcinoma cells. Cytokine 9:770.[Medline]
c signaling reveals a novel mechanism for cytokine receptor heterotypic desensitization. J. Clin. Invest. 108:1797.[Medline]
expression on human eosinophils. I. Loss of mIL-5 receptor on airway eosinophils and increased soluble IL-5 receptor in the airway after allergen challenge. J. Immunol. 169:6452.This article has been cited by other articles:
|
|
 |
|
  A. A. Khan, S. Martin, and B. Saha SEB-induced signaling in macrophages leads to biphasic TNF-{alpha} J. Leukoc. Biol., June 1, 2008; 83(6): 1363 - 1369. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. W. Johansson, E. A. B. Kelly, W. W. Busse, N. N. Jarjour, and D. F. Mosher Up-Regulation and Activation of Eosinophil Integrins in Blood and Airway after Segmental Lung Antigen Challenge J. Immunol., June 1, 2008; 180(11): 7622 - 7635. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Cheung, D. Marchant, E. K.-Y. Walker, Z. Luo, J. Zhang, B. Yanagawa, M. Rahmani, J. Cox, C. Overall, R. M. Senior, et al. Ablation of Matrix Metalloproteinase-9 Increases Severity of Viral Myocarditis in Mice Circulation, March 25, 2008; 117(12): 1574 - 1582. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Pazdrak, T. W. Young, S. Stafford, B. Olszewska-Pazdrak, C. Straub, V. Starosta, A. Brasier, and A. Kurosky Cross-Talk between ICAM-1 and GM-CSF Receptor Signaling Modulates Eosinophil Survival and Activation J. Immunol., March 15, 2008; 180(6): 4182 - 4190. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Flood-Page, C. Swenson, I. Faiferman, J. Matthews, M. Williams, L. Brannick, D. Robinson, S. Wenzel, W. Busse, T. T. Hansel, et al. A Study to Evaluate Safety and Efficacy of Mepolizumab in Patients with Moderate Persistent Asthma Am. J. Respir. Crit. Care Med., December 1, 2007; 176(11): 1062 - 1071. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Kang, D. H. Lee, H. Seo, J. S. Park, K. H. Nam, S. Y. Shin, C.-S. Park, and I. Y. Chung Regulation of Functional Phenotypes of Cord Blood Derived Eosinophils by {gamma}-Secretase Inhibitor Am. J. Respir. Cell Mol. Biol., November 1, 2007; 37(5): 571 - 577. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Barthel, N. N. Jarjour, D. F. Mosher, and M. W. Johansson Dissection of the Hyperadhesive Phenotype of Airway Eosinophils in Asthma Am. J. Respir. Cell Mol. Biol., September 1, 2006; 35(3): 378 - 386. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Lopez, V. del Pozo, T. Miguel, B. Sastre, C. Seoane, E. Civantos, E. Llanes, M. L. Baeza, P. Palomino, B. Cardaba, et al. Inhibition of Chronic Airway Inflammation and Remodeling by Galectin-3 Gene Therapy in a Murine Model J. Immunol., February 1, 2006; 176(3): 1943 - 1950. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Inoue, R. Kato, S. Fukuyama, A. Nonami, K. Taniguchi, K. Matsumoto, T. Nakano, M. Tsuda, M. Matsumura, M. Kubo, et al. Spred-1 negatively regulates allergen-induced airway eosinophilia and hyperresponsiveness J. Exp. Med., January 3, 2005; 201(1): 73 - 82. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J.C. de Haas, K. E. Veldkamp, A. Peschel, F. Weerkamp, W. J.B. Van Wamel, E. C.J.M. Heezius, M. J.J.G. Poppelier, K. P.M. Van Kessel, and J. A.G. van Strijp Chemotaxis Inhibitory Protein of Staphylococcus aureus, a Bacterial Antiinflammatory Agent J. Exp. Med., March 1, 2004; 199(5): 687 - 695. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. McMillan, J. Kearley, J. D. Campbell, X.-W. Zhu, K. Y. Larbi, J. M. Shipley, R. M. Senior, S. Nourshargh, and C. M. Lloyd Matrix Metalloproteinase-9 Deficiency Results in Enhanced Allergen-Induced Airway Inflammation J. Immunol., February 15, 2004; 172(4): 2586 - 2594. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Gregory, A. Kirchem, S. Phipps, P. Gevaert, C. Pridgeon, S. M. Rankin, and D. S. Robinson Differential Regulation of Human Eosinophil IL-3, IL-5, and GM-CSF Receptor {alpha}-Chain Expression by Cytokines: IL-3, IL-5, and GM-CSF Down-Regulate IL-5 Receptor {alpha} Expression with Loss of IL-5 Responsiveness, but Up-Regulate IL-3 Receptor {alpha} Expression J. Immunol., June 1, 2003; 170(11): 5359 - 5366. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Y. Liu, J. B. Sedgwick, M. E. Bates, R. F. Vrtis, J. E. Gern, H. Kita, N. N. Jarjour, W. W. Busse, and E. A. B. Kelly Decreased Expression of Membrane IL-5 Receptor {alpha} on Human Eosinophils: I. Loss of Membrane IL-5 Receptor {alpha} on Airway Eosinophils and Increased Soluble IL-5 Receptor {alpha} in the Airway After Allergen Challenge J. Immunol., December 1, 2002; 169(11): 6452 - 6458. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) cells was analyzed by flow cytometry, and
data are depicted as percentage of positive eosinophils (mean ± SEM).
*, p < 0.05 compared with percentage of
mIL-5R
, Reflect
medium only; •, represent treatment with the respective ligands (IL-5
or GM-CSF). Data are expressed as mean ± SEM. *,
p < 0.05, treatment vs 0-h baseline; 
