| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
*
The Lung Pharmacology Group, Department of Respiratory Medicine and Allergology, Institute of Heart and Lung Diseases, Göteborg University, Gothenburg, Sweden;
Department of Clinical Immunology, Karolinska Hospital, The Karolinska Institute, Stockholm, Sweden; and Departments of
Pathology and Molecular Medicine and
§
Medicine, McMaster University, Hamilton, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain. We conclude that the bone
marrow is activated by airway allergen exposure, and that newly
produced eosinophils contribute to a substantial degree to the airway
eosinophilia induced by allergen. Airway allergen exposure increases
the number of cells expressing IL-5-protein in the bone marrow. The
bone marrow, as well as the lung, are possible targets for
anti-IL-5-treatment. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recently, some studies have shown that the bone marrow is stimulated in asthmatic patients (13, 14, 15). Studies in dogs (16, 17, 18), mice (4, 19, 20, 21), and humans (15, 22, 23, 24) suggest that increased granulocytopoiesis is induced during exacerbation of asthma-like processes, such as inflammation induced by allergen exposure. However, the degree of the contribution of the bone marrow to the airway inflammatory process, and the mechanisms by which the bone marrow is stimulated, are still poorly understood. IL-5 has been shown to be crucial for the induction of airway eosinophilia, because mice lacking IL-5 (IL-5 knockout mice) fail to develop airway eosinophilia in response to allergen exposure unless the animals are reconstituted with IL-5 (25, 26). Furthermore, treatment of wild-type mice with a neutralizing anti-IL-5 Ab strongly reduces the eosinophilic inflammatory response induced by allergen (19, 27, 28, 29, 30, 31, 32).
The aim of this study was to determine the kinetics and
contribution of eosinophilopoiesis to the airway eosinophilia induced
by airway allergen exposure and to determine the role and site of
action of IL-5 in this response. To do this, we chose to use mice
sensitized and airway exposed to OVA. A thymidine analog,
5-bromo-2'-deoxyuridine (BrdU),3 which is incorporated
into DNA during the S phase of the cell cycle, was used to
pulse-label newly produced inflammatory cells (33) at
different time points during repeated allergen exposure. Intracellular
staining of IL-5 was performed in bone marrow cells and quantified by
flow cytometry. The capacity of bone marrow cells to release IL-5
protein was quantified in vitro after magnetic separation. The role and
site of action of IL-5 in the allergen-induced eosinophilia were
investigated by pretreatment with a neutralizing anti-IL-5 Ab,
given either by airway or systemic route. To further elucidate the
capacity of IL-5 to have effects in different tissues,
immunocytochemical staining of the IL-5R -chain in bone marrow and
airways was performed.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This study was approved by the Animal Ethics Committee in Gothenburg, Sweden, as well as by the Animal Research Ethics Board of McMaster University, Hamilton, Canada. Male BALB/c mice, 5–6 wk old, were purchased from B&K Universal AB (Sollentuna, Sweden) and Charles River Laboratories (Ottawa, Ontario, Canada). The mice were maintained under conventional animal housing conditions and provided with food and water ad libitum.
Sensitization and allergen exposure protocol
All mice were immunized on two different days (5 days apart) by i.p. injections of 0.5 ml alum-precipitated Ag containing 8 µg of OVA (Sigma-Aldrich Sweden AB, Tyresö, Sweden) bound to 4 mg of aluminum hydroxide (Sigma-Aldrich) in PBS. Eight days after the second sensitization, the animals were rapidly and briefly anesthetized using CO2 gas, and received intranasal (i.n.) administration with 100 µg of OVA in 25 µl of PBS, or PBS alone. These OVA or PBS exposures were performed on 2, 3, or 5 consecutive days (from days 1 to 2, 3, or 5). One day after the last of the OVA exposures, samples (bronchoalveolar lavage (BAL) fluid, bone marrow cells, and blood) were collected.
In a preliminary experiment, we examined the efficiency of i.n. administered solutions to reach the lungs by giving Evans blue dye (n = 3). Three min after i.n. administration of 25 µg of Evans blue dye in 25 µl of PBS, we isolated trachea, bronchi, lungs, esophagus, stomach, and nose tissue. The content of dye was quantified in each organ by spectrophotometry, after extraction of dye in formamide. The relative amount of dye recovered in each organs in relation to the total amount of given dye were; 30.7 ± 9.7% in trachea, bronchi, and lungs; 12.7 ± 10.2% in esophagus and stomach; and 25.7 ± 4.6% in nose.
In vivo protocol for evaluation of eosinophilopoiesis
To elucidate the kinetics of cellular inflammation, we labeled newly produced inflammatory cells using BrdU. Animals were given BrdU (Boehringer Mannheim Scandinavia, Bromma, Sweden) at a dose of 1 mg in 0.25 ml of PBS i.p. twice (8 h apart, total dose 2 mg/animal) on different occasions during the repeated allergen exposure period (day 1, 3, or 5). On each day, the first injection of BrdU was performed on awake mice 1 h before the OVA exposure, and the second injection was given 7 h after the exposure. The control animals received PBS i.p. at the same time points. Three different types of treatments were given.
First, to evaluate whether BrdU would influence the response to repeated i.n. OVA exposures, we exposed two groups of animals to OVA on five consecutive days (days 1 to 5), and one of these groups was injected with BrdU on two occasions, 8 h apart, on day 1. The other group was given PBS at corresponding time points.
Second, to determine the time course of airway inflammation, bone marrow eosinophilia, and BrdU staining after repeated allergen exposure, these responses were evaluated in four groups of animals, exposed to either OVA or PBS on two, three, or five consecutive days.
Third, to more closely determine the kinetics of the accumulation of new cells in the airways, six groups of animals were used for the examination of the effect of the timing of BrdU injection on BrdU staining in BAL and bone marrow cells. Three of these were given OVA i.n. on five consecutive days, and three were given PBS. BrdU was given i.p. to one OVA and one PBS group either on day 1, 3, or 5. Again, two injections of BrdU were given 8 h apart on each day. The groups given BrdU on day 1 are the same as in the second treatment procedure (see above). For each study group, cells were harvested 24 h after the last allergen exposure.
In vivo protocol for in vitro experiments (FACS and CD3+ cell enrichment)
Bone marrow cells were collected a day after the third of three OVA exposures on consecutive days. These cells were washed, and RBC were eliminated by centrifugation over a 65% Percoll gradient at 1500 rpm in room temperature for 30 min. Cells were washed two to three times in PBS, and further processed for flow cytometry or magnetic separation and subsequent cytokine-release experiments (see below). Bone marrow cells from three to eight animals were pooled for further processing and analysis. We chose to collect the cells after three allergen exposures because it was clear that the bone marrow had reached an activation state at this time, evident as increased numbers of eosin staining cells.
In vivo protocol for anti-IL-5 treatment
The effects of a neutralizing anti-IL-5 Ab treatment on the bone marrow and airway inflammatory responses induced by repeated allergen exposure were investigated in three series of experiments. Animals were divided into twelve groups, which were all exposed to OVA on five consecutive days. One hour before the OVA exposure, animals were given anti-mouse IL-5 mAb (clone TRFK-5; R&D Systems Europe, Abingdon, U.K.) or its isotype control, rat IgG1 (clone R3-34; PharMingen, San Diego, CA), by different routes (either by i.n. instillation, or by i.p. injection) and at different doses. First, six groups were pretreated i.p. with either TRFK-5 (1–100 µg/animal in 0.5 ml of PBS) or rat IgG1 (100 µg/animal in 0.5 ml of PBS). Second, four additional groups were repeatedly (total of five times) pretreated i.n. with either TRFK-5 (6–50 µg/animal in 25 µl of PBS) or rat IgG1 (20 µg/animal in 25 µl of PBS) 1 h before each OVA exposure on five consecutive days. Third, two separate groups were pretreated i.p. with either TRFK-5 (100 µg/animal) or rat IgG1 (100 µg/animal) and were also injected with BrdU (1 mg/animal x 2, 8 h apart, i.p.) on the same day (day 1).
Cell collection and sample processing
All samples were collected 24 h after the last OVA or PBS exposure. The animals were anesthetized with a mixture of xylazine (130 mg/kg) and ketamine (670 mg/kg) i.p. When the animals were in adequately deep anesthesia, the chest was opened and the animals were bled by penetration of the heart. After tracheostomy, BAL was performed by instilling 0.25 ml of PBS through the tracheal cannula, followed by gentle aspiration. This was repeated with 0.2 ml of PBS, and fluid from both lavages was pooled and kept on ice until further processing. Approximately 0.4 ml of the instilled fluid was consistently recovered. Then, one femur was removed, the attached muscle was gently scraped off and both ends of the femur were cut open. The bone marrow cells were removed by perfusion of the femur with 1.5 ml of PBS. The bone marrow cell suspension was also kept on ice until further processing. The total cell numbers in BAL and bone marrow samples were determined using standard hematologic procedures. Cytospins of BAL and bone marrow samples were prepared and stained with May-Grünwald-Giemsa stain for differential cell counts. Cell differential was determined by counting 300–500 cells using a light microscope (Zeiss Axioplan 2; Carl Zeiss, Jena, Germany). The cells were identified using standard morphological criteria, and mature vs immature eosinophils were determined by nuclear morphology, May-Grünwald-Giemsa staining properties, and cytoplasmic granulation, as previously described by Lee et al. (34). The cytospin preparations for immunocytochemistry were fixed with 4% paraformaldehyde in PBS for 20 min or with periodate-lysin-paraformaldehyde (PLP) for 10 min, and then washed in 15% sucrose in PBS for 10 min. After these procedures, the preparations were air-dried overnight, and stored at -80°C until further examination. BAL supernatant and serum samples were also stored at -80°C.
Immunocytochemical detection of BrdU-labeled cells
BrdU incorporated into cellular DNA in cytospin preparations of BAL and bone marrow was detected with immunocytochemistry, using a mouse mAb against BrdU. The paraformaldehyde-fixed cytospin preparations were washed with TBS and subjected to digestion in 0.1% trypsin and 0.1% CaCl2 in PBS at 37°C for 15 min. The slides were incubated in 4 M HCl for 15 min to denature the DNA, followed by neutralization and blocking of nonspecific binding sites with incubation buffer (0.5% BSA and 0.1% Tween 20 in PBS). The slides were then incubated with 1 U/ml of alkaline phosphatase-conjugated mouse anti-BrdU mAb (clone BMG6H8; Boehringer Mannheim) in incubation buffer at 37°C for 30 min. After washing, the alkaline phosphatase reaction was developed for 15 min, using Fast Red Substrate System (No. 1496549; Boehringer Mannheim). The slides were counterstained with Mayer’s hematoxylin for 1 min. All slides were evaluated with the light microscope (Zeiss Axioplan 2) in random fields of view by a "blinded" experimenter. Cells with any nuclear red staining were counted as BrdU-labeled cells. Three hundred cells in both BAL and bone marrow cytospin preparations were evaluated.
Immunocytochemical detection of bone marrow CD3+ cells
CD3+ bone marrow cells were identified
using a mAb against the specific cell surface marker CD3
for mouse T
lymphocytes. The PLP-fixed cytospin preparations were washed with TBS
and treated with 1:20 dilution of normal rabbit serum (No. X902; Dako,
Glostrup, Denmark) in incubation buffer (1% BSA and 0.1% saponin in
TBS) at room temperature for 15 min. The slides were incubated with 1.7
µg/ml of rat anti-CD3 Ab (clone CD3-12; Cedarlane
Laboratories, Hornby, Ontario, Canada) or isotype control, rat IgG1
(clone R3-34; PharMingen) in incubation buffer at 4°C overnight.
After washing with TBS-0.1% saponin, incubation with 1:100 dilution of
rabbit anti-rat Ig (No. Z494; Dako) and 2% normal mouse serum (No.
X910; Dako) was followed by a treatment with 1:50 dilution of rat
monoclonal APAAP complex (No. D488; Dako) at room temperature for each
10 min, and these were repeated twice. After washing with TBS, the
alkaline phosphatase reaction was developed for 15 min, using New
Fuchsin Substrate System (No. K698; Dako). The slides were
counterstained with Mayer’s hematoxylin for 1 min. All slides were
blindly evaluated on the light microscope (Zeiss Axioplan 2) in random
fields of view. Five hundred cells were counted.
FACS analysis of bone marrow cells
For the double-staining procedure (CD3 surface/IL-5
intracellular), the following mAbs and corresponding isotype-matched
control Abs were used: PE-conjugated anti-CD3 (clone 145-2C11)
and FITC-conjugated anti-IL-5 (clone TRFK-5) were both purchased
from PharMingen Canada (Mississauga, Ontario, Canada). Both Abs were
used in saturation concentrations. Isotype-matched control Abs were run
in parallel to determine unspecific binding, and this was used to
define the cut-off for positively labeled cells, which was 99% of the
cells labeled with respective control Ab.
Surface staining was performed as follows: To block unspecific binding,
cell suspensions were initially incubated at 4°C for 20 min with PBS
supplemented with 5% mouse/rat sera, and thereafter incubated at 4°C
for 30 min with PE-anti-CD3 or isotype control Ab, and then
washed twice. Subsequent intracellular staining was performed as
described previously, using a cell membrane permeabilization technique,
the FOG (paraformaldehyde-fixation with subsequent
n-octyl-ß-D-glucopyranoside
treatment)-method; Ref. 35). Surface-immunostained cells
were fixed in 4% paraformaldehyde at 20°C for 5 min, followed by one
wash in PBS and subsequent incubation with the detergent
n-octyl-ß-D-glucopyranoside (0.74%;
Sigma Chemical Company, St. Louis, MO) at 20°C for another 6 min,
followed by one wash. After incubation with PBS supplemented with 5%
mouse/rat sera (to block intracellular unspecific staining),
permeabilized cells were incubated with either FITC-anti-IL-5 or
corresponding isotype-matched control Ab at 4°C for 30 min, followed
by two washes. To confirm that accurate permeabilization was obtained,
intracellular binding of anti-vimentin Abs was measured. Over 90%
of cell populations subjected to permeabilization stained positive for
intracellular vimentin as previously described (35). The
cells were finally resuspended in PBS supplemented with 1%
paraformaldehyde and analyzed using a FACScan flow cytometer equipped
with an argon ion laser (Becton Dickinson Instrument Systems,
Mississauga, Ontario, Canada). The instrument was calibrated with
Standard Brite (Flow Cytometry Standards, San Juan, Puerto Rico) to
ensure the same fluorescence level in each experiment. After
compensation settings were established, a quadrant was set based on the
isotype controls, and the percentage of
CD3+/IL-5+ cells were
calculated. Off-line analysis was performed using the PC lysis software
supplied by Becton Dickinson.
In vitro culture of CD3+ cells and measurements of IL-5 release
Bone marrow from OVA-exposed mice (on three consecutive days)
was used, and cells from three to eight animals were pooled.
CD3+ cells were enriched from the bone marrow
using a magnetic cell sorting system (MACS; Miltenyi Biotec, Germany).
Bone marrow cells, depleted of RBC and granulocytes in a single cell
suspension in PBS (with 2 mM EDTA and 0.5% BSA), were labeled with a
biotinylated mAb directed to CD3+ lymphocytes
(CD3 145-2C11; PharMingen) in a concentration of 0.5 µg/ml. After
washing the cells, 10 µl of streptavidin magnetic microbeads
(MACS)/107 cells were added, according to the
manufacturer’s recommendation. The magnetic labeled
CD3+ cells were enriched on a positive selection
column over a magnetic field. The enrichment procedure resulted in
90% CD3+ cells, according to
immunocytochemical staining.
The enriched CD3+ cell fraction and also the cell fraction depleted of CD3+ cells (CD3-negative) were cultured in RPMI 1640 medium (Life Technologies, Täby, Sweden) complemented with 10% FCS, 1% penicillin-streptomycin, 1% sodium pyruvate, and 2 mM L-glutamine (all obtained from Sigma-Aldrich) in a concentration of 0.25 x 106 cells/200 µl of medium in a 96-well plate. Two different types of experiments were performed. First, the CD3-positive and -negative fractions were added directly to the wells together with calcium ionophore (end concentraction 1 µM) and PMA (end concentration 2 ng/ml), and incubated for 24 h in a humidified incubator at 37°C with 5% CO2. The culture media were taken off and stored at -80°C until analysis of IL-5 content. Second, unseparated bone marrow cells were added to the wells in a concentration of 2 x 106 cells/ml, and adherent cells were allowed to attach to the plastic for 2 h. The wells were washed several times with medium to remove unattached cells, and the CD3-positive and -negative fractions were added and incubated as above. The culture media were then collected. IL-5 in the culture media, BAL fluid, and serum were measured using a commercial available mouse IL-5 ELISA, using the manufacturer’s instructions (Endogen, Woburn, MA). The detection limit for IL-5 was 5 pg/ml.
Quantification of rat IgG1 in TRFK-5-treated mice
Levels of rat IgG1 in serum and BAL fluid from TRFK-5-treated animals were measured by the ELISA method. ELISA plates (Maxisorp F96; Nunc, Naperville, IL) were incubated with 50 µl of mouse anti-rat IgG1 mAb (clone B46–2; PharMingen) at a concentration of 2 µg/ml in binding solution (0.1 M Na2HPO4, adjusted to pH 9.0 with 0.1 M NaH2PO4) at 4°C for 18 h. After discarding the coating solution, the remaining binding sites on the plate were blocked with 10% FBS in TBS at room temperature for 2 h. The plates were then washed with TBS containing 0.05% Tween 20, and 100 µl of mouse serum or BAL supernatant, undiluted or diluted between 1:2 and 1:32 with ELISA buffer (10% FBS and 0.05% Tween 20 in TBS) were added to each well. The plates were incubated at 4°C for 18 h. After washing, 100 µl of biotin-conjugated mouse anti-rat IgG1 mAb (clone RG11/39.4; PharMingen) at a concentration of 1 µg/ml in ELISA buffer was added and incubated at room temperature for 1 h. After further washing, the plates were incubated with 1:2000 dilution of alkaline phosphatase-conjugated streptavidin (No. D396; Dako). The plate-bound alkaline phosphatase activity was detected with p-nitrophenyl phosphate substrate (Sigma 104; Sigma-Aldrich) and assessed with an ELISA reader (type 349; Labsystems, Stockholm, Sweden) at 405 nm. A standard curve was obtained with serial dilutions of purified rat IgG1 (clone R3–34; PharMingen), and results were expressed as nanograms of purified rat IgG1 per milliliter of serum or BAL supernatant. The detection limit for rat IgG1 was 8 ng/ml.
IL-5R -chain immunocytochemistry
For specific identification of the IL-5R -chain, a rabbit
anti-mouse polyclonal IL-5R -chain Ab (No. RTIL5RACabr; Research
Diagnostics, Flanders, NJ) was used. PLP-fixed BAL and bone marrow
cytospin preparations were washed in TBS and treated with 1:20 dilution
of normal swine serum (No. X901; Dako) in TBS supplemented with 1% BSA
(TBS-BSA) at room temperature for 15 min to avoid unspecific binding.
The slides were incubated with 0.1 µg/ml of anti-IL-5R -chain
Ab in TBS-BSA at room temperature for 1 h. After rinsing, slides
were incubated with a secondary Ab, biotinylated
F(ab')2 of swine anti-rabbit Igs (No. E431;
Dako), at a concentration of 2.1 µg/ml in TBS-BSA with 10% normal
mouse serum (No. X910; Dako) at room temperature for 30 min. After
further washing, slides were treated with 1:100 dilution of
streptavidin-ß-galactosidase conjugate (No. 1112481; Boehringer
Mannheim) at room temperature for 30 min. The ß-galactosidase
activity was visualized by a ß-Gal Staining set (No. 1828673;
Boehringer Mannheim) at 37°C for 40 min. The slides were
counterstained with Mayer’s hematoxylin for 1 min. The counterstaining
was kept fairly weak to avoid interaction with the specific staining.
The number of IL-5R -chain-positive cells was assessed by counting
of 300 cells on the light microscope (Zeiss Axioplan 2) in random
fields of view. To test specificity of the immunostaining, a control
peptide for competition (No. RTIL5RA-CP; Research Diagnostics) was
used. To the anti-IL-5R Ab, the peptide Ag were added at a 30 x
weight excess, and then preincubated at room temperature for 2 h.
The Ab-peptide solution was then added at the primary step and further
processed as described above. This peptide totally blocked the positive
staining, arguing for the specificity of the immunostaining of IL-5R
-subunit.
Statistical analysis
All data are expressed as mean ± SEM. Statistical analysis was conducted using nonparametric ANOVA (Kruskal-Wallis test) to evaluate variance among more than two groups. If a significant variance was found, an unpaired two-group test was used to determine significant differences between individual groups. Between two groups of animals, only an unpaired two-group analysis was performed, without a preceding Kruskal-Wallis test. p < 0.05 was considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Intranasal administration of allergen (OVA 100 µg per day) to
sensitized mice on two, three, or five consecutive days induced
increases in mainly eosinophils, and to a lesser degree neutrophils in
BAL, as compared with sensitized mice exposed to PBS (Table I
). The number of BAL eosinophils was
sequentially increased after two, three, and five consecutive days of
OVA exposure. Neutrophils were significantly increased in BAL on days 3
and 4 only.
|
Administration of BrdU i.p. to sensitized mice on day 1 of
allergen exposure did not affect the numbers or the profile of
inflammatory cells in either bone marrow or BAL after repeated OVA
exposure on five consecutive days (Table II).
|
). The number of unlabeled
granulocytes was not significantly different among these three
OVA-exposed groups, but increased in comparison with the PBS-exposed
group. The relative number of eosinophils in relation to the total
number of granulocytes was 44.4, 87.1, and 97.2% after two, three, and
five consecutive days of OVA exposure, respectively, showing that a
dominant portion of the granulocytes are eosinophils. Photomicrographs
of BrdU-stained BAL cells from one OVA- and one PBS-exposed mouse are
shown in Fig. 2.
|
|
The number of bone marrow cells staining positive for eosin was
markedly increased after i.n. administration of OVA (100 µg per day)
on three (10.9 ± 1.8% of total bone marrow cells,
p < 0.05 compared with PBS-exposed animals) and five
consecutive days (11.8 ± 1.9%, p < 0.01) in the
sensitized mice (Fig. 3). The number of
eosin-staining cells with immature morphology was more clearly
increased.
|
The number of BrdU-staining cells was significantly increased in
the BAL in animals injected with BrdU on day 1 or 3, but not when BrdU
was injected on day 5 (Fig. 4). In a
separate group of sensitized mice, BrdU was injected on both days 1 and
3 of OVA exposure, and the relative number of granulocytes stained for
BrdU in BAL was 78.0 ± 4.1%.
|
30–60% of the total cells incorporated BrdU
into their nuclei after BrdU injection on various timings during
repeated allergen exposure (Table III).
No significant difference in total bone marrow BrdU staining was
induced by OVA exposure as compared with PBS exposure.
|
For measurements of IL-5, and for localization of IL-5-producing
cells isolated from the bone marrow, we chose to examine samples
24 h after three exposures on consecutive days, because a maximal
bone marrow eosinophilia was observed at this time point (Fig. 3
). The
concentration of IL-5 was increased in both serum and BAL fluid after
repeated allergen exposure (OVA 100 µg on three consecutive days) in
sensitized mice (233 ± 87 and 152 ± 26 pg/ml in serum and
in BAL fluid, respectively) but undetectable in both compartments in
PBS-exposed mice.
CD3+ cells were detected in the bone marrow by
immunocytochemistry, using a mAb directed against CD3 (Fig. 5). The relative number of
CD3+ cells in the bone marrow was not
significantly increased in OVA- vs PBS-exposed mice (0.69 ± 0.16
and 0.49 ± 0.10%, respectively; p = 0.40).
|
A) revealed that OVA-exposed
mice had significantly more bone marrow cells expressing both
intracellular IL-5 staining and CD3 cell surface staining (Fig. 6B). However, non-CD3-cells in the same gate also showed
significantly increased intracellular staining for IL-5. Furthermore,
when the bone marrow cells were stimulated for 6 h with PMA and
calcium ionophore, a further increase in intracellular IL-5 staining
was observed, mainly in the OVA-exposed group, in both cell
compartments.
|
Effect of anti-IL-5 Ab on bone marrow and airway eosinophilia
The levels of rat IgG1 in serum and in BAL after i.p. or i.n.
administration of TRFK-5 are shown in Fig. 7. Intraperitoneal administration of
anti-IL-5 (TRFK-5; 1–100 µg/animal i.p. given on day 1)
dose-dependently attenuated the increase of eosinophils in both BAL and
bone marrow induced by repeated i.n. administration of allergen (OVA
100 µg on five consecutive days; Figs. 8A and 9A). The
attenuating effect of systemically administered anti-IL-5 was
associated with strongly attenuated BrdU-staining of BAL granulocytes
(2.8 ± 1.4 vs 41.0 ± 15.8 x 104
cells/ml BAL fluid in TRFK-5-treated and rat IgG1-treated mice,
respectively; p < 0.05).
|
|
B). This effect was associated with
some degree of inhibition of the bone marrow eosinophilia induced by
allergen (Fig. 9B), but only
at doses causing concentration of Ab in serum exceeding 50 ng/ml
(Fig. 7).
|
B), but the
effect of the i.p. treatment on BAL eosinophilia is much more prominent
(Fig. 8, A and B), arguing that the systemic
effect of anti-IL-5 can influence local airway eosinophilia.
However, 10 µg of i.p. TRFK-5 given once causes similar concentration
of serum rat IgG1 to 50 µg of i.n.TRFK-5 given five times, but the
effect of the i.n. treatment on BAL eosinophilia is slightly more
prominent, arguing that there is also an effect of anti-IL-5
locally in the airways. Staining of IL-5R in bone marrow and airways
Immunocytochemical staining of the murine IL-5R -chain
(CD
125) was prominent in cytospins of bone marrow cells (Fig. 10A) from sensitized and
OVA-exposed mice (OVA x 5), whereas no such staining could be
found in BAL eosinophils from the same mice (Fig. 10B),
despite prominent BAL eosinophilia. The relative number of cells in the
bone marrow staining positively for the IL-5R were 37.2 ± 11.4%
in nonsensitized nonexposed mice, 40.6 ± 3.8% in sensitized
OVA-exposed mice, and 39.3 ± 4.0% in sensitized PBS-exposed mice
(n = 5 per group, no significant variances
among groups). Furthermore, we detected no difference in the relative
number of cells with weak or intense staining (intensity scale 0–2).
For example, OVA-exposed animals had 16.3 ± 3.9% of bone marrow
cells with high intensity staining, and the corresponding value for
PBS-exposed animals was 16.5 ± 3.9%.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain is more prominent in bone marrow cells than in BAL
eosinophils. Furthermore, the inhibitory effect of systemically
administered anti-IL-5 (TRFK-5) on eosinophilic inflammation in the
airways is parallel with a strong reduction of the bone marrow
eosinophil content. Also, local treatment with anti-IL-5 delivered
directly to the airways revealed that anti-IL-5 has some inhibitory
effects on airway eosinophilia via local mechanisms in vivo. Overall,
these data suggest that the eosinophilopoiesis is involved in the
induction of airway eosinophilic inflammation induced by airway
allergen exposure, and that release of IL-5 from cells within the bone
marrow, acting on specific IL-5R on bone marrow cells, is involved in
this process.
Repeated allergen exposure results in an accumulation of eosinophils in
the airways over several days, both in humans (36) and in
sensitized mice (Table I). In the present experiments in a murine model
of allergic airway inflammation, eosinophils are already evident in the
airways 24 h after two allergen exposures on consecutive days, and
after the third exposure, newly produced eosinophils significantly
contribute to this inflammatory response (Table I and Fig. 1). We have
proven this by detecting increased numbers of BrdU-labeled granulocytes
in the airways at these time points. BrdU is incorporated into the
cells by replacing thymidine, and its presence is proven by
immunocytochemical staining of nuclei (33, 37, 38).
Importantly, we are unable to distinguish BrdU-stained eosinophils from
neutrophils, because the BrdU staining using Fast Red Substrate system
is incompatible with eosin staining. However, in these experiments, the
number of eosinophils in the airways is much larger than the number of
neutrophils, proving that most of the BrdU-stained granulocytes are
eosinophils. BrdU will stain all newly produced cells if present in the
microenvironment at the time of mitosis. Furthermore, BrdU can be
carried over to several new generations of cells (39).
Therefore, it seems likely that the strongly BrdU-positive cells that
we have detected in BAL were passing through their latest mitosis at a
time of high concentrations of circulating BrdU (Fig. 2). Cells with
weaker staining could be either cells produced at times of low
concentrations of BrdU or cells having passed through several mitoses
before having fully matured. The experiments using BrdU-injection at a
single time point suggest that at least 64% of the granulocytes
present in BAL after five allergen exposures on consecutive days are
newly produced. This may be a low estimation because when BrdU was
injected both on days 1 and 3 of allergen exposure, almost 80% of the
granulocytes in the airways were BrdU-positive.
Importantly, the BAL eosinophilia observed during the early phase of
allergen exposure is likely to involve a storage pool of eosinophils
that were produced before the first allergen exposure, because a
majority of BAL granulocytes after two exposures were not BrdU-positive
(Fig. 1). Also, these experiments suggest that the accumulation of a
majority of the newly produced eosinophils in the airways will take
>24 h during repeated allergen exposure because injection of BrdU
24 h before BAL resulted in few BrdU-labeled cells in BAL (Fig. 4). It is possible that some of these newly produced eosinophils may
have completed their proliferation and maturation outside the bone
marrow. Some studies have shown that eosinophil colony-forming activity
is increased in the peripheral blood of atopic asthmatic patients
during the exacerbation of asthma (22). Furthermore, the
number of circulating CD34+ cells are increased
in atopic subjects (14), and CD34+
cells can also be detected in the bronchial mucosa in asthmatic
subjects (40). Recently, a study has suggested that a
subset of eosinophils in fact may differentiate in peripheral tissue
such as the nasal mucosa (41) in a process that is highly
IL-5 dependent. Together, these findings suggest that some progenitors
will traffic from the bone marrow, via blood, to the allergen-exposed
tissue, and the eosinophil production may occur in several tissues
concomitantly, especially in conditions with substantial eosinophilic
inflammation.
In cells taken from the bone marrow, we did not detect any increased
frequency of BrdU staining in animals exposed to allergen. This argues
against an overall lineage-nonspecific stimulation of the bone marrow
by allergen exposure in this model (Table III), which is similar to
findings in dogs (18).
The stimulation of the bone marrow eosinophilopoiesis in vivo is also
evident by increased numbers of eosin-staining cells with immature
morphology within the bone marrow compartment (Fig. 3). This finding
confirms previous experiments using C57BL/6 mice (21) as
well as BALB/c mice (4, 20). The increased number of
immature eosinophils is clearly evident the day after three allergen
exposures, in line with the increased BrdU staining in the airways at
this time point, suggesting that the eosinophilopoietic response in the
bone marrow takes some days to develop. Also, these data suggest that
it is possible to use this time point to evaluate mechanisms of bone
marrow stimulation induced by repeated allergen exposure.
After three allergen exposures, we detected increased numbers of both
CD3+ cells and non-CD3 cells isolated from the
bone marrow that contained measurable levels of intracellular IL-5
protein (Fig. 6B). Furthermore, both bone marrow
CD3+ cells and non-CD3 cells were able to release
IL-5 protein in vitro in the presence of adherent mononuclear cells.
Together these data suggest that cells expressing IL-5 protein are
present in the bone marrow after repeated airway allergen exposure,
confirming a previous study showing increased expression of IL-5 mRNA
in the bone marrow after airway allergen exposure in the mouse
(42). Our data extend this observation by showing that
these cells can release IL-5 protein when stimulated in vitro. The
previous study has also suggested that there is an increased total
number of CD3+ cells within the bone marrow after
airway allergen exposure, which we also have seen in C57BL/6 mice (our
unpublished data) but were unable to confirm in these experiments using
BALB/c mice. We observed much fewer CD3+ cells in
the bone marrow compared with Minshall and colleagues, who used a
polyclonal anti-CD3 Ab, whereas we used a monoclonal
anti-CD3 Ab. Despite these differences in methodology, these
three studies suggest that any increase of the total number of T
lymphocytes in the bone marrow after airway allergen exposure may be
small. Overall, the data argue that airway allergen exposure enhances
IL-5 production in both CD3+ cells and non-CD3
cells present in the bone marrow and, therefore, that both of these
cell types may contribute to the stimulation of the eosinophilopoiesis.
Also, the CD3+ cells are rare in the bone marrow,
whereas other IL-5-producing cells are more frequent, again arguing for
a role of non-CD3 cells, such as CD34+ cells
(42) in the local IL-5 release in the bone marrow.
Even though we find pronounced production of IL-5 in bone marrow cells,
it is clear that IL-5 is produced in several tissue compartments,
including the lungs (43, 44, 45, 46). Also, in these experiments,
we were able to detect increased levels of IL-5 in both BAL fluid and
serum after repeated allergen exposure. It is possible that increased
serum level of IL-5 may originate from several tissues, including
lungs, local lymph nodes, and bone marrow. However, the exact site of
action of anti-IL-5 on allergen-induced airway eosinophilia has not
been clearly described in several previous studies in which
anti-IL-5 was administered systemically (27, 28, 29, 30, 31, 32). A
recent study has shown that the anti-IL-5 Ab delivered via the
airway route attenuates allergen-induced eosinophilia in BAL and lung
tissues (47). However, in their study it is difficult to
clearly distinguish the local effect of anti-IL-5 from the systemic
effect because i.n. administration of anti-IL-5 Ab resulted in a
high concentration of the Ab in the circulation, which was above the
reported serum concentration needed to inhibit BAL eosinophilia
(48) and also bone marrow eosinophilia in our study (Figs. 7A and 9). To elucidate more closely how an anti-IL-5 Ab
(TRFK-5) would block airway eosinophilia, we treated sensitized mice
with this Ab either by repeated i.n. instillation or by i.p.
injection.
TRFK-5 given i.p. resulted in dose-dependent concentrations of the Ab
in the circulation (Fig. 7A) and dose-dependently attenuated
allergen-induced BAL eosinophilia, as well as BrdU staining of BAL
granulocytes. This inhibitory effect is parallel with a strong
inhibition of the bone marrow eosinophilia. The importance of the
circulating IL-5 for airway eosinophilia is in line with one of our
previous studies, showing that airway eosinophilia can be induced in
sensitized IL-5 knockout mice by systemic reconstitution with IL-5, but
not by local lung reconstitution (26). Thus, systemic IL-5
seems to be important for the induction of airway eosinophilia, and
this effect is likely to involve stimulation of eosinophilopoiesis.
Intranasally administered TRFK-5 can also exert inhibitory effects on
airway eosinophilia, partly via local mechanisms. For example, a high
dose of locally administered TRFK-5 has a slightly more pronounced
effect on BAL eosinophilia than an i.p. dose causing a similar
concentration of Ab in serum (e.g., 50 µg x 5 i.n. vs 10
µg x 1 i.p.; Figs. 7 and 8). This argues that there is
some effect of anti-IL-5 given locally to the airways on airway
eosinophilia. In this case, the concentration of Ab was higher in BAL
fluid after i.n. treatment, showing that the treatment was
compartmentalized to some degree to the airways.
In an attempt to understand the site of action of IL-5 at the
receptor level, we stained both bone marrow and BAL cytospins for the
-chain of IL-5R using immunocytochemistry. Using this method, we
were able to detect strong positive staining only in bone marrow cells,
but not in BAL eosinophils. We could not detect IL-5Rs in lung tissue
of allergen-exposed mice, using immunohistochemistry, whereas the
staining was prominent in bone marrow tissue samples (our unpublished
data). Thus, in this model, it may be that eosinophils shed most of the
-chain of IL-5R before reaching the airways. This confirms several
previous studies showing quite low expression of IL-5R in mature
eosinophils (49, 50). The high expression of IL-5R
-chain in the bone marrow supports the concept that IL-5 is
intricately involved in eosinophilopoiesis, and also one important
effect of this cytokine is localized to immature eosinophils bearing
IL-5Rs. In humans, it has been shown that mRNA specific for the
-chain of IL-5R can be found also in airway tissue (40, 51), and it is possible that this mRNA expression leads to IL-5R
expression on the cell surface. Certainly, it is likely that airway
eosinophils do express small numbers of functionally active IL-5Rs,
which, however, are impossible to stain by immunocytochemical
techniques. This notion is supported by a previous study, showing
effects of IL-5 on the survival of BAL eosinophils cultured in vitro
(52). In addition, it has been reported that specific IL-5
transgenic expression in lung epithelium in mice results in pathologic
changes characteristic of asthma, including eosinophil infiltration
into the airways (53). However, in these animals there was
a substantial leakage of IL-5 into the circulation (mean serum
concentration 1696 pg/ml), and more than 20 times higher numbers of
eosinophils were found in the bone marrow as compared with wild-type
mice (53), supporting effects on eosinophilopoiesis by
IL-5. Our finding that repeated airway administration of anti-IL-5
could attenuate allergen-induced BAL eosinophilia is in line with a
role of IL-5 also at this level. Our data agree with and extend the
recent study by Shardonofsky and colleagues showing that respiratory
delivery of anti-IL-5 Ab is capable of inhibiting airway
eosinophilia and airway hyperresponsiveness in allergic mice
(47). Thus, local airway treatment with anti-IL-5 may
also have beneficial effects on eosinophilic airway inflammation but
larger total doses are required.
In the present mouse experiments, we could detect prominent staining of
IL-5R -chain in 16% of bone marrow cells and some staining in up
to 40% of bone marrow cells, which were independent of allergen
exposure. This high frequency of IL-5R staining in mice is in contrast
to that in humans, in which much smaller numbers of bone marrow cells
express this epitope. Furthermore, in humans this receptor also seems
to be up-regulated by allergen exposure in asthmatics developing a late
asthmatic reaction (15). This is in parallel with
increased number of CFUs producing eosinophils in bone marrow samples
taken after allergen exposure (24). Thus, allergen
exposure seems to have slightly different regulatory effects in the
bone marrow of allergic mice and asthmatic patients, although the
models are similar in some respects because bone marrow activation can
be observed in both species (24, 13, 14, 15, 20, 24).
IL-5 can exert more than one important effect in the bone marrow. The first event seen when exogenous IL-5 is given by i.v. administration is that the number of eosinophils in the bone marrow is reduced, and the number of eosinophils in the circulation is increased (54). A similar effect is also seen with a single-day allergen exposure (4, 19). Recent studies using an in situ bone marrow perfusion model have shown that IL-5 stimulates mobilization of eosinophils from the bone marrow (55, 56). Thus, after allergen exposure, IL-5 seems to be involved in the release of eosinophils from the bone marrow compartment into the circulation to facilitate a rapid induction of airway eosinophilia. This release of eosinophils induced by allergen can be blocked by anti-IL-5 (57). During repeated or chronic allergen exposure, the situation is more complex. IL-5 is involved in enhanced allergen-induced eosinophilopoiesis, but may also be involved in the release of more eosinophils into the circulation. However, one key effect of blocking IL-5-induced effects in these experiments seems to be the attenuation of eosinophilopoiesis because anti-IL-5 strongly blocked the increase in the numbers of bone marrow eosin-staining cells, as well as BrdU-positive BAL granulocytes, induced by repeated allergen exposure.
These experiments show that a very specific chain of events occurs during repeated allergen exposure, resulting in airway eosinophilia. These events include enhanced production of IL-5 in bone marrow cells, resulting in an enhanced eosinophilopoiesis in this tissue, and a prominent contribution of newly produced eosinophils in the airway inflammatory response. To block the effects of IL-5 at the receptor level, the bone marrow can be targeted because IL-5Rs are highly expressed in this tissue. Alternative therapeutic approaches to reduce allergen-induced airway eosinophilia may be to block mechanisms that cause increased IL-5 production both in the bone marrow and peripheral tissues, or to reduce the expression of IL-5Rs.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Jan Lötvall, The Lung Pharmacology Group, Department of Respiratory Medicine and Allergology, Institute of Heart and Lung Diseases, Göteborg University, Guldhedsgatan 10A, 413 46 Gothenburg, Sweden.
3 Abbreviations used in this paper: BrdU, 5-bromo-2'-deoxyuridine; i.n., intranasal(ly); BAL, bronchoalveolar lavage; PLP, periodate-lysin-paraformaldehyde.
Received for publication April 1, 1999. Accepted for publication July 12, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit expression on bone marrow-derived CD34+ cells from asthmatic subjects: a novel marker of progenitor cell commitment towards eosinophilic differentiation. J. Clin. Invest. 100:2466.[Medline]
), and unstained cells (
). The relative numbers of
BrdU-positive cells at different time points are 27.5, 49.6, and 64.3%
after two, three, and five OVA exposures, respectively. BrdU was
injected twice on the first day of OVA exposure, 1 h before and
7 h after the exposure. Data are shown as mean ±
SEM. Kruskall-Wallis test shows significant variance among groups. *,
p < 0.05; **, p < 0.01
compared with PBS x 5-exposed group; 
