Cell Adhesion Molecules Regulate Fibrotic

Process via Th1/Th2/Th17 Cell Balance in a

Bleomycin-Induced Scleroderma Model
This information is current as Ayumi Yoshizaki, Koichi Yanaba, Yohei Iwata, Kazuhiro
of July 22, 2021. Komura, Asako Ogawa, Yuichiro Akiyama, Eiji Muroi,

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

Toshihide Hara, Fumihide Ogawa, Motoi Takenaka,
Kazuhiro Shimizu, Minoru Hasegawa, Manabu Fujimoto,
Thomas F. Tedder and Shinichi Sato
J Immunol 2010; 185:2502-2515; Prepublished online 12
July 2010;
doi: 10.4049/jimmunol.0901778
[Link]

References This article cites 62 articles, 23 of which you can access for free at:
[Link]

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision
• No Triage! Every submission reviewed by practicing scientists
• Fast Publication! 4 weeks from acceptance to publication
*average

Subscription Information about subscribing to The Journal of Immunology is online at:

[Link]
Permissions Submit copyright permission requests at:
[Link]
Email Alerts Receive free email-alerts when new articles cite this article. Sign up at:
[Link]

The Journal of Immunology is published twice each month by

The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2010 by The American Association of
Immunologists, Inc. All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
 The Journal of Immunology

Cell Adhesion Molecules Regulate Fibrotic Process via

Th1/Th2/Th17 Cell Balance in a Bleomycin-Induced
Scleroderma Model

Ayumi Yoshizaki,* Koichi Yanaba,* Yohei Iwata,* Kazuhiro Komura,† Asako Ogawa,*
Yuichiro Akiyama,* Eiji Muroi,* Toshihide Hara,* Fumihide Ogawa,* Motoi Takenaka,*
Kazuhiro Shimizu,* Minoru Hasegawa,† Manabu Fujimoto,† Thomas F. Tedder,‡ and

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

Shinichi Satox
Mice s.c. injected with bleomycin, an experimental model for human systemic sclerosis, develop skin and lung fibrosis, which is
mediated by inflammatory cell infiltration. This process is highly regulated by multiple adhesion molecules and does not require Ag
sensitization. To assess the role of adhesion molecules in this pathogenetic process, bleomycin-induced fibrosis was examined in mice
lacking adhesion molecules. L-selectin and/or ICAM-1 deficiency inhibited skin and lung fibrosis with decreased Th2 and Th17
cytokines and increased Th1 cytokines. In contrast, P-selectin deficiency, E-selectin deficiency with or without P-selectin blockade,
or P-selectin glycoprotein ligand 1 (PSGL-1) deficiency augmented the fibrosis in parallel with increased Th2 and Th17 cytokines
and decreased Th1 cytokines. Furthermore, loss of L-selectin and/or ICAM-1 reduced Th2 and Th17 cell numbers in bronchoal-
veolar lavage fluid, whereas loss of P-selectin, E-selectin, or PSGL-1 reduced Th1 cell numbers. Moreover, Th1 cells exhibited higher
PSGL-1 expression and lower expression of LFA-1, a ligand for ICAM-1, whereas Th2 and Th17 cells showed higher LFA-1 and
lower PSGL-1 expression. This study suggests that L-selectin and ICAM-1 regulate Th2 and Th17 cell accumulation into the skin
and lung, leading to the development of fibrosis, and that P-selectin, E-selectin, and PSGL-1 regulate Th1 cell infiltration, resulting
in the inhibition of fibrosis. The Journal of Immunology, 2010, 185: 2502–2515.

S
ystemic sclerosis (SSc) is a connective tissue disease cytokines, especially fibrogenic Th2 and Th17 cytokines and TGF-
characterized by excessive extracellular matrix deposition b1, a major fibrogenic growth factor, which positively correlate
in the skin, lung, and other visceral organs with an auto- with disease severity (5, 6).
immune background (1). The presence of autoantibodies is a cen- In general, leukocyte recruitment into inflammatory sites is
tral feature of SSc, since anti-nuclear Abs are detected in .90% of achieved using constitutive or inducible expression of multiple cell
patients (2). SSc patients have autoantibodies that react to various adhesion molecules (7). L-selectin (CD62L), E-selectin (CD62E),
intracellular components, such as DNA topoisomerase I (topo I), and P-selectin (CD62P) primarily mediate leukocyte capture and
centromeric protein B, U1-ribonucleoprotein (RNP), and histone rolling on the endothelium (8). L-selectin is constitutively ex-
(2). Furthermore, abnormal activation of immune cells, including pressed by most leukocytes (8). Whereas P-selectin is rapidly
T cells, B cells, NK cells, and macrophages, has been identified in mobilized to the surface of activated endothelium or platelets,
SSc (3). A recent study has shown that skin and lung fibrosis is E-selectin expression is induced within several hours after activa-
ameliorated by treatment with cyclophosphamide, an immunosup- tion with inflammatory cytokines (8). The selectins share a highly
pressive agent, indicating that immune activation leads to fibrosis conserved N-terminal lectin domain that can interact with sialy-
through the stimulation of collagen production by fibroblasts (4). lated and fucosylated oligosaccharides such as sialyl Lewis X (9).
Indeed, SSc patients exhibit inflammatory cell infiltration, espe- Although various candidates have been identified as potential
cially CD4+ T cells, and elevated serum levels of various ligands for selectins, P-selectin glycoprotein ligand 1 (PSGL-1)
is the best characterized ligand, which is recognized by all three
*Department of Dermatology, Nagasaki University Graduate School of Biomedical selectins (10). PSGL-1 is a mucin-like, disulfide-linked homo-
Sciences, Nagasaki; †Department of Dermatology, Kanazawa University Graduate dimer expressed by all subsets of leukocytes and is a high-
School of Medical Science, Kanazawa; xDepartment of Dermatology, University of
Tokyo Graduate School of Medicine, Tokyo, Japan; and ‡Department of Immunol-
affinity ligand for E- and P-selectins (11). PSGL-1 has also been
ogy, Duke University Medical Center, Durham, NC 27710 shown to bind to L-selectin, but its affinity is lower than E- and
Received for publication June 5, 2009. Accepted for publication June 5, 2010. P-selectins (12). ICAM-1 (CD54) is a member of the Ig super-
This work was supported by a grant for research on intractable diseases from the family that is constitutively expressed not only on endothelial
Ministry of Health, Labor, and Welfare of Japan (to S.S. and Y.Y.) and National cells, but also on fibroblasts and epithelial cells (13). It can be up-
Institutes of Health Grants CA96547, CA105001, and AI56363 (to T.F.T.).
regulated transcriptionally by several proinflammatory cytokines,
Address correspondence and reprint requests to Dr. Shinichi Sato, Department of such as TNF-a, IFN-g, and IL-1 (13). ICAM-1 forms the counter-
Dermatology, University of Tokyo Graduate School of Medicine, 7-3-1 Hongo, Bun-
kyo-ku, Tokyo 113-8655, Japan. E-mail address: satos-der@[Link] receptor for the lymphocyte b2 integrins, such as LFA-1 (7). The
Abbreviations used in this paper: BAL, bronchoalveolar lavage; BLM, bleomycin; ICAM-1/LFA-1 interactions predominantly mediate firm adhesion
COL1A2, a2(I) collagen; CTL, control; HPF, high-power field; MFI, mean fluores- and transmigration of leukocytes at sites of inflammation (7). In-
cence intensity; PSGL-1, P-selectin glycoprotein ligand 1; RNP, ribonucleoprotein; hibition of LFA-1 attenuated intratracheal bleomycin treatment-
SSc, systemic sclerosis; topo I, topoisomerase I; WT, wild type.
induced pulmonary fibrosis. However, the studies investigating the
Copyright Ó 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 role of L-selectin and ICAM-1 in fibrosis are limited. A recent

[Link]/cgi/doi/10.4049/jimmunol.0901778
The Journal of Immunology 2503

study has shown that intratracheal bleomycin treatment-induced Materials and Methods
pulmonary fibrosis is inhibited in L-selectin2/2 mice and ICAM- Mice
12/2 mice (14). In contrast, another study has suggested that an
L-selectin-deficient (L-selectin2/2) mice were produced as described
antagonist of ICAM-1 does not attenuate intratracheal bleomycin elsewhere (32). ICAM-12/2, P-selectin2/2, E-selectin2/2, and PSGL-12/2
treatment-induced pulmonary fibrosis, although the same treat- mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Mice
ment decreases leukocyte infiltration in the bronchoalveolar la- lacking both L-selectin and ICAM-1 (L-selectin/ICAM-12/2) were
vage (BAL) (15). Thus, the in vivo contribution of L-selectin generated as described previously (18). All mice were backcrossed 10
generations onto the C57BL/6 genetic background. Mice used for
and ICAM-1 to fibrosis remains unclear. experiments were 6 wk old. Both body size and lung size were similar
Although these cell adhesion molecules play important roles in for mutant and wild-type (WT) mice (data not shown). All studies and pro-
leukocyte transmigration, their association to inflammation cedures were approved by the Committee on Animal Experimentation of
remains controversial in several inflammatory models. Inhibition or Nagasaki University Graduate School of Medical Science.
loss of L-selectin, ICAM-1, P-selectin, E-selectin, or PSGL-1 leads Bleomycin treatment
to a significant reduction in leukocyte rolling and emigration in
Bleomycin (Nippon Kayaku, Tokyo, Japan) was dissolved in PBS at
many inflammatory models, such as the tight-skin mouse model,

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

a concentration of 1 mg/ml and sterilized by filtration. Bleomycin or PBS
which is a genetic model for human SSc, immune complex depo- (300 mg) was injected s.c. into a single location on the shaved back of the
sition-induced tissue injury, contact hypersensitivity, intratracheal mice daily for 4 wk with a 27-gauge needle, as described previously (30).
bleomycin-induced pulmonary fibrosis, and peritonitis (14, 16–21). For blocking study, mAbs to P-selectin (RB40.34, rat IgG1, 30 mg per
In contrast, some studies have suggested that loss of P-selectin mouse; BD Biosciences, San Jose, CA) were injected i.v. 30 min be-
fore first bleomycin or PBS treatment and three times per week into
and/or E-selectin increases inflammatory response, such as ex- E-selectin2/2 mice, as described previously (16).
perimental glomerulonephritis, collagen-induced arthritis, and
bleomycin-induced pulmonary fibrosis models (22–25). Thus, Preparation of BAL fluid
E- and P-selectins regulate inflammatory response either posi- BAL cells were prepared as described elsewhere (23). Briefly, bleomycin-
tively or negatively according to the inflammatory models. Pre- or PBS-treated mice were sacrificed and both lungs were excised. BAL
vious studies have shown that these cell adhesion molecules also fluid was collected as follows: 1 ml of saline was instilled three times and
withdrawn from the lung via an intratracheal cannula. Total leukocyte
regulate Th1, Th2, and Th17 cell migration. Inhibition or loss of counts were performed using a hemocytometer in the presence of trypan
ICAM-1 and/or L-selectin reduces Th1, Th2, and Th17 cell mi- blue. Cell differential counts were determined after cytospin centrifugation
gration (26–28), while some studies have suggested that Th1 and with May-Giemsa staining. Neutrophils were identified morphologically in
Th2 cell immigration is induced by L-selectin and/or ICAM-1 the BAL cells, as described previously (23). A total of 200 cells were
deficiency or blockade (26, 27, 29). P-selectin, E-selectin, and/or counted from randomly chosen high-power microscopic fields for each
sample, and at least 10 mice of each group were examined.
PSGL-1 deficiency or inhibition reduces Th1 and Th2 cell infil-
tration (28). Similarly, Th17 cell infiltration is significantly in- Histopathologic assessment of dermal fibrosis
hibited by E- and P-selectin deficiency (29). Thus, these cell ad- Morphologic characteristics of skin sections were assessed under a light
hesion molecules regulate Th cell balance, according to the tissue microscope. All skin sections were taken from the para-midline, lower back
site and the nature of the inflammatory stimuli. region (the same anatomic site to minimize regional variations in thickness).
Recently, Yamamoto et al. (30) established a new mouse model Sections were stained with H&E. Dermal thickness, defined as the thick-
ness of skin from the top of the granular layer to the junction between the
of SSc using bleomycin treatment: the s.c. injection of bleomycin dermis and s.c. fat, was examined. Ten random measurements were taken
induces fibrosis in the dermis and lung, autoantibody production, per section. All of the sections were examined independently by two
and dermal and pulmonary inflammatory cell infiltration, all of investigators in a blinded fashion. The skin from male mice was generally
which closely mimic the features of human SSc. In this mouse thicker than that from female mice despite the bleomycin or PBS treatment
(data not shown). Since similar results were obtained when male or female
model, B cells play important roles in the development of fibrosis
mice were analyzed separately, only data from female mice were presented
(31). However, in the absence of CD19, which is a critical signal for skin thickness in this study. Mast cells were identified by toluidine blue
transduction molecule of B cells, bleomycin-induced fibrosis is staining. Cells containing metachromatic granules were counted in 10
not completely inhibited, with an ∼30% reduction (31). This random grids under high-magnification (3400) power fields of a light
finding suggests that other immune cells and molecules also play microscope.
important roles in the development of fibrosis. However, the con- Histopathologic assessment of lung fibrosis
tribution of cell adhesion molecules to disease manifestations,
Lungs were excised after 4 wk of treatment with bleomycin or PBS, pro-
including autoimmunity, and to the mechanism underlying Th1, cessed as previously described, and stained by H&E and Van Gieson stain to
Th2, and Th17 cell infiltration remains unknown in the bleomy- detect collagen. The severity of fibrosis was semiquantitatively assessed
cin-induced SSc model. In this study, we investigated the role of according to Ashcroft et al. (33). Briefly, the lung fibrosis was graded on
cell adhesion molecules in the development of fibrosis and auto- a scale of 0–8 by examining randomly chosen fields of the left middle lobe
at a magnification of 3100. The grading criteria were as follows: grade 0,
immunity induced by bleomycin using mice lacking L-selectin, normal lung; grade 1, minimal fibrous thickening of alveolar or bronchiolar
ICAM-1, P-selectin, E-selectin, and PSGL-1. According to our walls; grade 3, moderate thickening of walls without obvious damage to
results, L-selectin and/or ICAM-1 deficiency reduced dermal lung architecture; grade 5, increased fibrosis with definite damage to lung
sclerosis, pulmonary fibrosis, and autoimmunity induced by bleo- structure and formation of fibrous bands or small fibrous masses; grade 7,
mycin treatment with increased Th1 cell infiltration and de- severe distortion of structure and large fibrous areas; and grade 8, total
fibrous obliteration of fields. Grades 2, 4, and 6 represent intermediates
creased Th2 and Th17 cell infiltration. In contrast, P-selectin, between the aforementioned criteria. All of the sections were scored in-
E-selectin, or PSGL-1 deficiency augments disease manifestations dependently by two investigators in a blinded fashion.
induced by bleomycin treatment in parallel with decreased Th1
Immunohistochemical staining
cell infiltration and increased Th2 and Th17 cell infiltration.
These results suggest that L-selectin and ICAM-1 mainly regulate Frozen tissue sections of skin and BAL cells after cytospin centrifugation
Th2 and Th17 cell infiltration, leading to the development of were incubated with rat mAb specific for macrophages (F4/80; Serotec,
Oxford, U.K.), B220 (BD Pharmingen, San Diego, CA), and CD3 (clone
fibrosis, whereas P-selectin, E-selectin, and PSGL-1 mainly 145-2C11; BD Pharmingen). Rat IgG (SouthernBiotech, Birmingham, AL)
regulate Th1 cell infiltration, which results in the inhibition of was used as a control for nonspecific staining. Stained cells were counted in
fibrosis. 10 random grids under high-magnification (3400) power fields of a light
2504 CELL ADHESION MOLECULES REGULATE Th CELL BALANCE

microscope. Each section was examined independently by two investiga- Fibroblast proliferation and collagen synthesis with IL-4, IFN-g,
tors in a blinded fashion. and IL-17 stimulation
Determination of hydroxyproline content in the skin and lung Cultured dermal fibroblasts (1.2 3 104/well) were seeded into a 96-well
tissue plate. Fibroblasts were serum-starved for 12 h and then cultured for 24 h
with or without 10 ng/ml murine rIL-4 and rIFN-g and 50 ng/ml murine
Hydroxyproline is a modified amino acid uniquely found at a high per- rIL-17 (R&D Systems). Proliferation of cultured dermal fibroblasts was
centage in collagen. Therefore, the skin and lung tissue hydroxyproline quantified by a colorimetric BrdU cell proliferation ELISA kit (Roche
content was measured as a quantitative measure of collagen deposition as Applied Science, Indianapolis, IN). After 24 h of incubation with or with-
previously described (18). The punch biopsy (6 mm) samples obtained out rIL-4, rIFN-g, and rIL-17, BrdU (10 mM) was added to each well and
from shaved dorsal skin and the harvested right lung of each mouse were incubated for 24 h. To analyze mRNA expression of pro-a2(I) collagen
analyzed. A hydroxyproline standard solution of 0–6 mg/ml was used to (COL1A2) and TGF-b1, total RNA was isolated from fibroblasts shortly
generate a standard curve. after 24 h of incubation with or without murine rIL-4, rIFN-g, and rIL-17.
Anti-nuclear Ab analysis Isolation and polarization of splenic T cells
Anti-nuclear Abs were assessed by indirect immunofluorescence staining
Splenocytes were obtained from 6- to 8-wk-old WT mice. Unless stated
using sera diluted 1/50 and HEp-2 substrate cells (Medical and Biological

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

otherwise, all cell cultures were performed in RPMI 1640 supplemented
Laboratories, Nagoya, Japan) as described (31). Anti-nuclear Abs were
with 10% FCS, 2 mM glutamine, 100 U/ml penicillin, 100 mg/ml strep-
detected using FITC-conjugated F(ab9)2 fragments specific for mouse
tomycin, and 2 mM 2-ME. T cells were enriched with a mouse CD4+
IgG plus IgM plus IgA (SouthernBiotech).
T cell kit using an autoMACS isolator (Miltenyi Biotec, Bergisch Glad-
ELISA for serum cytokines, Igs, and autoantibodies bach, Germany). A total of .99% of these cells were CD4+ when tested
with anti-CD4 mAb (BD Biosciences; data not shown). Naive CD4+
Sera were obtained by a cardiac puncture after 4 wk of treatment with T cells were obtained by surface staining with anti-CD62L, anti-CD44,
bleomycin or PBS and were stored at 280˚C. Serum levels of IL- and anti-CD25 mAbs. The CD62L+CD442CD252 population was isolated
4, IL-6, IL-10, IL-17, IFN-g, TGF-b1, and TNF-a were assessed using by flow cytometry cell sorting with a FACSAria (BD Biosciences). Cells
specific ELISA kits according to the manufacturer’s protocol (R&D were activated by plate-bound anti-CD3 (5 mg/ml) and anti-CD28 (5
Systems, Minneapolis, MN). Serum Ig concentrations were assessed as mg/ml) mAbs for 3 d. The Th0 condition indicated the neutral condition
described (31) using affinity-purified mouse IgM, IgG1, IgG2a, IgG2b, (no exogenous cytokines and anti-cytokine Abs). The Th1 condition indi-
IgG3, and IgA (SouthernBiotech) to generate standard curves. The relative cates addition of IL-12 (10 ng/ml) and anti–IL-4 Ab (10 mg/ml). The Th2
Ig concentration of each sample was calculated by comparing the mean condition indicates addition of IL-4 (10 ng/ml) and anti–IL-12 Ab (10
OD obtained for duplicate wells to a semilog standard curve of titrated mg/ml). The Th17 condition indicates addition of IL-6 (10 ng/ml), TGF-b1
standard Ab using linear regression analysis. The specific ELISA kits were (5 ng/ml), anti–IFN-g mAb (10 mg/ml), and anti–IL-4 mAb (10 mg/ml).
used to measure anti-topo I (Medical and Biological Laboratories), anti- All cytokines were from R&D Systems. All anti-cytokine Abs were from
centromeric protein B (Funakoshi, Tokyo, Japan), and anti–U1-RNP (Med- BD Pharmingen.
ical and Biological Laboratories). These ELISA plates were incubated with
serum samples diluted 1/100. Relative levels of autoantibodies were de- Flow cytometry
termined for each group of mice using pooled serum samples. Sera were
diluted at log intervals (1/10–1/105) and assessed for relative autoantibody Abs used in this study included FITC-conjugated mAbs to IL-4 (Imgenex,
levels as above except that the results were plotted as OD versus dilution San Diego, CA), IFN-g (GeneTex, San Antonio, TX), and IL-17 (Novus
(log scale). The dilutions of sera giving half-maximal OD values were Biologicals, Littleton, CO); PE-conjugated mAbs to IL-17, L-selectin,
determined by linear regression analysis, thus generating arbitrary units LFA-1, and PSGL-1 (all from BD Biosciences); and cyanine 5–PE anti-
per milliliter values for comparison between sets of sera. CD4 mAb (LifeSpan BioSciences, Seattle, WA). Single-cell suspensions of
BAL cells or CD4+ T cells were incubated with the Abs for 30 min at 4˚C.
RNA isolation and real-time PCR The cells were washed and fixed with 1% paraformaldehyde in PBS.
IFN-g, IL-4, and IL-17 production of BAL lymphocytes and cultured
Total RNA was isolated from lower back skin and lung with RNeasy spin CD4+ T cells was determined by flow cytometric intracellular cytokine
columns (Qiagen, Crawley, U.K.). Total RNA from each sample was re- analysis, as previous described (36). Briefly, cells were suspended at 106
verse-transcribed into cDNA. Expression of IL-4, IL-6, IL-10, IL-17, IFN-g, per milliliter in RPMI 1640 containing 2 mM glutamine and incubated
TGF-b1, and TNF-a was analyzed by an ABI Prism 7000 sequence with 10 mg/ml brefeldin A (Calbiochem, San Diego, CA) for 2 h at 37˚C.
detector (Applied Biosystems, Foster City, CA) using a real-time PCR Samples were stained for cell surface markers PSGL-1, LFA-1, and CD4
quantification method according to the manufacturer’s instructions. Se- for 30 min at 4˚C. After permeabilizing with FACS permeabilizing solu-
quence-specific primers and probes were designed by Pre-Developed Taq- tion according to the manufacturer’s instructions (BD Pharmingen), the
Man assay reagents or Assay-On-Demand (Applied Biosystems). GAPDH cells were then stained for intracellular IFN-g, IL-4, and IL-17. Cells were
was used to normalize mRNA. Relative expression of real-time PCR prod- washed and analyzed on a FACScan flow cytometer (BD Pharmingen),
ucts was determined by using the DDCT method (31) to compare target with data have been analyzed from 105 cells. Positive and negative pop-
gene and housekeeping gene mRNA expression. ulations of cells were determined using unreactive isotype-matched mAbs
(Beckman Coulter, Brea, CA) as controls for background staining.
Fibroblast culture and stimulation
Skin samples of 1 cm3 were taken from the para-middle, lower back region of P-selectin, E-selectin, and ICAM-1 binding assay in polarized
WT mice. To obtain fibroblasts, the tissue was cut into 1-mm3 pieces, placed Th cells
in sterile plastic dishes, cultured in DMEM (Invitrogen, Gaithersburg, MD)
We performed P-selectin, E-selectin, and ICAM-1 binding assays, as de-
containing 10% heat-inactivated FCS, 100 U/ml penicillin (Invitrogen), and
scribed previously (37). Recombinant mouse P-selectin, E-selectin, or
100 mg/ml streptomycin (Invitrogen), and cultured at 37˚C in a 5% CO2
ICAM-1/Fc chimera proteins were obtained from R&D Systems. Th0,
humidified atmosphere. After 2–3 wk of incubation, the outgrowth of fibro-
Th1, Th2, or Th17 cells (5 3 104), which were isolated by flow cytometry
blasts was detached by brief trypsin treatment and recultured in the medium.
cell sorting with a FACSAria (BD Biosciences), were resuspended in 100
Confluent cultures of fibroblasts were serum-starved for 12 h and then
ml of RPMI 1640 medium. Cells were incubated with 0.3 mg of each
cultured with or without 10 ng/ml murine rIL-4 and rIFN-g and 50
chimera protein for 30 min at 4˚C and resuspended in 100 ml of RPMI
ng/ml murine rIL-17 (R&D Systems) for 24 h. We used FITC-conjugated
1640 medium containing FITC-labeled goat anti-mouse IgG (2 mg/ml;
anti-CD90.2 mAb (BD Biosciences) and PE-conjugated anti-CD45 mAb
Jackson ImmunoResearch Laboratories, West Grove, PA). After incubation
(Serotec) to differentiate fibroblasts and leukocytes (34). The CD452
for another 30 min at 4˚C, cells were analyzed on a FACScan flow cytom-
CD90+ cells were recognized as fibroblasts. The purity of fibroblasts con-
firmed with flow cytometry was .99%, with no leukocytes found in the eter (BD Pharmingen), with data having been analyzed from 105 cells.
harvested cells (data not shown). Furthermore, the purity of fibroblasts was Statistical analysis
assessed by microscopic examination of parallel samples grown on culture
slides (BD Biosciences) and stained with H&E (data not shown), as pre- All data are expressed as mean values 6 SD. The Mann-Whitney U test was
viously described (35). In each experiment, obtained fibroblasts were exam- used to determine the level of significance of differences between sample
ined at the same time and under the same conditions of cultures (e.g., cell means, and ANOVA followed by Bonferroni’s test was used for multiple
density, passage, and days after plating). comparisons.
The Journal of Immunology 2505

Results (p , 0.05; Fig. 1B). Additionally, bleomycin-treated E-selectin2/2

L-selectin and/or ICAM-1 loss attenuated the development of mice administrated with anti–P-selectin mAb and PSGL-12/2 mice
fibrosis induced by bleomycin, whereas P-selectin loss, showed further accumulation of the hydroxyproline compared
E-selectin loss with or without P-selectin blockade, or with bleomycin-treated P-selectin2/2 and E-selectin2/2 mice (p ,
PSGL-1 loss augmented it 0.05; Fig. 1B).
Similar results were obtained for the lung fibrosis score and the
Bleomycin was injected s.c. into the back of mice daily for
pulmonary hydroxyproline content (Fig. 2A, 2B). After 4 wk,
4 wk. Previous studies have shown that skin and lung fibrosis,
bleomycin-treated WT mice exhibited extensive inflammatory cell
epithelial injury, and inflammatory cell infiltration develop during
infiltration, fibrosis, granulomas, alveolar epithelial injury, and in-
the first 4 wk of bleomycin treatment peak in the fourth week and
creased hydroxyproline content (Fig. 2A, 2B). L-selectin or
begin to resolve 6 wk after the cessation of treatment (30, 31). In
ICAM-1 deficiency reduced such histological changes and hydrox-
this study, skin and lung fibrosis in mutant and WT mice treated
yproline content, whereas P-selectin, E-selectin, or PSGL-1 defi-
with either bleomycin or PBS was histopathologically assessed
ciency augmented it. Thus, skin and lung fibrosis induced by s.c.
1, 2, 3, and 4 wk after bleomycin treatment. The dermal thickness
bleomycin injection was inhibited by L-selectin or ICAM-1 defi-

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

(the thickness from the top of the granular layer to the junction
ciency and was exacerbated by P-selectin, E-selectin, or PSGL-1
between the dermis and s.c. fat) and lung fibrosis score showed
deficiency. Both L-selectin and ICAM-1 deficiency significantly
a time-dependent increase in bleomycin-treated mice (data not
inhibited skin and lung fibrosis relative to L-selectin or ICAM-1
shown), which is consistent with previous studies (30, 31). After
deficiency alone. Furthermore, the deterioration effect of PSGL-1
4 wk, bleomycin treatment induced significantly greater dermal deficiency or E-selectin deficiency with P-selectin blockade on skin
thickness relative to PBS treatment in mutant and WT mice (p , and lung fibrosis was greater than that of P-selectin or E-selectin
0.05; Fig. 1A, 1B). The dermal thickness was similar between deficiency alone.
nontreated and PBS-treated mice (data not shown). The dermal
thickness in bleomycin-treated WT mice significantly increased Leukocyte infiltration into skin and lung was inhibited by
by 2.3-fold compared with PBS-treated WT mice (p , 0.05; Fig. L-selectin and/or ICAM-1 loss, whereas it was enhanced by
1A, 1B). Bleomycin-treated L-selectin2/2 mice, ICAM-12/2 P-selectin loss, E-selectin loss with or without P-selectin
mice, and L-selectin/ICAM-12/2 mice showed moderate blockade, or PSGL-1 loss
thickening of dermal tissue that was 23, 26, and 49% thinner, The numbers of mast cells, macrophages, T cells, and B cells have
respectively, than that found in bleomycin-treated WT mice been reported to increase in sclerotic skin and fibrotic lung from
(p , 0.05). Moreover, skin thickness of bleomycin-treated human SSc patients and bleomycin-induced SSc mouse models
L-selectin/ICAM-12/2 mice was thinner than that of bleomycin- (31, 38). Therefore, the numbers of these immune cells were
treated L-selectin2/2 and ICAM-12/2 mice (p , 0.01). In contrast, assessed in skin and BAL fluid after 4 wk of bleomycin treatment
the dermal thickness in bleomycin-treated P-selectin2/2 mice, E- in mutant and WT mice. The skin and BAL numbers of mast cells,
selectin2/2 mice, E-selectin2/2 mice treated with anti–P-selectin neutrophils, macrophages, T cells, and B cells were greater in
mAb, and PSGL-12/2 mice was 15, 18, 28, and 31% thicker, bleomycin-treated mice than in PBS-treated mice (p , 0.005;
respectively, than that in bleomycin-treated WT mice (p , 0.05; Figs. 1C, 2C). In skin tissue, bleomycin-treated L-selectin2/2,
Fig. 1A, 1B). Bleomycin-treated E-selectin2/2 mice administrated ICAM-12/2, and L-selectin/ICAM-12/2 mice exhibited lower
with anti–P-selectin mAb and PSGL-12/2 mice showed the numbers of these cells compared with bleomycin-treated WT mice
increased dermal thickness compared with bleomycin-treated (p , 0.05), except for T cell numbers (Fig. 1C). In contrast, the
P-selectin2/2 and E-selectin2/2 mice (p , 0.05; Fig. 1A, 1B). numbers of mast cells, T cells, and B cells were greater in bleo-
Additionally, skin thickness of bleomycin-treated L-selectin2/2 mycin-treated P-selectin2/2 mice, E-selectin2/2 mice, E-selectin2/2
mice, ICAM-12/2 mice, and L-selectin/ICAM-12/2 mice was mice treated with anti–P-selectin mAb, and PSGL-12/2 mice than
significantly thinner than that of bleomycin-treated P-selectin2/2 in bleomycin-treated WT mice (p , 0.01; Fig. 1C).
mice, E-selectin2/2 mice, E-selectin2/2 mice treated with anti– In BAL cells, bleomycin-treated L-selectin2/2, ICAM-12/2,
P-selectin mAb, and PSGL-12/2 mice (p , 0.01; Fig. 1A, 1B). and L-selectin/ICAM-12/2 mice showed decreased numbers of
Masson trichrome staining revealed thickened collagen bundles in total leukocytes, including neutrophils and macrophages,
the skin from bleomycin-treated WT mice, which was reduced compared with bleomycin-treated WT mice (p , 0.05; Fig. 2C).
by L-selectin and/or ICAM-1 deficiency and was increased by Moreover, bleomycin-treated L-selectin/ICAM-12/2 mice showed
P-selectin deficiency, E-selectin deficiency with or without P- lower numbers of total leukocytes and neutrophils than did
selectin blockade, or PSGL-1 deficiency (data not shown). Cuta- bleomycin-treated L-selectin2/2 and ICAM-12/2 mice (p ,
neous fibrosis was also assessed by quantifying hydroxyproline 0.05; Fig. 2C). However, T cell and B cell numbers in BAL fluid
content of 10 mg of skin samples from mutant and WT mice were similar among bleomycin-treated WT, L-selectin2/2, ICAM-
(Fig. 1B). Although the hydroxyproline content in bleomycin- 12/2, and L-selectin/ICAM-12/2 mice. In contrast, the numbers of
treated WT mice was increased by 2.9-fold relative to that in total leukocytes, including neutrophils, macrophages, T cells, and
PBS-treated WT mice (p , 0.01), bleomycin-treated L-selectin2/2, B cells, significantly increased in bleomycin-treated P-selectin2/2
ICAM-12/2, and L-selectin/ICAM-12/2 mice reduced the mice, E-selectin2/2 mice, E-selectin2/2 mice treated with anti–P-
hydroxyproline content by 29, 31, and 41%, respectively, in selectin mAb, and PSGL-12/2 mice compared with those in
bleomycin-treated WT mice (p , 0.05). Moreover, the hydroxy- bleomycin-treated WT mice (p , 0.05). E-selectin2/2 mice
proline content of bleomycin-treated L-selectin/ICAM-12/2 mice treated with anti–P-selectin mAb and PSGL-12/2 mice exhibited
was significantly reduced compared with bleomycin-treated L- further increased numbers of neutrophils relative to bleomycin-
selectin2/2 and ICAM-12/2 mice (p , 0.01). In contrast, the treated P-selectin2/2 and E-selectin2/2 mice (p , 0.05). Thus,
hydroxyproline content was increased in bleomycin-treated P- inflammatory cell recruitment to BAL fluid was inhibited by L-
selectin2/2 mice (10% increase), E-selectin2/2 mice (9%), E- selectin or ICAM-1 deficiency. Furthermore, both L-selectin and
selectin2/2 mice treated with anti–P-selectin mAb (22%), and ICAM-1 deficiency inhibited this recruitment more strongly than
PSGL-12/2 mice (20%) compared with bleomycin-treated WT mice did L-selectin or ICAM-1 deficiency alone. In contrast, P-selectin
2506 CELL ADHESION MOLECULES REGULATE Th CELL BALANCE

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

FIGURE 1. Skin fibrosis (A, B) and the numbers of mast cells, macrophages, T cells, and B cells at the bleomycin-injected site of skin (C) from PBS-
treated (open bar) or BLM-treated (filled bar) WT mice, L-selectin2/2 mice (L2/2), ICAM-12/2 mice (ICAM-12/2), L-selectin/ICAM-12/2 mice (L/
ICAM-12/2), P-selectin2/2 mice (P2/2), E-selectin2/2 mice (E2/2), E-selectin2/2 mice treated with anti–P-selectin mAb (E2/2 + anti-P Ab), and PSGL-
12/2 mice (PSGL-12/2). A, Representative histological sections stained with H&E are shown (original magnification 340). Skin fibrosis was assessed by
quantitatively measuring dermal thickness and hydroxyproline content 4 wk after bleomycin treatment (B). Mast cells were identified by toluidine blue
staining; macrophages, T cells, and B cells were stained with F4/80, anti-CD3 mAb, and anti-B220 mAb, respectively (C). These results represent those
obtained with at least 10 mice in each group. The dermal thickness was measured under a light microscope. Cells were counted in 10 random grids under
magnification of 3400 high-power fields. Each histogram shows the mean (6SD) results obtained for 10 mice in each group. pp , 0.05; ppp , 0.01 versus
PBS-treated each group of mice. †p , 0.05; ††p , 0.01 versus BLM-treated WT mice. BLM, bleomycin; HPF, high-power field.

or E-selectin deficiency alone induced inflammatory cell recruit- 12/2, and L-selectin/ICAM-12/2 mice relative to bleomycin-
ment to BAL fluid. Moreover, this increasing effect of E-selectin treated WT mice, whereas serum levels of IFN-g were higher
deficiency with P-selectin blockade and PSGL-1 deficiency was in these mice than in bleomycin-treated WT mice (p , 0.05;
greater than that of P-selectin or E-selectin deficiency alone. Fig. 3A). In contrast, bleomycin-treated P-selectin2/2 mice, E-
selectin2/2 mice, E-selectin2/2 mice treated with anti–P-
Effect of cell adhesion molecule deficiency or blockade on selectin mAb, and PSGL-12/2 mice exhibited elevated levels of
cytokine production IL-4, IL-6, IL-17, and TGF-b1 compared with bleomycin-treated
It has been suggested that IL-4, IL-6, IFN-g, IL-17, TNF-a, TGF- WT mice (p , 0.05), whereas they showed lower levels of IFN-g
b1, and IL-10 production contributes to bleomycin-induced fibro- than did bleomycin-treated WT mice (p , 0.05; Fig. 3A). There
sis by regulating the collagen production by fibroblasts (30, 31, 39, were no significant differences in serum levels of TNF-a and
40). Therefore, in the serum, sclerotic skin, and fibrotic lung, the IL-10 between bleomycin-treated mutant and WT mice. Similar
production of these cytokines was assessed by ELISA and real- results were obtained for mRNA expression levels in the skin and
time PCR. In the serum, skin, and lung, bleomycin-treated mutant lung (data not shown), except that bleomycin-treated L-selectin/
and WT mice had elevated levels of IL-4, IL-6, IFN-g, IL-17, ICAM-12/2 mice showed lower expression levels of IL-4, IL-17,
TNF-a, TGF-b1, and IL-10 compared with PBS-treated mutant and TGF-b1 in the skin and lung and IL-6 in the lung than
and WT mice (p , 0.01). Serum levels of IL-4, IL-6, IL-17, and did L-selectin2/2 and ICAM-12/2 mice (p , 0.05). In lung
TGF-b1 were reduced in bleomycin-treated L-selectin2/2, ICAM- tissue, ICAM-12/2 mice exhibited lower IL-17 expression than
The Journal of Immunology 2507

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

FIGURE 2. Lung fibrosis (A, B) and the influx numbers of total leukocytes, including neutrophils, macrophages, T cells, and B cells, into BAL fluid (C)
from PBS-treated (open bar) or bleomycin-treated (filled bar) WT mice, L-selectin2/2 mice (L2/2), ICAM-12/2 mice (ICAM-12/2), L-selectin/ICAM-12/2
mice (L/ICAM-12/2), P-selectin2/2 mice (P2/2), E-selectin2/2 mice (E2/2), E-selectin2/2 mice treated with anti–P-selectin mAb (E2/2 + anti-P Ab), and
PSGL-12/2 mice (PSGL-12/2). A, Representative histological sections stained with H&E are shown (original magnification 3100). Lung fibrosis was
assessed by quantitatively measuring lung fibrosis score and hydroxyproline content 4 wk after bleomycin treatment (B). The BAL cell counts were as
described in Materials and Methods. These results were obtained from at least 10 mice in each group. Lung fibrosis score was measured under a light mi-
croscope. The differential BAL cells were counted in 10 random grids under magnification of 3400 high-power fields. Each histogram shows the mean (6SD)
results obtained for 10 mice of each group. pp , 0.05; ppp , 0.01 versus PBS-treated each group. †p , 0.05; ††p , 0.01 versus bleomycin-treated WT mice.
BLM, bleomycin; HPF, high-power field.
2508 CELL ADHESION MOLECULES REGULATE Th CELL BALANCE

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

FIGURE 3. Levels of IL-4, IL-6, IFN-g, IL-17, TNF-a, TGF-b1, and IL-10 in serum samples from WT mice, L-selectin2/2 mice (L2/2), ICAM-12/2
mice (ICAM-12/2), L-selectin/ICAM-12/2 mice (L/ICAM-12/2), P-selectin2/2 mice (P2/2), E-selectin2/2 mice (E2/2), E-selectin2/2 mice treated with
anti–P-selectin mAb (E2/2 + anti-P Ab), and PSGL-12/2 mice (PSGL-12/2) treated with either PBS (open bar) or bleomycin (filled bar). Serum samples
were obtained by a cardiac puncture 4 wk after treatment with either bleomycin or PBS. Serum cytokine levels were assessed using specific ELISA. Each
histogram shows the mean (6SD) results obtained for 10 mice of each group. ppp , 0.01 versus PBS-treated each group. †p , 0.05; ††p , 0.01 versus
BLM-treated WT mice.

did L-selectin2/2 mice (p , 0.05). Moreover, E-selectin2/2 mice Effect of Th1, Th2, and Th17 cytokines on dermal fibroblasts
treated with anti–P-selectin mAb and PSGL-12/2 mice exhibited We investigated the effect of IL-4, IFN-g, and IL-17 on function
higher expression levels of IL-4, IL-17, and TGF-b1 in the skin and of fibroblasts obtained from WT mice (Fig. 4). Stimulation with
lung and IL-6 in the lung than did P-selectin2/2 and E-selectin2/2 IL-4 and IL-17 significantly increased fibroblast proliferation com-
mice (p , 0.05). In contrast, E-selectin2/2 mice treated with anti– pared with media alone (p , 0.01). In contrast, IFN-g stimulation
P-selectin mAb and PSGL-12/2 mice exhibited a reduction in skin inhibited fibroblast proliferation compared with media alone (p ,
IFN-g expression that was significantly lower than that found in 0.05). COL1A2 and TGF-b1 mRNA expressions were quantified
P-selectin2/2 mice as well as in E-selectin2/2 mice (p , 0.05). by real-time PCR in cultured dermal fibroblasts. COL1A2 and
Thus, in the serum, sclerotic skin, and fibrotic lung, bleomycin TGF-b1 mRNA levels in fibroblasts significantly increased by
treatment induced the overexpression of various cytokines. IL-4 and IL-17 stimulation, whereas they decreased by IFN-g
L-selectin and/or ICAM-1 deficiency reduced expression levels stimulation compared with media alone (p , 0.01).
of fibrotic Th2 cytokines, such as IL-4 and IL-6, IL-17, and
TGF-b1, a major fibrogenic growth factor, whereas P-selectin
deficiency, E-selectin deficiency with or without P-selectin block- Infiltration of Th1, Th2, and Th17 cells was regulated by cell
ade, and PSGL-1 deficiency increased it. In contrast, L-selectin adhesion molecules in the bleomycin-induced SSc mouse model
and/or ICAM-1 deficiency increased expression of IFN-g, anti- We investigated Th1, Th2, and Th17 cell frequencies in BAL fluid
fibrotic Th1 cytokine, whereas P-selectin deficiency, E-selectin de- from bleomycin-treated mutant and WT mice (Fig. 5). Bleomycin-
ficiency with or without P-selectin blockade, and PSGL-1 deficiency treated mutant and WT mice exhibited significantly increased
decreased it. frequencies of Th1, Th2, and Th17 cells compared with PBS-

FIGURE 4. Proliferation and collagen synthesis of dermal fibroblasts obtained from WT mice. Cultured fibroblasts were serum starved for 12 h and then
cultured for 24 h with murine rIL-4 (10 ng/ml), rIFN-g (10 ng/ml), and rIL-17 (50 ng/ml). Total RNA from fibroblasts was extracted and reverse transcribed
to cDNA, and mRNA expression of COL1A2 and TGF-b1 was analyzed by real-time PCR. In proliferation assays, after 24 h incubation, BrdU (10 mM)
was added to each well and incubated for 24 h. BrdU incorporation in proliferating cells was quantified by ELISA. Each histogram shows the mean (6SD)
results obtained for six mice of each group. pp , 0.05; ppp , 0.01 versus fibroblasts cultured with media alone.
The Journal of Immunology 2509

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

FIGURE 5. Th1, Th2, and Th17 cell frequencies of BAL in PBS-treated (open bar) or bleomycin-treated (filled bar) WT mice, L-selectin2/2 mice (L2/2),
ICAM-12/2 mice (ICAM-12/2), L-selectin/ICAM-12/2 mice (L/ICAM-12/2), P-selectin2/2 mice (P2/2), E-selectin2/2 mice (E2/2), E-selectin2/2 mice
treated with anti–P-selectin mAb (E2/2 + anti-P Ab), and PSGL-12/2 mice (PSGL-12/2). We determined Th1, Th2, and Th17 cells by surface
CD4 expression and intracellular expression of IFN-g, IL-4, and IL-17 as previously described (36). BAL fluid was analyzed by flow cytometry after
4 wk of PBS or bleomycin treatment. These data are representative of three independent experiments (A). Percentages of Th1, Th2, and Th17 cells are shown
in the each quadrant. We also show summaries of Th1, Th2, and Th17 cell frequencies in each group (B). Each histogram shows the mean (6SD) results
obtained for 10 mice of each group. pp , 0.005; ppp , 0.001 versus PBS-treated each group. †p , 0.05; ††p , 0.01 versus bleomycin-treated WT mice.
BLM, bleomycin.

treated mutant and WT mice in the BAL fluid (p , 0.01; Fig. 5B) Expression of LFA-1 and PSGL-1 on polarized Th1, Th2, and
but not in homogenate lung parenchyma (data no shown). There Th17 cells
were no significant differences in Th1, Th2, and Th17 cell fre- To investigate how cell adhesion molecules regulate Th1, Th2, and
quencies between PBS-treated mutant and WT mice (Fig. 5B). Th17 cell infiltration, expression of LFA-1, a ligand of ICAM-1,
The population expressing IFN-g, IL-4, or IL-17 did not over- and PSGL-1 in polarized Th1, Th2, and Th17 cells and nonpolar-
lap (Fig. 5A), which is consistent with previous studies (36). Bleo- ized Th0 cells were analyzed by flow cytometry (Fig. 6). In Th0
mycin-treated L-selectin2/2, ICAM-12/2, and L-selectin/ICAM- cells obtained from the Th0 condition, IFN-g, IL-4, and IL-17
12/2 mice exhibited significantly reduced Th2 and Th17 cell expression was hardly detectable (Fig. 6A). In Th1, Th2, and
frequencies and significantly increased Th1 cell frequencies in Th17 cells obtained from the Th1, Th2, and Th17 conditions,
BAL fluid compared with bleomycin-treated WT mice (p , respectively, the population expressing IFN-g, IL-4, or IL-17 did
0.05; Fig. 5B). L-selectin/ICAM-12/2 mice displayed further not overlap (data not shown). Th1, Th2, and Th17 cells did not
reduction of Th2 and Th17 cell influx into BAL fluid relative express L-selectin (data not shown). Th1 cells showed lower
to L-selectin2/2 and ICAM-12/2 mice (p , 0.05). Th1 cell fre- LFA-1 expression and ICAM-1 binding ability compared with
quencies were significantly reduced in bleomycin-treated P- Th0 cells (p , 0.01 and p , 0.05, respectively; Fig. 6B, 6C).
selectin2/2 and E-selectin2/2 mice compared with bleomycin- In contrast, Th1 cells exhibited higher PSGL-1 expression and
treated WT mice (p , 0.05). The reducing effect of E-selectin P-selectin and E-selectin binding ability relative to Th0 cells
loss with P-selectin blockade and PSGL-1 loss on Th1 cell influx (p , 0.01, p , 0.001, and p , 0.05, respectively). Although
was greater than that of P-selectin or E-selectin deficiency alone LFA-1 expression and ICAM-1 binding ability increased in Th2
(p , 0.05). Bleomycin-treated P-selectin2/2 mice, E-selectin2/2 and Th17 cells compared with Th0 cells (p , 0.01), Th2 and Th17
mice with or without P-selectin blockade, and PSGL-12/2 mice cells exhibited lower expression levels of PSGL-1 and P-selectin
also exhibited significantly increased Th2 and Th17 cell fre- and E-selectin binding ability (p , 0.01). Additionally,
quencies in BAL fluid compared with bleomycin-treated WT mice LFA-1 expression levels and ICAM-1 binding ability in Th17 cells
(p , 0.05, respectively). Moreover, E-selectin2/2 mice treated were greater than those in Th2 cells (p , 0.05). There was no
with anti–P-selectin mAb and PSGL-12/2 mice displayed higher significant difference in the frequencies of LFA-1+ or PSGL-1+
frequencies of Th17 cells compared with P-selectin2/2 and E- cells among Th0, Th1, Th2, and Th17 cells (Fig. 6A). These
selectin2/2 mice (p , 0.05). Thus, these results suggest that results suggest that PSGL-1 is preferentially used to recruit Th1
L-selectin and/or ICAM-1 deficiency inhibits Th2 and Th17 cell cells into inflammatory lesions, whereas Th2 and Th17 cells dom-
influx into BAL fluid, whereas Th1 cell infiltration is induced inantly use LFA-1.
by L-selectin and/or ICAM-1 deficiency. P-selectin loss, E-
selectin loss with or without P-selectin blockade, and PSGL-1 Contribution of cell adhesion molecules to Ig production
loss inhibit Th1 cell influx and induce Th2 and Th17 cell in- Serum Ig levels in bleomycin-treated mutant and WT mice were
filtration. also investigated (Fig. 7). PBS-treated mutant mice had similar
2510 CELL ADHESION MOLECULES REGULATE Th CELL BALANCE

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

FIGURE 6. The LFA-1 and PSGL-1 expression levels (A, B) and ICAM-1, P-selectin, and E-selectin binding ability (C) in Th0, Th1, Th2, and Th17
cells. We determined Th0, Th1, Th2, and Th17 cells by surface CD4 expression and intracellular expression of IFN-g, IL-4, and IL-17 as previously
described (36). Polarized or nonpolarized splenic CD4+ T cells were analyzed by flow cytometry. These data are representative of three independent
experiments. Numbers indicate the percentage of cells in each quadrant. Histograms indicate MFI (6SD). MFI, mean fluorescence intensity.

IgM, IgG1, IgG2a, IgG2b, IgG3, and IgA levels to PBS-treated half-maximal OD values in ELISA generated arbitrary units per
WT mice (data not shown). Bleomycin treatment increased serum milliliter that could be directly compared between groups (values
IgM, IgG1, IgG2a, IgG2b, IgG3, and IgA levels compared with in parentheses of Fig. 8). Bleomycin-treated L-selectin2/2 and
PBS treatment (p , 0.05). L-selectin/ICAM-12/2 mice treated ICAM-12/2 mice had decreased IgM autoantibody levels to topo I
with bleomycin had decreased IgM and IgG3 levels compared with and U1-RNP and reduced IgG autoantibody levels to topo I relative
bleomycin-treated WT mice (p , 0.05), whereas the levels of other to bleomycin-treated WT mice (p , 0.05). L-selectin and ICAM-1
isotypes were similar among bleomycin-treated L-selectin2/2, deficiency decreased bleomycin-induced IgM autoantibody levels
ICAM-12/2, L-selectin/ICAM-12/2, and WT mice. P-selectin2/2 to topo I and U1-RNP, as well as IgG autoantibody levels to topo I
and E-selectin2/2 mice administered bleomycin had increased compared with bleomycin-treated WT mice (p , 0.05). In contrast,
IgG2b levels compared with bleomycin-treated WT mice (p , bleomycin-treated P-selectin2/2 and E-selectin2/2 mice showed
0.05), whereas there were no significant differences in the levels of increased IgM autoantibody levels to topo I and increased IgG
other isotypes among bleomycin-treated P-selectin2/2, E-selectin2/2, autoantibody levels to topo I relative to bleomycin-treated WT mice
and WT mice. In E-selectin2/2 mice treated with anti–P-selectin (p , 0.05). Bleomycin-treated E-selectin2/2 mice treated with
mAb and in PSGL-12/2 mice, bleomycin administration increased anti–P-selectin mAb and PSGL-12/2 mice had increased IgM
serum levels of all Ig isotypes, except IgA, compared with autoantibody levels to topo I, as well as IgG autoantibody levels
bleomycin-treated WT mice (p , 0.05). Thus, treatment with bleo- to topo I compared with bleomycin-treated WT mice (p , 0.05).
mycin induced hypergammaglobulinemia, which was inhibited Furthermore, IgG anti-topo I Ab production in E-selectin2/2 mice
by L-selectin and/or ICAM-1 loss, whereas the loss of P-selectin, treated with anti–P-selectin mAb and PSGL-12/2 mice was greater
E-selectin with or without P-selectin blockade, or PSGL-1 aug- than those in P-selectin2/2 and E-selectin2/2 mice (p , 0.05).
mented Ig production. Thus, bleomycin-induced production of various autoantibodies,
especially SSc-specific anti-topo I Ab, decreased with L-selectin
L-selectin and/or ICAM-1 loss inhibited autoantibody and/or ICAM-1 deficiency, whereas it increased by P-selectin loss,
production in bleomycin-treated mice, whereas P-selectin loss, E-selectin loss with or without P-selectin blockade, or PSGL-1 loss.
E-selectin loss with or without P-selectin blockade, and
PSGL-1 loss increased it Discussion
Anti-nuclear Abs were rarely detectable in PBS-treated mutant and Cell adhesion molecules play a critical role in the development of
WT mice (6%, 1/18). Anti-nuclear Abs with a homogeneous chro- several inflammatory diseases (7). In general, inhibition or loss of
mosomal staining pattern were detected in 47% (16/34) of bleomycin- cell adhesion molecules attenuates inflammatory response in many
treated WT mice, which was similar to that in bleomycin-treated in vivo experimental models (14, 16–18). Dermal sclerosis of the
P-selectin2/2 mice (41%, 14/34), E-selectin2/2 mice (44%, 15/ tight-skin mouse model, a genetic model for SSc, and pulmonary
34), E-selectin2/2 mice with blockade of P-selectin (53%, 18/34), fibrosis induced by intratracheal bleomycin treatment are almost
and PSGL-12/2 mice (53%, 18/34). In contrast, the frequencies completely eliminated by loss of both L-selectin and ICAM-1,
of anti-nuclear Ab positivity were lower in bleomycin-treated L- whereas loss of L-selectin or ICAM-1 alone results in less inhibi-
selectin2/2 (29%, 10/34), ICAM-12/2 (26%, 9/34), and L-selectin/ tion (14, 18). Similarly, cutaneous contact hypersensitivity response
ICAM-12/2 (12%, 4/34) mice. Autoantibody specificities were and TNF-a–induced leukocyte rolling are almost completely inhi-
further assessed by ELISA (Fig. 8). The dilution of sera giving bited by loss of both E-selectin and P-selectin, while loss of each
The Journal of Immunology 2511

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

FIGURE 7. Serum Ig levels in bleomycin-treated WT mice, L-selectin2/2 mice (L2/2), ICAM-12/2 mice (ICAM-12/2), L-selectin/ICAM-12/2 mice
(L/ICAM-12/2), P-selectin2/2 mice (P2/2), E-selectin2/2 mice (E2/2), E-selectin2/2 mice treated with anti–P-selectin mAb (E2/2 + anti-P Ab), and
PSGL-12/2 mice (PSGL-12/2). Serum Ig levels in PBS-treated WT mice were used as control Ig levels. Serum Ig levels were determined by isotype-
specific ELISA. Horizontal bars represent mean Ig levels. pp , 0.05; ppp , 0.01 versus each CTL group. †p , 0.05; ††p , 0.01 versus bleomycin-treated
WT mice. CTL, control.

molecule alone leads to less inhibition (19, 20). In the present levels of these cytokines in cell adhesion molecule-deficient mice
study, L-selectin and/or ICAM-1 deficiency inhibited the develop- treated with bleomycin (Fig. 3). Lack of L-selectin and/or
ment of dermal and pulmonary fibrosis with decreased inflamma- ICAM-1 reduced IL-4, IL-6, IL-17, and TGF-b1 expression and
tory cell infiltration in the bleomycin-induced SSc model (Figs. increased IFN-g expression in parallel with inhibited dermal
1, 2). The inhibitory effect of both L-selectin and ICAM-1 defi- sclerosis and pulmonary fibrosis. P-selectin loss, E-selectin loss
ciency was significantly greater than that of L-selectin or ICAM-1 with or without P-selectin blockade, or PSGL-1 loss reduced IFN-g
deficiency alone. Unexpectedly, P-selectin deficiency, E-selectin and increased IL-4, IL-6, IL-17, and TGF-b1, which was
deficiency with or without P-selectin blockade, and PSGL-1 de- accompanied by increased dermal and pulmonary fibrosis. Thus,
ficiency exacerbated inflammatory cell infiltration induced by bleo- the expression of these cell adhesion molecules alters the cytokine
mycin treatment, resulting in deteriorated dermal sclerosis and production, which may contribute to the development of skin and
pulmonary fibrosis, more severe histologic change, and increased lung fibrosis induced by bleomycin treatment.
dermal and pulmonary collagen deposition (Figs. 1, 2). The deteri- It is possible that loss of cell adhesion molecule function se-
oration effect of E-selectin deficiency with P-selectin blockade and lectively alters the trafficking pattern of fibrogenic Th2 and Th17
PSGL-1 deficiency was significantly greater than that of P-selectin cells and antifibrogenic Th1 cells to the skin and lung. This may
or E-selectin deficiency alone. Collectively, the results of the present result in differential production of cytokines, which may then
study are the first to reveal cooperative and deterioration roles directly or indirectly influence the development of dermal sclerosis,
of L-selectin and ICAM-1 and synergic and inhibitory roles pulmonary fibrosis, and inflammatory cell infiltration. Previous
of P-selectin, E-selectin, and PSGL-1 in the development of bleo- studies have demonstrated that loss of L-selectin and/or ICAM-1
mycin-induced fibrosis. function inhibits lung fibrosis induced by intratracheal bleomycin
Previous studies have demonstrated fibrotic effect of Th2 cyto- treatment (14), whereas loss of P-selectin and/or E-selectin func-
kines, such as IL-4 and IL-6 on dermal sclerosis and pulmonary fi- tion augments it (23). These studies suggest that the mechanisms
brosis (14, 18, 31). Th17 cytokines, such as IL-17, also have by which cell adhesion molecules regulate tissue fibrosis could be
a fibrogenic effect on dermal, pulmonary, and cardiac fibroblasts through the regulation of Th1 and Th2 cell infiltration. Two stud-
(41, 42). Indeed, SSc patients exhibit elevated serum levels of these ies have demonstrated that loss of L-selectin and/or ICAM-1 func-
Th2 and Th17 cytokines (5, 6, 43, 44). These cytokines promote tion inhibited Th2 cell immigration, inducing Th1 cell infiltration
collagen synthesis of fibroblasts via TGF-b1 and COL1A2 in the allergic lung disease model (46, 47). Additionally, loss of
signaling (41). Some studies have also shown that IFN-g, one of P-selectin, E-selectin, and/or PSGL-1 function prevents Th1 cell
the Th1 cytokines, has an antifibrotic effect on skin and pulmonary infiltration and induces Th2 cell immigration in the cutaneous
fibrosis (36, 45). Although these findings suggest that Th1, Th2, delayed-type hypersensitivity and allergic lung disease models
and Th17 responses are associated with disease activity in SSc (11, 48). Consistent with these findings, the results of this study
patients, the mechanism of fibrosis in SSc remains unclear. In show that lack of L-selectin and/or ICAM-1 inhibited Th2 cell
this study, we showed that IL-4 or IL-17 stimulation increased infiltration and induced Th1 cell immigration into BAL fluid,
proliferation and TGF-b1 and COL1A2 expression of dermal whereas P-selectin loss, E-selectin loss with or without P-
fibroblasts obtained from WT mice, whereas IFN-g inhibited it selectin blockade, or PSGL-1 loss decreased Th1 cell infiltration
(Fig. 4). The results of this study indicate differential expression and induced Th2 cell infiltration in the bleomycin-induced SSc
2512 CELL ADHESION MOLECULES REGULATE Th CELL BALANCE

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

FIGURE 8. Autoantibody levels in sera from bleomycin-treated WT mice, L-selectin2/2 mice (L2/2), ICAM-12/2 mice (ICAM-12/2), L-selectin/
ICAM-12/2 mice (L/ICAM-12/2), P-selectin2/2 mice (P2/2), E-selectin2/2 mice (E2/2), E-selectin2/2 mice treated with anti–P-selectin mAb (E2/2 +
anti-P Ab), and PSGL-12/2 mice (PSGL-12/2). Autoantibody levels in sera from PBS-treated WT mice were used as control. Relative autoantibody levels
were determined by Ig subclass-specific ELISA. Values in parentheses represent the dilutions of pooled sera giving half-maximal OD values in autoantigen-
specific ELISA, which were determined by linear regression analysis to generate arbitrary units per milliliter that could be directly compared between each
group of mice (n = 6 for each). Horizontal bars represent mean Ab levels. pp , 0.05; ppp , 0.01 versus each CTL group. †p , 0.05; ††p , 0.01 versus
bleomycin-treated WT mice. CTL, control.

model (Fig. 5). In contrast, other studies showed that L-selectin selectin blockade, and PSGL-1 loss induced Th17 cell migration
and/or ICAM-1 loss decreased Th1 cell infiltration and induced (Fig. 5). Consistent with these findings, a recent study has sug-
Th2 cell infiltration on inflamed endothelium induced by Th2- gested that loss of L-selectin and/or ICAM-1 function inhibited
delived factors (26, 27). Loss of P-selectin, E-selectin, and/or Th17 cell infiltration to the inflammatory sites in the experimental
PSGL-1 function attenuated Th2 cell tethering and rolling on colitis model (29, 50). Furthermore, P-selectin and E-selectin de-
endothelium and epithelium in the intestinal inflammation model ficiency induces Th17 cells in the neutrophila condition induced
(28, 49). Thus, it is likely that each cell adhesion molecule regu- by G-CSF treatment (51). Collectively, in the bleomycin-induced
lates Th cell balance in several ways, according to the tissue site SSc model, loss of L-selectin and/or ICAM-1 function may lead to
and nature of the inflammation stimuli. Additionally, loss of L- dominant Th1 cell infiltration in parallel with decreased skin and
selectin and/or ICAM-1 decreased Th17 cell infiltration, lung fibrosis, whereas P-selectin loss, E-selectin loss with or
whereas P-selectin loss, E-selectin loss with or without P- without P-selectin blockade, and PSGL-1 loss may induce
The Journal of Immunology 2513

dominant Th2 and Th17 cell infiltration that is accompanied by switching to IgG2a. In contrast, Th2 cells provided efficient help
deteriorated skin and lung fibrosis. for B cell activation and class switching to IgG1 (56, 57). IFN-g
Our flow cytometric analysis of polarized CD4+ T cells promotes IgG2a class switching, and IL-4 promotes IgG1 class
exhibited differential expression of LFA-1 and PSGL-1 and bind- switching (56–58). Therefore, concentrations of serum IgG2a and
ing ability for ICAM-1, P-selectin, and E-selectin, whereas the IgG1 reflect Th1 and Th2 responses in vivo (56–59). However, in
frequencies of LFA-1+ or PSGL-1+ cells were similar among this study, serum levels of each Ig isotype were associated with
Th0, Th1, Th2, and Th17 cells (Fig. 6). Previously, blocking tissue damage, such as inflammatory cell infiltration and fibrosis,
studies using anti–LFA-1 or PSGL-1 mAbs have suggested that rather than with Th1 and/or Th2 cytokine production. It has been
Th cell migration mainly depends on LFA-1 or PSGL-1, although previously hypothesized that immune responses to autoantigens
virtually all lymphocytes express LFA-1 and PSGL-1 (11, 46). are induced by cryptic self-epitopes that are generated by the
Although the mechanism by which Th cell infiltration is regulated modification of self-Ags (60). The exposure of cryptic self-
by LFA-1 and PSGL-1 remains unclear, our present study is the epitopes activates potentially autoreactive T cells that have not
first to suggest that expression intensity of LFA-1 and PSGL-1 previously encountered the cryptic self, thereby breaking T cell
contributes to Th cell infiltration. LFA-1 was highly expressed by tolerance. In this regard, bleomycin-induced tissue injury has been

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

Th2 and Th17 cells compared with Th0 cells, whereas Th1 cells shown to induce modification of the self-Ags, such as Fas-
showed decreased intensity levels of LFA-1 expression. Th1 cells dependent and/or oxygen species-induced metal-dependent cleav-
exhibited increased expression intensity of PSGL-1, whereas age of topo I (61). Moreover, apoptosis is detected in the skin of
Th2 and Th17 cells showed lower intensity levels of PSGL-1 ex- human SSc patients and bleomycin-induced SSc model mice,
pression. Similarly, ICAM-1 binding ability of Th2 and Th17 cells which were associated with the severity of tissue damage (61,
was higher than that of Th0 cells, while Th1 cells showed lower 62). Therefore, the production of anti-topo I Ab may be related
binding ability for ICAM-1. Th1 cells exhibited higher P-selectin to the modification of topo I by bleomycin-induced tissue injury.
and E-selectin binding ability, whereas Th2 and Th17 cells showed In this study, L-selectin and/or ICAM-1 deficiency reduced bleo-
lower binding ability for P-selectin and E-selectin. These expression mycin-induced tissue injury, which paralleled with decreased
and binding ability patterns may in part explain why Th2 and autoantibody and Ig production (Figs. 1, 2, 7, 8). In contrast,
Th17 cell infiltration was inhibited by ICAM-1 deficiency and P-selectin deficiency, E-selectin deficiency with or without P-
why Th1 cell migration was suppressed by E-selectin, P-selectin, selectin blockade, and PSGL-1 deficiency exacerbated bleomycin-
or PSGL-1 deficiency (Fig. 5). induced tissue injury, which was accompanied by increased antoan-
L-selectin and LFA-1 play important roles in Ag sensitization, tibody and Ig production (Figs. 1, 2, 7, 8). Thus, these regulations of
because these adhesion molecules mediate naive T cell migration tissue injury partially explain that cell adhesion molecule deficiency
into the draining lymph nodes (52, 53). Furthermore, these adhe- contribute to autoantibody and Ig production in the bleomycin-
sion molecules are one of the costimulatory molecules on the induced SSc model.
surface of APCs (52–54). In this study, we used the bleomycin- To date, there have been few studies addressing an in vivo role of
induced SSc model, which does not require Ag sensitization (30, cell adhesion molecules in SSc models. This is the first systematic
31). Therefore, we can exclude the possible role of these adhesion study to reveal relative contribution of adhesion molecules, L-
molecules on the sensitization phase. During T cell differentiation, selectin, ICAM-1, LFA-1, P-selectin, E-selectin, and PSGL-1 in
activated L-selectin is shed from the cell surface by proteolytic the development of dermal sclerosis and pulmonary fibrosis in
cleavage by an as yet unidentified membrane-bound metallopro- the bleomycin-induced SSc mouse model. These results provide
tease (“sheddase”) (55). Indeed, in this study, Th1, Th2, and Th17 additional clues to understanding the complexity of the pathogen-
cells did not express L-selectin. However, previous studies have esis of SSc.
shown that L-selectin plays important roles in passive Arthus re-
action and intratracheal bleomycin-induced pulmonary fibrosis, Acknowledgments
which do not also require Ag sensitization (14, 21). Additionally, We thank M. Yozaki, A. Usui, K. Shimoda, Y. Yamada, and M. Matsubara
rapidly activated L-selectin shedding enhances LFA-1 expression for technical assistance.
and the LFA-1/ICAM-1 binding ability (7, 55). Therefore,
LFA-1 expression intensity also reflects L-selectin activity, which
Disclosures
could account for the diminishing effect of L-selectin deficiency
The authors have no financial conflicts of interest.
on Th2 and Th17 immigration. Collectively, the results of this
study suggest that Th1 cell infiltration is mainly regulated by
PSGL-1 and PSGL-1 counterreceptors, P-selectin and E-selectin, References
whereas Th2 and Th17 cell infiltration is mainly controlled by 1. LeRoy, E. C., C. Black, R. Fleischmajer, S. Jablonska, T. Krieg, T. A. Medsger,
L-selectin, LFA-1, and an LFA-1 counterreceptor, ICAM-1. Fur- Jr., N. Rowell, and F. Wollheim. 1988. Scleroderma (systemic sclerosis): clas-
thermore, in bleomycin-treated L-selectin2/2, ICAM-12/2, and
sification, subsets and pathogenesis. J. Rheumatol. 15: 202–205.
2. Kuwana, M., Y. Okano, J. Kaburaki, and H. Inoko. 1996. Clinical correlations
L-selectin/ICAM-12/2 mice, the reduced rate of Th17 cell with HLA type in Japanese patients with connective tissue disease and anti-U1
infiltration was greater than those of Th2 cell infiltration. This small nuclear RNP antibodies. Arthritis Rheum. 39: 938–942.
3. Sato, S., M. Fujimoto, M. Hasegawa, and K. Takehara. 2004. Altered blood
may reflect that Th17 cell infiltration strongly depends on B lymphocyte homeostasis in systemic sclerosis: expanded naive B cells and
LFA-1 compared with Th2 cell infiltration. diminished but activated memory B cells. Arthritis Rheum. 50: 1918–1927.
4. Tashkin, D. P., R. Elashoff, P. J. Clements, J. Goldin, M. D. Roth, D. E. Furst,
As we reported previously, bleomycin treatment induced the E. Arriola, R. Silver, C. Strange, M. Bolster, et al; Scleroderma Lung Study
production of autoantibodies, especially SSc-specific anti-topo I Research Group. 2006. Cyclophosphamide versus placebo in scleroderma lung
Ab, and hypergammaglobulinemia, both of which are central fea- disease. N. Engl. J. Med. 354: 2655–2666.
5. Needleman, B. W., F. M. Wigley, and R. W. Stair. 1992. Interleukin-1, interleukin-
tures of human SSc (31). In this study, L-selectin and/or ICAM-1 2, interleukin-4, interleukin-6, tumor necrosis factor a, and interferon-g levels in
deficiency reduced these autoimmune abnormalities, whereas sera from patients with scleroderma. Arthritis Rheum. 35: 67–72.
P-selectin deficiency, E-selectin deficiency with or without P- 6. Murata, M., M. Fujimoto, T. Matsushita, Y. Hamaguchi, M. Hasegawa,
K. Takehara, K. Komura, and S. Sato. 2008. Clinical association of serum in-
selectin blockade, and PSGL-1 deficiency deteriorated it (Figs. terleukin-17 levels in systemic sclerosis: is systemic sclerosis a Th17 disease? J.
7, 8). Recent studies showed that Th1 cells supported Ig isotype Dermatol. Sci. 50: 240–242.
2514 CELL ADHESION MOLECULES REGULATE Th CELL BALANCE

7. Springer, T. A. 1994. Traffic signals for lymphocyte recirculation and leukocyte 32. Arbonés, M. L., D. C. Ord, K. Ley, H. Ratech, C. Maynard-Curry, G. Otten,
emigration: the multistep paradigm. Cell 76: 301–314. D. J. Capon, and T. F. Tedder. 1994. Lymphocyte homing and leukocyte rolling
8. Tedder, T. F., X. Li, and D. A. Steeber. 1999. The selectins and their ligands: and migration are impaired in L-selectin-deficient mice. Immunity 1: 247–260.
adhesion molecules of the vasculature. Adv. Mol. Cell Biol. 28: 65–111. 33. Ashcroft, T., J. M. Simpson, and V. Timbrell. 1988. Simple method of estimating
9. Varki, A. 1994. Selectin ligands. Proc. Natl. Acad. Sci. USA 91: 7390–7397. severity of pulmonary fibrosis on a numerical scale. J. Clin. Pathol. 41: 467–470.
10. McEver, R. P., and R. D. Cummings. 1997. Role of PSGL-1 binding to selectins 34. Domeij, H., T. Modéer, H. C. Quezada, and T. Yucel-Lindberg. 2005. Cell
in leukocyte recruitment. J. Clin. Invest. 100(Suppl.): S97–S103. expression of MMP-1 and TIMP-1 in co-cultures of human gingival fibroblasts
11. Borges, E., W. Tietz, M. Steegmaier, T. Moll, R. Hallmann, A. Hamann, and and monocytes: the involvement of ICAM-1. Biochem. Biophys. Res. Commun.
D. Vestweber. 1997. P-selectin glycoprotein ligand-1 (PSGL-1) on T helper 1 but 338: 1825–1833.
not on T helper 2 cells binds to P-selectin and supports migration into inflamed 35. Garcı́a-Ramallo, E., T. Marques, N. Prats, J. Beleta, S. L. Kunkel, and
skin. J. Exp. Med. 185: 573–578. N. Godessart. 2002. Resident cell chemokine expression serves as the major
12. Asa, D., L. Raycroft, L. Ma, P. A. Aeed, P. S. Kaytes, A. P. Elhammer, and mechanism for leukocyte recruitment during local inflammation. J. Immunol.
J. G. Geng. 1995. The P-selectin glycoprotein ligand functions as a common 169: 6467–6473.
human leukocyte ligand for P- and E-selectins. J. Biol. Chem. 270: 11662–11670. 36. Kimura, T., Y. Ishii, Y. Morishima, A. Shibuya, K. Shibuya, M. Taniguchi,
13. Dustin, M. L., R. Rothlein, A. K. Bhan, C. A. Dinarello, and T. A. Springer. M. Mochizuki, A. E. Hegab, T. Sakamoto, A. Nomura, and K. Sekizawa. 2004.
1986. Induction by IL 1 and interferon-g: tissue distribution, biochemistry, and Treatment with a-galactosylceramide attenuates the development of bleomycin-
function of a natural adherence molecule (ICAM-1). J. Immunol. 137: 245–254. induced pulmonary fibrosis. J. Immunol. 172: 5782–5789.
14. Hamaguchi, Y., Y. Nishizawa, M. Yasui, M. Hasegawa, Y. Kaburagi, K. Komura, 37. Gao, Y., N. Li, R. Fei, Z. Chen, S. Zheng, and X. Zeng. 2005. P-selectin-
T. Nagaoka, E. Saito, Y. Shimada, K. Takehara, et al. 2002. Intercellular adhe- mediated acute inflammation can be blocked by chemically modified heparin,

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021

sion molecule-1 and L-selectin regulate bleomycin-induced lung fibrosis. Am. J. RO-heparin. Mol. Cells 19: 350–355.
Pathol. 161: 1607–1618. 38. Silver, R. M., K. S. Miller, M. B. Kinsella, E. A. Smith, and S. I. Schabel. 1990.
15. Matsuse, T., S. Teramoto, H. Katayama, E. Sudo, H. Ekimoto, H. Mitsuhashi, Evaluation and management of scleroderma lung disease using bronchoalveolar
Y. Uejima, Y. Fukuchi, and Y. Ouchi. 1999. ICAM-1 mediates lung leukocyte lavage. Am. J. Med. 88: 470–476.
recruitment but not pulmonary fibrosis in a murine model of bleomycin-induced 39. Nakagome, K., M. Dohi, K. Okunishi, R. Tanaka, J. Miyazaki, and
lung injury. Eur. Respir. J. 13: 71–77. K. Yamamoto. 2006. In vivo IL-10 gene delivery attenuates bleomycin induced
16. Yanaba, K., K. Komura, M. Horikawa, Y. Matsushita, K. Takehara, and S. Sato. pulmonary fibrosis by inhibiting the production and activation of TGF-b in the
2004. P-selectin glycoprotein ligand-1 is required for the development of cuta- lung. Thorax 61: 886–894.
neous vasculitis induced by immune complex deposition. J. Leukoc. Biol. 76: 40. Braun, R. K., C. Ferrick, P. Neubauer, M. Sjoding, A. Sterner-Kock, M. Kock,
374–382. L. Putney, D. A. Ferrick, D. M. Hyde, and R. B. Love. 2008. IL-17 producing
17. Shimada, Y., M. Hasegawa, Y. Kaburagi, Y. Hamaguchi, K. Komura, E. Saito, gammadelta T cells are required for a controlled inflammatory response after
K. Takehara, D. A. Steeber, T. F. Tedder, and S. Sato. 2003. L-selectin or bleomycin-induced lung injury. Inflammation 31: 167–179.
ICAM-1 deficiency reduces an immediate-type hypersensitivity response by 41. Sakkas, L. I., I. C. Chikanza, and C. D. Platsoucas. 2006. Mechanisms of dis-
preventing mast cell recruitment in repeated elicitation of contact hypersensitiv- ease: the role of immune cells in the pathogenesis of systemic sclerosis. Nat.
ity. J. Immunol. 170: 4325–4334. Clin. Pract. Rheumatol. 2: 679–685.
18. Matsushita, Y., M. Hasegawa, T. Matsushita, M. Fujimoto, M. Horikawa, 42. Venkatachalam, K., S. Mummidi, D. M. Cortez, S. D. Prabhu, A. J. Valente, and
T. Fujita, A. Kawasuji, F. Ogawa, D. A. Steeber, T. F. Tedder, et al. 2007. In- B. Chandrasekar. 2008. Resveratrol inhibits high glucose-induced PI3K/Akt/
tercellular adhesion molecule-1 deficiency attenuates the development of skin ERK-dependent interleukin-17 expression in primary mouse cardiac fibroblasts.
fibrosis in tight-skin mice. J. Immunol. 179: 698–707. Am. J. Physiol. 294: H2078–H2087.
19. Frenette, P. S., T. N. Mayadas, H. Rayburn, R. O. Hynes, and D. D. Wagner. 43. Roumm, A. D., T. L. Whiteside, T. A. Medsger, Jr., and G. P. Rodnan. 1984.
1996. Susceptibility to infection and altered hematopoiesis in mice deficient in Lymphocytes in the skin of patients with progressive systemic sclerosis: quan-
both P- and E-selectins. Cell 84: 563–574. tification, subtyping, and clinical correlations. Arthritis Rheum. 27: 645–653.
20. Bullard, D. C., E. J. Kunkel, H. Kubo, M. J. Hicks, I. Lorenzo, N. A. Doyle, 44. Deleuran, B., and D. J. Abraham. 2007. Possible implication of the effector
C. M. Doerschuk, K. Ley, and A. L. Beaudet. 1996. Infectious susceptibility and CD4+ T-cell subpopulation TH17 in the pathogenesis of systemic scleroderma.
severe deficiency of leukocyte rolling and recruitment in E-selectin and P- Nat. Clin. Pract. Rheumatol. 3: 682–683.
selectin double mutant mice. J. Exp. Med. 183: 2329–2336. 45. Giacomelli, R., P. Cipriani, A. Fulminis, G. Barattelli, M. Matucci-Cerinic,
21. Yanaba, K., Y. Kaburagi, K. Takehara, D. A. Steeber, T. F. Tedder, and S. Sato. S. D’Alò, G. Cifone, and G. Tonietti. 2001. Circulating g/d T lymphocytes
2003. Relative contributions of selectins and intercellular adhesion molecule-1 to from systemic sclerosis (SSc) patients display a T helper (Th) 1 polarization.
tissue injury induced by immune complex deposition. Am. J. Pathol. 162: 1463– Clin. Exp. Immunol. 125: 310–315.
1473. 46. Lee, S. H., J. E. Prince, M. Rais, F. Kheradmand, C. M. Ballantyne, G. Weitz-
22. Bullard, D. C., J. M. Mobley, J. M. Justen, L. M. Sly, J. G. Chosay, C. J. Dunn, Schmidt, C. W. Smith, and D. B. Corry. 2008. Developmental control of integrin
J. R. Lindsey, A. L. Beaudet, and N. D. Staite. 1999. Acceleration and increased expression regulates Th2 effector homing. J. Immunol. 180: 4656–4667.
severity of collagen-induced arthritis in P-selectin mutant mice. J. Immunol. 163: 47. Matsuzaki, S., K. Shinozaki, N. Kobayashi, and K. Agematsu. 2005. Polarization
2844–2849. of Th1/Th2 in human CD4 T cells separated by CD62L: analysis by transcription
23. Horikawa, M., M. Fujimoto, M. Hasegawa, T. Matsushita, Y. Hamaguchi, factors. Allergy 60: 780–787.
A. Kawasuji, Y. Matsushita, T. Fujita, F. Ogawa, K. Takehara, et al. 2006. E- 48. Austrup, F., D. Vestweber, E. Borges, M. Löhning, R. Bräuer, U. Herz, H. Renz,
and P-selectins synergistically inhibit bleomycin-induced pulmonary fibrosis. R. Hallmann, A. Scheffold, A. Radbruch, and A. Hamann. 1997. P- and E-
Am. J. Pathol. 169: 740–749. selectin mediate recruitment of T-helper-1 but not T-helper-2 cells into
24. Rosenkranz, A. R., D. L. Mendrick, R. S. Cotran, and T. N. Mayadas. 1999. P- inflammed tissues. Nature 385: 81–83.
selectin deficiency exacerbates experimental glomerulonephritis: a protective 49. Bonder, C. S., M. U. Norman, T. Macrae, P. R. Mangan, C. T. Weaver,
role for endothelial P-selectin in inflammation. J. Clin. Invest. 103: 649–659. D. C. Bullard, D. M. McCafferty, and P. Kubes. 2005. P-selectin can support both
25. He, X., T. R. Schoeb, A. Panoskaltsis-Mortari, K. R. Zinn, R. A. Kesterson, Th1 and Th2 lymphocyte rolling in the intestinal microvasculature. Am. J.
J. Zhang, S. Samuel, M. J. Hicks, M. J. Hickey, and D. C. Bullard. 2006. Pathol. 167: 1647–1660.
Deficiency of P-selectin or P-selectin glycoprotein ligand-1 leads to accelerated 50. Bendjelloul, F., P. Rossmann, P. Malý, V. Mandys, M. Jirkovská, L. Prokesová,
development of glomerulonephritis and increased expression of CC chemokine L. Tucková, and H. Tlaskalová-Hogenová. 2000. Detection of ICAM-1 in ex-
ligand 2 in lupus-prone mice. J. Immunol. 177: 8748–8756. perimentally induced colitis of ICAM-1-deficient and wild-type mice: an immu-
26. Savage, N. D., S. H. Harris, A. G. Rossi, B. De Silva, S. E. Howie, G. T. Layton, nohistochemical study. Histochem. J. 32: 703–709.
and J. R. Lamb. 2002. Inhibition of TCR-mediated shedding of L-selectin 51. Stark, M. A., Y. Huo, T. L. Burcin, M. A. Morris, T. S. Olson, and K. Ley. 2005.
(CD62L) on human and mouse CD4+ T cells by metalloproteinase inhibition: Phagocytosis of apoptotic neutrophils regulates granulopoiesis via IL-23 and IL-
analysis of the regulation of Th1/Th2 function. Eur. J. Immunol. 32: 2905–2914. 17. Immunity 22: 285–294.
27. Thauland, T. J., Y. Koguchi, S. A. Wetzel, M. L. Dustin, and D. C. Parker. 2008. 52. Van Seventer, G. A., Y. Shimizu, K. J. Horgan, and S. Shaw. 1990. The
Th1 and Th2 cells form morphologically distinct immunological synapses. J. LFA-1 ligand ICAM-1 provides an important costimulatory signal for T cell
Immunol. 181: 393–399. receptor-mediated activation of resting T cells. J. Immunol. 144: 4579–4586.
28. Mangan, P. R., D. O’Quinn, L. Harrington, C. S. Bonder, P. Kubes, D. F. Kucik, 53. Catalina, M. D., M. C. Carroll, H. Arizpe, A. Takashima, P. Estess, and
D. C. Bullard, and C. T. Weaver. 2005. Both Th1 and Th2 cells require P-selectin M. H. Siegelman. 1996. The route of antigen entry determines the requirement
glycoprotein ligand-1 for optimal rolling on inflamed endothelium. Am. J. for L-selectin during immune responses. J. Exp. Med. 184: 2341–2351.
Pathol. 167: 1661–1675. 54. Xu, J., I. S. Grewal, G. P. Geba, and R. A. Flavell. 1996. Impaired primary T cell
29. Lim, H. W., J. Lee, P. Hillsamer, and C. H. Kim. 2008. Human Th17 cells share responses in L-selectin-deficient mice. J. Exp. Med. 183: 589–598.
major trafficking receptors with both polarized effector T cells and FOXP3+ 55. Rizoli, S. B., O. D. Rotstein, and A. Kapus. 1999. Cell volume-dependent
regulatory T cells. J. Immunol. 180: 122–129. regulation of L-selectin shedding in neutrophils: a role for p38 mitogen-
30. Yamamoto, T., S. Takagawa, I. Katayama, K. Yamazaki, Y. Hamazaki, activated protein kinase. J. Biol. Chem. 274: 22072–22080.
H. Shinkai, and K. Nishioka. 1999. Animal model of sclerotic skin, I: Local 56. Liblau, R. S., S. M. Singer, and H. O. McDevitt. 1995. Th1 and Th2 CD4+
injections of bleomycin induce sclerotic skin mimicking scleroderma. J. Invest. T cells in the pathogenesis of organ-specific autoimmune diseases. Immunol.
Dermatol. 112: 456–462. Today 16: 34–38.
31. Yoshizaki, A., Y. Iwata, K. Komura, F. Ogawa, T. Hara, E. Muroi, M. Takenaka, 57. Snapper, C. M., and J. J. Mond. 1993. Towards a comprehensive view of im-
K. Shimizu, M. Hasegawa, M. Fujimoto, et al. 2008. CD19 regulates skin and munoglobulin class switching. Immunol. Today 14: 15–17.
lung fibrosis via Toll-like receptor signaling in a model of bleomycin-induced 58. Snapper, C. M., F. D. Finkelman, and W. E. Paul. 1988. Differential regulation of
scleroderma. Am. J. Pathol. 172: 1650–1663. IgG1 and IgE synthesis by interleukin 4. J. Exp. Med. 167: 183–196.
The Journal of Immunology 2515

59. Ishizaki, K., A. Yamada, K. Yoh, T. Nakano, H. Shimohata, A. Maeda, 61. Yamamoto, T., and K. Nishioka. 2004. Possible role of apoptosis in the
Y. Fujioka, N. Morito, Y. Kawachi, K. Shibuya, et al. 2007. Th1 and type 1 pathogenesis of bleomycin-induced scleroderma. J. Invest. Dermatol. 122: 44–
cytotoxic T cells dominate responses in T-bet overexpression transgenic mice 50.
that develop contact dermatitis. J. Immunol. 178: 605–612. 62. Sgonc, R., M. S. Gruschwitz, H. Dietrich, H. Recheis, M. E. Gershwin, and
60. Rosen, A., and L. Casciola-Rosen. 1999. Autoantigens as substrates for apoptotic G. Wick. 1996. Endothelial cell apoptosis is a primary pathogenetic event un-
proteases: implications for the pathogenesis of systemic autoimmune disease. derlying skin lesions in avian and human scleroderma. J. Clin. Invest. 98: 785–
Cell Death Differ. 6: 6–12. 792.

Downloaded from [Link] at ULB/Biblio/Fac de Med/Erasme/Bruxelles on July 22, 2021