ARTHRITIS & RHEUMATISM
Vol. 65, No. 2, February 2013, pp 492–502
Alpha2-Antiplasmin Regulates the Development of
Dermal Fibrosis in Mice by Prostaglandin F2␣
Synthesis Through Adipose Triglyceride
Lipase/Calcium-Independent Phospholipase A2
Yosuke Kanno,1 Eri Kawashita,1 Akiko Kokado,1 Kiyotaka Okada,2
Shigeru Ueshima,3 Osamu Matsuo,2 and Hiroyuki Matsuno1
Objective. Systemic sclerosis (SSc) is character-
ized by fibrosis of the skin and visceral organs. Patients
with SSc have enhanced plasma levels of the plasmin–
␣2-antiplasmin (␣2AP) complex, and we recently impli-
cated ␣2AP in the development of fibrosis through
transforming growth factor ␤ (TGF␤) production. This
study was undertaken to clarify how ␣2AP induces
TGF␤ production and the development of fibrosis.
Methods. To clarify the detailed mechanism by
which ␣2AP induces TGF␤ production, we focused on
adipose triglyceride lipase (ATGL)/calcium-independent
phospholipase A2 (iPLA2) and examined whether ATGL/
iPLA2 is associated with ␣2AP-induced TGF␤ produc-
tion. The mouse model of bleomycin-induced SSc was
used to evaluate the role of ␣2AP in the development of
fibrosis. Dermal thickness and collagen content were
determined in mouse skin treated with phosphate buff-
ered saline or bleomycin. Moreover, we cultured SSc-
like fibroblasts from the bleomycin-treated mouse skin
and examined the production of TGF␤ and prostaglan-
din F2␣ (PGF2␣).
Results. We found that ␣2AP binding to ATGL
promoted PGF2␣ synthesis through iPLA2 in fibro-
Supported by the Ministry of Education, Culture, Sports,
Science, and Technology, Japan (Grants-in-Aid for Young Scientists
B: 23790107 to Dr. Kanno).
1
Yosuke Kanno, PhD, Eri Kawashita, PhD, Akiko Kokado,
Hiroyuki Matsuno, PhD: Doshisha Women’s College of Liberal Arts,
Kyoto, Japan; 2Kiyotaka Okada, PhD, Osamu Matsuo, PhD: Kinki
University School of Medicine, Ohnohigashi, Osakasayama, Osaka,
Japan; 3Shigeru Ueshima, PhD: Kinki University School of Medicine,
Ohnohigashi, Osakasayama, Osaka, Japan, and Kinki University
School of Agriculture, Nara, Nara, Japan.
Address correspondence to Yosuke Kanno, PhD, Depart-
ment of Clinical Pathological Biochemistry, Faculty of Pharmaceutical
Science, Doshisha Women’s College of Liberal Arts, 97-1 Kodo,
Kyo-tanabe 610-0395, Kyoto, Japan. E-mail: ykanno@dwc.doshisha.ac.jp.
Submitted for publication February 22, 2012; accepted in
revised form October 18, 2012.
492
DOI 10.1002/art.37767
© 2013, American College of Rheumatology
blasts, and the PGF2␣ synthesis that was promoted by
␣2AP induced TGF␤ production in fibroblasts. In ad-
dition, the neutralization of ␣2AP attenuated the pro-
duction of TGF␤ and PGF2␣ in SSc-like fibroblasts
from mice. The ␣2AP deficiency attenuated bleomycin-
induced fibrosis and PGF2␣ synthesis, while the admin-
istration of PGF2␣ to ␣2AP-deficient mice facilitated
␣2AP deficiency–attenuated fibrosis.
Conclusion. These findings suggest that ␣2AP
regulates the development of fibrosis by PGF2␣ synthe-
sis through ATGL/iPLA2. The inhibition of ␣2AP-
initiated pathways might provide a novel therapeutic
approach to fibrotic diseases.
Systemic sclerosis (SSc) affects the skin and the
internal organs, resulting in tissue fibrosis. Although the
disease process involves immunologic mechanisms, vas-
cular damage, and activation of fibroblasts, the patho-
genesis of SSc remains to be further elucidated. Fibrotic
diseases are characterized by excessive scarring due to
excessive production, deposition, and contraction of
the extracellular matrix (ECM). This process usually
occurs over many months and years, and can lead to
organ dysfunction or death. Although it is known that
the multiple biologic actions of transforming growth
factor ␤ (TGF␤) contribute to the central role that it
plays in many fibrotic diseases (1), the regulation and
mechanism responsible for TGF␤ production in fibrotic
diseases remain poorly understood.
The ␣2-antiplasmins (␣2AP) are serpins with a
molecular weight of 65–70 kd (2) that rapidly inactivate
plasmin, resulting in the formation of a stable inactive
complex, plasmin–␣2AP (3). Many studies have shown
that the levels of the plasmin–␣2AP complex in plasma
are elevated in fibrotic diseases, including SSc, diabetic
nephropathy, liver cirrhosis, and rheumatoid arthritis
REGULATION OF DEVELOPMENT OF DERMAL FIBROSIS BY ␣2-ANTIPLASMIN
(4–7). In addition, we have shown that ␣2AP is associ-
ated with the development of fibrosis through TGF␤
production (8,9). However, the detailed mechanism by
which ␣2AP induces TGF␤ production and the devel-
opment of fibrosis is unclear.
Phylogenetically, ␣2AP is most closely related to
the noninhibitory serpin pigment epithelium–derived
factor (PEDF) (10). The structures of ␣2AP (11) and
PEDF (12) are very similar, and they both have 3
␤-sheets and 9 ␣-helices. It has recently been reported
that adipose triglyceride lipase (ATGL), which has been
described as a member of the calcium-independent
phospholipase A2 (iPLA2)/nutrin/patatin-like phospho-
lipase domain–containing protein 2 family, is a receptor
for PEDF, and that PEDF binding stimulates the enzy-
matic PLA2 of ATGL (13).
Release of arachidonic acid (AA) from mem-
brane glycerophospholipids is catalyzed by PLA2 en-
zymes, and AA is sequentially metabolized to prosta-
glandin G2 (PGG2) and then to PGH2 by cyclooxygenase
1 (COX-1) and/or COX-2. PGH2 is then converted to
various bioactive PGs (thromboxane A2, PGD2, PGE2,
PGF2␣ and PGI2) (14). The loss of PLA2 suppressed
bleomycin-induced fibrosis (15). In addition, PGF2␣,
which is one of the PGs that stimulates TGF␤ produc-
tion (16), is associated with the development of fibrosis
(17). PLA2/PGF2␣ plays an important role in the patho-
genesis of fibrotic diseases.
In this study, we focused on the PEDF receptor,
ATGL, and found that ␣2AP promoted PGF2␣ synthesis
through ATGL/iPLA2. In addition, we demonstrated
that the ␣2AP-promoted PGF2␣ synthesis regulates
TGF␤ production and the development of fibrosis.
MATERIALS AND METHODS
Animals. Deficient mice were generated by homolo-
gous recombination using embryonic stem cells, as previously
described (18,19). All experiments were performed in accor-
dance with institutional guidelines and the Guide for the Care
and Use of Laboratory Animals published by the US National
Institutes of Health.
Reagents. The ␣2AP was purchased from Calbiochem.
Other chemical substances were obtained from Sigma.
Induction of dermal fibrosis. We induced dermal
fibrosis in mice as previously described (8). Briefly, dermal
fibrosis was induced in 8-week-old male C57BL/6J mice by
administration of phosphate buffered saline (PBS) alone,
bleomycin alone, or bleomycin and the iPLA2-specific inhibitor
bromoenol lactone (BEL; Cayman Chemical) (n ⫽ 7 mice per
group). Bleomycin was dissolved in PBS at 1 mg/ml. A total of
100 ␮l of PBS, 100 ␮l of bleomycin, or 100 ␮l of bleomycin and
5 mg/kg of BEL was administered subcutaneously into the
shaved backs of the mice. In an additional experiment, dermal
fibrosis was induced in 8-week-old male ␣2AP-deficient
493
(␣2AP⫺/⫺) mice and 8-week-old wild-type mice (␣2AP⫹/⫹) by
administration of PBS alone, bleomycin alone, or bleomycin
and PGF2␣ (n ⫽ 4 ␣2AP⫺/⫺ mice and n ⫽ 4 wild-type mice per
treatment group). A total of 100 ␮l of PBS, 100 ␮l of bleomycin
alone, or 100 ␮l of bleomycin and 400 ␮g/kg of PGF2␣ was
administered subcutaneously into the shaved backs of the
mice. In both experiments, administration in the same site was
carried out daily for up to 3 weeks. At the end of different
observation periods, the mice were killed at the indicated times
using an overdose of pentobarbital, and a skin sample was
carefully collected from each mouse. These samples were used
for the protein preparations. For extraction of the protein, skin
samples, including the scab and the complete epithelial mar-
gins, were trimmed to specimens measuring 7–8 mm in diam-
eter, placed immediately in liquid nitrogen, and stored at
⫺80°C until used.
Cell culture. NIH3T3 fibroblasts (1 ⫻ 105) were
seeded into dishes measuring 35 mm in diameter and main-
tained in 2 ml of Dulbecco’s modified Eagle’s medium
(DMEM) containing 10% fetal calf serum (FCS) at 37°C in a
humidified atmosphere with 5% CO2/95% air. After 6 days,
the medium was replaced with serum-free DMEM. The cells
were then used for experiments.
Primary cultured cells. Normal fibroblasts and SSc-
like fibroblasts were obtained from the skin of wild-type mice
as previously described (20). SSc-like fibroblasts were also
obtained from the bleomycin-treated skin of wild-type mice.
Bleomycin administration in the same site was carried out daily
for up to 3 weeks.
Binding assay for ␣2AP and ATGL. Immunopre-
cipitation was performed after incubating 800 ng of ␣2AP or
bovine serum albumin (BSA; negative control) with 200 ␮g of
fibroblast lysates for 30 minutes. Immunoprecipitation was
performed using an Immunoprecipitation Kit (Invitrogen).
ATGL small interfering RNA (siRNA) experiments.
NIH3T3 cells were transfected with ATGL siRNA (Santa Cruz
Biotechnology) using Lipofectamine 2000 (Invitrogen) ac-
cording to the recommendations of the manufacturer. A
nonspecific siRNA was used as the control. Three days after
transfection, the cells were stimulated with ␣2AP and then
evaluated by Western blot analysis.
Enzyme-linked immunosorbent assay (ELISA). PGF2␣
levels in the medium or mouse skin samples were measured
using a PGF2␣ enzyme immunoassay kit (Enzo Life Sciences).
The absorbance of the ELISA samples was measured at
405 nm using an EL340 Biokinetic Reader (BioTek Instru-
ments).
Western blot analysis. Western blot analysis was un-
dertaken as previously described (21). We detected TGF␤,
ATGL, phospho-serine, phospho–ERK-1/2, phospho-JNK,
ERK-1/2, and JNK by incubation with anti-TGF␤ antibody
(R&D Systems), anti-ATGL antibody (Santa Cruz Biotech-
nology), anti–phospho-serine antibody (Invitrogen), anti–
phospho–ERK-1/2 antibody (Cell Signaling Technology),
anti–phospho-JNK antibody (Cell Signaling Technology),
anti–ERK-1/2 antibody (Cell Signaling Technology), and anti-
JNK antibody (Cell Signaling Technology), respectively, fol-
lowed by incubation with horseradish peroxidase–conjugated
antibody to rabbit IgG (Amersham Pharmacia Biotech).
Immunohistochemical staining for ␣2AP, TGF␤,
␣-smooth muscle actin (␣-SMA), and phospho-Smad2/3.
Specimens from the central area of the mouse skin treated with
494 KANNO ET AL

Figure 1. Transforming growth factor ␤ (TGF␤) production is induced by ␣2-antiplasmin (␣2AP) binding and activation of adipose triglyceride
lipase (ATGL). A, Western blot analysis (left) and densitometric analysis (right) of ATGL expression in the skin of wild-type mice treated with
phosphate buffered saline (PBS) or bleomycin (Bleo). B, Immunoprecipitation (IP) and immunoblotting (IB) of fibroblast lysates treated with ␣2AP
or bovine serum albumin (BSA). Lysates were immunoprecipitated with mouse IgG or an anti-ATGL antibody, followed by immunoblotting with
an anti-␣2AP antibody and an anti-ATGL antibody. C, IP and IB (left) and densitometric analysis (right) of phospho-serine expression in cell lysates
treated with ␣2AP for the indicated times. Lysates were immunoprecipitated with an anti-ATGL antibody, followed by immunoblotting with
anti–phospho-serine antibody. D, Western blot analysis (left) and densitometric analysis (right) of ATGL expression in fibroblasts transfected with
control or ATGL small interfering RNA (siRNA), confirming the specific depletion of ATGL. E, Western blot analysis (left) and densitometric
analysis (right) of TGF␤ expression in fibroblasts transfected with control or ATGL siRNA and stimulated with 200 nM ␣2AP for 24 hours. In A, C,
D, and E, values are the mean ⫾ SEM (n ⫽ 3). ⴱ ⫽ P ⬍ 0.01.

PBS or bleomycin were excised by using a punch biopsy
instrument (Techne). Paraffin-embedded sections were la-
beled with anti-␣2AP, TGF␤ ␣-SMA, or phospho-Smad2/3
primary antibody, then secondarily labeled with Cy3-
conjugated anti-rabbit IgG (Molecular Probes). The signals
were then detected using a laser scanning microscope. The
stained images obtained from separate fields on the specimens
(n ⫽ 4) were analyzed using ImageJ software.
Measurement of dermal thickness. The dermal thick-
ness (distance from the epidermal–dermal junction to the
dermal–muscle junction) was measured in mouse skin sections
(n ⫽ 6 mice).
REGULATION OF DEVELOPMENT OF DERMAL FIBROSIS BY ␣2-ANTIPLASMIN 495

Measurement of collagen content in mouse skin sec- ing to ATGL triggered the activation of ATGL. Mass
tions by Sircol biochemical assay. The collagen content was spectrometry analysis detected 2 phosphorylated serine
measured as previously described (22). Briefly, collagen con- residues in murine ATGL (25). We showed that ␣2AP in-
tent was assessed using sirius red staining. This approach was
chosen because it accurately reflects the collagen content duced the phosphorylation of serine in ATGL (Figure 1C).
assessed with a hydroxyproline assay and allows areas of Moreover, to clarify whether ␣2AP induces TGF␤
localized collagen accumulation to be specifically evaluated. In production through ATGL, we investigated the effects
these assays, sections of mouse skin were stained with sirius red of ␣2AP-induced TGF␤ production on the reduction of
as described by Junqueira et al (23). After deparaffinization, ATGL expression by siRNA. We confirmed that ATGL
the skin sections were treated with 0.2% phosphomolybdic acid
expression was attenuated by siRNA (Figure 1D). In ad-
for 5 minutes. Next, the skin sections were stained with 0.1%
sirius red for 90 minutes and 0.01N HCl for 2 minutes. The red dition, the siRNA-induced reduction in ATGL expres-
staining was then detected using a laser scanning microscope. sion attenuated ␣2AP-induced TGF␤ production (Figure
In each section, the sirius red–positive area was measured in 7 1E). These data suggest that ␣2AP interacts with ATGL,
randomly chosen fields and was expressed as a percent of that and that this interaction induces TGF␤ production.
observed in the PBS-treated mice. Activation of iPLA2 by ␣2AP induces TGF␤
Calcium-independent PLA2 activity. The activity of
production. ATGL has enzymatic iPLA2 activity (13).
iPLA2 in the mouse skin samples was determined as described
by Smani et al (24). The activity of iPLA2 was measured using Therefore, we examined whether iPLA2 activation is
a modified kit that was originally designed for cytosolic PLA2 associated with the ␣2AP-induced TGF␤ production in
(cPLA2) using a cPLA2 assay kit (Cayman Chemical). To fibroblasts by using an iPLA2-specific inhibitor, BEL.
detect the activity of iPLA2, and not of cPLA2, phospholipase BEL attenuated the ␣2AP-induced TGF␤ production in
activity was assayed by incubating the samples with the sub- fibroblasts (Figure 2A). It has been demonstrated that
strate arachidonoyl Thio-PC for 1 hour at 20°C in a modi-
fied calcium-free buffer (300 mM NaCl, 0.5% Triton X-100,
60% glycerol, 4 mM EGTA, 10 mM HEPES [pH 7.4], and 2 mg/
ml BSA). The reaction was stopped by the addition of 5,5⬘-
dithiobis(2-nitrobenzoic acid) for 5 minutes, and absorbance
was determined at 405 nm using an EL340 Biokinetic Reader.
The activity of iPLA2 was expressed in absorbance per milli-
grams of protein.
Statistical analysis. All data are expressed as the
mean ⫾ SEM. The statistical significance of the effect of each
treatment was determined by analysis of variance followed by
Student-Newman-Keuls test. P values less than 0.01 were
considered significant.

RESULTS
Binding and activation of ATGL by ␣2AP in-
duces TGF␤ production. In previous studies, we showed
that ␣2AP induced TGF␤ production (8,9). To clarify
the detailed mechanism underlying how ␣2AP induces
TGF␤ production, we focused on PEDF, which is closely
related to ␣2AP, and the PEDF receptor, ATGL, and
examined whether ATGL acts as a receptor for ␣2AP.
First, we confirmed that ATGL is present in the
skin of PBS-treated or bleomycin-treated wild-type mice
by a Western blot analysis. We observed that ATGL was
present in the skin, and that the expression of ATGL in
Figure 2. TGF␤ production is induced by ␣2AP through the activa-
the skin samples from bleomycin-treated mice was tion of calcium-independent phospholipase A2. A, Western blot ana-
higher than in those from PBS-treated mice (Figure 1A). lysis (left) and densitometric analysis (right) of TGF␤ production in
Second, to determine the ability of ␣2AP to bind ATGL, fibroblasts pretreated with DMSO or 10 ␮M bromoenol lactone (BEL)
we performed immunoprecipitation of ATGL from fi- for 60 minutes and then stimulated with 200 nM ␣2AP for 24 hours.
B, Western blot analysis (left) and densitometric analysis (right) of
broblast lysates in the presence and absence of recom-
TGF␤ production in fibroblasts pretreated with DMSO or 10 ␮M BEL
binant ␣2AP by using control IgG or an anti-ATGL for 60 minutes and then stimulated with 100 ng/ml pigment
antibody. We showed that ␣2AP interacted with ATGL epithelium–derived factor (PEDF) for 24 hours. Values are the
(Figure 1B). Third, we investigated whether ␣2AP bind- mean ⫾ SEM (n ⫽ 3). ⴱ ⫽ P ⬍ 0.01. See Figure 1 for other definitions.
496 KANNO ET AL

Figure 3. Prevention of bleomycin-induced fibrosis through inhibition of calcium-independent phospholipase A2. A, Representative hematoxylin-
stained and sirius red–stained skin sections from wild-type mice treated with PBS, bleomycin alone, or bleomycin and bromoenol lactone (BEL).
Arrows indicate the dermal thickness (from the epidermal–dermal junction to the dermal–muscle junction). Bars ⫽ 200 ␮m. B, Dermal thickness
in skin sections from wild-type mice treated as indicated. C, Collagen content, measured by sirius red staining, in the skin of wild-type mice treated
as indicated. The sirius red–positive area is expressed as a percent of that observed in the PBS-treated mice. D, Western blot analysis (left) and
densitometric analysis (right) of TGF␤ and ␣2AP expression in the skin of wild-type mice treated as indicated. E, Paraffin-embedded sections of
the skin of wild-type mice treated as indicated and stained with antibodies to ␣2AP, TGF␤, ␣-smooth muscle actin (␣-SMA), and phospho-Smad2/3.
The intensity of the staining was quantitatively evaluated as described in Materials and Methods. Bars ⫽ 200 ␮m. Values in B and C are the mean ⫾
SEM (n ⫽ 6); values in D and E are the mean ⫾ SEM (n ⫽ 4). ⴱ ⫽ P ⬍ 0.01. NS ⫽ not significant (see Figure 1 for other definitions).

PEDF binding stimulates the iPLA2 activity of ATGL
(13). We also showed that PEDF induced TGF␤ pro-
duction in fibroblasts, and that the PEDF-induced
TGF␤ production was attenuated by BEL (Figure 2B).
These data suggest that activation of iPLA2 induced
TGF␤ production.
Prevention of bleomycin-induced fibrosis through
inhibition of iPLA2. To verify the potential of iPLA2 as
a novel target for antifibrotic therapies in vivo, we
examined the effects of the iPLA2 inhibitor BEL on
bleomycin-induced fibrosis. The administration of BEL
attenuated bleomycin-induced fibrotic changes, such as
increased dermal thickness and collagen production, in
wild-type mice (Figures 3A–C). In addition, BEL at-
tenuated bleomycin-induced ␣-SMA (a hallmark of the
myofibroblast phenotype), TGF␤, and Smad2/3 phos-
phorylation (Figures 3D and E). However, BEL did not
attenuate bleomycin-induced ␣2AP expression (Figures
3D and E).
Activation of iPLA2 by ␣2AP induces PGF2␣
synthesis. PG synthesis is catalyzed by PLA2 activation
(26). PGF2␣ stimulates TGF␤ production (16) and is
associated with the development of fibrosis (17). There-
fore, we examined whether ␣2AP promoted PGF2␣
synthesis in fibroblasts. We observed that ␣2AP pro-
moted PGF2␣ synthesis, and that the ␣2AP-promoted
REGULATION OF DEVELOPMENT OF DERMAL FIBROSIS BY ␣2-ANTIPLASMIN
Figure 4. Activation of calcium-independent phospholipase A2 by
␣2-antiplasmin (␣2AP) induces prostaglandin F2␣ (PGF2␣) synthesis.
A, PGF2␣ expression in fibroblasts pretreated with DMSO or 10 ␮M
bromoenol lactone (BEL) for 60 minutes and then stimulated with
200 nM ␣2AP for 24 hours. B, PGF2␣ expression in fibroblasts
pretreated with DMSO, 10 ␮M SC560, or 10 ␮M NS398 for 60 minutes
and then stimulated with 200 nM ␣2AP for 24 hours. The PGF2␣
content in conditioned medium of fibroblasts was measured by
enzyme-linked immunosorbent assay as described in Materials and
Methods. Values are the mean ⫾ SEM (n ⫽ 4). ⴱ ⫽ P ⬍ 0.01;
ⴱⴱ ⫽ P ⬍ 0.05.
PGF2␣ synthesis was attenuated by BEL (Figure 4A).
These data suggest that ␣2AP promotes PGF2␣ synthesis
through iPLA2. Two isoforms of COX, COX-1 and
COX-2, are involved in the synthesis of PGs. We exam-
ined whether each of these isoforms was associated
with the ␣2AP-promoted PGF2␣ synthesis by using
selective inhibitors of COX-1 (SC560) and COX-2
(NS398). Both SC560 and NS398 reduced the ␣2AP-
497
promoted PGF2␣ synthesis, although SC560 had a
stronger effect (Figure 4B).
PGF2␣ synthesis by ␣2AP induces TGF␤ produc-
tion. We confirmed that PGF2␣ induces TGF␤ produc-
tion in fibroblasts (Figure 5A). We also examined
the PGF2␣-stimulated phosphorylation of ERK-1/2
and JNK and found that PGF2␣ activated both the
ERK-1/2 and JNK pathways in fibroblasts (Figure 5B).
In addition, we examined whether the ERK-1/2 and
JNK pathways were associated with PGF2␣-induced
TGF␤ production in fibroblasts by using a specific
inhibitor of MEK (PD98059) and a specific inhibitor of
JNK (SP600125). PD98059 and SP600125 attenuated
the PGF2␣-induced TGF␤ production in fibroblasts
(Figure 5C). Moreover, we examined whether ␣2AP-
promoted PGF2␣ synthesis is associated with TGF␤
production by using a selective FP receptor antagonist,
AL8810 (27). AL8810 attenuated the ␣2AP-induced
TGF␤ production in fibroblasts (Figure 5D). These data
suggest that ␣2AP induced TGF␤ production through
PGF2␣ synthesis. We also examined the effect of the
selective inhibitors of COX-1 (SC560) and COX-2
(NS398) on ␣2AP-induced TGF␤ production in fibro-
blasts. Both SC560 and NS398 reduced ␣2AP-induced
TGF␤ production, although SC560 had a stronger effect
(Figure 5E).
Role of ␣2AP in the production of TGF␤ and
PGF2␣ in SSc. In a previous study, we showed that the
expression of ␣2AP in the skin of mice with bleomycin-
induced SSc was significantly higher than that in the
skin of control mice (9). To clarify the role of ␣2AP
in the production of TGF␤ and PGF2␣ in SSc, we
cultured SSc-like fibroblasts from the skin of mice with
bleomycin-induced SSc and examined the production
of TGF␤ and PGF2␣. Although ␣2AP did not induce
the production of TGF␤ and PGF2␣ in the SSc-like
fibroblasts, the production of TGF␤ and PGF2␣ in the
SSc-like fibroblasts was increased dramatically com-
pared to that in normal fibroblasts (results are avail-
able from the corresponding author upon request). In
addition, the expression of ␣2AP in the SSc-like fibro-
blasts was increased dramatically compared to that in
normal fibroblasts (results are available from the corre-
sponding author upon request), and the neutralization
of ␣2AP attenuated the production of TGF␤ and PGF2␣
in the SSc-like fibroblasts (results are available from
the corresponding author upon request). These data
suggest that ␣2AP regulates the production of TGF␤
and PGF2␣ in SSc.
Effects of PGF 2 ␣ on the attenuation of
bleomycin-induced fibrosis by ␣2AP deficiency. In a
previous study, we showed that ␣2AP deficiency pro-
498 KANNO ET AL

Figure 5. TGF␤ production is induced by ␣2AP synthesis of prostaglandin F2␣ (PGF2␣). A, Western blot analysis (left) and densitometric analysis
(right) of TGF␤ production in fibroblasts stimulated with PGF2␣ (0, 0.1, or 1 ␮M) for 24 hours. B, Western blot analysis (left) and densitometric
analysis (right) of phosphorylation of ERK-1/2 and JNK in fibroblasts stimulated with 1 ␮M PGF2␣ for the indicated periods. C, Western blot analysis
(left) and densitometric analysis (right) of TGF␤ expression in fibroblasts pretreated with DMSO, 30 ␮M SP600125, or 30 ␮M PD98059 for 60
minutes and then stimulated with 1 ␮M PGF2␣ for 24 hours. D, Western blot analysis (left) and densitometric analysis (right) of TGF␤ expression
in fibroblasts pretreated with DMSO or 5 ␮M AL8810 for 60 minutes and then stimulated with 200 nM ␣2AP for 24 hours. E, Western blot analysis
(left) and densitometric analysis (right) of TGF␤ production in fibroblasts pretreated with DMSO, 10 ␮M SC560, or 10 ␮M NS398 for 60 minutes
and then stimulated with 200 nM ␣2AP for 24 hours. Values are the mean ⫾ SEM (n ⫽ 3). ⴱ ⫽ P ⬍ 0.01. See Figure 1 for other definitions.

tected mice against bleomycin-induced fibrosis (8,9).
To clarify whether the protection against bleomycin-
induced fibrosis in ␣2AP⫺/⫺ mice is associated with
iPLA2 activation and PGF2␣ synthesis, we measured
iPLA 2 activity and PGF 2 ␣ levels in the skin of
bleomycin-treated ␣2AP⫹/⫹ and ␣2AP⫺/⫺ mice. The
iPLA2 activity and PGF2␣ levels were lower in the skin of
␣2AP⫺/⫺ mice than ␣2AP⫹/⫹ mice (Figures 6A and B).
Next, we confirmed the effects of PGF2␣ on the
bleomycin-induced fibrotic changes in ␣2AP⫺/⫺ mice
by administering PGF2␣. Histologic examination of the
skin of ␣2AP⫺/⫺ mice treated with bleomycin and those
treated with PGF2␣ demonstrated a considerable in-
crease in dermal thickness in the mice treated with
PGF2␣ (Figure 6C). Quantitative analysis showed that
the dermal thickness and collagen content of the skin of
REGULATION OF DEVELOPMENT OF DERMAL FIBROSIS BY ␣2-ANTIPLASMIN 499

Figure 6. Effects of prostaglandin F2␣ (PGF2␣) on the attenuation of bleomycin-induced fibrosis by ␣2AP deficiency. A, Calcium-independent
phospholipase A2 (iPLA2) activity in skin sections from PBS-treated or bleomycin-treated ␣2AP⫹/⫹ and ␣2AP⫺/⫺ mice, measured as described in
Materials and Methods. B, PGF2␣ content in skin sections from bleomycin-treated ␣2AP⫹/⫹ and ␣2AP⫺/⫺ mice, measured by enzyme-linked
immunosorbent assay as described in Materials and Methods. C, Representative hematoxylin-stained and sirius red–stained skin sections from
␣2AP⫹/⫹ and ␣2AP⫺/⫺ mice treated with PBS, bleomycin alone, or bleomycin and PGF2␣. Arrows indicate the dermal thickness (from the
epidermal–dermal junction to the dermal–muscle junction). Bars ⫽ 200 ␮m. D, Dermal thickness in skin sections from ␣2AP⫹/⫹ and ␣2AP⫺/⫺ mice
treated as indicated. E, Collagen content, measured by sirius red staining, in the skin of ␣2AP⫹/⫹ and ␣2AP⫺/⫺ mice treated as indicated. The sirius
red–positive area is expressed as a percent of that observed in the PBS-treated ␣2AP⫹/⫹ mice. F, Western blot analysis (left) and densitometric
analysis (right) of TGF␤ expression in the skin of ␣2AP⫹/⫹ and ␣2AP⫺/⫺ mice treated as indicated. G, Paraffin-embedded sections of the skin of
␣2AP⫹/⫹ and ␣2AP⫺/⫺ mice treated as indicated and stained with antibodies to TGF␤, ␣-smooth muscle actin (␣-SMA), and phospho-Smad2/3. The
intensity of the staining was quantitatively evaluated as described in Materials and Methods. Bars ⫽ 200 ␮m. Values in A, B, E, and G are the mean ⫾
SEM (n ⫽ 4); values in D are the mean ⫾ SEM (n ⫽ 6); values in F are the mean ⫾ SEM (n ⫽ 3). ⴱ ⫽ P ⬍ 0.01; ⴱⴱ ⫽ P ⬍ 0.05. Abs ⫽ absorbance;
NS ⫽ not significant (see Figure 1 for other definitions).

␣2AP⫺/⫺ mice treated with bleomycin and PGF2␣ was with bleomycin alone (Figures 6C–E). In addition, the
increased compared with that of ␣2AP⫺/⫺ mice treated skin of ␣2AP⫺/⫺ mice treated with both bleomycin and
500
PGF2␣ showed increased expression of TGF␤, ␣-SMA,
and phospho-Smad2/3 compared to the skin of ␣2AP⫺/⫺
mice treated with bleomycin alone (Figures 6F and G).
These data suggest that PGF2␣ is associated with the
protection of mice against bleomycin-induced fibrosis
that results from ␣2AP deficiency.
DISCUSSION
TGF␤ has long been known to promote fibrosis
by the induction of ECM synthesis and contraction in
fibroblasts. In previous studies, we showed that ␣2AP
is associated with the development of fibrosis through
TGF␤ production (8,9). However, the mechanisms by
which ␣2AP regulated TGF␤ production and the de-
velopment of fibrosis were not precisely understood.
The structures of ␣2AP (11) and PEDF (12) are very
similar. In addition, some studies have demonstrated
that PEDF inhibits vascular endothelial growth factor
(VEGF) expression and angiogenesis (28–30), and we
previously showed that ␣2AP deficiency induced VEGF
expression and angiogenesis (20). These findings suggest
that ␣2AP and PEDF might have the same function.
Moreover, PEDF is involved in fibrosis (31). Therefore,
in the present study, we focused on PEDF and the
PEDF receptor, ATGL, and found that ␣2AP could
bind and activate ATGL. In addition, the reduction of
ATGL by siRNA attenuated ␣2AP-induced TGF␤ pro-
duction. These data suggest that ␣2AP induces TGF␤
production through ATGL.
The ligand binding to ATGL has been reported
to stimulate enzymatic iPLA2 activity (13). The ␣2AP-
induced TGF␤ production observed in this study was
attenuated by the inhibition of iPLA2. We also showed
that the inhibition of iPLA2 attenuated the bleomycin-
induced production of TGF␤, and protected mice
against fibrosis. However, inhibition of iPLA2 did not
attenuate bleomycin-induced ␣2AP expression. These
data suggest that ␣2AP is an upstream regulator of
iPLA2, and that the ␣2AP-mediated activation of iPLA2
is associated with TGF␤ production and the develop-
ment of fibrosis.
It has been reported that PGF2␣ stimulates
TGF␤ production (16) and plays a pivotal role in the
development of fibrosis (17). The synthesis of PGF2␣ is
involved in PLA2 activation and 2 isoforms of COX:
COX-1 and COX-2. We showed that ␣2AP-mediated
activation of iPLA2 is associated with PGF2␣ synthesis by
using a specific iPLA2 inhibitor. In addition, the selective
inhibitors of COX-1 and COX-2 reduced ␣2AP-
promoted PGF2␣ synthesis. Moreover, we showed that
␣2AP-induced TGF␤ production was attenuated by a
KANNO ET AL
specific iPLA2 inhibitor, a selective inhibitor of COX,
and a selective FP receptor antagonist. These data
suggest that ␣2AP induces TGF␤ production through
iPLA2 activation, COX, and PGF2␣ synthesis.
The effect of a specific COX-1 inhibitor was
significantly stronger than that of a specific COX-2
inhibitor. PGF2␣ has been reported to induce COX-2
expression (32,33). PGF2␣ synthesis was promoted by
␣2AP through iPLA2/COX-1, and ␣2AP-promoted
PGF2␣ induced TGF␤ production. At the same time, the
␣2AP-promoted PGF2␣ might induce COX-2 expres-
sion, and thereby enhance PGF2␣ synthesis. The en-
hancement of PGF2␣ synthesis might then lead to the
further induction of TGF␤ production. We also showed
that PGF2␣ induced TGF␤ production through the
ERK-1/2 and JNK pathways. Both the ERK-1/2 and
JNK pathways are associated with ␣2AP-induced TGF␤
production (9). These data suggest that ␣2AP-mediated
PGF 2 ␣ synthesis might induce TGF ␤ production
through the ERK-1/2 and JNK pathways, and regulate
the development of fibrosis.
Furthermore, we showed that the production of
TGF␤ PGF2␣, and ␣2AP in bleomycin-treated SSc-like
fibroblasts was significantly higher than that in normal
fibroblasts, and the neutralization of ␣2AP attenuated
SSc-enhanced TGF␤ and PGF2␣ synthesis. On the other
hand, in a mouse model of SSc, the levels of PGF2␣ in
the ␣2AP⫺/⫺ mice decreased compared with those in
wild-type mice, and the administration of PGF2␣ to
␣2AP⫺/⫺ mice facilitated TGF␤ production and the
development of fibrosis. These data suggest that the
enhancement of ␣2AP in SSc promotes PGF2␣ synthesis,
which then facilitates TGF␤ production and the devel-
opment of fibrosis.
Plasmin is known to activate matrix metallo-
proteinases (MMPs) and a number of growth factors,
including TGF␤ and hepatocyte growth factor (HGF).
Plasmin itself and plasmin-activated MMPs can degrade
most ECM proteins, including collagen, which is the
major proteinaceous component of fibrotic tissue (34).
In addition, plasmin-activated HGF has been shown to
contribute to antifibrosis (35,36). Moreover, the overex-
pression of urokinase plasminogen activator (37,38), and
the deletion of plasminogen activator inhibitor 1 (PAI-1)
(39) attenuated the development of fibrosis. Conversely,
the overexpression of PAI-1 led to the development of a
more pronounced fibrotic response (39). Therefore,
␣2AP might facilitate the development of fibrosis not
only by producing TGF␤, but also by inhibiting plasmin
activity. These findings indicate that ␣2AP expression
may represent a target that can be used to suppress the
development of fibrosis.
REGULATION OF DEVELOPMENT OF DERMAL FIBROSIS BY ␣2-ANTIPLASMIN
In conclusion, in the present study, ␣2AP acti-
vated PLA2 through ATGL and then promoted PGF2␣
synthesis. The ␣2AP-promoted PGF2␣ synthesis regu-
lated TGF␤ production and the development of fibrosis.
These findings may provide new insights into the de-
velopment of clinical therapies for the prevention of
fibrosis.
AUTHOR CONTRIBUTIONS
All authors were involved in drafting the article or revising it
critically for important intellectual content, and all authors approved
the final version to be published. Dr. Kanno had full access to all of the
data in the study and takes responsibility for the integrity of the data
and the accuracy of the data analysis.
Study conception and design. Kanno.
Acquisition of data. Kawashita, Kokado, Okada.
Analysis and interpretation of data. Ueshima, Matsuo, Matsuno.
REFERENCES
1. Border WA, Noble NA. Transforming growth factor ␤ in tissue
fibrosis. N Engl J Med 1994;331:1286–92.
2. Collen D. Identification and some properties of a new fast-reacting
plasmin inhibitor in human plasma. Eur J Biochem 1976;69:
209–16.
3. Lijnen HR, De Cock F, Van Hoef B, Schlott B, Collen D.
Characterization of the interaction between plasminogen and
staphylokinase. Eur J Biochem 1994;224:143–9.
4. Yagame M, Eguchi K, Suzuki D, Machimura H, Takeda H, Inoue
W, et al. Fibrinolysis in patients with diabetic nephropathy deter-
mined by plasmin-␣2 plasmin inhibitor complexes in plasma. J
Diabet Complications 1990;4:175–8.
5. Ohmoto K, Yamamoto S, Ideguchi S, Yamamoto R, Takatori K,
Ohumi T, et al. Clinical significance of thrombin-antithrombin III
complex and plasmin-␣2 plasmin inhibitor complex in chronic liver
diseases. In Japanese. Nihon Shokakibyo Gakkai Zasshi 1990;87:
1837–45.
6. Kawakami M, Kawagoe M, Harigai M, Hara M, Hirose T, Hirose
W, et al. Elevated plasma levels of ␣2-plasmin inhibitor–plasmin
complex in patients with rheumatic diseases: possible role of
fibrinolytic mechanism in vasculitis. Arthritis Rheum 1989;32:
1427–33.
7. Jinnin M, Ihn H, Yamane K, Asano Y, Yazawa N, Tamaki K.
Plasma plasmin-␣2-plasmin inhibitor complex levels are increased
in systemic sclerosis patients with pulmonary hypertension. Rheu-
matology (Oxford) 2003;42:240–3.
8. Kanno Y, Kuroki A, Okada K, Tomogane K, Ueshima S, Matsuo
O, et al. Alpha2-antiplasmin is involved in the production of
transforming growth factor ␤1 and fibrosis. J Thromb Haemost
2007;5:2266–73.
9. Kanno Y, Kawashita E, Minamida M, Kaneiwa A, Okada K,
Ueshima S, et al. Alpha2-antiplasmin is associated with the
progression of fibrosis. Am J Pathol 2010;176:238–45.
10. Irving JA, Pike RN, Lesk AM, Whisstock JC. Phylogeny of the
serpin superfamily: implications of patterns of amino acid conser-
vation for structure and function. Genome Res 2000;10:1845–64.
11. Law RH, Sofian T, Kan WT, Horvath AJ, Hitchen CR, Langen-
dorf CG, et al. X-ray crystal structure of the fibrinolysis inhibitor
␣2-antiplasmin. Blood 2008;111:2049–52.
12. Tombran-Tink J, Aparicio S, Xu X, Tink AR, Lara N, Sawant S,
et al. PEDF and the serpins: phylogeny, sequence conservation,
and functional domains. J Struct Biol 2005;151:130–50.
13. Notari L, Baladron V, Aroca-Aguilar JD, Balko N, Heredia R,
Meyer C, et al. Identification of a lipase-linked cell membrane
501
receptor for pigment epithelium-derived factor. J Biol Chem
2006;281:38022–37.
14. Ueno N, Takegoshi Y, Kamei D, Kudo I, Murakami M. Coupling
between cyclooxygenases and terminal prostanoid synthases.
Biochem Biophys Res Commun 2005;338:70–6.
15. Nagase T, Uozumi N, Ishii S, Kita Y, Yamamoto H, Ohga E, et al.
A pivotal role of cytosolic phospholipase A2 in bleomycin-induced
pulmonary fibrosis. Nat Med 2002;8:480–4.
16. Hou X, Arvisais EW, Jiang C, Chen DB, Roy SK, Pate JL, et al.
Prostaglandin F2␣ stimulates the expression and secretion of
transforming growth factor B1 via induction of the early growth
response 1 gene (EGR1) in the bovine corpus luteum. Mol
Endocrinol 2008;22:403–14.
17. Oga T, Matsuoka T, Yao C, Nonomura K, Kitaoka S, Sakata D,
et al. Prostaglandin F2␣ receptor signaling facilitates bleomycin-
induced pulmonary fibrosis independently of transforming growth
factor-␤. Nat Med 2009;15:1426–30.
18. Okada K, Lijnen HR, Dewerchin M, Belayew A, Matsuo O,
Collen D, et al. Characterization and targeting of the murine
␣2-antiplasmin gene. Thromb Haemost 1997;78:1104–10.
19. Lijnen HR, Okada K, Matsuo O, Collen D, Dewerchin M.
Alpha2-antiplasmin gene deficiency in mice is associated with
enhanced fibrinolytic potential without overt bleeding. Blood
1999;93:2274–81.
20. Kanno Y, Hirade K, Ishisaki A, Nakajima K, Suga H, Into T, et al.
Lack of ␣2-antiplasmin improves cutaneous wound healing via
over-released vascular endothelial growth factor-induced angio-
genesis in wound lesions. J Thromb Haemost 2006;4:1602–10.
21. Kanno Y, Into T, Lowenstein CJ, Matsushita K. Nitric oxide
regulates vascular calcification by interfering with TGF-␤ signal-
ling. Cardiovasc Res 2008;77:221–30.
22. Kanno Y, Kaneiwa A, Minamida M, Kanno M, Tomogane K,
Takeuchi K, et al. The absence of uPAR is associated with the
progression of dermal fibrosis. J Invest Dermatol 2008;128:2792–7.
23. Junqueira LC, Bignolas G, Brentani RR. Picrosirius staining plus
polarization microscopy, a specific method for collagen detection
in tissue sections. Histochem J 1979;11:447–55.
24. Smani T, Zakharov SI, Csutora P, Leno E, Trepakova ES,
Bolotina VM. A novel mechanism for the store-operated calcium
influx pathway. Nat Cell Biol 2004;6:113–20.
25. Kim SC, Chen Y, Mirza S, Xu Y, Lee J, Liu P, et al. A clean, more
efficient method for in-solution digestion of protein mixtures
without detergent or urea. J Proteome Res 2006;5:3446–52.
26. Tithof PK, Roberts MP, Guan W, Elgayyar M, Godkin JD.
Distinct phospholipase A2 enzymes regulate prostaglandin E2 and
F2␣ production by bovine endometrial epithelial cells. Reprod
Biol Endocrinol 2007;5:16.
27. Griffin BW, Klimko P, Crider JY, Sharif NA. AL-8810: a novel
prostaglandin F2␣ analog with selective antagonist effects at the
prostaglandin F2␣ (FP) receptor. J Pharmacol Exp Ther 1999;290:
1278–84.
28. Zhang SX, Wang JJ, Gao G, Parke K, Ma JX. Pigment epithelium-
derived factor downregulates vascular endothelial growth factor
(VEGF) expression and inhibits VEGF-VEGF receptor 2 binding
in diabetic retinopathy. J Mol Endocrinol 2006;37:1–12.
29. Ek ET, Dass CR, Choong PF. Pigment epithelium-derived factor:
a multimodal tumor inhibitor. Mol Cancer Ther 2006;5:1641–6.
30. Dawson DW, Volpert OV, Gillis P, Crawford SE, Xu H, Benedict
W, et al. Pigment epithelium-derived factor: a potent inhibitor of
angiogenesis. Science 1999;285:245–8.
31. Cosgrove GP, Brown KK, Schiemann WP, Serls AE, Parr JE,
Geraci MW, et al. Pigment epithelium-derived factor in idiopathic
pulmonary fibrosis: a role in aberrant angiogenesis. Am J Respir
Crit Care Med 2004;170:242–51.
32. Ueno T, Fujimori K. Novel suppression mechanism operating in
early phase of adipogenesis by positive feedback loop for en-
502
hancement of cyclooxygenase-2 expression through prostaglandin
F2␣ receptor mediated activation of MEK/ERK-CREB cascade.
FEBS J 2011;278:2901–12.
33. Sales KJ, Grant V, Jabbour HN. Prostaglandin E2 and F2␣ activate
the FP receptor and up-regulate cyclooxygenase-2 expression via
the cyclic AMP response element. Mol Cell Endocrinol 2008;285:
51–61.
34. Parks WC, Mecham RP. Matrix metalloproteinases. San Diego:
Academic Press; 1998. p. 1–14.
35. Bauman KA, Wettlaufer SH, Okunishi K, Vannella KM, Stoolman
JS, Huang SK, et al. The antifibrotic effects of plasminogen
activation occur via prostaglandin E2 synthesis in humans and
mice. J Clin Invest 2010;120:1950–60.
36. Hattori N, Mizuno S, Yoshida Y, Chin K, Mishima M, Sisson TH.
KANNO ET AL
The plasminogen activation system reduces fibrosis in the lung by
a hepatocyte growth factor-dependent mechanism. Am J Pathol
2004;164:1091–8.
37. Sisson TH, Hanson KE, Subbotina N, Patwardhan A, Hattori N,
Simon RH. Inducible lung-specific urokinase expression reduces
fibrosis and mortality after lung injury in mice. Am J Physiol Lung
Cell Mol Physiol 2002;283:L1023–32.
38. Sisson TH, Hattori N, Xu Y, Simon RH. Treatment of bleomycin-
induced pulmonary fibrosis by transfer of urokinase-type plasmin-
ogen activator genes. Hum Gene Ther 1999;10:2315–23.
39. Eitzman DT, McCoy RD, Zheng X, Fay WP, Shen T, Ginsburg D,
et al. Bleomycin-induced pulmonary fibrosis in transgenic mice
that either lack or overexpress the murine plasminogen activator
inhibitor-1 gene. J Clin Invest 1996;97:232–7.