Nutritionally Induced Obesity Is Attenuated in Transgenic Mice Overexpressing
Plasminogen Activator Inhibitor-1
H. Roger Lijnen, Erik Maquoi, Pierre Morange, Gabor Voros, Berthe Van Hoef, Francis Kopp,
Désiré Collen, Irène Juhan-Vague and Marie-Christine Alessi
Arterioscler Thromb Vasc Biol. 2003;23:78-84; originally published online October 31, 2002;
doi: 10.1161/01.ATV.0000044457.60665.DD
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272
Greenville Avenue, Dallas, TX 75231
Copyright © 2002 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/23/1/78
Data Supplement (unedited) at:
http://atvb.ahajournals.org/content/suppl/2003/01/08/23.1.78.DC1.html
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the
Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for
which permission is being requested is located, click Request Permissions in the middle column of the Web
page under Services. Further information about this process is available in the Permissions and Rights
Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013
Nutritionally Induced Obesity Is Attenuated in Transgenic
Mice Overexpressing Plasminogen Activator Inhibitor-1
H. Roger Lijnen,* Erik Maquoi,* Pierre Morange, Gabor Voros, Berthe Van Hoef, Francis Kopp,
Désiré Collen, Irène Juhan-Vague, Marie-Christine Alessi
Objective—The objective of this study was to investigate the role of plasminogen activator inhibitor-1 (PAI-1) in adipose
tissue development in vivo.
Methods and Results—Transgenic (Tg) mice overexpressing murine PAI-1 under control of the adipocyte promoter aP2
and wild-type (WT) controls were kept on standard food (SFD) or on high-fat diet (HFD) for 15 weeks. The body weight
and the weight of the isolated subcutaneous and gonadal fat deposits of the Tg mice kept on the HFD were significantly
lower than those of the WT mice. The number of adipocytes in the adipose tissue was similar for Tg and WT mice on
the HFD, but adipocyte hypotrophy and a significantly lower ratio of stroma cells/adipocytes were observed in the Tg
mice. A significant negative correlation (P⬍0.01) was observed between expression of preadipocyte factor-1, which
blocks adipocyte differentiation, and adipose tissue weight. Fasting insulin and total cholesterol levels on the HFD were
lower in Tg than in WT mice.
Conclusions—High circulating PAI-1 levels attenuate nutritionally induced obesity. This may be related to modifications
in adipose tissue cellularity affecting weight and plasma metabolic parameters. (Arterioscler Thromb Vasc Biol. 2003;
23:78-84.)
Key Words: plasminogen activator inhibitor-1 䡲 obesity 䡲 adipose tissue 䡲 adipocyte
D
evelopment of obesity is associated with extensive
remodeling of adipose tissue and involves adipogenesis,
angiogenesis, and extracellular matrix (ECM) proteolysis.1
The plasminogen/plasmin (fibrinolytic) system, which contributes to tissue remodeling by degradation of ECM and
basement membrane components or activation of latent
growth factors,2,3 may thereby play a role in the regulation of
adipose tissue growth.4,5 In mice6,7 and in humans,8 –15 a
significant correlation was observed between body mass
index or the amount of visceral fat and plasma levels of
plasminogen activator inhibitor-1 (PAI-1), the main physiological inhibitor of tissue-type (t-PA) and urokinase-type
(u-PA) plasminogen activator.3 Several groups have reported
PAI-1 synthesis by murine adipocyte cell lines,16 –20 human
adipose tissue explants,14,21,22 and primary cultures of human
adipocytes.23,24 In human or murine fat, PAI-1 production
was also observed in stroma cells.16,25 Adipose tissue PAI-1
mRNA levels were enhanced in obese individuals15,26 and
positively correlated with the volume and lipid content of fat
cells.15 A potential role of PAI-1 in development of obesity
was suggested by the findings that PAI-1– deficient mice kept
on a high fat diet developed more rapidly adipose tissue than
their lean counterparts27 and that transgenic (Tg) mice overexpressing a stable human PAI-1 variant had virtually no
intraperitoneal fat.28 In contrast, disruption of the PAI-1 gene
in genetically obese and diabetic ob/ob mice reduced adiposity and improved the metabolic profile.29 To additionally
elucidate the role of PAI-1 in adipose tissue development, we
have studied nutritionally induced obesity in Tg mice overexpressing murine PAI-1.
Methods
Generation and Characterization of
Transgenic Mice
For the construction of the transgene and the generation and
characterization of the Tg mice, including PAI-1 and Pref-1 mRNA
determination, please refer to http://atvb.ahajournals.org.
Diet Model
Five-week-old male mice weighing ⬇20 g were kept in microisolator
cages on a 12-hour day/night cycle and fed water and a standard fat
diet (SFD) (n⫽16 or 18 for wild-type [WT] or Tg mice) or high fat
diet (HFD) (n⫽16 or 20 for WT or Tg mice) at libitum. The SFD
(KM-04-k12, Muracon, Carfil) contains 4% (wt/wt) fat and 13% kcal
as fat, corresponding to a caloric value of 10.9 kJ/g. The HFD (TD
Received October 3, 2002; revision accepted October 17, 2002.
From the Center for Molecular and Vascular Biology (H.R.L., E.M., G.V., B.V.H., D.C.), University of Leuven, Belgium; Haematology Laboratory
(P.M., I.J.-V., M.C.A.), CHU Timone and INSERM EPI 9936, Marseille, France; and Histology Laboratory (F.K.), CHU Timone, Marseille, France.
*These authors contributed equally to this study.
Correspondence to H. Roger Lijnen, Center for Molecular and Vascular Biology, University of Leuven, Campus Gasthuisberg, O&N, Herestraat 49,
B-3000 Leuven, Belgium. E-mail [email protected]
© 2003 American Heart Association, Inc.
Arterioscler Thromb Vasc Biol. is available at http://www.atvbaha.org
DOI: 10.1161/01.ATV.0000044457.60665.DD
78
Downloaded from http://atvb.ahajournals.org/
by guest on February 21, 2013
Lijnen et al
88137, Harlan Teklad) contains 21% (wt/wt) fat coming from milk
fat and 49% (wt/wt) carbohydrate coming from sucrose and corn
starch, yielding 42% kcal as fat, and a caloric value of 20.1 kJ/g.
Mice were weighed every week. Food intake was measured for
two 24-hour periods at 5 or 6 weekly intervals during the diet period
and expressed as gram per 24 hours (55 to 70 measurements were
performed using 5 to 7 mice in each group). To evaluate physical
activity, mice were placed in a separate cage equipped with a turning
wheel linked to a computer to register full turning cycles in 48-hour
periods at 2 or 3 weekly intervals during the diet; data are
represented as number of cycles per 12 hours (at night) and are
mean⫾SEM of 20 to 30 measurements using 5 to 7 mice in each
group. The data were first averaged per mouse and are reported as
mean⫾SEM of the number of animals studied (6 or 7 for WT mice
on SFD or HFD and 5 or 6 for Tg mice on SFD or HFD).
After 15 weeks of SFD or HFD, following overnight fasting, the
mice were killed by IP injection of 60 mg/kg Nembutal (Abbott
Laboratories). All animal experiments were approved by the local
ethical committee and were performed in accordance with the
guiding principles of the American Physiological Society and the
International Society on Thrombosis and Hemostasis.30
PAI-1 and Obesity
79
measured at different time intervals (0 to 7 hours) and expressed as
ng per ␮g DNA.
Morphometric and Immunohistochemical Analysis
The areas of at least 200 adipocyte sections were measured and the
volume of each cell was calculated as described elsewhere.27 The
number of fat cells in the tissue was calculated as follows: weight of
total pad/mean fat cell lipid content (mean volume⫻lipid density).
On 10 randomly selected fields, the number of stroma cell nuclei and
the number of adipocytes, based on morphological criteria, were
determined and the results expressed as a ratio.
Staining of blood vessels was performed on paraffin sections using
biotinylated Bandeiraea (Griffonia) Simplicifolia BSI lectin (SigmaAldrich).34 For each animal, at least 12 randomly selected fields in 4
different sections were analyzed by computer-assisted image analysis, and data were averaged.
Statistical Analysis
Data are expressed as mean⫾SEM. Statistical significance between
groups was evaluated using nonparametric t testing. Values of
P⬍0.05 were considered statistically significant.
Blood Collection and Tissue Preparation
Blood was collected from the retroorbital sinus with or without
addition of trisodium citrate (final concentration 0.01 mol/L); plasma
and serum were stored at ⫺20°C.
Intraabdominal (gonadal [GON]) and inguinal subcutaneous (SC)
fat pads were removed and weighed. One portion was immediately
frozen at ⫺80°C for protein extraction; other portions were used for
histology and immunocytochemistry. Therefore, 10-␮m cryosections
and 7-␮m paraffin sections were prepared.
Other organs, including lungs, kidneys, liver, spleen, and heart,
were also removed, weighed, and stored at ⫺80°C.
Hematologic and Metabolic Parameters
PAI-1 antigen levels in plasma or tissue extracts were determined
with an ELISA specific for murine PAI-1.31 PAI-1 activity levels in
plasma were determined after incubation with excess human t-PA for
20 minutes at 37° and measurement of t-PA/PAI-1 complex by
ELISA using the anti-PAI-1 mAb-16F11 for coating and the antihuman t-PA mAb-62E8 for tagging.
For histopathologic examination, organs were removed (2 WT
males at 13 weeks of age, 2 Tg males at 12 and 23 weeks of age each,
and 3 Tg females at 19 to 20 weeks of age), and sections were stained
with H&E or with an antiserum against murine fibrinogen/fibrin, as
described elsewhere.32
White blood cells, red blood cells, platelets, hemoglobin, and
hematocrit levels were determined using standard laboratory assays.
Blood glucose concentrations were measured using Glucocard strips
(Menarini Diagnostics). Triglycerides, free fatty acids, alanine aminotransferase (ALT), and total cholesterol were evaluated using
routine clinical assays. Insulin was measured with a monoclonal
anti-rat insulin radioimmunoassay (Linco Research, Inc).
Fibrinolytic Parameters
Extraction of adipose tissue (⬇250 mg/mL) or of different organs
was performed as described,33 and the protein concentration of the
supernatant was determined (BCA assay, Pierce); extracts were
stored at ⫺80°C.
In situ zymography with cryosections of adipose tissue on fibrin
overlays was performed at 37°C for 48 hours without or with
addition to the gel of neutralizing antibodies against murine t-PA
(final concentration, 200 ␮g/mL) or of purified murine PAI-1 (final
concentration, 20 ␮g/mL). Lysis of the substrate gel was determined
by computer-assisted image analysis (expressed in arbitrary units)
and normalized to the section area.33
Adipose tissue explants were prepared and incubated as described
elsewhere14; PAI-1 antigen levels secreted into the medium were
Results
Generation and Characterization of
Transgenic Mice
A construct containing 5.4 kb of the adipocyte-specific aP2
promoter driving the full-length murine PAI-1 cDNA was
used to generate FVB Tg mice with PAI-1 overexpression. Of
17 live offspring from 4 different females, 4 (3 males and 1
female) had integrated the transgene. These were backcrossed with WT C57/Bl6 mice, and the Tg or WT littermates
in the offspring were identified by polymerase chain reaction
(not shown), and the plasma PAI-1 antigen level was measured by ELISA. The offspring of breeding pairs of WT and
Tg mice comprised 49% WT mice (49% male and 51%
female) and 51% Tg mice (43% male and 57% female). Mice
from 3 or 2 different founders were used in the WT or Tg
groups, respectively.
Histopathologic examination of H&E-stained crosssections through different organs of 12- to 23-week-old male
and female WT or Tg mice did not show any apparent
abnormalities or differences. The tissue sections included
spleen, liver, lung, kidney, heart, stomach, brain, eye, reproductive organs (testis and epididymis or uterus), and SC and
GON adipose tissue. Immunostaining with an antiserum
against fibrin(ogen) did not reveal significant fibrin deposits
in the organs or tissues. White and red blood cells, platelets,
hemoglobin, and hematocrit levels were comparable for the
male WT and Tg mice used in the diet study (not shown).
Strong expression of the transgene was confirmed in
adipose tissue, but weak expression in heart, kidney, and liver
and stronger expression in lung and spleen of Tg mice were
also observed, whereas the transgene was not detected in WT
tissues (Figure 1). Adipose tissue explants of mice on SFD
revealed higher PAI-1 antigen production for Tg compared
with WT mice (1.6⫾0.40 versus 0.43⫾0.13 ng/␮g DNA after
7 hours incubation for GON tissue [P⫽0.056], and 1.4⫾0.42
versus 0.49⫾0.15 ng/␮g DNA for SC tissue [P⫽0.016]),
whereas no significant difference was observed for mice on
HFD (data not shown).
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013
80
Arterioscler Thromb Vasc Biol.
January 2003
Figure 1. Determination of Tg-PAI-1 in GON
and SC adipose tissue and organs of WT and
Tg mice kept on SFD or HFD. Expression of
28S rRNA is shown as an internal standard,
and the ratio Tg/28S is indicated below each
lane. Lane (⫺) indicates a blank control without sample.
PAI-1 antigen levels in extracts of SC adipose tissue of Tg
mice kept on SFD and in both SC and GON tissue of Tg mice
on HFD were slightly elevated compared with those of WT
mice, whereas significantly elevated levels were observed in
lung and spleen of Tg mice, both on SFD and HFD (Table 1).
Plasma PAI-1 antigen levels at the start of the study (5
weeks of age) were about 10-fold higher in the Tg mice
compared with the WT littermates (2.1⫾0.20 or 3.0⫾0.18
ng/mL for WT mice included in SFD or HFD groups, with
corresponding values of 21⫾3.9 or 33⫾4.8 ng/mL for the Tg
mice). At the end of the study (20 weeks of age), plasma
PAI-1 antigen levels for both WT and Tg mice had not
significantly changed on the SFD, whereas after 15 weeks of
the HFD, they were 2- to 3-fold increased, resulting in 7-fold
higher levels of Tg compared with WT mice (Table 1). PAI-1
activity on a molar basis corresponded to 30% or 25% of the
antigen for WT mice on SFD or HFD and to 17% or 36% for
Tg mice on SFD or HFD.
Overexpression of PAI-1 Does Not Affect Adipose
Tissue Development in Mice on SFD
The food intake of WT and Tg mice was comparable
(4.6⫾0.1 g/24 hours versus 4.4⫾0.2 g/24 hours), whereas
physical activity at night was lower for WT mice
(14 100⫾1750 versus 28 400⫾2380 cycles/12 hours,
P⫽0.004). Still, after 15 weeks of SFD, the body weight and
the weight of the isolated SC or GON fat pads were not
significantly different for WT or Tg mice (Figure 2 and Table
2). The weight of other organs was also comparable for both
groups (not shown). The number and diameter of adipocytes
in gonadal adipose tissue was comparable for WT and GON
adipose tissue of Tg mice, whereas the ratio of stroma cells
versus adipocytes was significantly lower in the GON adipose tissue of Tg mice (Table 2). Expression of Pref-1 mRNA
in gonadal adipose tissue of Tg mice on SFD was significantly higher than in that of WT mice (ratio Pref-1 mRNA/
18S rRNA of 0.64⫾0.07 versus 0.13⫾0.005; P⬍0.01). Staining of blood vessels with the Bandeiraea Simplicifolia lectin
did not reveal significant differences between both genotypes
with respect to stained area, blood vessel size, and density
(Table 3).
Fibrin overlay revealed somewhat lower overall in situ
fibrinolytic activity in cryosections of SC or GON adipose
tissue of Tg compared with WT mice kept on SFD (ratio of
lysis area over section area of 0.48⫾0.08 versus 1.2⫾0.15 for
SC tissue [mean⫾SEM of 5 or 6 determinations, P⬍0.005],
with corresponding values of 0.38⫾0.12 versus 0.61⫾0.16
for GON tissue). Addition of murine PAI-1 or of polyclonal
rabbit antiserum against murine t-PA to the gel resulted in
inhibition of 98% of the fibrinolytic activity (data not shown).
Thus, on SFD, Tg mice with elevated plasma PAI-1 and
decreased adipose tissue fibrinolytic activity had fewer
stroma cells and higher Pref-1 expressions in adipose
tissue. Adipose tissue development was not significantly
affected, but histological tissue modifications were
observed.
Overexpression of PAI-1 Impairs Obesity in Mice
on HFD
The intake of the HFD was somewhat higher for WT than for
Tg mice (4.1⫾0.1 g/24 hours versus 3.5⫾0.1 g/24 hours), and
the physical activity at night was comparable (19 300⫾3080
TABLE 1. PAI-1 Antigen Levels in Plasma and Tissue Extracts of WT or Tg Mice
Kept on SFD or HFD for 15 Weeks
SFD
HFD
WT
Tg
WT
Tg
Plasma,* ng/mL
2.0⫾0.30
24⫾3.8†
11⫾2.7
80⫾16†
SC adipose,‡ ng/g
15⫾2.4
58⫾15
33⫾7.7
54⫾4.7
GON adipose,‡ ng/g
17⫾2.8
15⫾3.1
30⫾5.5
45⫾13
Lung,‡ ng/g
59⫾8.7
210⫾54§
46⫾2.7
375⫾58储
Kidney,‡ ng/g
62⫾3.5
79⫾5.5
87⫾5.0
99⫾8.8
Liver,‡ ng/g
54⫾4.6
79⫾17
88⫾14
85⫾11
Spleen,‡ ng/g
18⫾1.9
140⫾18†
20⫾4.1
240⫾44†
Heart,‡ ng/g
97⫾6.2
90⫾11
81⫾12
94⫾8.2
*Expressed as ng/mL plasma, mean⫾SEM of 16 to 20 determinations.
†P⬍0.0005 vs WT on the same diet.
‡Expressed as ng/g tissue, mean⫾SEM of 6 to 8 determinations.
§P⬍0.05 vs WT on the same diet.
储P⬍0.005 vs WT on the same diet.
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013
Lijnen et al
PAI-1 and Obesity
81
Figure 2. Evolution of the body weight of WT
(open symbols) or Tg (closed symbols) mice kept
on SFD (squares) or HFD (circles). Data are
mean⫾SEM of 16 to 20 experiments. *P⬍0.05,
**P⬍0.005, and ***P⬍0.001 versus WT on HFD.
versus 23 100⫾1870 cycles/12 hours). Because the body
weight gain on the HFD was significantly lower in the Tg
mice, these data suggest a lower feeding efficiency (weight
gain/calories consumed).
The HFD induced marked obesity in both the WT and Tg
mice. However, weight gain was lower in the Tg mice,
resulting in a significantly lower body weight after 15 weeks
of HFD (Figure 2 and Table 2). Also, the weight of the
isolated SC or GON fat pads was lower for the Tg mice. The
weight of other organs was comparable, except for a higher
liver weight in WT than in Tg mice (2210⫾150 versus
1500⫾110 mg, P⫽0.001); this corresponds with higher ALT
levels in WT than in Tg mice on HFD (150⫾31 versus
63⫾11 IU/mL, P⬍0.05). However, triglyceride levels in the
liver of WT mice on HFD were not significantly elevated
(1220⫾127 versus 840⫾190 mg/dL in Tg mice; P⫽0.41).
The number of adipocytes in the entire fat depots of WT and
Tg mice was very similar for the GON tissue and was
TABLE 2. Effect of PAI-1 Overexpression in Mice on Body Weight Gain and on
Adipose Tissue Weight and Cellularity After 15 Weeks of SFD or HFD
SFD
WT
HFD
Tg
WT
Tg
Body weight at start,* g
22⫾0.56
20⫾2.0
21⫾0.31
20⫾0.34
Body weight at end,† g
33⫾0.91
31⫾1.7
45⫾1.4
37⫾0.98‡
Weight gain, g
9.4⫾0.92
11⫾0.68
24⫾1.2
17⫾0.88‡
SC fat, g
0.24⫾0.037
0.24⫾0.027
1.69⫾0.15
1.05⫾0.12§
GON fat, g
0.49⫾0.073
0.42⫾0.049
1.62⫾0.15
1.22⫾0.11
ND
ND
9.70⫾1.28
16⫾2.3
5.64⫾0.45
5.05⫾0.62
Adipocyte number, ⫻106
SC
GON
4.92⫾0.76
6.49⫾0.88
Adipocyte diameter, ␮m
SC
GON
ND
62⫾4.1
ND
55⫾3.1
74⫾3.5
52⫾3.4§
85⫾2.3
80⫾3.3
Stroma cells/adipocytes
SC
GON
ND
ND
2.9⫾0.15
1.6⫾0.09‡
1.5⫾0.10
1.0⫾0.16储
3.0⫾0.29
1.7⫾0.16§
Data are mean⫾SEM of 12 to 20 determinations; ND indicates not determined.
*Body weight at 5 weeks of age.
†Body weight after 15 weeks of SFD or HFD.
‡P⬍ 0.0005 vs WT on the same diet.
§P⬍0.005 vs WT on the same diet.
储P⬍0.05 vs WT on the same diet.
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013
82
Arterioscler Thromb Vasc Biol.
January 2003
TABLE 3. Characterization of Blood Vessels in Gonadal
Adipose Tissue of WT or Tg Mice Kept on SFD or HFD for
15 Weeks
SFD
WT
HFD
Tg
WT
Tg
Stained area*
2.0⫾0.44
1.9⫾0.21
2.7⫾0.39
2.1⫾0.11
Vessel density†
280⫾56
270⫾47
200⫾34
330⫾54
74⫾4.8
77⫾5.5
140⫾19
Vessel size‡
68⫾6.2§
Data are mean ⫾ SEM of 5 to 10 determinations.
*Area stained with the Bandeiraea Simplicifolia lectin, in percent of total
area.
†Number of vessels per mm2 tissue.
‡Average vessel size in ␮m2.
§P⬍0.01 vs WT on the same diet.
somewhat higher in the SC tissue of the Tg mice (P⫽0.05).
Significant adipocyte hypotrophy was observed in the SC
tissue, and the ratio of stroma cells versus adipocytes was
lower in both SC and GON adipose tissue of the Tg mice
(Table 2). Expression of Pref-1 mRNA in gonadal adipose
tissue of Tg mice on HFD was about 2-fold higher than in that
of WT mice (ratio Pref-1 mRNA/18S rRNA of 0.31⫾0.09
versus 0.14⫾0.05; P⫽0.13).
Lectin staining did not reveal significant differences between WT and Tg mice on HFD, except for a higher average
blood vessel size in the WT mice (Table 3). Fibrin overlay
confirmed somewhat lower overall in situ fibrinolytic activity
in SC and GON adipose tissue of Tg compared with WT mice
(ratio of lysis area over section area of 0.37⫾0.078 versus
0.52⫾0.16 for SC tissue, with corresponding values of
0.29⫾0.058 versus 0.49⫾0.088 for GON tissue; mean⫾SEM
of 5 or 6 determinations).
After 15 weeks of HFD, WT mice presented with higher
glucose and insulin levels compared with the mice on SFD.
However, Tg mice on HFD had lower plasma levels of insulin
(0.68⫾0.11 versus 1.6⫾0.28 ng/mL, P⬍0.005) and total
cholesterol (3.8⫾0.26 versus 5.3⫾0.21 mmol/L, P⬍0.001)
than WT controls on HFD, whereas glucose (4.5⫾0.30 versus
4.5⫾0.42 mmol/L), triglyceride (1.5⫾0.11 versus 1.6⫾
0.16 mmol/L), and free fatty acids (0.89⫾0.05 versus
1.2⫾0.08 mmol/L) levels were comparable.
Discussion
To elucidate the role of PAI-1 in development of obesity, Tg
mice were generated with overexpression of murine PAI-1
under control of the adipocyte-specific promoter aP2. Strong
expression of the transgene was observed in adipose tissue,
lung, and spleen, and weak expression in heart, kidney, and
liver. Other studies using the aP2 promoter have previously
reported low-level expression of transgenes in other tissues,
including muscle, heart, spleen, and thymus.35–37 In this
study, 5-week-old male Tg and WT mice of the same genetic
background were kept on SFD or HFD for 15 weeks.
For both WT and Tg mice on SFD or HFD for 15 weeks,
the body weight was positively correlated with the weight of
the SC and GON adipose tissue and with the liver weight (all
P⬍0.0001). Tg mice on HFD had a lower body weight than
WT mice. This may be explained partially by the reduced
weight of SC and GON adipose tissue; it should be kept in
mind that not all fat deposits were recovered from the body.
Furthermore, the liver weight in the WT mice on HFD was
significantly higher than in the Tg mice, and ALT levels were
significantly elevated,38 although triglyceride levels were
only moderately enhanced. The weight of other organs was
comparable, suggesting that the observed differences in body
fat are not aspecific, although effects of PAI-1 overexpression
on general well being cannot be excluded.
The main findings of this study are that high circulating
plasma PAI-1 levels or reduced fibrinolytic activity in adipose tissue do not affect adipose tissue development in mice
kept on standard chow but result in a reduction of nutritionally induced obesity. Indeed, average body and fat pad
weights of mice on SFD were not affected by the ⬇12-fold
higher average PAI-1 levels in Tg mice, and no significant
correlation was observed between plasma PAI-1 antigen
levels and body weight or SC or GON fat pad weight. The
number and diameter of adipocytes in adipose tissue were not
affected, but the ratio of stroma cells versus adipocytes was
lower in the Tg mice. The HFD induced obesity and adipocyte hypertrophy in both WT and Tg mice. However, for Tg
mice on HFD, the 7-fold higher PAI-1 levels compared with
WT mice were associated with significantly lower body and
fat pad weights. These findings are in agreement with a
previous study showing more rapid adipose tissue development in PAI-1– deficient mice on HFD.27 Furthermore, significant adipocyte hypotrophy was observed in the SC adipose tissue of Tg mice on HFD, and the ratio of stroma cells
versus adipocytes was significantly lower both in SC and
GON adipose tissue of Tg compared with WT mice. In our
previous study with PAI-1– deficient mice, we found that the
ratio stroma cells versus adipocytes was also lower than in
WT mice.27 In this study, however, only a small effect on
weight gain and a decrease in endothelial cell number was
observed, whereas in the present study, a strong effect on
weight gain was observed but no difference in endothelial
cells. Overexpression of PAI-1 may thus differentially effect
cell populations compared with its deficiency.
Analysis of blood vessels did not reveal significant differences, except for a somewhat higher average vessel size but
lower density in adipose tissue of WT mice on HFD.
Overexpression of PAI-1 thus seems to modify the cellularity
of adipose tissue, however without significantly affecting
angiogenesis. Whether there is a direct relation between
modification of the stroma and adipocyte hypotrophy requires
additional investigation. In apparent contrast with our data,
disruption of the PAI-1 gene in genetically obese and diabetic
ob/ob mice resulted in reduced adiposity.29 Our model is
different, because it is based on nutritionally induced obesity.
The difference may be related to the absence of leptin in the
ob/ob mice. The data of the present study are in agreement
with our previous study in PAI-1– deficient mice27 and with a
recent report that in vivo plasminogen deficiency (inducing
hypofibrinolysis, as in our PAI-1 Tg mice) reduces fat
accumulation.42
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013
Lijnen et al
The Pref-1 gene product is abundant in preadipocytes but
absent in adipocytes, and constitutive expression of Pref-1
blocks differentiation.39 Overall, for WT and Tg samples on
SFD and HFD, a significant negative correlation was observed between Pref-1 mRNA expression in gonadal adipose
tissue and tissue weight (P⬍0.01), compatible with an inverse relationship with adipocyte differentiation. Interestingly, Pref-1 expression was significantly higher in GON
adipose tissue of Tg compared with WT mice, and ⬇2-fold
higher Pref-1 mRNA levels in adipose tissue of Tg mice on
HFD corresponded with a 25% reduction of GON adipose
tissue weight. This may indicate a lower degree of adipocyte
differentiation in the Tg adipose tissue.40,41 No correlations
were observed between Pref-1 mRNA levels and adipocyte
size or the ratio stroma cells/adipocytes. Interestingly, it was
recently reported that differentiation of plasminogendeficient stromal cells into mature adipocytes was reduced
compared with WT cells.42 This supports our findings that
reduced fibrinolytic activity may impair differentiation. Thus,
overexpression of PAI-1 may result in enhanced expression
of antidifferentiation factors such as Pref-1, which may
contribute to reduce tissue mass. A similar mechanism was
proposed to explain that overexpression of nuclear SREBP-1c
in adipose tissue interferes with adipocyte differentiation and
development of insulin resistance.40
Metabolic parameters were also affected by PAI-1 overexpression. The attenuation of obesity observed in TG mice on
HFD was indeed accompanied by attenuated hyperinsulinemia, with significantly lower insulin levels compared with
WT mice on HFD.
Taken together, PAI-1 overexpression had little effect on
adipose tissue development in mice kept on normal chow, but
attenuated nutritionally induced obesity. This is the first
experimental demonstration that high plasma PAI-1 levels are
associated with lower body weight in obese subjects. It may
be explained by an effect on the cellularity of the adipose
tissue (adipocyte hypotrophy, decreased number of stroma
cells, or effects on adipocyte differentiation), contributing to
a reduction of adipose tissue weight with modulation of
metabolic parameters.
Acknowledgments
This study was supported financially by grants from the Interuniversity Attraction Poles (IAP, project P5/02) and the Flemish Fund for
Scientific Research (FWO, project G.0112.02). Skillful technical
assistance by M.F. Bonzi, L. Frederix, A. De Wolf, and M. Verdier
is gratefully acknowledged.
References
1. Crandall DL, Hausman GJ, Kral JG. A review of the microcirculation of
adipose tissue: anatomic, metabolic, and angiogenic perspectives. Microcirculation. 1997;4:211–232.
2. Carmeliet P, Collen D. Development and disease in proteinase-deficient
mice: role of the plasminogen, matrix metalloproteinase and coagulation
system. Thromb Res. 1998;91:255–285.
3. Lijnen HR. Plasmin and matrix metalloproteinases in vascular
remodeling. Thromb Haemost. 2001;86:324 –333.
4. Juhan-Vague I, Alessi MC. Regulation of fibrinolysis in the development
of atherothrombosis: role of adipose tissue. Thromb Haemost. 1999;82:
832– 836.
PAI-1 and Obesity
83
5. Samad F, Loskutoff DJ. Hemostatic gene expression and vascular disease
in obesity: insights from studies of genetically obese mice. Thromb
Haemost. 1999;82:742–747.
6. Samad F, Loskutoff DJ. Tissue distribution and regulation of plasminogen activator inhibitor-1 in obese mice. Mol Med. 1996;2:568 –582.
7. Shimomura I, Funahashi T, Takahashi M, Maeda K, Kotani K, Nakamura
T, Yamashita S, Miura M, Fukada Y, Takemura K, Man Z, Toide K,
Nakayama N, Fukuda Y, Lin M-C, Wetterau J-R, Matsuzawa Y.
Enhanced expression of PAI-1 in visceral fat: possible contributor to
vascular disease in obesity. Nat Med. 1996;2:800 – 803.
8. Vague P, Juhan-Vague I, Aillaud MF, Badier C, Viard R, Alessi MC,
Collen D. Correlation between blood fibrinolytic activity, plasminogen
activator inhibitor level, plasma insulin level, and relative body weight in
normal and obese subjects. Metabolism. 1986;35:250 –253.
9. Mavri A, Stegnar M, Krebs M, Sentocnik JT, Geiger M, Binder BR.
Impact of adipose tissue on plasma plasminogen activator inhibitor-1 in
dieting obese women. Arterioscler Thromb Vasc Biol. 1999;19:
1582–1587.
10. Janand-Delenne B, Chagnaud C, Raccah D, Alessi MC, Juhan-Vague I,
Vague P. Visceral fat as a main determinant of plasminogen activator
inhibitor 1 level in women. Int J Obes Relat Metab Disord. 1998;22:
312–317.
11. Giltay EJ, Elbers JM, Gooren LJ, Emeis JJ, Kooistra T, Asscheman H,
Stehouwer CD. Visceral fat accumulation is an important determinant of
PAI-1 levels in young, nonobese men and women: modulation by
cross-sex hormone administration. Arterioscler Thromb Vasc Biol. 1998;
18:1716 –1722.
12. Mykkanen L, Ronnemaa T, Marniemi J, Haffner SM, Bergman R, Laakso
M. Insulin sensitivity is not an independent determinant of plasma plasminogen activator inhibitor-1 activity. Arterioscler Thromb. 1994;14:
1264 –1271.
13. Cigolini M, Targher G, Bergamo Andreis IA, Tonoli M, Agostino G, De
Sandre G. Visceral fat accumulation and its relation to plasma hemostatic
factors in healthy men. Arterioscler Thromb Vasc Biol. 1996;16:
368 –374.
14. Alessi MC, Peiretti F, Morange P, Henry M, Nalbone G, Juhan-Vague I.
Production of plasminogen activator inhibitor 1 by human adipose tissue:
possible link between visceral fat accumulation and vascular disease.
Diabetes. 1997;46:860 – 867.
15. Eriksson P, Reynisdottir S, Lönngvist F, Stemme V, Arner P. Adipose
tissue secretion of plasminogen activator inhibitor-1 in non obese individuals. Diabetologia. 1998;41:65–71.
16. Samad F, Yamamoto K, Loskutoff DJ. Distribution and regulation of
plasminogen activator inhibitor-1 in murine adipose tissue in vivo:
induction by tumor necrosis factor-␣ and lipopolysaccharide. J Clin
Invest. 1996;97:37– 46.
17. Morange PE, Aubert J, Peiretti F, Lijnen HR, Vague P, Verdier M, Negrel
R, Juhan-Vague I, Alessi MC. Glucocorticoids and insulin promote plasminogen activator inhibitor 1 production by human adipose tissue.
Diabetes. 1999;48:890 – 895.
18. Sakamoto T, Woodcock-Mitchell J, Marutsuka K, Mitchell JJ, Sobel BE,
Fujii S. TNF-alpha and insulin, alone and synergistically, induce plasminogen activator inhibitor-1 expression in adipocytes. Am J Physiol.
1999;276:C1391–C1397.
19. Samad F, Schneiderman J, Loskutoff D. Expression of fibrinolytic genes
in tissues from human atherosclerotic aneurysms and from obese mice.
Ann N Y Acad Sci. 1997;811:350 –358.
20. Lundgren CH, Brown SL, Nordt TK, Sobel BE, Fujii S. Elaboration of
type-1 plasminogen activator inhibitor from adipocytes: a potential
pathogenic link between obesity and cardiovascular disease. Circulation.
1996;93:106 –110.
21. Morange PE, Alessi MC, Verdier M, Casanova D, Magalon G,
Juhan-Vague I. PAI-1 produced ex vivo by human adipose tissue is
relevant to PAI-1 blood level. Arterioscler Thromb Vasc Biol. 1999;19:
1361–1365.
22. Cigolini M, Tonoli M, Borgato L, Frigotto L, Manzato F, Zeminian S,
Cardinale C, Camin M, Chiaramonte E, De Sandre G, Lunardi C.
Expression of plasminogen activator inhibitor-1 in human adipose tissue:
a role for TNF-alpha? Atherosclerosis. 1999;143:81–90.
23. Crandall DL, Groeling TM, Busler DE, Antrilli TM. Release of PAI-1 by
human preadipocytes and adipocytes independent of insulin and IGF-1.
Biochem Biophys Res Commun. 2000;279:984 –988.
24. Gottschling-Zeller H, Rohrig K, Hauner H. Troglitazone reduces plasminogen activator inhibitor-1 expression and secretion in cultured human
adipocytes. Diabetologia. 2000;43:377–383.
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013
84
Arterioscler Thromb Vasc Biol.
January 2003
25. Bastelica D, Morange P, Berthet B, Borghi H, Lacdroi O, Grino M,
Juhan-Vague I, Alessi M-C. Stromal cells are the main plasminogen
activator inhibitor-1-producing cells in human fat: evidence of differences
between visceral and subcutaneous deposits. Arterioscler Thromb Vasc
Biol. 2002;22:173–178.
26. Mavri A, Alessi M-C, Bastelica D, Geel-Georgelin O, Fina F, Sentocnik
JT, Stegnar M, Juhan-Vague I. Subcutaneous abdominal, but not femoral
fat expression of plasminogen activator inhibitor-1 (PAI-1) is related to
plasma PAI-1 levels and insulin resistance and decreases after weight
loss. Diabetologia. 2001;44:2025–2031.
27. Morange PE, Lijnen HR, Alessi MC, Kopp F, Collen D, Juhan-Vague I.
Influence of PAI-1 on adipose tissue growth and on metabolic parameters
in a murine model of diet-induced obesity. Arterioscler Thromb Vasc
Biol. 2000;20:1150 –1154.
28. Èren M, Su M, Atkinson J, King L, Declerck P, Vaughan DE. Phenotypic
derangements associated with overexpression of plasminogen activator
inhibitor-1 (PAI-1) in transgenic mice. Arterioscler Thromb Vasc Biol.
2001;21:695. Abstract 230.
29. Schafer K, Fujisawa K, Konstantinides S, Loskutoff J. Disruption of the
plasminogen activator inhibitor 1 gene reduces the adiposity and
improves the metabolic profile of genetically obese and diabetic ob/ob
mice. FASEB J. 2001;15:1840 –1842.
30. Giles AR. Guidelines for the use of animals in biomedical research.
Thromb Haemost. 1987;58:1078 –1084.
31. Declerck PJ, Verstreken M, Collen D. Immunoassay of murine t-PA,
u-PA and PAI-1 using monoclonal antibodies raised in gene-inactivated
mice. Thromb Haemost. 1995;74:1305–1309.
32. Lijnen HR, Okada K, Matsuo O, Collen D, Dewerchin M. ␣2-Antiplasmin
gene-deficiency in mice is associated with enhanced fibrinolytic potential
without overt bleeding. Blood. 1999;93:2274 –2281.
33. Lijnen HR, Van Hoef B, Lupu F, Moons L, Carmeliet P, Collen D.
Function of plasminogen/plasmin and matrix metalloproteinase systems
after vascular injury in mice with targeted inactivation of fibrinolytic
system genes. Arterioscler Thromb Vasc Biol. 1998;18:1035–1045.
34. Laitinen L. Griffonia Simplicifolia lectins bind specifically to endothelial cells
and some epithelial cells in mouse tissues. Histochem J. 1987;19:225–234.
35. Qiu J, Ogus S, Lu R, Chehab FF. Transgenic mice overexpressing leptin
accumulate adipose mass at an older, but not younger, age. Endocrinology. 2001;142:348 –358.
36. Massiéra F, Bloch-Faure M, Ceiler D, Murakami K, Fukamizu A, Gasc
J-M, Quignard-Boulangé A, Negrel R, Ailhaud G, Seydoux J, Meneton P,
Teboul M. Adipose angiotensinogen is involved in adipose tissue growth
and blood pressure regulation. FASEB J. 2001;15:2727–2729.
37. Moitra J, Mason MM, Olive M, Krylov D, Gavrilova O, Marcus-Samuels
B, Feigenbaum L, Lee E, Aoyama T, Eckhaus M, Reitman ML, Vinson
C. Life without white fat: a transgenic mouse. Genes Dev. 1998;12:
3168 –3181.
38. Tazawa Y, Noguchi H, Nishinomiya F, Takada G. Serum alanine aminotransferase activity in obese children. Acta Paediatr. 1997;86:238–241.
39. Wolf G. The molecular mechanism of the stimulation of adipocyte differentiation by a glucocorticoid. Nutr Rev. 1999;57:324 –326.
40. Shimomura I, Hammer RE, Richardson JA, Ikemoto S, Bashmakov Y,
Goldstein JL, Brown MS. Insulin resistance and diabetes mellitus in
transgenic mice expressing nuclear SREBP-1c in adipose tissue: model
for congenital generalized lipodystrophy. Genes Dev. 1998;12:
3182–3194.
41. Zhou Y-T, Wang Z-W, Higa M, Newgard CB, Unger RH. Reversing
adipocyte differentiation: implications for treatment of obesity. Proc Natl
Acad Sci U S A. 1999;96:2391–2395.
42. Hoover-Plow J, Ellis J, Yuen L. In vivo plasminogen deficiency reduces
fat accumulation. Thromb Haemost. 2002;87:1011–1019.
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013
1
(To be published on-line only)
I. Construction of the transgene
The promoter/enhancer of the mouse adipocyte P2 (aP2) gene as a 5.4 kb DNA fragment
in the plasmid pBluescript SKII(+) (1) was a kind gift of Dr. B.M. Spiegelman (Dana
Farber Cancer Institute, Boston, MA). The mouse PAI-1 cDNA was obtained as a 1.3 kb
PvuII-DraI fragment in pBSmr1 (2). The plasmid that directs fat-specific PAI-1
expression was constructed by inserting the mouse PAI-1 cDNA into the SmaI site of the
pSKII + vector downstream of the aP2 promoter/enhancer. A XbaI-HindIII fragment
containing the SV40 polyadenylation signal was inserted downstream of the PAI-1
cDNA. A 7.2 kb aP2-PAI-1 construct containing the aP2 promoter, PAI-1 cDNA and the
polyadenylation site, was obtained free of vector sequences by ClaI-SacII digestion and
gel purification.
II. Generation and characterization of transgenic mice
The aP2-PAI-1 construct was micro-injected in the pronucleus of fertilized zygotes from
FVB mice and transferred to pseudopregnant females. Offspring carrying the transgene
were identified by PCR on genomic DNA extracted from tail tips with two pairs of
transgene-specific
(5’-GCAAGCAAGGGCTGAAGACATC-3’;
CAAGGGCTGAAGACATCTGC-3’)
and
aP2-specific
5’(5’-
CCACAATGAGGCAAATCCATAAGG-3’; 5’-CATTGCCAGGGAGAACCAAA-3’)
oligonucleotides, which produce 373-bp and 265-bp DNA fragments. PCR amplification
of
the
mouse
b -globin
gene
CCTTGAGGCTGTCCAAGTGATTCAGGCCATCG-3’
(reverse,
and
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013
forward,
5’5’-
2
CCAATCTGCTCACACAGGATAGAGAGGGCAGG-3’ primers) was used as an
internal positive control.
Total RNA from different mouse tissues was prepared using the TRIzol reagent (Life
Technologies, Gaithersburg, MD), treated with RNase-free DNase I using RNeasy mini
spin columns (Qiagen, Valencia, CA) and RNA concentrations were determined with the
RiboGreen RNA quantification kit (Molecular Probes Europe, Leiden, The Netherlands).
To determine transgene PAI-1transcripts, a specific primer pair was designed: forward
(5'-GCAAGTGATGGAGCCTTGACA-3')
and
reverse
(5'-
TTCACTGCATTCTAGTTGTGGTTTG-3'). The expression level of the mRNA was
determined by semi-quantitative RT-PCR using the GeneAmp Thermostable RNA PCR
kit (Applied Biosystems) as described previously (3) and was normalized to the 28S
rRNA level.
Preadipocyte factor 1 (Pref-1) mRNA levels were quantified by real-time RT-PCR. Total
RNA (100 ng) was reverse transcribed for 1 h at 48°C using the Taqman Reverse
Transcription kit and 2.5 µmol/L random hexamers (Applied Biosystems, Foster City,
CA). Quantitative real-time PCR was performed in the ABI-prism 7700 sequence
detector, using the Taqman PCR core reagent kit (Applied Biosystems). Each reaction
contained 300 nmol/L of both forward (5’-AACCATGGCAGTGCATCTG-3’) and
reverse (5’-AGCATTCGTACTGGCCTTTC-3’) primers and 100 nmol/L of FAMlabeled fluorogenic probe (5’-FAM-AAATAGACGTTCGGGCTTGCACCTC-TAMRA3’) designated using the Primer Express software (Applied Biosystems). Samples were
normalized using the housekeeping gene 18S rRNA. Quantitative real-time PCR for 18S
rRNA was performed using the Taqman ribosomal RNA control (Applied Biosystems)
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013
3
according to the manufacturer’s protocol. Each PCR amplification was performed in
duplicate wells, using the following conditions: 2 min at 50°C and 10 min at 94°C,
followed by 40 two-temperature cycles (15 s at 95°C and 1 min at 60°C). To quantify the
results, serial dilutions of reverse transcribed total RNA extracted from a tissue with
known expression of the target were used for calibration.
III. References
1. Ross SR, Graves RA, Greenstein A, Platt KA, Shyu H-L, Mellovitz B, Spiegelman
BM. A fat-specific enhancer is the primary determinant of gene expression for
adipocyte P2 in vivo. Proc Natl Acad Sci USA. 1990; 87: 9590-9594.
2. Carmeliet P, Kieckens L, Schoonjans L, Ream B, Van Nuffelen A, Prendergast G,
Cole M, Bronson R, Collen D, Mulligan RC. Plasminogen activator inhibitor 1 genedeficient mice. I. Generation by homologous recombination and characterization. J
Clin Invest.1993; 92: 2746-2755.
3. Maquoi E, Munaut C, Colige A, Collen D, Lijnen HR. Modulation of the expression
of murine matrix metalloproteinases and their tissue inhibitors with obesity. Diabetes.
2002; 51: 1093-1101.
Downloaded from http://atvb.ahajournals.org/ by guest on February 21, 2013