(Circulation. 2001;103:3123.)
© 2001 American Heart Association, Inc.
Basic Science Reports |
From the Department of Cardiovascular Medicine and the Laboratory of Molecular and Cellular Pathology (H.S.), CREST, Hokkaido University School of Medicine, Sapporo, Japan, and Department of Medicine, University of Vermont, Burlington (B.E.S.).
Correspondence to Burton E. Sobel, MD, Department of Medicine, University of Vermont College of Medicine, Colchester Research Facility, 208 S Park Dr, Colchester, VT 05446. E-mail burton.sobel{at}vtmednet.org
| Abstract |
|---|
|
|
|---|
Methods and ResultsTo determine whether hypofibrinolysis in blood and tissues and its potential sequelae could be attenuated pharmacologically, we studied genetically modified obese mice. By 10 weeks of age, obese mice exhibited increases in left ventricular weight and glucose and immunoreactive insulin in blood. PAI-1 activity in blood measured spectrophotometrically was significantly elevated as well. The difference compared with values in lean controls widened by 20 weeks of age. Perivascular fibrosis in coronary arterioles and small coronary arteries was evident in obese mice 10 and 20 weeks of age, paralleling increases in PAI-1 and tissue factor expression evident by immunohistochemical image analysis, in situ hybridization, and reverse transcriptionpolymerase chain reaction. Inhibition of ACE activity initiated in obese mice 10 weeks of age and continued for 20 weeks arrested the increase in PAI-1 activity in blood and in cardiac PAI-1 and tissue factor mRNA as well as coronary perivascular fibrosis.
ConclusionsThus, inhibition of proteo(fibrino)lysis and augmented tissue factor expression in the heart precede and may contribute to the coronary perivascular fibrosis seen with obesity and insulin resistance. Furthermore, inhibition of ACE activity can attenuate all 3 phenomena.
Key Words: fibrinolysis coronary disease diabetes mellitus insulin
| Introduction |
|---|
|
|
|---|
The present study was designed to characterize expression of proteins involved in the fibrinolytic and coagulation systems in relation to coronary microvascular changes in genetically obese mice (ob/ob) that develop insulin resistance and noninsulin-dependent diabetes mellitus10 and to determine whether the inhibitor of ACE activity, known to lower PAI-1 concentrations,11 12 attenuates PAI-1 expression, perivascular fibrosis, or both in obese mice.
| Methods |
|---|
|
|
|---|
In animals 10 weeks of age (10 control and 10 obese mice) and in animals 20 weeks of age (10 control, 10 obese, and 7 temocapril-treated), anesthesia was induced with ether. Blood samples were drawn from the left ventricle. The hearts were perfused with PBS, excised, rinsed in PBS, and weighed. A portion of the left ventricle was embedded into O.C.T. compound (Miles), frozen in liquid nitrogen, stored at -80°C, cut into sections 10 µm thick with a cryostat, and stained with elasticaMassons trichrome. The remaining tissue was frozen in liquid nitrogen and stored at -80°C. Glucose and PAI-1 activity in blood were measured as described previously.5 Insulin was assayed by ELISA specific for mouse insulin (Shibayagi). Left ventricular myocardiumembedded coronary luminal areas, vessel walls, and perivascular fibrosis were quantified as described previously.13
Immunohistochemistry
PAI-1 and TF were detected immunohistochemically with
the streptavidinbiotin immunoperoxidase method
(Nichirei).14
In brief, frozen sections (10 µm) fixed in acetone were immersed in
3% H2O2 for 15 minutes
and incubated with PBS containing 10% rabbit serum (for PAI-1) and 5%
goat serum (for TF) to block nonspecific staining. The sections
were rinsed with PBS and incubated with anti-mouse PAI-1 sheep IgG
(American Diagnostica) or with
anti-human TF guinea pig
IgG15 diluted 1:250. After a
washing with PBS, the sections were incubated with biotinylated
anti-sheep rabbit IgG (Chemicon) for PAI-1 and anti-guinea pig goat IgG
(Vector) for TF. They were again washed with PBS, incubated with
peroxidase-labeled streptavidin, reacted with 3,3'-diaminobenzidine,
and counterstained with hematoxylin. Control staining was performed
with isotype IgG as the primary antibody. Qualitative assessment of
intensity of staining was performed with a visual grading scale (0 to
++++). Cardiac sections were characterized by comparison with grading
scale sections by 2 observers blinded to the categories of the
animals.16 Quantitative
image analysis of the sections was performed as previously
described16 by observers
blinded with respect to the groups of animals. The
immunohistochemically stained slides were viewed with a light
microscope, and digital gray-scale images were acquired with a CCD
camera with constant settings. Image collection and analysis
were performed with Microcomputer Imaging Device software
(Imaging Research). Gray-scale values (pixel
intensities) within the regions of interest were plotted as histograms,
and minimum, maximum, and mean pixel intensity values were calculated
with conventional software for comparison of intensities of
immunoperoxidase reaction products. Data are expressed as intensity
units above values with isotope IgG used as the primary antibody and as
a control for comparison. For identification of
endothelial cells, immunohistochemical staining was
performed with antifactor VIIIrelated antigen polyclonal antibody
(Nichirei).
Expression of PAI-1 and TF mRNA
Total RNA isolated from left ventricular
myocardium was used for first-strand cDNA synthesis. The
reverse transcriptionpolymerase chain reaction (RT-
PCR) with selected primers was used for amplification of PAI-1, TF, and ß-actin mRNA as previously described.17 PAI-1 and TF sequences were amplified in a thermal cycler (Perkin-Elmer) for 35 cycles. The quality of RNA preparation and cDNA synthesis was verified by amplifying DNA coding ß-actin, a housekeeping protein, under the same conditions.
RT-PCR products were visualized on 2% agarose gels with ethidium bromide. Signals were digitized and evaluated with an optical scanner (GT-9500, Seiko) with density measured with the use of an NIH image program in the public domain (Research Services Branch, NIH). In situ hybridization was performed as previously described.18 In brief, an 856-bp fragment of mouse PAI-1 (nucleotides 475 to 1330) derived from mouse PAI-1 cDNA was amplified by PCR and subcloned to pBluescript II SK (+) vector (Stratagene). Sense and antisense probes were prepared by linearizing the constructs with appropriate restriction enzymes and transcribed in vitro in the presence of digoxigenin-UTP and digoxigenin-labeling mixture with either T7 or T3 RNA polymerase (Roche Diagnostic). Frozen sections were fixed in 4% paraformaldehyde/PBS and treated with 0.25% acetic anhydride. Sections were prehybridized with 200 µL hybridization buffer (5xSSC, 50% formamide, 5xDenhardts solution, and 500 µg/mL tRNA) and hybridized with 40 ng of digoxigenin-labeled riboprobe in 100 µL of hybridization buffer at 70°C for 16 hours. They were washed with 5xSSC at 70°C for 10 minutes, 0.2xSSC at 70°C for 60 minutes; treated with 0.1 mol/L maleic acid buffer for 5 minutes and 1% blocking reagent containing maleic buffer for 60 minutes; and incubated with alkaline phosphataselabeled anti-digoxigenin antibody (1:5000) for 60 minutes (Roche Diagnostic). Signals were detected with X-Phosphate/nitro blue tetrazolium. Serial sections were hybridized with sense probe as negative controls.
Statistical Analysis
Comparisons between groups were performed by ANOVA
with multiple comparisons (Fishers protected least significance
t test) with Stat View (Abacus
Concepts). Morphological data were analyzed with
nonparametric statistics (Kruskal-Wallis for ANOVA
designs). Results expressed are mean±SD. A value of
P<0.05 was considered
significant.
| Results |
|---|
|
|
|---|
|
Glucose and Insulin Concentrations and PAI-1
Activity in Blood
Increased concentrations of glucose in blood were
evident in obese mice 10 weeks (231±15 versus 103±7 mg/dL in control,
P<0.01) and 20 weeks (242±16
versus 116±9 mg/dL, P<0.01)
of age. Immunoreactive insulin was increased in obese mice 10 weeks
(2.98±0.49 versus 0.56±0.19 ng/mL,
P<0.01) and 20 weeks
(3.03±0.36 versus 1.02±0.35 ng/mL,
P<0.01) of age as well.
Temocapril did not affect the blood glucose or insulin concentrations
in obese mice (results not shown). PAI-1 activity was higher in plasma
from obese mice 10 weeks of age (5.5±3.2 arbitrary units [AU]/mL)
than in controls (2.0±0.9 AU/mL,
P<0.05). A difference was also
evident in animals 20 weeks of age (3.5±1.9 versus 2.0±0.3 AU/mL,
P<0.05). Temocapril
significantly attenuated the increase in PAI-1 activity in obese mice
20 weeks of age (1.2±0.6 AU/mL,
P<0.05).
Morphometric Analysis
Considerable fibrosis was seen around coronary
arterioles and small arteries in obese mice 10 and 20 weeks of age
(Figure 2
), with a significant increase in the
fibrosis-to-lumen and fibrosis-to-wall ratios at 10 and 20 weeks of age
compared with values in controls. Treatment with temocapril attenuated
these increases.
|
Immunohistochemistry
Although only faint immunoreactivity of PAI-1 and TF
was detected in microvessels of control mice 20 weeks of age, strong
signals were seen in endothelium in obese mice 20 weeks
of age
(Figure 3
). Immunoreactivity of PAI-1 and TF was less in the
specimens from obese mice treated with temocapril from 10 to 20 weeks
of age than in those from untreated obese mice. As shown in
Figure 3
and in the
Table
,
samples from obese mice exhibited significantly more PAI-1 than those
from lean controls (3.32±0.62 versus 1.2±0.39 AU,
P<0.05). Temocapril attenuated
the increase (2.00±0.85 AU,
P<0.005). Samples from obese
mice exhibited significantly more TF (3.58±0.53 AU) than those
from lean control mice (1.15±0.35 AU,
P<0.05)
(Figure 3
, Table
).
Temocapril attenuated this increase as well (1.68±0.51 AU,
P<0.005). Quantitative
(automated image analysis) results were consistent with
these observations visually
(Table
).
Thus, as shown in the
Table
and
Figure 4
, PAI-1 and TF were increased in the obese mouse
samples, and temocapril attenuated the increases.
|
|
|
The abundance of PAI-1 or of TF relative to ß-actin in
hearts from obese and lean mice is shown in
Figure 5
. PAI-1 mRNA was markedly increased in obese
compared with lean mice 10 weeks of age. TF mRNA was increased as well,
but less markedly. PAI-1 mRNA increased further in obese mice 20 weeks
of age, as did TF mRNA. Treatment with temocapril significantly reduced
the magnitude of the increase in PAI-1 mRNA and TF mRNA in obese mice,
but not to control levels. Expression of ß-actin mRNA was not
altered. In situ hybridization demonstrated PAI-1 mRNA in medial layers
within the arteries
(Figure 6
), consistent with our
immunohistochemical results
(Figures 3
and 4
). No positive signals were observed in the
sense control sections.
|
|
| Discussion |
|---|
|
|
|---|
Fibrinolytic system activity is diminished in diabetic and insulin-resistant nondiabetic human subjects,5 19 which may be attributable to several mechanisms, including hyperinsulinemia and oxidative stress.20 21 22 Tumor necrosis factor, which is increased in blood and adipose tissue with diabetes, increases PAI-1 production.23 These factors may lead to impaired fibrinolysis well before the onset of diabetes.
Coronary Remodeling in Obese
Mice
We found that obese mice exhibit increased perivascular
fibrosis, perhaps reflecting the above alteration in local
concentrations of fibrinolytic and coagulation system proteins. The
increased PAI-1 is hypothesized to have attenuated generation of
plasmin and reduced proteolysis of
matrix.9 The increased TF is
hypothesized to have induced procoagulant activity, thereby providing a
provisional fibrin matrix augmenting cell
migration.8 24
Although blood pressure was not measured in animals in this study, leptin, the product of the ob gene, increases blood pressure25 and is deficient in ob/ob mice. Furthermore, the blood pressure of obese-hyperglycemic C57BL/6 mice produced by breeding mice heterozygous for the obesity gene is normal.26 Thus, it is unlikely that elevated blood pressure and its amelioration by temocapril accounted for our results.
Diabetic patients develop a paucity of coronary collaterals, indicative of limited angiogenesis.27 28 Because altered characteristics of left ventricular capillaries in streptozotocin-induced (insulin-deficient) diabetic rats have been reported,29 in a preliminary study we characterized capillary profiles in obese mice. Remodeling appeared to be constrained in the obese animals (unpublished observations).
By augmenting PAI-1 expression, angiotensin influences fibrinolytic and TGF-ß activity.11 12 It can also inhibit insulin signaling.30 In the present study, inhibition of ACE activity attenuated PAI-1 expression and microvascular remodeling. Thus, angiotensin II may increase expression of procoagulant and hypo(proteo)fibrinolytic effects in the heart associated with insulin resistance, which is potentially ameliorated by inhibition of ACE activity with attenuation of perivascular fibrosis in obese mice by reducing expression of PAI-1 and TF.31 32 33 34 35 36 These mechanisms may contribute to the favorable effects of inhibition of ACE activity seen in patients at high cardiovascular risk.37
| Acknowledgments |
|---|
Received October 19, 2000; revision received February 14, 2001; accepted February 22, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Fujita, N. Maeda, M. Sonoda, K. Ohashi, T. Hibuse, H. Nishizawa, M. Nishida, A. Hiuge, A. Kurata, S. Kihara, et al. Adiponectin Protects Against Angiotensin II-Induced Cardiac Fibrosis Through Activation of PPAR-{alpha} Arterioscler. Thromb. Vasc. Biol., May 1, 2008; 28(5): 863 - 870. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. D. Abel, S. E. Litwin, and G. Sweeney Cardiac Remodeling in Obesity Physiol Rev, April 1, 2008; 88(2): 389 - 419. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Thalmann and C. A. Meier Local adipose tissue depots as cardiovascular risk factors Cardiovasc Res, September 1, 2007; 75(4): 690 - 701. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Musaad and E. N. Haynes Biomarkers of Obesity and Subsequent Cardiovascular Events Epidemiol. Rev., May 10, 2007; (2007) mxm005v1. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Takeshita, M. Hayashi, S. Iino, T. Kondo, Y. Inden, M. Iwase, T. Kojima, M. Hirai, M. Ito, D. J. Loskutoff, et al. Increased Expression of Plasminogen Activator Inhibitor-1 in Cardiomyocytes Contributes to Cardiac Fibrosis after Myocardial Infarction Am. J. Pathol., February 1, 2004; 164(2): 449 - 456. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. E. Sobel, D. J. Taatjes, and D. J. Schneider Intramural Plasminogen Activator Inhibitor Type-1 and Coronary Atherosclerosis Arterioscler. Thromb. Vasc. Biol., November 1, 2003; 23(11): 1979 - 1989. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Yu, K. Ohmori, I. Kondo, L. Yao, T. Noma, T. Tsuji, K. Mizushige, and M. Kohno Correlation of functional and structural alterations of the coronary arterioles during development of type II diabetes mellitus in rats Cardiovasc Res, November 1, 2002; 56(2): 303 - 311. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |