Vascular Biology |
From the Institut National de la Santé et de la Recherche Médicale (INSERM) U541 (F.T., Z.M., S.L., B.E., A.T.), and the Department of Pathology (M.W.), Institut Fédératif de Recherche Circulation Lariboisière, Hôpital Lariboisière, Paris, France.
Correspondence to Alain Tedgui, PhD, INSERM U541, 41, Bd de la Chapelle, 75475 Paris Cedex 10, France. E-mail tedgui{at}infobiogen.fr
| Abstract |
|---|
|
|
|---|
Key Words: wall shear stress vascular remodeling matrix metalloproteinases
| Introduction |
|---|
|
|
|---|
| Methods |
|---|
|
|
|---|
Fourteen animals were left untreated: 4 were killed on day 3, and 10 were killed on day 15. To examine the role of MMPs in adaptive arterial enlargement, 9 rabbits were given a daily dose of 200 mg DOX, a nonspecific inhibitor of MMP expression or activity13 by oral gavage; 3 were killed on day 3 after creation of the AVF; and 6 were killed on day 15. Twelve rabbits received BB-94 (5 mg/kg BID), a specific MMP inhibitor,11 by intraperitoneal injection; 6 were killed on day 3 after creation of the AVF; and 6 were killed on day 15. In addition, to test the hypothesis that flow-induced NO may activate MMPs, 6 rabbits were given L-NAME (500 mg/L) in their drinking water and were killed on day 3 only, because MMPs were no longer activated by day 15. Treatments with DOX, BB-94, and L-NAME were begun 3 days before creation of the AVF and were continued until the animals were killed.
After 3 or 15 days, rabbits were anesthetized again by using the procedure described above, and 500 U heparin was given intravenously. Carotid arteries and jugular veins of both sides were exposed. The right CCAs served as nonoperated controls. Left and right external arterial diameters were measured exactly in situ by using a focusing eyepiece and a bow compass.4
Hemodynamic
Measurements
In 3 untreated, 3 DOX-treated, and 3 BB-94treated
rabbits, arterial pressure was measured after 15 days by
using a catheter introduced via the femoral artery and connected to a
Statham model P231D pressure transducer (Gould Instruments). Blood flow
velocity was measured in all animals, except in 3 BB-94 and 2
L-NAMEtreated rabbits, on
day 3 by using a perivascular ultrasonic flow velocity transducer
placed around the vessel wall, as described
earlier.4 The ultrasound
probe was connected to a range-gated Doppler
velocimeter (8 MHz, Alvar). Pulsatile and mean
Doppler shifts were recorded on a polygraph recorder
(Gould). Blood flow was measured in the right CCA and in the left CCA
proximal to the AVF.
WSS Calculation
WSS was calculated for right and left CCAs. After
assumption of laminar flow, WSS in dynes per square centimeter was
calculated by using the Poiseuille formula,
WSS=4µQ/
r3 ,
were µ is blood velocity (taken to be 0.0035 poise),
Q is blood flow (in mL/s), and
r is the radius (in
cm).
Gelatin Zymography
Zymographic analysis was performed in all
animals that were killed after 3 days (4 controls, 3 DOX-treated, 6
BB-94treated, and 6
L-NAMEtreated rabbits)
and in 4 control and 3 DOX-treated rabbits killed after 15
days.
After excision from the animals, vessels were washed with cold Hanks buffer and quickly frozen and powdered in LN2 by using a mortar and pestle. The powders were resuspended in ice-cold lysis buffer (20 mmol/L Tris-HCl, pH 7.5; 5 mmol/L EGTA; 150 mmol/L NaCl; 20 mmol/L glycerophosphate; 10 mmol/L NaF; 1 mmol/L sodium orthovanadate; 1% Triton X-100; 0.1% Tween 20; 1 µg/mL aprotinin; 1 mmol/L PMSF; 0.5 mmol/L N-tosyl-L-phenylalanine chloromethylketone; and 0.5 mmol/L N(a)-p-tosyl-L-lysine chloromethylketone) at a ratio of 0.3 mL/10 mg wet weight. Extracts were incubated on ice for 15 minutes and then centrifuged (12 000g, 15 minutes, 4°C). The detergent-soluble supernatant fractions were retained, and protein concentrations in samples were equalized by using a Bio-Rad protein assay. Samples were mixed in an SDSpolyacrylamide gel electrophoresis loading buffer (lacking reducing agents), applied to a 10% SDS-polyacrylamide gel containing 1 mg/mL gelatin (Bio-Rad), and separated by electrophoresis. Subsequently, SDS was removed from the gels by 2 washes (15 minutes) with 2.5% Triton X-100, and the gels were incubated overnight at 37°C in zymography buffer (50 mmol/L Tris, pH 7.5, and 10 mmol/L CaCl2) and stained with Coomassie brilliant blue (Serva). Gelatinolytic activity was visualized as clear areas of lysis in the gel. Densitometric analysis was performed by using NIH Image software.
Histological Studies
In 3 control, 3 DOX-treated, and 3 BB-94treated
animals examined on day 15, both carotid arteries were cannulated at
the level of the thoracic aorta and ligated. The jugular veins were
dissected free, and the end of the vein close to the carotid-jugular
anastomosis was ligated. Animals were then killed with an overdose of
sodium pentobarbital. Blood was washed out by perfusion with normal
saline solution, followed by 3% glutaraldehyde
solution in phosphate buffer. A constant pressure of 100 cm
H2O was maintained for 45 minutes. The CCAs were
then excised and stored in the same solution. Samples were stained with
orcein to reveal elastic fibers or probed with an anti-nitrotyrosine
antibody (clone 1A6, 20 µg/mL; Upstate Biotechnology) as an
indication of peroxynitrite
formation.14
Vascular Reactivity Studies
To ensure that DOX and BB-94 treatments had not
impaired endothelial function, the
endothelium-dependent relaxations to acetylcholine
(ACh) were examined in aortic rings from 3 untreated, 3 DOX-treated,
and 3 BB-94treated rabbits, as previously
described.4 The absence of
endothelium-dependent relaxations to ACh was checked in
3 L-NAMEtreated rabbits.
Responses to ACh (10-8 to
10-4 mol/L) were tested after precontraction
of the aorta with phenylephrine
(10-6 mol/L). Maximal relaxations and ACh
concentrations inducing 50% of maximal relaxation (ie,
IC50) were recorded.
Statistical Analysis
Results are expressed as mean±SEM. A 1-way ANOVA was
constructed to evaluate the effects of DOX and BB-94 treatments.
Multiple comparisons were done by using the Bonferroni test. Intragroup
comparisons between left and right CCAs were done by using a paired
Students t test. Differences
were considered significant at a value of
P<0.05.
| Results |
|---|
|
|
|---|
DOX and BB-94 treatments had no effect on endothelium-dependent relaxations. The maximal ACh-induced relaxations were 81±3% in control animals, 79±3% in DOX-treated rabbits, and 82±4% in BB-94treated rabbits. The IC50 values were similar (0.32±0.05 µmol/L in control animals, 0.039±0.04 µmol/L in DOX-treated animals, and 0.036±0.05 µmol/L in BB-94treated rabbits). By contrast, L-NAME treatment abolished endothelium-dependent relaxations.
Blood Flow
Blood flow measured in the left CCA at baseline before
creation of the AVF did not differ between untreated and DOX- or
BB-94treated animals (24.7±1.1 mL/min versus 26.3±1.8 and 24.1±0.9
mL/min, respectively). Flow measurements on day 15 revealed that the
AVF resulted in large increases in blood flow in left CCAs compared
with right CCAs
(Table
I). These increases were not significantly different
between untreated and DOX- or BB-94treated animals (4.4-fold versus
3.2-fold, P=0.09, and 3.4-fold,
P=0.10, respectively). Mean
blood flow in the right CCA was similar in both
groups.
|
Morphological Analysis
The mean arterial diameter of the CCAs
measured at baseline before creation of the AVF did not differ among
untreated, DOX-treated, and BB-94treated animals (1.93±0.03,
1.94±0.03, and 2.05±0.03 mm, respectively). In untreated
rabbits, the diameter of the flow-loaded left CCA increased by
13.6±1.9% 3 days after construction of the AVF compared with the
right CCA (P<0.01) and by
40.7±7.5% on day 15
(P<0.0001;
Table
II). These increases were abolished in DOX-treated
rabbits (3.6±1.8% on day 3 and 10.4±5.7% on day 15) and in
BB-94treated rabbits (2.2±1.6% on day 3 and 15.2±8.6% on day
15).
|
Calculated WSS
At 3 days postoperatively, calculated WSS values in the
left CCA of untreated and DOX- or BB-94treated animals were
significantly increased compared with the contralateral CCA
(Table
III), even though these increases were lower in
control than in DOX- or BB-94treated rabbits. Owing to the lack of
arterial enlargement, WSS values in left CCAs remained as
high at 15 days as they were at 3 days in DOX-treated animals
(51.1±5.1 versus 48.3±3.2 dyne/cm2,
respectively) and in BB-94treated animals (34.8±6.1 versus 39.5±3.5
dyne/cm2, respectively), whereas it
progressively decreased with time in control animals (30.3±4.7 versus
23.1±2.1 dyne/cm2 on days 3 and 15,
respectively) as a result of continuous adaptive
enlargement.
|
Zymographic Analysis
Gelatin zymographic analysis of protein
extracts from the CCAs obtained at baseline revealed a major lytic band
at 72 kDa, consistent with the pro-form of MMP-2
(Figure
1). A minor lytic band at 62 kDa, corresponding to
the active form of MMP-2, was barely detectable in these arteries.
Three days after construction of the AVF, gelatin zymography of protein
extracts from left CCAs of untreated rabbits showed a significant
increase in the 72- and 62-kDa gelatinolytic
activity
(Figure
1) and the appearance of a lytic band at 92 kDa,
consistent with the pro-form of MMP-9. Fifteen days after
construction of the AVF, zymographic activity in left CCAs had
decreased to baseline values
(Figure
1).
|
In DOX-treated rabbits, the increase in all
gelatinolytic bands (62-, 72-, and 92-kDa
activities) observed on day 3 was significantly reduced
(Figure
2), whereas in BB-94 or
L-NAMEtreated animals,
only the active forms were reduced
(Figure
3).
|
|
Histological
Analysis
In accordance with the morphometric findings,
histological analysis showed IEL fragmentation
and elastic fiber degradation in the left CCAs of untreated rabbits but
not in DOX- or BB-94treated animals
(Figure
4). Furthermore, in untreated and BB-94treated
animals, nitrotyrosine staining revealed the presence of peroxynitrite
in the medial and endothelial layers of flow-loaded
CCAs but neither in the control right CCAs nor in the left CCAs of
L-NAMEtreated rabbits
(Figure
5).
|
|
| Discussion |
|---|
|
|
|---|
Vascular responses to altered blood flow are endothelium dependent.3 5 In large arteries, the acute vasodilatation that accompanies increased flow is mediated by endothelium-derived NO15 and initiated very quickly.4 However, the endothelial mechanisms responsible for the vascular adaptation to chronically altered flow are not fully elucidated. In cases of chronically enhanced blood flow, the resulting arterial enlargement is far beyond the capability of distension owing to vasoactive dilatation; it involves structural remodeling.2 4 16 Blood flowinduced arterial enlargement tends to return shear stress values toward normal levels. Our present findings are in agreement with this paradigm. WSS values in flow-loaded left CCAs were abnormally high 3 days after creation of the AVF but then decreased, so that after 15 days they were no longer significantly different from baseline or right CCA values, as we previously found in the same model after 1 month.4
Previous microscopic and ultrastructural studies of the arterial wall proximal to an AVF have shown extensive tears and fragmentation of the IEL.17 18 19 Jones et al20 showed in rabbits that the CCA proximal to a carotid-jugular AVF exhibited transversely oriented tears of the IEL by day 5 postoperatively. These elastic tears increased in number and width with postoperative time. Nevertheless, at no time were inflammatory cells observed in the vicinity of the elastic tissue tears. Inflammatory cells could therefore not be held accountable for this process. Wong and Langille21 have demonstrated enlarged fenestrae in the IEL of flow-loaded rabbit CCAs by using laser scanning confocal microscopy. In agreement with those studies, we observed extensive IEL fragmentation and elastic fiber degradation in the media of flow-loaded arteries that were fixed at physiological distension. We hypothesize that this IEL fragmentation and elastin degradation could result from increased MMP activity in the arterial wall. Indeed, in the present study, we found an early upregulation and activation of MMP-2 and MMP-9 3 days after construction of the AVF between the left CCA and the external jugular vein. Increased MMP expression and activity were associated with an increase in WSS. This finding does not exclude the possibility that other MMPs were induced and/or activated by the chronic increase in blood flow and may have contributed to matrix degradation. In a recent report, Karwowski et al22 overlooked the role of MMP-2 and MMP-9 in blood flowinduced arterial enlargement because their MMP activity analysis was conducted only at 21 days after AVF creation. Similarly, in our study, MMP activity was no longer enhanced at 15 days.
To examine whether MMP activation was indeed responsible for the adaptive vascular remodeling, rabbits were treated with DOX, a compound with known inhibitory effects on MMP expression or activation,13 or with the synthetic specific MMP inhibitor BB-94.11 Recent studies have reported that RS-113,456, a specific, nonselective MMP inhibitor, diminished flow-mediated arterial enlargement in a rat AVF model, but the effect on vascular MMP activity was not documented.22 23 In our study, treatment with DOX significantly reduced proMMP-9 and proMMP-2 upregulation, as well as MMP-2 activation. BB-94 had no effect on the flow-induced increase in the proenzymes but markedly reduced MMP-2 activation. These findings are consistent with previous studies showing that DOX may act by unknown mechanisms on both MMP expression and activation,24 whereas BB-94 only inhibits MMP activation by binding of its hydroxamic acid group to the catalytically active zinc atom.11 As a result of MMP inhibition, arterial wall remodeling with elastin degradation and subsequent arterial enlargement were prevented. This supports our hypothesis that blood flowinduced MMP activation is critical for the occurrence of blood flowinduced vascular enlargement. Indeed, if MMP-induced IEL fenestrations and elastic fiber degradation follow increased blood flow, then this results in greater arterial distensibility leading to enhanced arterial diameter. As arterial caliber gradually increases, WSS diminishes and the stimulus for MMP production/activation fades. This process will continue until the arterial caliber is such that WSS is normalized.
The present study does not allow for identification of cells that elaborate MMPs. Nevertheless, endothelial cells, which are the sensors of increased blood flow, can release MMPs,25 26 27 and it has been reported that increased shear stress might stimulate MMP production by endothelial cells.28 Also, endothelial cells exposed to elevated shear stress might secrete factors that trigger MMP production by smooth muscle cells. These latter cells are known to be capable of producing MMPs in response to cytokines29 or growth factors, including fibroblast growth factor-2.30 Interestingly, expression of fibroblast growth factor-2 has been shown to be enhanced by fluid shear stress in cultured endothelial cells31 and in an in vivo model of AVF.32 MMPs are secreted as inactive zymogens (pro-MMPs) that require activation in the extracellular compartment. In previous studies, we4 and others33 revealed that long-term blockade of NO synthesis prevents arterial enlargement in response to increased blood flow, yielding an incomplete regulation of WSS, and we suggested that NO may participate in the remodeling process by activating pro-MMPs. Indeed, it has been shown that NO or peroxynitrite, which results from simultaneous production of NO and superoxide anions, activates MMPs.12 34 In agreement with these results, we found that chronic inhibition of the NO/L-arginine pathway by L-NAME treatment prevented MMP-2 activation in flow-loaded CCAs. Furthermore, we detected nitrotyrosine staining, indicative of the presence of peroxynitrite, in endothelial cells and throughout the media of flow-loaded CCAs of untreated and BB-94treated rabbits, but not in L-NAMEtreated animals. We therefore believe that NO (or peroxynitrite) is a significant activator of pro-MMPs in blood flowinduced vascular remodeling. Our finding does not rule out the possibility that other enzymatic systems are involved in triggering MMP activation in response to increased blood flow, in particular, the plasminogen/plasmin system that has been shown to be a significant activator of pro-MMPs in vivo.35
In conclusion, we have shown that (1) MMP-2 and MMP-9 are upregulated and activated in the arterial wall in response to chronic increases in blood flow; (2) MMP activation is required for adaptive arterial remodeling (IEL fragmentation and arterial enlargement) to occur; and (3) NO participates in flow-induced MMP activation.
| Acknowledgments |
|---|
Received March 2, 2000; accepted September 6, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Bouvet, S. Moreau, J. Blanchette, D. de Blois, and P. Moreau Sequential Activation of Matrix Metalloproteinase 9 and Transforming Growth Factor {beta} in Arterial Elastocalcinosis Arterioscler. Thromb. Vasc. Biol., May 1, 2008; 28(5): 856 - 862. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Lehoux Redox signalling in vascular responses to shear and stretch Cardiovasc Res, July 15, 2006; 71(2): 269 - 279. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. G. Lee, S. Dahi, R. Mahimkar, N. L. Tulloch, M. A. Alfonso-Jaume, D. H. Lovett, and R. Sarkar Intronic regulation of matrix metalloproteinase-2 revealed by in vivo transcriptional analysis in ischemia PNAS, November 8, 2005; 102(45): 16345 - 16350. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. K. Wakhloo, B. B. Lieber, R. Siekmann, D. J. Eber, and M. J. Gounis Acute and Chronic Swine Rete Arteriovenous Malformation Models: Hemodynamics and Vascular Remodeling AJNR Am. J. Neuroradiol., August 1, 2005; 26(7): 1702 - 1706. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. von Offenberg Sweeney, P. M Cummins, Y. A Birney, J. P Cullen, E. M Redmond, and P. A Cahill Cyclic strain-mediated regulation of endothelial matrix metalloproteinase-2 expression and activity Cardiovasc Res, September 1, 2004; 63(4): 625 - 634. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Konner, B. Nonnast-Daniel, and E. Ritz The Arteriovenous Fistula J. Am. Soc. Nephrol., June 1, 2003; 14(6): 1669 - 1680. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2000 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |