Articles |
From the Division of Cardiology, Department of Medicine (K.M.C., K.G.P., B.H.A.) and the Department of Surgery (G.J.F., M.G.D., P.-O.H.), Duke University Medical Center, Durham, NC, and the Cardiovascular Division, Department of Medicine, Yale University, New Haven, Conn (M.D.E.).
Correspondence to Brian H. Annex, MD, Durham VA Medical Center, 508 Fulton St (111A), Durham NC 27705.
| Abstract |
|---|
|
|
|---|
Key Words: intimal hyperplasia vascular surgery thrombosis inflammation
| Introduction |
|---|
|
|
|---|
TF is a transmembrane glycoprotein that binds to and activates factor VII.6 7 By catalyzing the rate-limiting step in the extrinsic coagulation cascade, TF is the major determinant of thrombin production in vivo.8 Immunohistochemical and in situ hybridization studies have shown that TF protein and mRNA are abundant in the adventitia of normal blood vessels.9 10 However, TF is consistently undetectable in the intima and endothelium of normal blood vessels, so that TF remains sequestered from circulating blood unless vessel injury occurs.9 10 In diseased vessels, however, TF expression has been demonstrated in carotid endarterectomy specimens10 and coronary artery lesions sampled by directional atherectomy.11
TF appears to play an important role in the response to vessel injury. Arterial balloon injury results in rapid upregulation of TF mRNA expression in medial smooth muscle, and TF mediates the increased procoagulant activity that has been observed in balloon-injured arterial segments.12 13 Coronary ECs express TF as a consequence of postischemic reperfusion injury.14 TF expression can be induced or upregulated in many cell types by agents or physiological stimuli that may play a role in atherosclerosis and in response to arterial injury.12 13 14 15 16 In addition, thrombin generation by TF also has pleiotropic actions,17 including stimulation of SMC proliferation,18 19 recruitment of inflammatory cells,20 and activation of ECs.21
The purpose of this study was to examine the changes in TF protein expression in a well established interposition rabbit jugular vein/carotid artery bypass graft model.22 23 In this model, venous bypass grafts undergo well characterized changes that can be divided into two temporal phases. In early vein grafts (postoperative days 1 to 3), nondenuding endothelial injury, inflammation, limited fibrin deposition, and platelet aggregation occur, but occlusive thrombosis is not a feature of this model.23 Prior studies using electron microscopy have demonstrated that polymorphonuclear leukocytes are the predominant cell type in the inflammatory infiltrate.23 In late vein grafts (postoperative days 14 to 28), the inflammatory infiltrate resolves, the functional integrity of the endothelium is restored, and SMC proliferation results in intimal hyperplasia.5 22 We hypothesized that modulation of TF protein might be an important feature of the injury response of veins to bypass grafting and might provide insights into the pathogenesis of vein graft failure.
| Methods |
|---|
|
|
|---|
Tissue Processing
After overnight fixation, tissues were prepared for both
paraffin-embedded and frozen sections. Vessel segments were either
dehydrated in a graded series of alcohol and embedded in paraffin
(Paraplast, Oxford Labware) or placed in 30% sucrose in PBS for 2
hours, embedded in OCT compound (Miles Scientific), and frozen in
liquid N2. Sections (6 µm) were prepared on
silane-coated glass microscope slides. Samples of rabbit skin and lung
served as control tissues for histologic and immunohistochemical stains
and were handled in a similar manner.
Immunohistochemistry
Immunohistochemistry for TF was performed using the murine
monoclonal anti-rabbit TF antibody AP-1 as previously
described.24 In brief, sections were deparaffinized in
xylene, rehydrated in a graded series of alcohol, and equilibrated in
PBS. Blocking solution (10% horse serum in PBS) was applied for 1 hour
at room temperature or overnight at 4°C. AP-1 was diluted in blocking
solution to 4 µg/mL and applied to tissue sections for 1 hour at
37°C. Incubation with the primary antibody was followed by sequential
incubation with biotinylated anti-mouse IgG and ABC reagent
according to the manufacturer's specifications (Vectastain ABC kit,
Vector Laboratories). Levamisole was added to block
endogenous alkaline phosphatase activity, and immune
complexes were localized with the use of the chromogenic
alkaline phosphatase substrate Vector Red (Vector Laboratories). The
sections were counterstained with hematoxylin, dehydrated, and mounted
with Permount (Fisher Scientific). In all experiments, rabbit skin was
included as a positive control. Two different negative controls were
used. First, for each vessel examined, a serial section was incubated
with a nonsense murine IgG monoclonal antibody or with a murine IgG
monoclonal antibody against human TF (TF9-9C3,9 11 ).
Second, we demonstrated the specificity of the AP-1 antibody by the
elimination of TF immunostaining when AP-1 antibody was
incubated overnight with rabbit brain thromboplastin (Organon Teknika)
Immunostaining for TF did not differ between
paraffin-embedded and frozen sections. Paraffin-embedded sections were
therefore used for all analysis owing to improved preservation
of tissue architecture and morphology.
To identify the cell types associated with TF expression in rabbit vessels, two approaches were used. First, serial sections were examined by using previously characterized mouse monoclonal antibodies. SMCs, ECs, leukocytes, or tissue macrophages were stained with antibodies directed against human smooth muscle actin (HHF35, Dako), human vWF (American Diagnostica), rabbit CD18 (Serotec), or rabbit RAM11 (Dako), respectively, at the supplier's recommended concentration. The anti-CD18 antibody recognizes the ß2 chain of the leukocyteadherence glycoprotein complex.25 This integrin is present on activated monocytes and neutrophils and mediates endothelial adhesion and migration.26 We confirmed the specificity of the CD18 antibody by the consistent absence of staining in sections of control veins and carotid artery. RAM11 was originally raised against alveolar macrophages and has been widely used to detect macrophages in rabbit arteries and vein grafts.27 28 Accordingly, histologic sections of rabbit lung were used as a positive control. Immunostaining with these antibodies was performed as described for the AP-1 antibody; alternatively, sections were pretreated with 3% H2O2 and biotinylated secondary antibodies, followed by an avidin-biotin-peroxidase conjugate and NiCl2-enhanced 3,3' diaminobenzidine (Vector Laboratories) to yield a black reaction product. In addition, early (day 1 or 3) vein grafts underwent cytochemical stains with PAS, which may be used to differentiate leukocyte types; the cytoplasm of PAS-positive cells stain red.29 Therefore, immunostaining for TF, CD18, or RAM11 with the peroxidase method was followed by histologic staining with PAS (Sigma Chemical Co).
Analysis of Tissue Sections
For analysis, vessels were categorized as control
artery, control vein, early vein graft (day 1 or 3 after grafting), or
late vein graft (day 14 or 28 after grafting). The intima, media, and
adventitia of representative vessel sections were
defined by using sections stained with hematoxylin-eosin and Masson's
trichrome as previously described.22 23 The adventitia of
all vessels showed abundant TF staining. To compare TF protein
expression between vessel categories, TF immunostaining
in the intima of the vessel was graded as present or absent in each
vessel quadrant. The numbers of quadrants with TF
immunostaining were compared by a
2 test. A value of P<.05 was
considered significant.
TF Activity and Western Blot Analysis
Protein was extracted from frozen tissue samples by grinding the
tissue to a fine powder in a mortar and pestle in liquid
N2, followed immediately by sonication in 0.5 mL ice-cold
Tris-saline buffer (0.05 mol/L Tris, pH 7.4, 150 mmol/L NaCl) as
previously described.9 Insoluble debris was pelleted in a
microcentrifuge at 14 000g for 5 minutes. Total
protein concentration in the supernatant was determined with the
Bradford assay (Bio-Rad Laboratories).
TF activity in vessel protein extracts was determined by a one-step recalcification assay as previously described.30 In brief, duplicate portions of vessel protein extract were added to citrate-anticoagulated, pooled, normal human platelet-poor plasma. The time taken to produce a fibrin clot after addition of CaCl2 at 37°C was determined with a fibrometer (BBL Fibrosystems). TF activity in vessel extracts was calculated by comparison with standard dilutions of a partially purified rabbit brain thromboplastin preparation (100 000 mU thromboplastin activity per milliliter).
For Western blot analysis, equal samples of vessel protein extract (50 µg total protein) were separated by SDSpolyacrylamide gel electrophoresis and electrophoretically transferred to a nitrocellulose membrane. The membrane was blocked in 5% nonfat milk and incubated with AP-1 (2 µg/mL) for 1 hour at room temperature. Protein bands were visualized with an anti-mouse IgG horseradish peroxidaseconjugated antibody followed by chemiluminescence (Dupont-NEN).
| Results |
|---|
|
|
|---|
TF Expression in Control Artery, Control Veins, and Early Vein
Grafts
Localization of TF Protein
Representative histologic and immunohistochemical
sections of rabbit control vein, early vein grafts, and control carotid
artery are shown in Fig 1
. In control veins (the
contralateral external jugular vein not used in the bypass operation),
intense TF staining was present throughout the adventitia and SMCs
of the thin media but not in ECs of the intima (Fig 1A
). In control
veins (n=6), TF immunostaining in the intima was
detected in only 5 of 24 vessel quadrants. In contrast, all early vein
grafts (n=18) displayed TF staining in the intima (72 of 72 vessel
quadrants, P<.001 versus control veins; Fig 1B
and 1C
).
|
Surgical manipulation alone was not responsible for the increased TF
expression in early vein grafts. The wall of the carotid artery segment
that formed the surgical anastomosis in early vein grafts demonstrated
TF staining in the adventitia but not in the smooth muscle media or
intima (Fig 1F
) and was similar to control carotid artery (the
contralateral artery not used in the bypass operation; Fig 1D
).
Identification of Cell Types Associated With TF Protein in Early
Vein Grafts
Serial sections were immunostained for TF and with a
series of antibodies against different cell types (Fig 2
). RAM11-positive macrophages were present
in the adventitia but not the intima and did not colocalize with TF
staining (Fig 2D
). HHF35-positive SMCs were present in the media,
which was thinned and partially disrupted (Fig 2E
). TF staining mainly
colocalized with areas characterized by CD18-positive leukocyte
infiltration (see insets to Fig 2A
and 2B
). Indeed, the entire vessel
wall of early vein grafts was heavily infiltrated by CD18-positive
leukocytes, which were adherent to the endothelial
surface and formed aggregates in the subendothelial
space (Figs 1B
and 2B
). In many places the integrity of the
endothelium was disrupted by infiltrating CD18-positive
leukocytes (Fig 2B
and 2C
). The nuclei of the majority of the cells in
the inflammatory infiltrate were multilobed (Fig 1B
and 1C
). In
addition, we found that these cells were PAS-positive and that CD18 and
TF colocalized in these PAS-positive cells (not shown). These results
demonstrate that increased TF protein expression in early vein grafts
is spatially associated with PAS-positive, CD18-positive
leukocytes.
|
TF Expression in Late Vein Grafts
There was a marked contrast in TF expression between late vein
grafts and early vein grafts. In late vein grafts (day 14 or 28),
intimal hyperplasia had developed, which was clearly defined on
histologic sections and had stained heavily for SMCs (Fig 3C
).22 23 The luminal surface was lined
with an intact endothelium (Fig 3D
). RAM11-positive
macrophages and CD18-positive leukocytes were rarely detected
in late vein grafts. Staining for TF was present in the adventitia
and to a lesser extent in SMCs of the media, but the intima was
practically devoid of TF staining (Fig 3B
). In late vein grafts (n=8),
TF staining in the intima was present in only 6 of 32 vessel
quadrants (P<.001 versus early vein grafts, P=NS
versus control vein).
|
TF Activity and Protein Expression in Control Veins and Vein
Grafts
TF activity in whole-vessel protein extracts was similar in
control veins and early and late vein grafts (mean±SD: control vein,
0.27±0.07; early vein grafts, 0.25±0.10; and late vein grafts,
0.42±0.08; all in units per milligram protein). Western blot
analysis of vessel total protein revealed no differences in the
quantity of TF protein (as a proportion of total protein) between
control veins, early vein grafts, and late vein grafts (data not
shown).
| Discussion |
|---|
|
|
|---|
Increased TF Protein Expression in the Intima of Early Vein
Grafts
The first major finding of this study was that TF protein was
increased in the intima of early vein grafts compared with control
veins. Importantly, the wall of the carotid artery that formed the
arteriovenous anastomosis with early vein grafts did not show TF
expression in the media or intima. Thus, increased TF expression in the
intima of early vein grafts appears to be a specific characteristic of
the injury sustained by a vein after arterial
bypass grafting and is not a general result of surgical
manipulation.
Our results showed that TF protein was present in the intima of vein grafts 24 hours after surgery and lasted for at least 3 days thereafter. This time course is prolonged compared with arterial injury models, wherein TF mRNA, protein, and procoagulant activity are detectable within 2 hours of balloon injury but return to baseline after 24 hours.12 13 The prolonged increase in TF expression in the intima of vein grafts may be the result of sustained vessel injury in veins subjected to the arterial circulation in contrast with the single episode of balloon injury in arteries. Also, in arterial injury, TF expression is upregulated in medial smooth muscle,12 whereas SMCs in early vein grafts have not been found to be consistently associated with TF expression. These findings may result from the fundamental differences between arterial balloon injury and venous bypass grafting, or they may reflect biological differences between arteries and veins.
In contrast to arterial bypass conduits, acute thrombosis is a significant cause of vein graft failure.2 3 As a key initiator of the extrinsic coagulation cascade, TF plays a pivotal role in intravascular thrombosis and mediates the increased procoagulant activity after arterial injury.8 11 12 13 In the model used in our study, endothelial fibrin deposition and platelet aggregation occur in early vein grafts.23 Therefore, TF may play a role in thrombosis in vein grafts, which supports the hypothesis that strategies aimed at inhibiting TF may be beneficial in reducing acute thrombosis in venous bypass grafts.31
TF Protein Expression in Late Vein Grafts With Intimal
Hyperplasia
The second major finding of this study was that TF protein was not
present in the intimal hyperplasia of late vein grafts. TF protein
expression therefore precedes the development of intimal hyperplasia in
vein grafts. Intimal hyperplasia is the universal response of veins
that are placed in the arterial circulation and is the
precursor of accelerated vein graft
atherosclerosis.32 33 TF per se or
TF-mediated thrombin generation may play a role in the development of
intimal hyperplasia after arterial
injury.6 12 13 Thrombin is a potent mediator of
platelet aggregation, resulting in the release of smooth muscle
mitogens such as platelet-derived growth factor.17 34
In addition, thrombin is a direct mitogen for
SMCs.17 18 19
The complex intravascular coagulation cascade offers several potential sites for pharmacological or molecular strategies to modulate the effects of thrombin on intimal hyperplasia.31 However, although TF is the major source of thrombin, which likely plays an important role in the development of intimal hyperplasia, TF may also have effects independent of thrombin production that can affect the response to vessel injury. In tumors, overexpression of TF increases vascular endothelial growth factor activity and thereby appears to modulate EC growth and development, despite the inhibition of thrombin.35 36 This model of venous bypass grafting provides the opportunity to study in vivo the effects of inhibiting TF or thrombin and to observe these effects on the development of intimal hyperplasia.
Cell Types Associated With TF Expression in Early Vein
Grafts
The third major finding of the present study was the close
association of CD18-positive leukocytes with areas of increased TF
expression in the intima of early vein grafts. Some of these leukocytes
had multilobed nuclei and a majority were PAS-positive. A number of
different cell types in the vessel wall have been shown to express TF
when activated in vitro or in models of arterial
injury, including ECs, SMCs, and monocytes.6 7 12 13 14 15 16 Our
findings do not exclude the possibility of some TF expression in early
vein grafts by ECs, RAM11-negative macrophages, or vascular
SMCs. Nevertheless, CD18-positive leukocytes were consistently
associated with areas of intense TF staining and appeared to be
responsible for the increased TF expression in the intima of early vein
grafts.
Mechanism of TF Expression by CD18-Positive Leukocytes in Vein
Grafts
The results of this study further support the link between
endothelial injury and the pathogenesis of intimal
hyperplasia.4 22 23 The CD18 complex on leukocytes
interacts specifically with integrins such as intercellular adhesion
molecule 1, which is present on ECs only after
activation.26 There is evidence that TF expression by
mononuclear cells is signaled by endothelial adhesion
through E-selectin37 and P-selectin,38
suggesting that TF expression by CD18-positive leukocytes in vein
grafts results from leukocyte adherence to and migration across the
injured endothelium.
Some limitations of this study need to be considered. First, spatial changes in TF protein expression were localized by immunohistochemistry, which is subject to inherent limitations. However, all studies were performed with negative and positive controls, and the antibody used in this study (AP-1) has been previously characterized and functions as an inhibitor of TF activity.14 24 We further showed the specificity of this antibody by demonstrating a single dominant protein band on Western blots of total vessel protein extracts. Second, owing to the overwhelming TF protein expression in the adventitia, we were not able to demonstrate differences in overall vessel TF protein or procoagulant activity in early vein grafts compared with control vein or late vein grafts. In arteries, a thick muscular media permits enzymatic separation of the intima/media from the adventitia and the quantification of proteins in different regions.14 In control veins and early vein grafts, the extremely thin media would make this approach difficult.
| Selected Abbreviations and Acronyms |
|---|
|
|
| Acknowledgments |
|---|
| Footnotes |
|---|
Received April 23, 1996; accepted September 26, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Schachner, G. Laufer, and J. Bonatti In vivo (animal) models of vein graft disease. Eur. J. Cardiothorac. Surg., September 1, 2006; 30(3): 451 - 463. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Schachner Pharmacologic inhibition of vein graft neointimal hyperplasia J. Thorac. Cardiovasc. Surg., May 1, 2006; 131(5): 1065 - 1072. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Tabuchi, M. Shichiri, A. Shibamiya, T. Koyama, F. Nakazawa, J. Chung, S. Hirosawa, and M. Sunamori Non-viral in vivo thrombomodulin gene transfer prevents early loss of thromboresistance of grafted veins Eur. J. Cardiothorac. Surg., November 1, 2004; 26(5): 995 - 1001. [Abstract] [Full Text] [PDF] |
||||
![]() |
Q. Dai, J. Huang, B. Klitzman, C. Dong, P. J. Goldschmidt-Clermont, K. L. March, J. Rokovich, B. Johnstone, E. J. Rebar, S. K. Spratt, et al. Engineered Zinc Finger-Activating Vascular Endothelial Growth Factor Transcription Factor Plasmid DNA Induces Therapeutic Angiogenesis in Rabbits With Hindlimb Ischemia Circulation, October 19, 2004; 110(16): 2467 - 2475. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Kwei, G. Stavrakis, M. Takahas, G. Taylor, M. J. Folkman, M. A. Gimbrone Jr, and G. Garcia-Cardena Early Adaptive Responses of the Vascular Wall during Venous Arterialization in Mice Am. J. Pathol., January 1, 2004; 164(1): 81 - 89. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Baetta, M. Camera, C. Comparato, C. Altana, M. D. Ezekowitz, and E. Tremoli Fluvastatin Reduces Tissue Factor Expression and Macrophage Accumulation in Carotid Lesions of Cholesterol-Fed Rabbits in the Absence of Lipid Lowering Arterioscler. Thromb. Vasc. Biol., April 1, 2002; 22(4): 692 - 698. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Y. Kim, P. L. Walinsky, F. D. Kolodgie, C. Bian, J. L. Sperry, C. B. Deming, E. A. Peck, J. G. Shake, G. B. Ang, R. H. Sohn, et al. Early Loss of Thrombomodulin Expression Impairs Vein Graft Thromboresistance: Implications for Vein Graft Failure Circ. Res., February 8, 2002; 90(2): 205 - 212. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. E. J. West, T. J. Guzik, E. Black, and K. M. Channon Enhanced Superoxide Production in Experimental Venous Bypass Graft Intimal Hyperplasia : Role of NAD(P)H Oxidase Arterioscler. Thromb. Vasc. Biol., February 1, 2001; 21(2): 189 - 194. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. Schonbeck, F. Mach, G. K. Sukhova, M. Herman, P. Graber, M. R. Kehry, and P. Libby CD40 Ligation Induces Tissue Factor Expression in Human Vascular Smooth Muscle Cells Am. J. Pathol., January 1, 2000; 156(1): 7 - 14. [Abstract] [Full Text] [PDF] |
||||
![]() |
G.C. Hughes, B. H. Annex, B. Yin, A. M. Pippen, P. Lin, A. P. Kypson, K. G. Peters, J. E. Lowe, and K. P. Landolfo Transmyocardial laser revascularization limits in vivo adenoviral-mediated gene transfer in porcine myocardium Cardiovasc Res, October 1, 1999; 44(1): 81 - 90. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Y. Kim, P. L. Walinsky, F. D. Kolodgie, C. Bian, J. L. Sperry, C. B. Deming, E. A. Peck, J. G. Shake, G. B. Ang, R. H. Sohn, et al. Early Loss of Thrombomodulin Expression Impairs Vein Graft Thromboresistance: Implications for Vein Graft Failure Circ. Res., February 8, 2002; 90(2): 205 - 212. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Baetta, M. Camera, C. Comparato, C. Altana, M. D. Ezekowitz, and E. Tremoli Fluvastatin Reduces Tissue Factor Expression and Macrophage Accumulation in Carotid Lesions of Cholesterol-Fed Rabbits in the Absence of Lipid Lowering Arterioscler. Thromb. Vasc. Biol., April 1, 2002; 22(4): 692 - 698. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1997 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |