Articles |
the Institute of Cardiovascular Research, Bayer AG, Wuppertal, Germany.
Correspondence to Dr Christoph Gerdes, Bayer AG, Institute of Cardiovascular Research, D-42096 Wuppertal, Germany.
| Abstract |
|---|
|
|
|---|
Key Words: restenosis animal models arterial injury thrombin r-hirudin
| Introduction |
|---|
|
|
|---|
Besides the action of thrombin in the early thrombotic phase, its potential involvement in later stages of the restenotic process is also suggested by its direct chemoattractant and mitogenic effects on SMCs.9 10 11 12 These effects occur via a receptor-mediated pathway,13 and increased expression of the thrombin receptor has been demonstrated in human atherosclerotic tissue.14
Animal studies have confirmed the presence of early thrombin-dependent events immediately after balloon angioplasty of porcine carotid arteries by the reduction of fibrin and platelet deposition with the direct thrombin inhibitor r-hirudin.15 16 In a study of atherosclerotic rabbit arteries, short-term treatment with r-hirudin for 2 hours reduced restenosis 4 weeks after angioplasty.17 We compared the effects of r-hirudin in different animal models of arterial injury to identify variations among species, between models of single and double arterial injury, and between short- and long-term r-hirudin treatment.
| Methods |
|---|
|
|
|---|
Rabbit Experiments
Male New Zealand White rabbits (Interfauna, Huntingdon, England; body weight, 2.5 to 3.5 kg) maintained on a normal rabbit chow diet were anesthetized with ketamine (8 mg/kg IV; Ketavet, Parke-Davis) and xylazine (0.4 mg/kg IV; Rompun, Bayer). Additional anesthesia was achieved by administration of ketamine (35 mg/kg IM) and xylazine (4 mg/kg IM). The injury procedure was performed with a 3F Fogarty balloon catheter and was analogous to that in rats, resulting in denudation of the endothelial layer. In the single-injury experiments the rabbits were killed with an overdose of pentobarbital 4 weeks after the injury procedure. In the double-injury model, left carotid arteries underwent a second dilatation-induced injury 4 weeks after the initial one. For the second injury a 3.5-mm Gruntzig angioplasty catheter (ACS, Danimed) was introduced via the external carotid artery and inflated in the previously denuded area twice for 30 seconds each to 5 to 6 atm. Previous experiments have shown disruption of preexisting plaque and a restimulation of neointimal growth by this technique.18 The vessels were harvested 4 weeks after the second injury.
Minipig Experiments
Male Gottingen miniature pigs (Lars Ellegard, Dalmose, Denmark; body weight, 27 to 36 kg) were fed an atherogenic diet containing 1% cholesterol and 10% dried egg yolk starting 9 months before and continuing throughout the course of the experiment. Total plasma cholesterol levels were monitored enzymatically (Boehringer Mannheim kit). The minipigs were anesthetized with azaperone (8 mg/kg IM; Stresnil, Janssen) and metomidate (Hypnodil, Janssen; 10 mg/kg IP). Whenever necessary, supplementary metomidate was given intravenously. The minipigs received heparin (100 U/kg IV) for anticoagulation. The right common carotid artery was surgically exposed and opened by a distal 5-mm longitudinal incision. A 5F Fogarty balloon catheter was introduced proximally into the carotid artery, inflated with saline, and withdrawn five times under near-maximal distention of the arterial wall. Previous experiments revealed complete endothelial denudation and disruption of medial layers by this procedure. The catheter was removed, the incisions were closed by surgical sutures, and the animal was allowed to recover after finishing its respective r-hirudin or control treatment. Postoperative antibiotic protection was provided by penicillin (Tardomyocel, Bayer; 500 000 IU IM TID). Pigs received aspirin (500 mg/d PO) until the end of the experiment. Four weeks after injury the pigs were killed by exsanguination after captive bolt stunning. All experiments conformed to the German federal animal protection law and were approved by the legal district authority.
r-Hirudin Treatment
r-Hirudin (supplied by Rhein Biotech) was dissolved in sterile physiological saline. In the first experiment 14 rats received short-term treatment with r-hirudin, consisting of an intravenous bolus of 1 mg/kg immediately before injury, followed by an infusion of 1 mg · kg-1 · h -1 IV for 2 hours via the left jugular vein, and concluded by a single subcutaneous injection of 6 mg/kg 2 hours after injury. Seven control animals were administered saline via the same protocol. In the second experiment r-hirudin was delivered as a long-term treatment for 3 (n=10) or 14 (n=14) days by infusion of 1 mg · kg-1 · h -1 IV via osmotic pumps (Alzet 2ML2, Alza Corp). The pumps were implanted under the dorsal skin and connected to a flexible plastic tube that had been introduced into the left jugular vein. Ten control animals received saline infusions under the same protocol.
Rabbits were administered the short-term treatment only (1 mg/kg IV bolus, 1 mg · kg-1 · h-1 IV infusion for 2 hours, and 6 mg/kg SC injection), which was given to 10 animals in the single-injury experiment (control, n=10). In the double-injury experiment (n=7; control, n=24) the rabbits underwent r-hirudin treatment after the second dilatation-induced injury.
Minipigs (n=10; control, n=18) received the short-term r-hirudin treatment (1 mg/kg IV bolus, 1 mg · kg-1 · h -1 IV infusion for 2 hours, and 6 mg/kg SC injection) at the time of injury.
The level of thrombin inhibition was monitored by determination of the aPTT (Boehringer Mannheim aPTT kit) at different times during the r-hirudin treatment. To minimize stress in the rats induced by withdrawal of excess blood, sampling for aPTT determinations was done in a parallel group of 12 r-hirudintreated rats.
Tissue Analysis
After the animals were killed the injured vessels were flushed in situ with buffered saline containing 0.7% (wt/vol) NaNO2 to achieve maximal vasodilation and standardize vessel wall tone. Carotid arteries were subsequently perfused with 10% (vol/vol) buffered formalin at 100 mm Hg pressure. After excision the injured vessels were divided into segments (1 segment per rat, 2 per rabbit, and 4 per minipig carotid artery), stored in 10% (vol/vol) formalin, and embedded in OCT medium for cryosectioning. Frozen sections were stained with resorcin-fuchsin, van Gieson's picrofuchsin, and hematoxylin.
Morphometric analysis was performed with a videomicroscope and a computerized image processing system (IMPAC). After the lumen boundary and internal and external elastic laminae for each section were manually traced, cross-sectional areas of lumen, neointima, media, and total vessel were calculated and the neointima to media ratio was derived. In previous experiments, specific segments with the most pronounced neointima formation had been identified for each species. These segments were selected for further evaluation by averaging the morphometric data from 5 sections in each segment. Vessels with occlusive thrombi were excluded from further analysis.
Statistics
All data are presented as mean±SEM. Statistical comparisons between groups were performed with unpaired t tests and data were rejected as not significantly different if P>.05.
| Results |
|---|
|
|
|---|
|
In rabbits treated with r-hirudin, aPTT peaked at 95.3±9.4 seconds in the single-injury experiment and at 133.8±13.9 seconds in the double-injury experiment from baseline levels of 50.3±5.9 and 60.9±4.4 seconds, respectively.
In minipigs, short-term r-hirudin treatment produced a maximum aPTT elongation of 80.5±6.8 seconds from a baseline of 22.4±1.3 seconds. The atherogenic diet fed to the minipigs resulted in an increase of total plasma cholesterol from
2 mmol/L (baseline) to 11.9±0.4 mmol/L (mean±SEM, n=20) within 4 weeks after the diet was started.
Occurrence of Thrombotic Occlusions
Occlusive thrombi in injured arteries were observed in rabbits and minipigs. In the rabbit single-injury experiment, 1 vessel from an r-hirudintreated animal was occluded. In the rabbit double-injury experiment, no occlusions occurred in the r-hirudintreated group, but 63% (15 of 24) of injured carotid arteries were occluded in the control group. In minipigs the thrombosis rate in r-hirudintreated animals was 40% (4 of 10) and in control animals 56% (5 of 9).
Morphometry
Morphometric parameters are summarized in the Table
. In rats short-term r-hirudin treatment did not significantly change the cross-sectional areas of any vessel compartment (Fig 2
). However, a 3- or 14-day infusion of r-hirudin in rats resulted in reductions of 25% (P=.007) or 27% (P=.003), respectively, in neointimal area, which corresponded to significant reductions in the neointima to media ratio. A slight (18%) reduction in total vessel area was observed in rats treated with r-hirudin for 3 days after injury.
|
|
In rabbits, short-term r-hirudin treatment reduced neointimal area by 59% (single injury, P=.05; Fig 3
) and 44% (double-injury model, P=.02). In addition, a significant increase (47%) in luminal cross-sectional area was observed in r-hirudintreated rabbits in the double-injury experiment.
|
In minipigs there were no significant changes in morphometric parameters after short-term treatment with r-hirudin, although a trend toward a reduction in lumen size and an increase in neointima area was seen (Fig 4
).
|
| Discussion |
|---|
|
|
|---|
Mechanisms of Thrombin in Vascular Injury
Activation of platelets by thrombin is accomplished via proteolytic cleavage of the membrane-bound thrombin receptor.13 This receptor has also been identified on SMCs, where its stimulation results in cellular proliferation and migration,9 10 11 12 and on monocytes, which respond to thrombin stimulation with chemotaxis.24 Increased expression of the thrombin receptor has been demonstrated in human atherosclerotic tissue14 and in lesions formed after experimental arterial injury.25 These multiple functions of thrombin suggest its potential involvement in several phases of the restenotic process after vessel reopening procedures.26
Choice of r-Hirudin Treatment Regimen
The direct thrombin inhibitor r-hirudin has proved to be highly effective in preventing platelet and fibrin deposition in stented porcine coronary arteries and after balloon injury of porcine carotid arteries.16 23 In an atherosclerotic rabbit balloon angioplasty model, r-hirudin reduced restenosis when infused for only 2 hours.17 We took a similar approach but gave an additional subcutaneous injection to achieve an
2-fold aPTT elongation for at least 4 hours.
Animal Models
To our knowledge this is the first published report on the long-term effects of r-hirudin in a rat arterial injury model. After short-term treatment had not shown an effect on morphometric parameters with respect to vessel response, we decided to treat the rats with a continuous infusion of r-hirudin for 3 or 14 days. Both treatment groups displayed a similar reduction in neointimal area (25% and 27%, respectively), indicating that there is a time frame for neointima reduction by thrombin inhibition of
3 days after injury. Although highly significant, the size of the effect is only moderate and might be explained by a slightly lower aPTT prolongation than the 2-fold increase that we had aimed for.
In the rabbit experiments, our results confirm the findings of other groups with r-hirudin and the irreversible thrombin inhibitor D-Phe-Ala-Pro-Arg chloromethyl ketone.17 27 Sarembock et al17 were the first to find angiographic and histomorphometric evidence of reduced stenosis by r-hirudin in an atherosclerotic rabbit femoral artery angioplasty model. In that study, however, histomorphometric parameters were expressed as percent lumen stenosis only. In our experiments neointimal area was reduced by r-hirudin in both the single- and the double-injury model in rabbits. However, an increase in vessel lumen area was observed in the double-lesion model only. This finding is of particular interest because preexisting atherosclerotic plaque in experimental arterial injury is an important feature when experimental data are extrapolated to the clinical situation.19 20
Another characteristic of the double-lesion model is the high rate of thrombotic occlusion compared with only 1 case in the single-injury experiment. This observation corresponds to different types of injury, with endothelial denudation only in the single-lesion model versus disruption of the plaque and media in the double-lesion model. Treatment with r-hirudin eliminated thrombosis in the double-injury experiment, compared with a 63% reocclusion rate in the placebo-treated group.
The findings in our rat and rabbit models suggest interspecies differences in the time of susceptibility to thrombin inhibition after arterial injury. The narrowest time frame obviously exists in rabbits, in which treatment for a few hours resulted in the most marked reduction in neointimal area. This may suggest that thrombin acts primarily in the acute thrombotic phase after injury. There is evidence, however, that thrombin generation persists in deendothelialized rabbit aortic walls for as long as 10 days after injury.28
In rats susceptibility to thrombin inhibition lasts
3 days. The lack of further reduction in neointimal area by extending r-hirudin treatment for >3 days may suggest that the role of thrombin in this species is limited to the acute phase of thrombocyte deposition. As has been observed in thrombocytopenic rats,29 a reduction in platelet deposition results in less release of mediators (probably platelet-derived growth factor30 ) that influence the migration but not proliferation of SMCs. However, with regard to the absence of an intimal layer in normal rat arteries and the importance of SMC migration, the rat model must be considered with caution as a model for human restenosis.
The differences in responsiveness to r-hirudin treatment between rats and rabbits may be partly explained by differential activity of the coagulation system or by variable thrombogenicity of the injured vessel wall surface. Fingerle et al29 described a persistent thrombogenic subendothelium in rats 48 hours after injury, whereas data for rabbits suggest that 8 to 10 hours after injury the exposed subendothelium becomes inactivated and unresponsive to platelets.31
Our results in the minipig model did not reveal any positive effect of short-term r-hirudin treatment on morphometric parameters. We cannot exclude the possibility that any inhibitory effects of r-hirudin might have been masked by the heparin plus aspirin treatment given to all animals. However, there was a trend toward a reduction of thrombotic occlusion (probably due to deep medial injury) from 56% in the control group to 40% in the r-hirudintreated group. In porcine arterial injury models, contradictory effects of r-hirudin treatment have been published. In a carotid artery model of deep balloon injury, Webster et al32 observed a significant reduction in fibrinogen deposition after a 2-week infusion of r-hirudin. The reduction in intimal area and number of proliferating cells, however, did not achieve statistical significance. In a minipig coronary stent model, Unterberg et al33 recently found a significant reduction in mean intimal thickness that was comparable with that achieved with two intravenous boluses alone and continued with subcutaneous treatment for 2 weeks. We did not evaluate such continued treatment, but the data of Unterberg et al33 suggest positive effects of r-hirudin therapy, with anticoagulation levels comparable to those achieved in our study. However, their model involved injury of minipig coronary arteries with oversized stents, which may not be exactly comparable to our carotid artery balloon injury procedure. This is important with regard to the predictive value of pig models of restenosis, which have been considered superior to various small-animal models,19 20 34 because of their anatomic and pathophysiological correlates to the human situation as well as their susceptibility to atherosclerosis. For example, in both porcine models and the clinical situation, angioplasty characteristically results in deep injury with disruption of subendothelial layers.35 Indeed, many agents that have demonstrated antiproliferative effects in rodent models later proved to have negative effects in both pigs and humans.36 37 38 39
Clinical Correlations
A recent clinical trial40 investigated the effects of adjunctive treatment with r-hirudin in PTCA patients. At 7 months neither clinical nor angiographic end points differed significantly from those in heparin-treated control subjects. However, the incidence of early clinical events was significantly reduced when r-hirudin was given intravenously on the first day followed by subcutaneous treatment for another 3 days after PTCA. Another trial compared the use of bivalirudin (Hirulog), another direct thrombin inhibitor, with heparin during PTCA.41 Bivalirudin did not improve clinical outcome except for a reduction of acute ischemic complications in a subgroup of patients with postinfarction angina.
The observation of improvement in early clinical events by thrombin inhibition during PTCA raises the issue of potentially beneficial effects of prolonged thrombin inactivation after angioplasty. Restenosis development in humans usually takes several months compared with assessment periods of only a few weeks in current animal models. Thus, long-term improvements in clinical outcome might require prolonged thrombin inactivation.
The problem of prolonged thrombin inhibition, however, is its interference with the blood coagulation system. Although neither r-hirudin nor bivalirudin increased the rate of bleeding complications compared with heparin-treated control groups in clinical trials of post-PTCA restenosis,40 41 r-hirudin treatment has revealed the risk of bleeding complications in other studies,42 43 thus indicating the need for closely monitoring blood coagulation during treatment with thrombin inhibitors.
In conclusion, the interspecies variations observed with identical r-hirudin treatments recommend the use of a combination of animal models before initiating large-scale clinical trials of potential antirestenotic treatments. A definitive answer to the question of how much of the observed effects of r-hirudin can be attributed to inhibition of receptor activation rather than of blood coagulation presumably awaits the discovery of potent, selective receptor antagonist compounds.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
Received December 5, 1995;
revision received March 20, 1996;
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Bea, J. Kreuzer, M. Preusch, S. Schaab, B. Isermann, M. E. Rosenfeld, H. Katus, and E. Blessing Melagatran Reduces Advanced Atherosclerotic Lesion Size and May Promote Plaque Stability in Apolipoprotein E- Deficient Mice Arterioscler. Thromb. Vasc. Biol., December 1, 2006; 26(12): 2787 - 2792. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. de Waard, E. K. Arkenbout, P. Carmeliet, V. Lindner, and H. Pannekoek Plasminogen Activator Inhibitor 1 and Vitronectin Protect Against Stenosis in a Murine Carotid Artery Ligation Model Arterioscler. Thromb. Vasc. Biol., December 1, 2002; 22(12): 1978 - 1983. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Andrade-Gordon, C. K. Derian, B. E. Maryanoff, H.-C. Zhang, M. F. Addo, W.-m. Cheung, B. P. Damiano, M. R. D'Andrea, A. L. Darrow, L. de Garavilla, et al. Administration of a Potent Antagonist of Protease-Activated Receptor-1 (PAR-1) Attenuates Vascular Restenosis Following Balloon Angioplasty in Rats J. Pharmacol. Exp. Ther., July 1, 2001; 298(1): 34 - 42. [Abstract] [Full Text] |
||||
![]() |
C. Patterson, G. A. Stouffer, N. Madamanchi, and M. S. Runge New Tricks for Old Dogs : Nonthrombotic Effects of Thrombin in Vessel Wall Biology Circ. Res., May 25, 2001; 88(10): 987 - 997. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. H. Erlich, E. M. Boyle, J. Labriola, J. C. Kovacich, R. A. Santucci, C. Fearns, E. N. Morgan, W. Yun, T. Luther, O. Kojikawa, et al. Inhibition of the Tissue Factor-Thrombin Pathway Limits Infarct Size after Myocardial Ischemia-Reperfusion Injury by Reducing Inflammation Am. J. Pathol., December 1, 2000; 157(6): 1849 - 1862. [Abstract] [Full Text] [PDF] |
||||
![]() |
W.-m. Cheung, M. R. D’Andrea, P. Andrade-Gordon, and B. P. Damiano Altered Vascular Injury Responses in Mice Deficient in Protease-Activated Receptor-1 Arterioscler. Thromb. Vasc. Biol., December 1, 1999; 19(12): 3014 - 3024. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Takada, H. Tanaka, T. Yamada, O. Ito, M. Kogushi, M. Yanagimachi, T. Kawamura, T. Musha, F. Yoshida, M. Ito, et al. Antibody to Thrombin Receptor Inhibits Neointimal Smooth Muscle Cell Accumulation Without Causing Inhibition of Platelet Aggregation or Altering Hemostatic Parameters After Angioplasty in Rat Circ. Res., May 19, 1998; 82(9): 980 - 987. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. W. Courtman, S. M. Schwartz, and C. E. Hart Sequential Injury of the Rabbit Abdominal Aorta Induces Intramural Coagulation and Luminal Narrowing Independent of Intimal Mass : Extrinsic Pathway Inhibition Eliminates Luminal Narrowing Circ. Res., May 19, 1998; 82(9): 996 - 1006. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Gallo, A. Padurean, V. Toschi, J. Bichler, J. T. Fallon, J. H. Chesebro, V. Fuster, and J. J. Badimon Prolonged Thrombin Inhibition Reduces Restenosis After Balloon Angioplasty in Porcine Coronary Arteries Circulation, February 17, 1998; 97(6): 581 - 588. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. C. Nichols, D. A. Bellinger, R. L. Reddick, G. G. Koch, J. L. Sigman, G. Erickson, T. du Laney, T. Johnson, M. S. Read, and T. R. Griggs von Willebrand Factor Does Not Influence Atherogenesis in Arteries Subjected to Altered Shear Stress Arterioscler. Thromb. Vasc. Biol., February 1, 1998; 18(2): 323 - 330. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Capron, J. Jarnet, D. Heudes, D. Joseph-Monrose, and P. Bruneval Repeated Balloon Injury of Rat Aorta : A Model of Neointima With Attenuated Inhibition by Heparin Arterioscler. Thromb. Vasc. Biol., September 1, 1997; 17(9): 1649 - 1656. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |