- Email: [email protected]
Intimal hyperplasia is reduced by ornithine decarboxylase inhibition
JOURNAL
OF SURGICAL
RESEARCH
50,634-637
lntimal Hyperplasia ERIC D. ENDEAN,
(1991)
Is Reduced by Ornithine Decarboxylase
M.D., JOHN F. KISPERT, M.D., KURTIS W. MARTIN,
M.D., AND WILLIAM
Inhibition
O’CONNOR,
M.D.
Departments of Surgery and Pathology, University of Kentucky Medical Center and Veterans Administration Medical Center, Lexington, Kentucky 40536 Presented
at the Annual
Meeting
of the Association
for Academic
INTRODUCTION
Intimal hyperplasia (IH) is a significant cause of restenosis after carotid endarterectomy or bypass grafting which can result in failure of the operation [l]. These lesions represent a hyperplastic response of smooth muscle cells and/or fibroblasts in response to an arterial injury. Because this response involves cellular proliferation, it is hypothesized that induction of ornithine decarboxylase (ODC), to increase polyamine synthesis, is a basic biochemical step needed for the formation of IH. While polyamines (putrescine, spermidine, and spermine) are felt to be key requirements for regulation of 634 Inc. reserved.
14-17,199O
METHODS
Inc.
0022.4804/91$1.50 Copyright 0 1991 by Academic Press, All rights of reproduction in any form
Texas, November
cell growth and differentiation [2], their physiologic function is poorly understood at a molecular level. Polyamines are synthesized from ornithine through the action of ODC, which generally serves as the rate-limiting enzyme. ODC activity is known to increase in tissues undergoing cell division and has been used as a biochemical marker of cell proliferation [3,4]. This enzyme can be irreversibly blocked with the suicide inhibitor, cu-difluoromethylornithine (DFMO) (Fig. 1). This study was undertaken to determine if administration of DFMO to block ODC activity, and thereby reduce polyamine synthesis, would inhibit the formation of IH in rabbits which were subjected to a deendothelialization injury.
Polyamines are intracellular cations that are thought to play a role in the regulation of cell growth and differentiation. This study was undertaken to determine if inhibition of ornithine decarboxylase (ODC), the ratelimiting enzyme of polyamine synthesis, suppresses formation of intimal hyperplasia (IH) after arterial injury. Twenty New Zealand white rabbits underwent balloon catheter deendothelialization of a common carotid artery. Treated animals (n = 10) were given a-difluoromethylornithine (DFMO), an inhibitor of ODC, ad lib in drinking water as a 2% solution. DFMO was begun 3 days prior to surgery and continued until vessel harvest. Vessels were perfusion-fixed at harvest, 2 (n = 10) and 4 (n = 10) weeks postoperatively. All arteries remained patent. There were no histologic differences in the IH between treated and untreated animals. The intima and media surface areas on serial arterial cross sections were determined using computer-assisted planimetry. There was a significant difference in the IH surface area of injured arteries between untreated and DFMO-treated animals at both 2 (17.6 i 2.0 vs 3.0 * 1.6 pm’; P i 0.001) and 4 weeks (27.4 t 5.6 vs 7.1 + 1.8 pm2; P G 0.008). No differences were seen in medial thickness. We conclude that ODC inhibition reduces early development of IH after arterial deendothelialization. These data support the hypothesis that polyamines may be cellular messengers involved in the regulation of IH formation. o 1991 Academic Press,
Surgery, Houston,
Twenty New Zealand white rabbits were randomized to receive oral DFMO (n = 10) or serve as controls (n = 10). DFMO was begun 3 days prior to surgery and continued until vessel harvest. The DFMO was administered through the drinking water as a 2% solution provided ad lib. Animals which have been treated with DFMO ingest varying amounts of drug (0.18-1.28 g/kg/ day), but have averaged 0.5 g/kg/day. The rabbits were anesthetized with ketamine (22 mg/kg) and acepromazine (1.1 mg/kg) maintaining anesthesia with 2% halothane. The bifurcations of both carotid arteries were exposed through a midline neck incision. The animals were given 1,000 U heparin sodium iv. One common carotid artery was cannulated through the external carotid artery with a 2F embolectomy catheter. The catheter was passed retrograde into the aortic arch, inflated, and withdrawn through the common carotid artery such that slight resistance against the arterial wall was generated. This was repeated two additional times, rotating the catheter 90” between passes to assure complete endothelial denudation. Pilot studies using this model had demonstrated, by scanning electron microscopy, that this method of inducing arterial injury resulted in complete endothelial denudation. The catheter was then removed and the external carotid artery was ligated. The contralateral carotid artery underwent a similar dissection with ligation of the external carotid artery, but without
ENDEAN
ET AL.:
INTIMAL
HYPERPLASIA
REDUCTION
TABLE
Ornithine
S-Adenosylmethionine I
* I
4 1
group St-Methylthioadenosine
Intimal
Putrescine
Decarbo xylated S-Adenosylmethionine /I I aminopropyl 2 or 3 b
c
1
L
2
NI-
1 Spermine
Control DFMO treated ‘“,I-
1 Area in DFMO-Treated Animals
Surface area (Mm2 + SEM) Acetylspermidine
Sperm!dine$ 3
Hyperplasia Surface and Untreated
6 \
635
BY ODC INHIBITION
2 weeks*
4 weeks**
17.6 + 2.0 (n = 5) 3.0 t 1.6 (n = 5)
27.4 -t 5.6 (n = 5) 7.1 * 1.8 (n = 5)
Acetylspermine
FIG. 1. Biosynthetic pathway for polyamine synthesis as adapted from Pegg and McCann [2]. (1) Ornithine decarboxylase, (2) spermidine synthase, (3) spermine synthase, (4) S-adenosylmethionine decarboxylase, (5) spermidine N’-acetyltransferase, (6) polyamine oxidase.
* P < 0.0001, ** P i 0,008; Student’s
unpaired
dent’s unpaired t test. Significance < 0.05 level.
t test.
was accepted at the P
RESULTS introduction of the embolectomy catheter. The wounds were closed and the animals recovered. Animal care complied with the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals (NIH Publication 80-23, revised 1985). The common carotid arteries were removed 2 (n = 10) and 4 (n = 10) weeks postoperatively. The rabbits were anesthetized and both carotid arteries were exposed from the aortic arch to their bifurcations after which the animals were given heparin sodium iv (1,000 U). The aortic arch was cannulated and the carotid arteries were perfused with normal saline at 100 mm Hg pressure until all blood in the carotid arteries and jugular veins was flushed out of the vessels. The arteries were perfusionfixed for 15 min at 100 mm Hg using 1% paraformaldehyde and 2.5% glutaldehyde in 0.1 M phosphate buffer, pH 7.4. Each carotid artery was removed from its origin to its bifurcation and immersed in the same fixative until the tissues were completely fixed. The common carotid arteries were divided into five equal segments and cross sections were obtained from each area. These sections were mounted and stained with hematoxylin and eosin for histologic examination. IH was defined as that tissue between the lumen and the internal elastic lamina. The tissue containing smooth muscle cells lying between the internal elastic lamina and the adventitia comprised the media layer of the arterial wall. Crosssectional surface areas of IH and media were determined by computer-assisted planimetry, using a camera lucida and a digitizing pad. The measured surface areas from each of the five sections of the artery were averaged to obtain an overall mean area for each artery. These overall means were used for comparison with other arteries. The slides were reviewed by a pathologist (WO’C) who was blinded as to treatment group to characterize the morphology of the IH tissue. Statistical analysis of intima and media surface areas between treatment groups was carried out using the Stu-
All injured arteries, in both the DFMO treatment groups and untreated groups, were patent at the time of vessel harvest. Visual inspection of the injured vessels, when harvested, revealed gross differences between DFMO-treated and untreated animals. The injured arteries of untreated animals appeared to be scarred and sclerotic and were firm to palpation. The injured arteries of DFMO-treated animals, however, retained an appearance similar to that of the contralateral uninjured arteries. There was a significant difference (Table 1) in the measured IH surface area which formed on injured arteries taken from untreated as compared to DFMO-treated animals at 2 (17.6 +- 2.0 vs 3.0 * 1.6 pm2; P < 0.001) and 4 weeks (27.4 * 5.6 vs 7.12 1.8 pm2; P < 0.008). No IH was found on histologic sections taken from uninjured arteries. No significant differences in media surface areas were found in injured arteries removed from untreated and DFMO-treated animals at either 2 or 4 weeks. In addition, no differences in media thickness were found between 2 and 4 weeks within each treatment group (Table 2).
TABLE Media
2
Surface Area in DFMO-Treated and Untreated Animals Surface Area (pm’ f SEM)
2 weeks ,r 4L4.4;.
Control
N.S. DFMO
treated
L47.6 k 3.8 (n = 5) -N.S.'
4 weeks N.S. -, 55.5 f 9.5 (n = 5) 1 N.S.
38.3 2 4.4-1 (n = 5)
636
JOURNAL
OF SURGICAL
RESEARCH:
Morphologically, the IH tissue contained concentric or eccentric cellular layers on the luminal aspect of the internal elastic lamina. Most cells were spindle shaped with round or oval nuclei. The extracellular matrix consisted of intervening ground substance and collagen. A few monocytes and rare polymorphonuclear cells were noted. No morphologic differences in cell appearance or cell density in the IH tissue were detected between DFMO-treated and untreated specimens using light microscopy.
DISCUSSION While there are a number of methods which are available to induce an arterial injury, an embolectomy catheter has been widely used experimentally to denude the endothelium of a vessel [5-81. This injury results in the formation of IH. We have documented in pilot studies that uniform deendothelialization did occur after passing an embolectomy catheter through a carotid artery. Additionally, in the present study, we found no significant differences in media cross-sectional surface area in injured vessels either at time of vessel harvest or between treated and untreated animals. This indicates that no significant injury was produced in the media with our method of inducing arterial injury. Intimal hyperplasia forming after arterial injury results from proliferation of smooth muscle cells and/or fibroblasts. Polyamines are thought to be essential for cell differentiation and growth and it was hypothesized that they would be essential factors required for IH formation. ODC is the initial enzyme in the polyamine biosynthetic pathway and generally serves as the ratelimiting enzyme for polyamine synthesis. Other reports have demonstrated a rise in ODC activity following an arterial injury [5, 91. This increase in ODC activity occurs prior to the appearance of IH. In the current study we inhibited ODC through the oral administration of DFMO. The results showed a significant decrease in IH which formed on injured arteries in DFMO-treated animals. This biologic response to ODC inhibition by DFMO administration supports the hypothesis that polyamines may be biochemical agents required for IH formation after arterial wall injury. DFMO was the agent used in this study to inhibit ODC. DFMO is a suicide inhibitor of the enzyme ODC, and it covalently binds to the active site of the enzyme, rendering it permanently inactive. ODC has a short half-life resulting in significant turnover and ongoing synthesis of the enzyme. Administration of an inhibitor thus needs to have sustained levels over long periods of time. We chose to give the DFMO as a solution in the drinking water, relying on the animals’ natural oral intake to achieve adequate levels of the drug. We chose this method, which has been used by other investigators [l&14], in order to simplify drug administration. The
VOL.
50, NO. 6, JUNE
1991
average amount of drug ingested by the animals in this study is similar to what has been reported [12, 131. While there was a significant difference in IH formation between untreated and treated animals at both 2 and 4 weeks, complete suppression of IH formation in the DFMO-treated animals did not occur. The percentage increase in IH surface area which occurred between 2 and 4 weeks was similar in both treatment groups (63% in untreated and 59% in DFMO-treated). This observation suggests that inhibition of the early burst of ODC activity after arterial injury may be key for preventing the formation of thick IH. Since DFMO administration did not totally abolish IH formation, formation of IH may proceed through other mechanisms, but at a slower pace. However, if the amount of IH which forms early, in response to a burst of ODC activity, is significantly decreased, restenosis may be minimized. Currently, the role of growth factors, especially those released by activated platelets, in initiating and sustaining IH formation through stimulation of arterial smooth muscle cell proliferation is undergoing extensive investigation [ 151. Multiple trials using different anti-platelet regimens in both animal models and in clinical series have attempted to demonstrate inhibition of IH formation. These studies have had varied results [ 16-201. Platelet inhibition alone, as has been the focus of many trials, may be insufficient for reducing IH which forms in response to growth factors. Other studies have suggested that the effect of growth factors on cell division may be dependent on ODC activity. Quiescent NIH 3T3 cells, stimulated to divide by the addition of PDGF, were noted to have an increase in ODC activity [21]. This increased activity was inhibited in a dose-dependent fashion by DFMO. Similarly, Thyberg and Fredholm [ 221 demonstrated that DFMO abolished the rise in DNA synthesis seen in cultured rat aortic smooth muscle cells when DFMO was added within 4 hr of PDGF administration. These studies suggest that induction of ODC with subsequent production of polyamines may be a component of the cellular mechanism through which PDGF stimulates cell division. Consequently, these in vitro studies suggest that blockade of ODC may be a method of altering IH formation which occurs in response to growth factors irrespective of the cell type responsible for growth factor synthesis. Because cell types, other than platelets, are known to release growth factors, ODC blockade could have a broader effect on IH formation than platelet inhibition alone. Recent work by Nishida et al. [9] has demonstrated that increased polyamine synthesis after arterial deendothelialization may be a transient phenomena. They have shown that ODC activity peaks 6 hr after injury and that arterial polyamine levels were noted to peak 2 days after injury with return to baseline levels within 7 days of injury. It may be that changes in polyamine content occur early after injury, when there is a burst of ODC activity with resultant increases in polyamine sy’n-
ENDEAN
ET AL.:
INTIMAL
thesis. The effect of DFMO on IH formation observed in the current study may have been a result of inhibition of this early burst of ODC activity. If this is correct, an early and short course of DFMO to inhibit the ODC activity burst following an arterial injury might prove to be clinically beneficial for reduction of IH. In summary, this is the first study to demonstrate that blockade of the enzyme ODC, using the suicide inhibitor DFMO, results in reduction of early IH which forms in response to an arterial injury. This study suggests that ODC may play a role in early biochemical interactions linking arterial injury to IH formation. However, it was noted that IH did form on the injured arteries of DFMOtreated animals. Furthermore, between 2 and 4 weeks, the rate of IH development was similar between the DFMO-treated and untreated groups. This suggests that other mechanisms contribute to IH formation and studies will be needed to determine the long-term effect of ODC inhibition on IH formation. Further studies will also be necessary to define the duration of DFMO treatment required to sustain this biologic response and to verify, biochemically, a decrease in arterial wall polyamine content as a result of ODC blockade. The precise molecular mechanism(s) through which ODC and polyamines exert their effects also await further elucidation. ACKNOWLEDGMENTS This work was supported in part by VA Medical Research Funds. The authors thank Dr. E. H. W. Bohme and Dr. P. P. McCann of the Merrell Dow Research Institute, Cincinnati, OH, for the generous gift of DFMO used in this study.
REFERENCES 1.
2. 3.
4.
5.
6.
7.
REDUCTION
HYPERPLASIA
Chervu, A., and Moore, W. S. An overview of intimal hyperplasia. Surg. Gynecol. Obstet. 171: 433, 1990. Pegg, A. E., and McCann, P. P. Polyamine metabolism and function. Am. J. Physiol. 243 (Cell Physiol. 12): C212, 1982. Fozard, J. R., and Koch-Weser, J. Pharmacological consequences of inhibition of polyamine biosynthesis with DL-adifluoromethylornithine. Trends Pharmacol. Sci. 3: 107, 1982. Heby, O., Gray, J. W., Lindl, P. A., Marton, L. J., and Wilson, C. B. Changes in L-ornithine decarboxylase activity during the cell cycle. Biochem. Biophys. Res. Commun. 71: 99, 1976. Majesky, M. W., Schwartz, S. M., Clowes, M. M., and Clowes, A. W. Heparin regulates smooth muscle S phase entry in the injured rat carotid artery. Circ. Res. 61: 296, 1987. Clowes, A. W., Reidy, M. A., and Clowes, M. M. Kinetics of cellular proliferation after arterial injury I. Smooth muscle growth in the absence of endothelium. Lab. Inuest. 49: 327, 1983. Chervu, A., Moore, W. S., Quinones-Baldrich, W. J., and Henderson, T. Efficacy of corticosteroids in suppression of intimal hyperplasia. J. Vast. Surg. 10: 129, 1989.
BY ODC INHIBITION
637
8. Schwartz, T. H., Dobrin, P. B., Mrkvicka,
9.
lo.
“’
12.
R., Skowron, L., and Cole, M. B. Early myointimal hyperplasia after balloon catheter embolectomy: Effect of shear forces and multiple withdrawals. J. Vast. Surg. 7: 495, 1988.
Nishida, K., Abiko, T., Ishihara, M., and Tomikawa, M. Arterial injury-induced smooth muscle cell proliferation in rats is accompanied by increase in polyamine synthesis and level. A therosclerosis 83: 119, 1990. Olson, J. W., Atkinson, J. E., Hacker, A. D., Altiere, R. J., and Gillespie, M. N. Suppression of polyamine biosynthesis prevents monocrotaline-induced pulmonary edema and arterial medial thickening. Tonicol. Appl. Pharmacol. 81: 91, 1985. Gillespie, M. N., Dyer, K. K., Olson, J. W., O’Connor, W. N., and Altiere, R. J. a-difluromethylornithine, an inhibitor of polyamine synthesis, attenuates monocrotoline-induced pulmonary vascular hyperresponsiveness in isolated perfused rat lungs. Res. Commun. Chem. Pathol. Pharmacol. 50: 365, 1985. Fozard, J. R., Part, M-L,, Prakash, N. J., Grove, J., Schechter, P. J., Sjoerdsma, A., and Koch-Weser, J. L-ornithine decarboxylase: An essential role in early mammalian embryogenesis. Science 208: 505, 1980.
r3. Fozard, J. R., Part, M-L., Prakash, N. J., and Grove, J. Inhibition of murine embryonic development by a-difluoromethylornithine, an irreversible inhibitor of ornithine decarboxylase. Eur. J. Pharmacol. 65: 379, 1980. 14.
Olson, J. W., Hacker, A. D., Altiere, R. J., and Gillespie, M. N. Polyamines and the development of monocrotaline-induced pulmonary hypertension. Am. J. Physiol. 247 (Heart Circ. Physiol.
15.
Ross, R., Raines, E. W., and Bowen-Pope, D. F. The biology of platelet-derived growth factor. Cell 46: 155, 1986.
16.
McCann, R. L., Hagen, P-O., and Fuchs, J. C. A. Aspirin and dipyridamole decrease intimal hyperplasia in experimental vein grafts. Ann. Surg. 191: 238, 1980.
16):H682,1984.
17
Hagen, P-O., Wang, Z-G., Mikat, E. M., and Hackel, D. B. Antiplatelet therapy reduces aortic intimal hyperplasia distal to small diameter vascular prostheses (PTFE) in nonhuman primates. Ann. Surg. 195: 328, 1982.8. 18. McCready, R. A., Price, M. A., Kryscio, R. J., Hyde, G. L., Mattingly, S. S., and Griffen, W. 0. Failure of antiplatelet therapy with ibuprofen (Motrin) to prevent neointimal fibrous hyperplasia. J. Vast. Surg. 2: 205, 1985. 19. Radic, Z. S., O’Malley, M. K., Mikat, E. M., Makhoul, R. G., McCann, R. L., Cole, C. W., and Hagen, P-O. The role of aspirin and dipyridamole on vascular DNA synthesis and intimal hyperplasia following deendothelialization. J. Surg. Res. 41: 84, 1986. 20. Quinones-Baldrich, W. J., Ziomek, S., Henderson, T., and Moore, W. S. Patency and intimal hyperplasia: The effect of aspirin on small arterial anastomosis. Ann. Vast. Surg. 2: 50, 1988. 21. Vicenzi, E., Bianchi, M., Salmona, M., Donati, M. B., Poggi, A., and Ghezzi, P. Platelet-derived growth factor induces ornithine decarboxylase activity in NIH 3T3 cells. Biochem. Biophys. Res. Commun. 127: 843,1985.
22. Thyberg, J., and Fredholm,
B. B. Induction boxylase activity and putrescine synthesis muscle cells stimulated with platelet-derived CellRes. 170: 160, 1987.
of ornithine decarin arterial smooth growth factor. Exp.
OF SURGICAL
RESEARCH
50,634-637
lntimal Hyperplasia ERIC D. ENDEAN,
(1991)
Is Reduced by Ornithine Decarboxylase
M.D., JOHN F. KISPERT, M.D., KURTIS W. MARTIN,
M.D., AND WILLIAM
Inhibition
O’CONNOR,
M.D.
Departments of Surgery and Pathology, University of Kentucky Medical Center and Veterans Administration Medical Center, Lexington, Kentucky 40536 Presented
at the Annual
Meeting
of the Association
for Academic
INTRODUCTION
Intimal hyperplasia (IH) is a significant cause of restenosis after carotid endarterectomy or bypass grafting which can result in failure of the operation [l]. These lesions represent a hyperplastic response of smooth muscle cells and/or fibroblasts in response to an arterial injury. Because this response involves cellular proliferation, it is hypothesized that induction of ornithine decarboxylase (ODC), to increase polyamine synthesis, is a basic biochemical step needed for the formation of IH. While polyamines (putrescine, spermidine, and spermine) are felt to be key requirements for regulation of 634 Inc. reserved.
14-17,199O
METHODS
Inc.
0022.4804/91$1.50 Copyright 0 1991 by Academic Press, All rights of reproduction in any form
Texas, November
cell growth and differentiation [2], their physiologic function is poorly understood at a molecular level. Polyamines are synthesized from ornithine through the action of ODC, which generally serves as the rate-limiting enzyme. ODC activity is known to increase in tissues undergoing cell division and has been used as a biochemical marker of cell proliferation [3,4]. This enzyme can be irreversibly blocked with the suicide inhibitor, cu-difluoromethylornithine (DFMO) (Fig. 1). This study was undertaken to determine if administration of DFMO to block ODC activity, and thereby reduce polyamine synthesis, would inhibit the formation of IH in rabbits which were subjected to a deendothelialization injury.
Polyamines are intracellular cations that are thought to play a role in the regulation of cell growth and differentiation. This study was undertaken to determine if inhibition of ornithine decarboxylase (ODC), the ratelimiting enzyme of polyamine synthesis, suppresses formation of intimal hyperplasia (IH) after arterial injury. Twenty New Zealand white rabbits underwent balloon catheter deendothelialization of a common carotid artery. Treated animals (n = 10) were given a-difluoromethylornithine (DFMO), an inhibitor of ODC, ad lib in drinking water as a 2% solution. DFMO was begun 3 days prior to surgery and continued until vessel harvest. Vessels were perfusion-fixed at harvest, 2 (n = 10) and 4 (n = 10) weeks postoperatively. All arteries remained patent. There were no histologic differences in the IH between treated and untreated animals. The intima and media surface areas on serial arterial cross sections were determined using computer-assisted planimetry. There was a significant difference in the IH surface area of injured arteries between untreated and DFMO-treated animals at both 2 (17.6 i 2.0 vs 3.0 * 1.6 pm’; P i 0.001) and 4 weeks (27.4 t 5.6 vs 7.1 + 1.8 pm2; P G 0.008). No differences were seen in medial thickness. We conclude that ODC inhibition reduces early development of IH after arterial deendothelialization. These data support the hypothesis that polyamines may be cellular messengers involved in the regulation of IH formation. o 1991 Academic Press,
Surgery, Houston,
Twenty New Zealand white rabbits were randomized to receive oral DFMO (n = 10) or serve as controls (n = 10). DFMO was begun 3 days prior to surgery and continued until vessel harvest. The DFMO was administered through the drinking water as a 2% solution provided ad lib. Animals which have been treated with DFMO ingest varying amounts of drug (0.18-1.28 g/kg/ day), but have averaged 0.5 g/kg/day. The rabbits were anesthetized with ketamine (22 mg/kg) and acepromazine (1.1 mg/kg) maintaining anesthesia with 2% halothane. The bifurcations of both carotid arteries were exposed through a midline neck incision. The animals were given 1,000 U heparin sodium iv. One common carotid artery was cannulated through the external carotid artery with a 2F embolectomy catheter. The catheter was passed retrograde into the aortic arch, inflated, and withdrawn through the common carotid artery such that slight resistance against the arterial wall was generated. This was repeated two additional times, rotating the catheter 90” between passes to assure complete endothelial denudation. Pilot studies using this model had demonstrated, by scanning electron microscopy, that this method of inducing arterial injury resulted in complete endothelial denudation. The catheter was then removed and the external carotid artery was ligated. The contralateral carotid artery underwent a similar dissection with ligation of the external carotid artery, but without
ENDEAN
ET AL.:
INTIMAL
HYPERPLASIA
REDUCTION
TABLE
Ornithine
S-Adenosylmethionine I
* I
4 1
group St-Methylthioadenosine
Intimal
Putrescine
Decarbo xylated S-Adenosylmethionine /I I aminopropyl 2 or 3 b
c
1
L
2
NI-
1 Spermine
Control DFMO treated ‘“,I-
1 Area in DFMO-Treated Animals
Surface area (Mm2 + SEM) Acetylspermidine
Sperm!dine$ 3
Hyperplasia Surface and Untreated
6 \
635
BY ODC INHIBITION
2 weeks*
4 weeks**
17.6 + 2.0 (n = 5) 3.0 t 1.6 (n = 5)
27.4 -t 5.6 (n = 5) 7.1 * 1.8 (n = 5)
Acetylspermine
FIG. 1. Biosynthetic pathway for polyamine synthesis as adapted from Pegg and McCann [2]. (1) Ornithine decarboxylase, (2) spermidine synthase, (3) spermine synthase, (4) S-adenosylmethionine decarboxylase, (5) spermidine N’-acetyltransferase, (6) polyamine oxidase.
* P < 0.0001, ** P i 0,008; Student’s
unpaired
dent’s unpaired t test. Significance < 0.05 level.
t test.
was accepted at the P
RESULTS introduction of the embolectomy catheter. The wounds were closed and the animals recovered. Animal care complied with the Principles of Laboratory Animal Care and the Guide for the Care and Use of Laboratory Animals (NIH Publication 80-23, revised 1985). The common carotid arteries were removed 2 (n = 10) and 4 (n = 10) weeks postoperatively. The rabbits were anesthetized and both carotid arteries were exposed from the aortic arch to their bifurcations after which the animals were given heparin sodium iv (1,000 U). The aortic arch was cannulated and the carotid arteries were perfused with normal saline at 100 mm Hg pressure until all blood in the carotid arteries and jugular veins was flushed out of the vessels. The arteries were perfusionfixed for 15 min at 100 mm Hg using 1% paraformaldehyde and 2.5% glutaldehyde in 0.1 M phosphate buffer, pH 7.4. Each carotid artery was removed from its origin to its bifurcation and immersed in the same fixative until the tissues were completely fixed. The common carotid arteries were divided into five equal segments and cross sections were obtained from each area. These sections were mounted and stained with hematoxylin and eosin for histologic examination. IH was defined as that tissue between the lumen and the internal elastic lamina. The tissue containing smooth muscle cells lying between the internal elastic lamina and the adventitia comprised the media layer of the arterial wall. Crosssectional surface areas of IH and media were determined by computer-assisted planimetry, using a camera lucida and a digitizing pad. The measured surface areas from each of the five sections of the artery were averaged to obtain an overall mean area for each artery. These overall means were used for comparison with other arteries. The slides were reviewed by a pathologist (WO’C) who was blinded as to treatment group to characterize the morphology of the IH tissue. Statistical analysis of intima and media surface areas between treatment groups was carried out using the Stu-
All injured arteries, in both the DFMO treatment groups and untreated groups, were patent at the time of vessel harvest. Visual inspection of the injured vessels, when harvested, revealed gross differences between DFMO-treated and untreated animals. The injured arteries of untreated animals appeared to be scarred and sclerotic and were firm to palpation. The injured arteries of DFMO-treated animals, however, retained an appearance similar to that of the contralateral uninjured arteries. There was a significant difference (Table 1) in the measured IH surface area which formed on injured arteries taken from untreated as compared to DFMO-treated animals at 2 (17.6 +- 2.0 vs 3.0 * 1.6 pm2; P < 0.001) and 4 weeks (27.4 * 5.6 vs 7.12 1.8 pm2; P < 0.008). No IH was found on histologic sections taken from uninjured arteries. No significant differences in media surface areas were found in injured arteries removed from untreated and DFMO-treated animals at either 2 or 4 weeks. In addition, no differences in media thickness were found between 2 and 4 weeks within each treatment group (Table 2).
TABLE Media
2
Surface Area in DFMO-Treated and Untreated Animals Surface Area (pm’ f SEM)
2 weeks ,r 4L4.4;.
Control
N.S. DFMO
treated
L47.6 k 3.8 (n = 5) -N.S.'
4 weeks N.S. -, 55.5 f 9.5 (n = 5) 1 N.S.
38.3 2 4.4-1 (n = 5)
636
JOURNAL
OF SURGICAL
RESEARCH:
Morphologically, the IH tissue contained concentric or eccentric cellular layers on the luminal aspect of the internal elastic lamina. Most cells were spindle shaped with round or oval nuclei. The extracellular matrix consisted of intervening ground substance and collagen. A few monocytes and rare polymorphonuclear cells were noted. No morphologic differences in cell appearance or cell density in the IH tissue were detected between DFMO-treated and untreated specimens using light microscopy.
DISCUSSION While there are a number of methods which are available to induce an arterial injury, an embolectomy catheter has been widely used experimentally to denude the endothelium of a vessel [5-81. This injury results in the formation of IH. We have documented in pilot studies that uniform deendothelialization did occur after passing an embolectomy catheter through a carotid artery. Additionally, in the present study, we found no significant differences in media cross-sectional surface area in injured vessels either at time of vessel harvest or between treated and untreated animals. This indicates that no significant injury was produced in the media with our method of inducing arterial injury. Intimal hyperplasia forming after arterial injury results from proliferation of smooth muscle cells and/or fibroblasts. Polyamines are thought to be essential for cell differentiation and growth and it was hypothesized that they would be essential factors required for IH formation. ODC is the initial enzyme in the polyamine biosynthetic pathway and generally serves as the ratelimiting enzyme for polyamine synthesis. Other reports have demonstrated a rise in ODC activity following an arterial injury [5, 91. This increase in ODC activity occurs prior to the appearance of IH. In the current study we inhibited ODC through the oral administration of DFMO. The results showed a significant decrease in IH which formed on injured arteries in DFMO-treated animals. This biologic response to ODC inhibition by DFMO administration supports the hypothesis that polyamines may be biochemical agents required for IH formation after arterial wall injury. DFMO was the agent used in this study to inhibit ODC. DFMO is a suicide inhibitor of the enzyme ODC, and it covalently binds to the active site of the enzyme, rendering it permanently inactive. ODC has a short half-life resulting in significant turnover and ongoing synthesis of the enzyme. Administration of an inhibitor thus needs to have sustained levels over long periods of time. We chose to give the DFMO as a solution in the drinking water, relying on the animals’ natural oral intake to achieve adequate levels of the drug. We chose this method, which has been used by other investigators [l&14], in order to simplify drug administration. The
VOL.
50, NO. 6, JUNE
1991
average amount of drug ingested by the animals in this study is similar to what has been reported [12, 131. While there was a significant difference in IH formation between untreated and treated animals at both 2 and 4 weeks, complete suppression of IH formation in the DFMO-treated animals did not occur. The percentage increase in IH surface area which occurred between 2 and 4 weeks was similar in both treatment groups (63% in untreated and 59% in DFMO-treated). This observation suggests that inhibition of the early burst of ODC activity after arterial injury may be key for preventing the formation of thick IH. Since DFMO administration did not totally abolish IH formation, formation of IH may proceed through other mechanisms, but at a slower pace. However, if the amount of IH which forms early, in response to a burst of ODC activity, is significantly decreased, restenosis may be minimized. Currently, the role of growth factors, especially those released by activated platelets, in initiating and sustaining IH formation through stimulation of arterial smooth muscle cell proliferation is undergoing extensive investigation [ 151. Multiple trials using different anti-platelet regimens in both animal models and in clinical series have attempted to demonstrate inhibition of IH formation. These studies have had varied results [ 16-201. Platelet inhibition alone, as has been the focus of many trials, may be insufficient for reducing IH which forms in response to growth factors. Other studies have suggested that the effect of growth factors on cell division may be dependent on ODC activity. Quiescent NIH 3T3 cells, stimulated to divide by the addition of PDGF, were noted to have an increase in ODC activity [21]. This increased activity was inhibited in a dose-dependent fashion by DFMO. Similarly, Thyberg and Fredholm [ 221 demonstrated that DFMO abolished the rise in DNA synthesis seen in cultured rat aortic smooth muscle cells when DFMO was added within 4 hr of PDGF administration. These studies suggest that induction of ODC with subsequent production of polyamines may be a component of the cellular mechanism through which PDGF stimulates cell division. Consequently, these in vitro studies suggest that blockade of ODC may be a method of altering IH formation which occurs in response to growth factors irrespective of the cell type responsible for growth factor synthesis. Because cell types, other than platelets, are known to release growth factors, ODC blockade could have a broader effect on IH formation than platelet inhibition alone. Recent work by Nishida et al. [9] has demonstrated that increased polyamine synthesis after arterial deendothelialization may be a transient phenomena. They have shown that ODC activity peaks 6 hr after injury and that arterial polyamine levels were noted to peak 2 days after injury with return to baseline levels within 7 days of injury. It may be that changes in polyamine content occur early after injury, when there is a burst of ODC activity with resultant increases in polyamine sy’n-
ENDEAN
ET AL.:
INTIMAL
thesis. The effect of DFMO on IH formation observed in the current study may have been a result of inhibition of this early burst of ODC activity. If this is correct, an early and short course of DFMO to inhibit the ODC activity burst following an arterial injury might prove to be clinically beneficial for reduction of IH. In summary, this is the first study to demonstrate that blockade of the enzyme ODC, using the suicide inhibitor DFMO, results in reduction of early IH which forms in response to an arterial injury. This study suggests that ODC may play a role in early biochemical interactions linking arterial injury to IH formation. However, it was noted that IH did form on the injured arteries of DFMOtreated animals. Furthermore, between 2 and 4 weeks, the rate of IH development was similar between the DFMO-treated and untreated groups. This suggests that other mechanisms contribute to IH formation and studies will be needed to determine the long-term effect of ODC inhibition on IH formation. Further studies will also be necessary to define the duration of DFMO treatment required to sustain this biologic response and to verify, biochemically, a decrease in arterial wall polyamine content as a result of ODC blockade. The precise molecular mechanism(s) through which ODC and polyamines exert their effects also await further elucidation. ACKNOWLEDGMENTS This work was supported in part by VA Medical Research Funds. The authors thank Dr. E. H. W. Bohme and Dr. P. P. McCann of the Merrell Dow Research Institute, Cincinnati, OH, for the generous gift of DFMO used in this study.
REFERENCES 1.
2. 3.
4.
5.
6.
7.
REDUCTION
HYPERPLASIA
Chervu, A., and Moore, W. S. An overview of intimal hyperplasia. Surg. Gynecol. Obstet. 171: 433, 1990. Pegg, A. E., and McCann, P. P. Polyamine metabolism and function. Am. J. Physiol. 243 (Cell Physiol. 12): C212, 1982. Fozard, J. R., and Koch-Weser, J. Pharmacological consequences of inhibition of polyamine biosynthesis with DL-adifluoromethylornithine. Trends Pharmacol. Sci. 3: 107, 1982. Heby, O., Gray, J. W., Lindl, P. A., Marton, L. J., and Wilson, C. B. Changes in L-ornithine decarboxylase activity during the cell cycle. Biochem. Biophys. Res. Commun. 71: 99, 1976. Majesky, M. W., Schwartz, S. M., Clowes, M. M., and Clowes, A. W. Heparin regulates smooth muscle S phase entry in the injured rat carotid artery. Circ. Res. 61: 296, 1987. Clowes, A. W., Reidy, M. A., and Clowes, M. M. Kinetics of cellular proliferation after arterial injury I. Smooth muscle growth in the absence of endothelium. Lab. Inuest. 49: 327, 1983. Chervu, A., Moore, W. S., Quinones-Baldrich, W. J., and Henderson, T. Efficacy of corticosteroids in suppression of intimal hyperplasia. J. Vast. Surg. 10: 129, 1989.
BY ODC INHIBITION
637
8. Schwartz, T. H., Dobrin, P. B., Mrkvicka,
9.
lo.
“’
12.
R., Skowron, L., and Cole, M. B. Early myointimal hyperplasia after balloon catheter embolectomy: Effect of shear forces and multiple withdrawals. J. Vast. Surg. 7: 495, 1988.
Nishida, K., Abiko, T., Ishihara, M., and Tomikawa, M. Arterial injury-induced smooth muscle cell proliferation in rats is accompanied by increase in polyamine synthesis and level. A therosclerosis 83: 119, 1990. Olson, J. W., Atkinson, J. E., Hacker, A. D., Altiere, R. J., and Gillespie, M. N. Suppression of polyamine biosynthesis prevents monocrotaline-induced pulmonary edema and arterial medial thickening. Tonicol. Appl. Pharmacol. 81: 91, 1985. Gillespie, M. N., Dyer, K. K., Olson, J. W., O’Connor, W. N., and Altiere, R. J. a-difluromethylornithine, an inhibitor of polyamine synthesis, attenuates monocrotoline-induced pulmonary vascular hyperresponsiveness in isolated perfused rat lungs. Res. Commun. Chem. Pathol. Pharmacol. 50: 365, 1985. Fozard, J. R., Part, M-L,, Prakash, N. J., Grove, J., Schechter, P. J., Sjoerdsma, A., and Koch-Weser, J. L-ornithine decarboxylase: An essential role in early mammalian embryogenesis. Science 208: 505, 1980.
r3. Fozard, J. R., Part, M-L., Prakash, N. J., and Grove, J. Inhibition of murine embryonic development by a-difluoromethylornithine, an irreversible inhibitor of ornithine decarboxylase. Eur. J. Pharmacol. 65: 379, 1980. 14.
Olson, J. W., Hacker, A. D., Altiere, R. J., and Gillespie, M. N. Polyamines and the development of monocrotaline-induced pulmonary hypertension. Am. J. Physiol. 247 (Heart Circ. Physiol.
15.
Ross, R., Raines, E. W., and Bowen-Pope, D. F. The biology of platelet-derived growth factor. Cell 46: 155, 1986.
16.
McCann, R. L., Hagen, P-O., and Fuchs, J. C. A. Aspirin and dipyridamole decrease intimal hyperplasia in experimental vein grafts. Ann. Surg. 191: 238, 1980.
16):H682,1984.
17
Hagen, P-O., Wang, Z-G., Mikat, E. M., and Hackel, D. B. Antiplatelet therapy reduces aortic intimal hyperplasia distal to small diameter vascular prostheses (PTFE) in nonhuman primates. Ann. Surg. 195: 328, 1982.8. 18. McCready, R. A., Price, M. A., Kryscio, R. J., Hyde, G. L., Mattingly, S. S., and Griffen, W. 0. Failure of antiplatelet therapy with ibuprofen (Motrin) to prevent neointimal fibrous hyperplasia. J. Vast. Surg. 2: 205, 1985. 19. Radic, Z. S., O’Malley, M. K., Mikat, E. M., Makhoul, R. G., McCann, R. L., Cole, C. W., and Hagen, P-O. The role of aspirin and dipyridamole on vascular DNA synthesis and intimal hyperplasia following deendothelialization. J. Surg. Res. 41: 84, 1986. 20. Quinones-Baldrich, W. J., Ziomek, S., Henderson, T., and Moore, W. S. Patency and intimal hyperplasia: The effect of aspirin on small arterial anastomosis. Ann. Vast. Surg. 2: 50, 1988. 21. Vicenzi, E., Bianchi, M., Salmona, M., Donati, M. B., Poggi, A., and Ghezzi, P. Platelet-derived growth factor induces ornithine decarboxylase activity in NIH 3T3 cells. Biochem. Biophys. Res. Commun. 127: 843,1985.
22. Thyberg, J., and Fredholm,
B. B. Induction boxylase activity and putrescine synthesis muscle cells stimulated with platelet-derived CellRes. 170: 160, 1987.
of ornithine decarin arterial smooth growth factor. Exp.