Neutralizing Antibodies Directed Against Osteopontin Inhibit Rat Carotid Neointimal Thickening After Endothelial Denudation
Osteopontin is an arginine-glycine-aspartate-containing acidic glycoprotein with adhesive and migratory activities in vitro. We previously showed that osteopontin was highly expressed in injured rat arteries as well as in human atherosclerotic plaques. In contrast, uninjured blood vessels make very little osteopontin. In this report, we have investigated the role of osteopontin in rat neointima formation using neutralizing antibodies. Rats were treated with either nonimmune or antiosteopontin antibody and subjected to endothelial denudation of the carotid artery by using a balloon catheter. Two weeks after injury, intimal areas and cell numbers were significantly decreased (33% and 31%, respectively) in the antiosteopontin group compared with the nonimmune IgG group. No differences in carotid medial areas or cell numbers were observed. Intimal and medial replication rates, as measured by continuous bromodeoxyuridine infusion during the final week of the experimental protocol, were not significantly different between the two groups. No gross histological changes were noted in the intimas formed in the presence of either neutralizing or nonimmune antibody. In addition, no difference in early carotid medial cell replication rate was observed when antibodies were infused for 4 days after angioplasty. These data demonstrate for the first time a functional role for osteopontin in the process of carotid neointimal thickening in vivo and suggest that osteopontin plays an active role in the remodeling processes important for human atherosclerotic and restenotic lesion development.
Reprint requests to Dr Cecilia M. Giachelli, PhD, Pathology Dept, University of Washington, Vascular Biology, Box 357335, Seattle, WA 98195. E-mail firstname.lastname@example.org.
- Received January 17, 1996.
- Revision received May 21, 1996.
Intimal thickening is a hallmark of atherosclerotic and restenotic vascular occlusive lesions. The steps involved in formation of the neointima have been studied in detail in several animal models and include proliferation of smooth muscle cells (SMCs) in the tunica media and their migration to the tunica intima.1 The intimal SMC is phenotypically distinguishable from the medial SMC in gene expression pattern as well as proliferative, migratory, and matrix-generating potentials.2 3 In the rat, intimal SMCs proliferate and deposit extracellular matrix such that a fibrous, muscular lesion is ultimately formed. These early steps are believed to be involved in development of human atherosclerotic and restenotic vascular lesions as well.
Osteopontin is a 66-kD secreted glycoprotein containing the cell adhesion motif RGD.4 We discovered the expression of osteopontin in intimal SMC using a differential cloning strategy aimed at identifying genes specific to the intimal SMC.5 Subsequently, we showed that angioplasty of either the rat aorta or carotid artery caused a dramatic increase in osteopontin mRNA and protein synthesis in SMCs invading the neointima. The spatial and temporal pattern of osteopontin expression coincided with alterations in SMC phenotype, invasion of the intima, and proliferation. Transforming growth factor-β, basic fibroblast growth factor, and angiotensin II (all growth factors implicated in the initiation of neointimal thickening) were able to induce osteopontin mRNA and protein synthesis in cultured SMCs.6 Furthermore, osteopontin is abundant in human atherosclerotic and restenotic lesions although virtually absent in normal arteries.6 7 8 9 Taken together, these data strongly implicate osteopontin in neointima formation both in animals and in humans.
A potential role for osteopontin in neointima formation was suggested by its functional properties in vitro. Osteopontin is an adhesive and chemotactic stimulus for SMCs.10 11 Receptors for osteopontin on SMCs include the αvβ3, αvβ1, and αvβ5 integrins.12 Recently, integrin antagonists have been reported to block intimal thickening in animal models,13 14 thereby suggesting that cell adhesive interactions are required for intima formation. The aim of the current study was to critically test the hypothesis that osteopontin was functionally involved in neointimal lesion formation using the rat carotid artery injury model. Our data show that treatment of rats with an antibody directed against osteopontin caused a significant reduction in neointimal thickening and intimal cell number after arterial balloon catheter denudation. The mechanism of this inhibition appeared exclusive of effects on medial and intimal proliferation and overall matrix deposition as assessed by histological staining, consistent with a role for osteopontin in early accumulation of SMCs in the arterial intima.
Materials and Reagents
Preparation of Anti-Osteopontin and Nonimmune Antibodies
A neutralizing antibody (OP199) was prepared by immunizing a goat with rat osteopontin as previously described.10 Neutralizing plasma as well as plasma from a nonimmunized goat were prepared, and IgGs were purified identically by caprylic acid precipitation and sepharose column chromatography.15 Antibody was injected in 0.01 mol/L phosphate buffer containing 0.9% saline. OP199 has been characterized previously and specifically inhibits cell adhesion and migration to osteopontin with no cross-reactivity with vitronectin or fibronectin.10
Three-month-old male Sprague-Dawley rats (weight, ≈400 g; B&K Universal, Redmond, Wash) were used for these studies. All animals were allowed rat chow and water ad libitum. The following surgical procedures were all approved by the University of Washington Animal Care Committee and were followed in accordance with institutional guidelines. On the day of surgery, animals were anesthetized with ketamine hydrochloride (50 mg/kg), xylazine (5 mg/kg), and acepromazine (1 mg/kg) administered intramuscularly. The left common carotid artery was deendothelialized by passage of a 2F Fogarty embolectomy catheter, which was inserted into the external carotid, passed down to the aortic arch, inflated, and drawn with a twisting motion up to the carotid bifurcation, where the balloon was deflated. This procedure was performed three times to ensure complete denudation before the catheter was removed and the external carotid was tied off, leaving blood flow through the internal carotid artery.
Immediately before the balloon injury, the animal was injected via the tail vein with 1 mL of either OP199 or nonimmune IgG (each at a concentration of 50.9 mg/mL). Under the same anesthesia, an osmotic pump (model 2ML1, ALZA Corp) containing either OP199 or nonimmune IgG was implanted subcutaneously in the back, and a catheter running from the pump was inserted into the right jugular vein to allow for continuous intravenous delivery.
In the first protocol, antibodies were infused for 7 days. On day 7 after the balloon injury, the IgG pumps were removed and osmotic pumps (model 2001) containing the thymidine analogue 5-bromo-2′-deoxyuridine (BrDu, 30 mg/mL; Sigma Chemical Co) were inserted subcutaneously. The resulting continuous delivery of this drug cumulatively labeled all cells replicating during the second week after balloon injury. On day 10 after ballooning, each animal was given an additional intravenous injection (via tail vein) of 0.5 mL OP199 or nonimmune IgG.
Fourteen days after the carotid injury, animals were again anesthetized as described above. One milliliter of blood was taken from the femoral vein for serum preparation, and animals were then injected intravenously with 0.5 mL of 5% Evan's Blue dye to allow identification of deendothelialized areas of artery. Approximately 5 minutes later, animals were euthanatized with an intraperitoneal injection of pentobarbital and were perfused retrogradely via the abdominal aorta with 4% paraformaldehyde in 0.1 mol/L phosphate buffer for 5 minutes at the animal's blood pressure. Segments of injured and uninjured carotid arteries were excised from the middle third portion of each vessel that clearly showed no endothelial regrowth on the basis of Evans Blue dye uptake. These specimens were immersion fixed in 4% paraformaldehyde overnight and routinely processed and paraffin embedded.
In the second protocol, antibodies were infused at the same concentration as for protocol 1 (50.9 mg/mL) but for 4 days after angioplasty. On day 3, BrDu pellets (Boehringer-Mannheim) were implanted subcutaneously to label cells replicating in the media in the final 18 hours of treatment. Blood samples, tissue fixation, and processing were identical to that described above for protocol 1.
Histochemistry and Morphometry
A segment from the middle third of each carotid artery showing complete endothelial denudation was examined. Four sections (5-μm thick) per carotid were examined, with the distance between sections being ≥50 μm so that replicate countings of the same nuclei would not be a concern. Staining was performed for BrDu with the use of a specific monoclonal antibody (IgG2, kind gift of Mat Daemen) and an indirect avidin-biotin immunoperoxidase method, as previously described.16 Positively stained cells were visualized with 3,3′-diaminobenzidine, and slides were counterstained with hematoxylin. BrDu stained cells and total cell numbers (estimated by counting nuclei) were counted per vessel cross section (4 sections/rat). Replicative indexes were calculated by use of the equation: Replication Index=BrDu-Positive Cells per 4 Cross Sections/Total Cells per 4 Cross Sections×100.
Additional arterial cross sections (4 sections/rat) were stained with orcein to highlight elastic tissue and used for measurement of intimal and medial areas. Measurements were taken with the use of a video-microscopy–based morphometry system and Optimas software (Bioscan).
Assessment of Tissue Antibody Penetration
Access and penetration of antibodies into vascular tissues were assessed by immunochemistry. The presence of goat IgG was detected in carotid cross sections with the use of the Vector ABC ELITE kit for goat according to the manufacturer's instructions. For both OP199- and normal goat IgG–treated rats, staining was observed in the intimal, medial, and adventitial regions of the vessel, indicating equivalent penetration and deposition of both of the antibodies at the injury sites (data not shown).
Monocytes-macrophages were identified with the monoclonal antibody ED1 (Harlan Bioproducts for Science), which specifically identifies all cells of monocyte-macrophage lineage, as previously described.17 Four sections from each vessel (50 μm apart) were counted to determine macrophage content of the vessel wall.
Values are given as mean±SE. Comparison of the two group means were made by use of unpaired, two-tailed Student's t test. A value of P<.05 was considered significant.
Anti-Osteopontin Antibodies Inhibit Rat Carotid Neointimal Thickening
We10 previously described the isolation and extensive characterization of a goat polyclonal antibody, OP199, that was capable of specifically neutralizing osteopontin-mediated adhesion and migration in vascular SMCs in vitro and that did not cross-react with vitronectin, fibronectin, platelet-derived growth factor, or secreted SMC products. This antibody was used in the present study to examine the role of osteopontin in rat carotid neointimal thickening.
In the first experimental protocol, neutralizing or nonimmune antibodies (in purified IgG form) were administered by bolus intravenous injection (50.9 mg) immediately before ballooning, followed by continuous intravenous infusion (11 mg per rat per day) via osmotic minipump for 7 days after ballooning. A final bolus injection (25.5 mg) was administered on day 10, and animals were euthanatized on day 14. This time course of antibody administration was chosen because osteopontin levels are elevated early (within 6 hours) after balloon injury, and peak levels occur between days 7 and 10.6 Both nonimmune goat IgG and OP199 were deposited to a similar extent in all layers of the blood vessel as assessed by immunochemistry with the use of an anti-goat IgG antibody (data not shown).
After euthanasia, both injured (left) and uninjured (right) carotid arteries were excised and embedded in paraffin, and histological area measurements were made. As shown in Fig 1A⇓, ballooned rats treated with anti-osteopontin antibody had significantly less neointimal thickening than nonimmune antibody–treated rats (.18±.02 mm2 versus .27±.03 mm2; P=.027). This represents a 33% decrease in intimal area from nonimmune IgG–treated rats. In contrast, medial areas (Fig 1B⇓) of the ballooned carotid arteries were not different between the groups (.16±.01 mm2 versus .14±.01 mm2; P=.148). Likewise, no change in medial area was observed in the right, unmanipulated carotid from the same rats, indicating that there was no systemic toxic effect of the antibodies during the course of the experiment (Fig 1C⇓). Areas of left carotid intima and media and right carotid media in the nonimmune IgG group were comparable to those in rats infused with 0.9% saline or Ringer's solution (data not shown).
Morphology of Antibody-Treated Vessels
Injured left carotid and uninjured right carotid artery sections were stained with orcein, which detects elastin, and representative sections are shown in Fig 2⇓. Consistent with the area measurements, intimal thickness was suppressed in anti-osteopontin–treated rats compared with rats treated with nonimmune IgG (compare Fig 2a and 2d⇓⇓). Medial thickness in either the injured left carotid (Fig 2a and 2d⇓⇓) or uninjured right carotid artery (Fig 2c and 2f⇓⇓) was comparable between groups. Higher-power inspection of the injured carotid arteries (Fig 2b and 2e⇓⇓) indicated that whereas intimal areas were reduced in the anti-osteopontin–treated rats, no obvious change in tissue morphology and matrix architecture as determined by histological orcein staining was evident. In fact, considerable matrix deposition, as measured by elastin staining, was evident in both treatment groups.
Anti-Osteopontin Antibodies Decrease Total Intimal Cell Number
Mechanistically, the decreased intimal area observed in the anti-osteopontin–treated rats could be due to a decrement in the number of SMCs in the intima or to a decrease in the amount of matrix generated by these cells. The histological studies described above (Fig 2⇑), however, suggested that a gross change in matrix deposition as detected by orcein staining of elastin most likely was not responsible for the decreased area. To examine cellularity of the vessels, the number of cells in the intima and media of the injured or uninjured arteries was estimated by counting nuclei per cross section. As shown in Table 1⇓, there was a highly significant decrease in the number of cells making up the intima in the anti-osteopontin– compared with the nonimmune antibody–treated rats (1064±73 versus 1522±95 cells/cross section; P=.002). The decrease could not be accounted for by a difference in macrophage content between the two groups because very few macrophages were detected within the neointimas of either group with the use of ED1 immunohistochemical staining (nonimmune-antibody group [n=7], 5.8±2 macrophages/cross section versus anti-osteopontin group [n=8], 4.5±2.3 macrophages/cross section; P=.25). No significant change in cell number in the injured left or uninjured right carotid media was found (388±27 versus 382±36 cells/cross section and 285±20 versus 310±26 cells/cross section, respectively). The 31% decrease in cell number in the intima was consistent with the 33% decrease in intimal area in the anti-osteopontin–treatment group, suggesting that the decrease in intimal area was due to a decrease in cell number.
Anti-Osteopontin Antibodies Do Not Suppress Intimal Cell Growth
The decrease in cell number in the intimas of anti-osteopontin–treated rats could be due to decreased proliferation of intimal SMCs or to inhibition of smooth muscle migration into the intima. To distinguish between these possibilities, intimal SMC proliferation was measured. Replication indexes were determined by continuous BrDu infusion between days 7 and 14 in the experimental groups, followed by BrDu immunochemistry and counting. As shown in Table 2⇓, there was no significant difference in the percentage of labeled cells in the left carotid intimas between the anti-osteopontin– and nonimmune IgG–treated rats (57.5±6% versus 55.6±5%, respectively). Likewise, no significant differences were seen between the treatment groups in left or right carotid medial replication indexes under these labeling conditions (17.6±4% versus 8.9±2%, and 2.5±1% versus 3.4±3%, respectively). These data suggest that inhibition of intimal thickening and decreased intimal SMC numbers in anti-osteopontin antibody–treated rats cannot be explained by a decrease in continued growth of intimal cells after SMC entry into the intima.
Anti-Osteopontin Antibodies Do Not Suppress Medial Cell Growth
In the second experimental protocol, neutralizing or nonimmune antibodies were administered by bolus intravenous injection (50.9 mg) immediately before ballooning, followed by continuous intravenous infusion (11 mg per rat per day) via osmotic minipump for 4 days after ballooning. BrDu pellets were implanted 18 hours before the animals were euthanatized on day 4 to label any cells replicating in the injured media to ascertain whether anti-osteopontin–antibody treatment might affect early medial replication rates. As in protocol 1, replication indexes in the carotid media were determined by BrDu immunochemistry and counting. As shown in Table 2⇑, no statistically significant differences in medial replication rates were observed between nonimmune- and anti-osteopontin–treated rats (5.7±5.8% [n=6] versus 13.4±9.3% [n=5], respectively; P=.13). Likewise, no difference in total carotid medial cell numbers was observed between nonimmune- and anti-osteopontin–treated rats (301±37 cells/cross section [n=6] versus 288±63 cells/cross section [n=5], respectively; P=.68) in this experiment. These data suggest that inhibition of neointima formation by anti-osteopontin antibodies is probably not due to inhibition of early medial cell replication.
We and others6 7 8 9 have previously correlated osteopontin overexpression with neointima formation in both experimental animal and human vascular diseases. The present study demonstrates that rat carotid neointimal lesions induced by balloon catheter denudation can be reduced by in vivo neutralization of osteopontin. These data provide the first evidence of a causal role for osteopontin in the pathogenesis of neointimal lesions.
The mechanism by which inhibition of osteopontin affects intimal lesion formation is suggested by the present studies. We observed little change in the histological character of the intimal matrix after antibody treatment. In contrast, the intimas from anti-osteopontin–treated rats contained 33% fewer cells than intimas from control rats. This did not appear to be due to changes in the macrophage content of the vessels but rather to a decrease in the SMC content of the neointima. However, no effect was seen on cumulative intimal SMC replication indexes from day 7 to day 14. Because the majority of intimal cell replication occurs during this time,1 the effect of the anti-osteopontin antibody on cell number is most likely not due to an effect on intimal SMC proliferation. Likewise, anti-osteopontin treatment did not appear to affect either early or late carotid medial replication rates. These findings are consistent with the lack of effect of osteopontin on SMC proliferation observed in vitro.11 Although it is possible that the effect of OP199 might be due to an increase in complement-mediated lysis of osteopontin-producing cells, this is highly unlikely because (1) both nonimmune and OP199 antibodies deposited to similar extents in the injured blood vessels, and (2) there was no decrease in medial cell numbers even though osteopontin expression occurs in medial SMCs during the first 4 days after injury.6
Given our understanding of the early events of neointima formation in the rat, the most likely interpretation of the data is that inhibition of vascular osteopontin decreases the number of cells migrating from the media to the intima, thereby decreasing the total intimal cell number without changing intimal replication rate. This might be due to (1) a direct block of SMC migration by the anti-osteopontin antibody or (2) a decrease in the number of cells available to migrate from the media. The first possibility is supported by the findings that osteopontin is a chemotactic and migratory stimulus for SMCs in vitro10 11 and appears to be absolutely required for invasion of SMC into a collagenous matrix.18 Because early carotid medial replication rates were unchanged in anti-osteopontin–treated rats, the second possibility could be true if there were an increased apoptosis of medial SMCs early after injury. However, the experimental design used in the present study does not allow us to distinguish between these two possibilities.
Although a number of treatments have been shown to block SMC proliferation or migration, only a minority have been shown to affect the longer-term outcome of intimal thickening. For example, although inhibition of matrix metalloproteinase completely blocks SMC migration into the neointima, there is no effect on intimal size at 14 days (Reference 19 and M. Bendeck and M. Reidy, oral communication, November 15, 1996). Likewise, antibody directed against basic fibroblast growth factor almost completely blocks medial SMC replication 48 hours after injury, yet the intimas that form at 8 days are not significantly different from controls.20 Inhibition of platelet-derived growth factor or transforming growth factor-β, on the other hand, leads to inhibition of neointimal thickening in the rat carotid injury model by ≈30% to 40%.21 22 It is interesting that these two cytokines are also good inducers of osteopontin in vascular SMCs.6 23 On the basis of these findings, it is interesting to speculate that part of the mechanism from injury to neointima formation requires cytokine induction of osteopontin, as well as other adhesive and/or extracellular matrix molecules.
The injured vessel wall is exposed to a large number of circulating adhesive ligands, such as vitronectin, fibronectin, thrombospondin, or fibrinogen, which no doubt participate in vessel repair. Moreover, the balloon-injured vessel itself is also a source of adhesive ligands, such as the embryonic fibronectins EIIIA and EIIIB,24 tenascin,25 and thrombospondin,26 which may also be involved in providing the correct adhesive context for arterial repair and remodeling. The fact that we could block intimal thickening by inhibiting osteopontin suggests that the function of osteopontin cannot be completely replaced by these circulating or locally produced factors. These data suggest that a unique function may be subserved by osteopontin during neointima formation. In support of this concept, vascular SMCs whose osteopontin levels were effectively reduced by overexpressing an antisense-osteopontin construct were unable to invade collagen gels even in the presence of serum.18 Thus, osteopontin may be uniquely required for vascular SMC functions important in neointima formation.
This work was supported by NIH grants HL-18645, HL-03174, and DK-47659. Lucy Liaw was supported by NIH training grant HL-07312. Denis deBlois was supported by a fellowship award from the Medical Research Council of Canada. Dr Giachelli is an Established Investigator of the American Heart Association. We thank Alicia Momberg, Patti Polinsky, and Robin Najar for their expert technical assistance, and Dr Charles Murry for helpful discussions.
Reidy MA, Fingerle J, Majesky MW. Proliferation of vascular smooth muscle cells in vivo. In: Suckling KE, Groot PHE, eds. Hyperlipidaemia and Atherosclerosis. London, UK: Academic Press; 1988:149-164.
Majesky MW, Giachelli CM, Reidy MA, Schwartz SM. Rat carotid neointimal smooth muscle cells re-express a developmentally regulated phenotype during repair of arterial injury. Circ Res. 1992;71:759-768.
Giachelli CM, Majesky MW, Schwartz SM. Developmentally regulated cytochrome P450IA1 expression in cultured rat vascular smooth muscle cells. J Biol Chem. 1991;266:3981-3986.
Giachelli CM, Schwartz SM, Liaw L. Molecular and cellular biology of osteopontin: potential role in cardiovascular disease. Trends Cardiovasc Med. 1995;5:88-95.
Giachelli CM, Bae N, Lombardi DM, Majesky MW, Schwartz SM. Molecular cloning and characterization of 2B7, a rat mRNA which distinguishes smooth muscle cell phenotypes in vitro and is identical to osteopontin (secreted phosphoprotein I, 2aR). Biochem Biophys Res Commun. 1991;177:867-873.
Giachelli C, Bae N, Almeida M, Denhardt D, Alpers CE, Schwartz SM. Osteopontin expression is elevated during neointima formation in rat arteries and in human atherosclerotic plaques. J Clin Invest. 1993;92:1686-1696.
O'Brien ER, Garvin MR, Stewart DK, Hinohara T, Simpson JB, Schwartz SM, Giachelli CM. Osteopontin is synthesized by macrophage, smooth muscle and endothelial cells in primary and restenotic human coronary atherosclerotic plaques. Arterioscler Thromb. 1994;14:1648-1656.
Shanahan CS, Weissberg PL, Metcalfe JC. Isolation of gene markers of differentiated and proliferating vascular smooth muscle cells. Circ Res. 1993;73:193-204.
Liaw L, Almeida M, Downey W, Hart CE, Schwartz SM, Giachelli CM. Osteopontin promotes vascular cell adhesion and spreading and is chemotactic for smooth muscle cells in vitro. Circ Res. 1994;74:214-224.
Yue TL, McKenna J, Ohlstein EH, Farach-Carson MC, Butler WT, Johanson K, McDevitt P, Feuerstein GZ, Stadel JM. Osteopontin-stimulated vascular smooth muscle cell migration is mediated by b3 integrin. Exp Cell Res. 1995;214:459-464.
Liaw L, Skinner MP, Raines EW, Ross R, Cheresh DA, Schwartz SM, Giachelli CM. The adhesive and migratory effects of osteopontin are mediated via distinct cell surface integrins: role of αvβ3 in smooth muscle cell migration to osteopontin in vitro. J Clin Invest. 1995;95:713-724.
Slepian MJ, Massia SP. Local delivery of a cyclic RGD peptide inhibits neointimal hyperplasia following balloon injury. Circulation. 1994;88(suppl 4):I-372. Abstract.
van-Kleef EM, Smits JF, De Mey JG, Cleutjens JP, Lombardi DM, Schwartz SM, Daemen MJ. Alpha 1 adrenoreceptor blockade reduces the angiotensin II induced vascular SMC DNA synthesis in the rat thoracic aorta and carotid artery. Circ Res. 1992;70:1122-1127.
Bendeck MP, Zempo N, Clowes AW, Galardy RE, Reidy MA. Smooth muscle cell migration and matrix metalloproteinase expression after arterial injury in the rat. Circ Res. 1994;75:539-545.
Lindner V, Reidy MA. Proliferation of smooth muscle cells after vascular injury is inhibited by an antibody against basic fibroblast growth factor. Proc Natl Acad Sci U S A. 1991;88:3739-3743.
Ferns GA, Raines EW, Sprugel KH, Motani AS, Reidy MA, Ross R. Inhibition of neointimal smooth muscle accumulation after angioplasty by an antibody to PDGF. Science. 1992;253:1129-1132.
Wolf YG, Rasmussen LM, Ruoslahti E. Antibodies against transforming growth factor b1 suppress intimal hyperplasia in a rat model. J Clin Invest. 1994;93:1172-1178.
Dubin D, Peters JH, Brown LF, Logan B, Kent KC, Berse B, Berven S, Cercek B, Sharifi BG, Pratt RE, Dzau VJ, Van de water L. Balloon catheterization induces arterial expression of embryonic fibronectins. Arterioscler Thromb Vasc Biol. 1995;15:1958-1967.