Aortic wall cell proliferation via basic fibroblast growth factor gene transfer limits progression of experimental abdominal aortic aneurysm

Aortic wall cell proliferation via basic fibroblast growth factor gene transfer limits progression of experimental abdominal aortic aneurysm Katsuyuki Hoshina, MD,a,b,c Hiroyuki Koyama, MD,a,b,c Tetsuro Miyata, MD, PhD,b Hiroshi Shigematsu, MD, PhD,b Tsuyoshi Takato, MD, PhD,c Ronald L. Dalman, MD,d and Hirokazu Nagawa, MD, PhD,b Tokyo, Japan; and Palo Alto, Calif Objective: Our previous study demonstrated that high flow conditions stimulated cell proliferation in the aortic wall in a rat model of abdominal aortic aneurysm (AAA), and we speculated that there is a possible relation between medial cell density and aortic wall integrity. In the present study we delivered the basic fibroblast growth factor (bFGF) gene to the aortic wall of a rat AAA model and evaluated the effects of growth factor– enhanced smooth muscle cell (SMC) proliferation on aneurysm progression. Methods: AAA was induced in rats by means of infusion of porcine pancreatic elastase. Immediately after elastase infusion the abdominal aorta was filled with an expression plasmid vector containing the bFGF gene (bFGF group) or LacZ gene (control group); then gene transfer to the aortic wall was carried out with an in vivo electroporation method. The animals were killed 7 days after treatment, and the aneurysm was measured. The numbers of SMCs, macrophages, and endothelial cells were counted with immunostaining, and cell replication was evaluated with bromodeoxyuridine (BrdU) staining. Results: Aneurysm diameter in the bFGF group was significantly smaller than that in the control group (4.6 ⴞ 0.3 mm vs 6.5 ⴞ 1.4 mm; P < .01). The numbers of medial SMCs and BrdU-incorporated cells in the bFGF group were significantly greater than those in the control group (SMC, 101 ⴞ 34 per high-power field [hpf] vs 80 ⴞ 31/hpf; P < .05, BrdU, 107 ⴞ 63/hpf vs 50 ⴞ 33/hpf; P < .05), whereas no difference was detected in the numbers of macrophages and endothelial cells between the 2 groups. Conclusions: Delivery of bFGF to the aortic wall induced significant enhancement of medial SMC proliferation, without an increase in inflammatory infiltration, then successfully limited aneurysm enlargement. These findings suggest that increased medial cellularity inhibits aneurysm formation, which possibly offers a clue for developing a new strategy for treatment of AAAs. ( J Vasc Surg 2004;40:512-8.) Clinical Relevance: Small abdominal aortic aneurysms with intima and media not completely damaged would be a good indication for this therapeutic gene transfer. The walls of large abdominal aortic aneurysms usually exhibit few viable intimal or medial cells, and it is thus not possible to induce cell proliferation. Among various components of the aortic wall, we focused on local cellularity as one of the protective mechanisms against aneurysm expansion. Although it is difficult to directly apply the present method using electroporation to abdominal aorta in human beings because of its risk for complication from electrical damage, advances in gene delivery techniques might realize the strategy of this study in the future.

Abdominal aortic aneurysm (AAA) is recognized as a life-threatening vascular disease, but the pathologic mechanisms of AAA formation are not well understood. Morphologic analyses of clinical specimens of the aneurysm wall demonstrate some characteristic findings associated with AAAs, including chronic infiltration of inflammatory cells, degradation of extracellular matrix, and apoptosis of smooth muscle cells (SMCs).1,2 In particular, because extracellular matrix sustains the strength and elasticity of the aortic wall, many studies have focused on the mechanisms From the Department of Vascular Regeneration,a and the Division of Vascular Surgery, Department of Surgery,b Graduate School of Medicine, The University of Tokyo, the Division of Tissue Engineering,c The University of Tokyo Hospital, and the Division of Vascular Surgery,d Stanford University, Palo Alto. Competition of interest: none. Reprint requests: Dr Hiroyuki Koyama, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. (e-mail: [email protected]). 0741-5214/$30.00 Copyright © 2004 by The Society for Vascular Surgery. doi:10.1016/j.jvs.2004.06.018

512

of matrix degradation, and matrix metalloproteinases (MMPs), released from inflammatory and other cells, have been considered to have an important role in matrix degradation and aneurysm formation.3,4 MMP inhibition by drugs or gene disruption prevents development of AAAs in experimental animal models,5,6 and clinical studies show evidence indicating upregulated activity of MMPs in patients with AAAs.7,8 However, recent studies have revealed that increased MMP-9 activity does not necessarily predict AAA progression in all cases.9,10 We previously evaluated the effects of wall shear stress and relative wall strain on aneurysm formation, and reported that high wall shear stress and relative wall strain suppressed enlargement of the aorta in a rat elastaseinduced AAA model.9-11 Further, histologic analysis revealed marked proliferation of medial SMCs in the aortic wall subjected to high wall shear stress and relative wall strain, accompanied by upregulation of mitogenic gene expression.9 Because SMCs in the vascular wall are an

JOURNAL OF VASCULAR SURGERY Volume 40, Number 3

Hoshina et al 513

Fig 1. In vivo gene transfer to aortic wall. a, A pair of electrode plates was positioned on the isolated abdominal aorta. b, After elastase infusion, plasmid solution was pooled for 20 minutes, followed by electroporation. Small bubbles appeared along the electrode during electric pulsation.

important source of matrix molecules such as elastin and collagen, we hypothesized that enhanced cellularity of the medial layer stabilized aortic integrity. Basic fibroblast growth factor (bFGF) possesses potent mitogenic activity for vascular SMCs in vivo.12,13 In the present study, therefore, we delivered the gene for bFGF to the aortic wall of a rat AAA model by means of an in vivo electroporation technique and evaluated whether cell proliferation in the aortic wall is associated with AAA formation. MATERIAL AND METHODS Plasmid. Plasmid pCAcchbFGFcs23 was constructed by inserting a modified human bFGF complementary DNA with the secretory signal sequence into the pCAGGS expression vector. The cDNA of modified human bFGF fused with the signal sequence of interleukin-2 was obtained from plasmid pTB1079 (donated by Dr. K. Igarashi, Biotechnology Research Laboratories, Takeda Chemical Industries, Osaka, Japan). The control plasmid for the animal experiment was pCAZ3, which contains Escherichia coli LacZ in the same expression plasmid. Plasmids were grown in E coli JM109, and were prepared with Qiagen EndoFree Mega Kits. In vitro study. To evaluate expression and secretion of bFGF from pCAcchbFGFcs23-treated rat cells, the plasmid pCAcchbFGFcs23 (15 ␮g) was transfected into 1.5 ⫻ 106 cultured rat vascular SMCs (from thoracic aorta, passage 4) by a method that used polyethylenimine (Ex Gen 500; MBI Fermentas). After transfection, SMCs were cultured in 3 mL of Dulbecco modified Eagle minimum

Fig 2. In vitro study for confirmation of plasmid transfer. Plasmid containing modified human basic fibroblast gene factor (bFGF) gene was transfected into cultured rat smooth muscle cells (SMCs). Culture medium of pCAcchbFGFcs23-treated SMCs (lanes 5-7) showed 2 forms of bFGF (18 and 22 kD), whereas medium samples of nontransfected SMCs (lanes 2-4) showed no band. Lane 1, molecular size marker.

essential medium (Gibco BRL) containing 2% fetal bovine serum in a 6-cm dish. The culture medium was changed at 24-hour intervals, and the medium was harvested at 3 days after transfection. Culture medium of non-transfected SMCs was used as control medium. Each culture medium (50 ␮L) was subjected to Western blot analysis with a mouse monoclonal antibody against bovine bFGF (1:500; Santa Cruz Biotechnology) according to a protocol described previously.14 This analysis was repeated twice. Rat AAA model. Male Sprague-Dawley rats (3-4 months; Saitama Rabbitry), fed a normal diet, were used in all experiments. Surgical procedures were performed under sterile conditions with a stereomicroscope, and all proto-

514 Hoshina et al

JOURNAL OF VASCULAR SURGERY September 2004

Fig 3. In vivo expression of transfected gene. a, Frozen section of aortic wall 2 days after elastase infusion/ electroporation of LacZ (⫻100). X-gal staining showed successful transfection (blue, arrow). b, Western blot analysis revealed that pCAcchbFGFcs23-treated samples (lanes 2-4) showed higher expression of bFGF than pCAZ3-treated samples (lanes 5-7). Lane 1, Molecular size marker. c, Densities of the bands on Western blot were quantified. Values are shown as mean ⫾ SD. *P ⬍ .05). bFGF, pCAcchbFGFcs23-treated samples. Cont, pCAZ3-treated samples.

cols conformed to the Guide for Care and Use of Laboratory Animals (National Institutes of Health Publ. No. 85-23, revised 1996). After anesthesia with intraperitoneal injection of 50 mg/kg of sodium pentobarbital (Abbott), a 10-mm segment of the infrarenal aorta was exposed and completely isolated from adjacent connective tissue, and the maximum diameter of the exposed aorta was measured with electronic microcalipers in systole. Elastase was infused according to the method of Anidjar and Dobrin with modifications as described.15,16 In brief, PE-10 tubing was advanced into the infrarenal aorta from the right deep femoral artery. The proximal and distal parts of the infrarenal aorta were clamped by means of temporary ligation, and subsequently 15 units of type 1 porcine pancreatic elastase (E-1250; Sigma) in 2 mL of normal saline solution was infused over 2 hours through the PE-10 tube. In vivo gene transfer to aortic wall. After elastase infusion, residual elastase solution was washed out with 2 mL of saline solution, and pCAcchbFGFcs23 (120 ␮g)

diluted with 300 ␮L of phosphate-buffered saline solution was gently infused within physiologic pressure through the PE-10 tube, and the isolated aorta was incubated for 20 minutes (bFGF group). A pair of stainless steel electrode plates (10 ⫻ 2 mm) were then positioned on the left and right sides of the infrarenal aortic wall (Fig 1, a and b), and electric pulses were delivered with an electric pulse generator (ECM 830; BTX); 5 square wave pulses of 15 V were followed by 5 pulses of opposite polarity at a rate of 1 pulse per second, with pulse duration of 50 ms.17 One group of rats was subjected to incubation with pCAZ3 in the same manner as control (control group). To confirm in vivo gene transfer to the aortic wall, pCAZ3-treated rats were killed 2 days after gene transfer, and frozen sections of the aorta were analyzed with X-gal staining. Further, pCAcchbFGFcs23-treated or pCAZ3-treated rats (n ⫽ 3 for each treatment) were killed on day 2 to assess in vivo expression of bFGF. Fresh aortic samples were rinsed with cold phosphate-buffered saline solution and lysed in lysis buffer (1%

JOURNAL OF VASCULAR SURGERY Volume 40, Number 3

Hoshina et al 515

Fig 4. Maximum diameter of infrarenal aorta (a), and luminal area index (b). Average maximum diameter at day 7 was greater in bFGF group than in control group, whereas no difference was detected in luminal area index. Values are shown as mean ⫾ SD. *P ⬍ .01, NS, no significant difference. Cont, control group; bFGF, bFGF group.

sodium dodecylsulfate, 100 mmol/L of sodium chloride, 62.5 mmol/L of Tris [pH 7.6], 1 mmol/L of phenylmethylsulfonyl fluoride, and 10 mg/mL of leupeptin), and the protein concentration was measured with a bicinchonic acid assay (Pierce Biotechnology). Equal amounts (20 ␮g total protein) of each lysate were subjected to Western blot analysis with anti-bFGF antibody. The density of the bands on Western blots was quantified as described.18 Total values of bFGF expression were determined by summing density values from all bFGF bands, and were used for statistical analysis. Measurement of aneurysm and histologic findings. Seven days after operation the maximum diameter of the aneurysm was measured with electronic microcalipers, and then the rats were killed. For histologic analysis animals of both groups were perfused with 4% paraformaldehyde solution in 0.1 mol/L of phosphate buffer (pH 7.4) through the common iliac artery at 120 mm Hg, and the infrarenal aorta was excised. After 1-hour immersion in the same fixative, the aortic sample was divided into 2 segments (proximal and distal segments) and embedded in paraffin. One set of sections (4 ␮m) was stained with hematoxylin, and the luminal area (AL) of the aortic section was measured as described.9 To standardize the values of AL, the area inside the external elastic lamina (AEEL) was measured, and the luminal area index was calculated as AL/ AEEL. Other sets of sections were immunostained for smooth muscle actin, macrophages, and CD31. A mono-

clonal antibody against human ␣-smooth muscle actin (1: 400; Sigma-Aldrich), rat macrophage (1:100; Chemicon), or mouse CD31 (20 ␮g/mL; Santa Cruz Biotechnology) was applied after blocking with 1% goat serum. Subsequent incubation with biotinylated anti-mouse immunoglobulin G antibody for 30 minutes and application of an avidin-biotin complex method kit (Vector Laboratories) were performed. Control slides were stained with a matching concentration of non-immune mouse immunoglobulin G rather than primary antibodies. Cell count was carried out on each slide at ⫻200 magnification. To measure the SMC density in the aortic medial layer, ␣-smooth muscle actin-positive cells per hpf were counted in 4 hpf per section (2 pairs of randomly selected opposing hpf) and averaged.9 The density of macrophages in the aortic media was evaluated in the same manner. For evaluation of de-endothelialization of the aortic lumen, CD31-positive cells on the luminal surface (endothelial cells) were counted. The average of each value from 2 sections per animal was used for statistical analysis. Cell proliferation. At 1 hour before sacrifice, rats received an intraperitoneal injection of 5-bromo-2=deoxyuridine (BrdU; 50 mg/kg) for analysis of cell proliferation. Replicating cells in each section were stained with a monoclonal antibody against BrdU (1:50; DAKO) as described.9 BrdU-positive nuclei in the entire circumference of the aortic media were counted as an index of aortic cell

516 Hoshina et al

JOURNAL OF VASCULAR SURGERY September 2004

Fig 5. Photomicrographs of aneurysm wall from control group (a) and bFGF group (b) at 7 days after treatment (⫻200). c, Sections were immunostained with anti-␣-smooth muscle actin antibody. ␣-Smooth muscle actin-positive cells stain brown. Medial smooth muscle cell density was evaluated by counting ␣-smooth muscle actin-positive cells in 4 high-power fields per section and averaging these values. d, Density of macrophages in aortic media was quantified in the same manner. e, CD31-positive cells on the luminal surface (endothelial cells) were counted. Values are shown as mean ⫾ SD. *P ⬍ .01. NS, No significant difference. Cont, Control group; bFGF, bFGF group.

proliferation.9 The average of each value from 2 sections per animal was used for statistical analysis. Statistical analysis. To prevent bias in the evaluation, a single student helper performed all histologic cell counting in blinded fashion. Results were expressed as mean ⫾ SD. Comparisons between groups were made with either paired or unpaired 2-tailed t tests, with significance determined at P ⬍ .05. RESULTS In vitro study. Western blot analysis showed that pCAcchbFGFcs23-treated rat SMCs secreted 2 forms of bFGF (18 and 22 kD) into the culture medium at 3 days after transfection, whereas no expression of bFGF was detected in the culture medium of non-transfected SMCs (Fig 2). In vivo expression of transduced gene. To determine in vivo expression of the transduced gene, pCAZ3treated aortic samples were analyzed with X-gal staining,

and abundant expression of LacZ was shown as blue staining (Fig 3, a). Western blot analysis detected bFGF expression in rat aortic wall incubated with pCAcchbFGFcs23 and also pCAZ3, and expression in pCAcchbFGFcs23treated samples was significantly higher than that in pCAZ3-treated samples (Fig 3, b and c). Another form (21 kD) of bFGF was detected in the aortic wall in vivo. Measurement of aneurysm diameter. Aneurysmal changes in the infrarenal aorta were observed in all rats at 7 days after elastase treatment and gene transfer. Maximum diameter of the aneurysms in the bFGF group was significantly smaller than that in the control group (Fig 4, a). Luminal area index showed no significant difference between the 2 groups (Fig 4, b). Cell number. SMCs were stained with anti-␣-smooth muscle actin antibody, and the number of medial SMCs was significantly increased in samples from the bFGF group compared with the control group 7 days after operation (Fig 5, a-c). In contrast, counting of macrophages in the aortic media

JOURNAL OF VASCULAR SURGERY Volume 40, Number 3

Hoshina et al 517

showed no significant difference between the bFGF and control groups (Fig 5, d). Furthermore, few scattered endothelial cells were detected in any samples of aortic wall, and there was no significant difference between the groups (Fig 5, e). Medial cell proliferation. Immunostaining with antiBrdU antibody demonstrated proliferating cells in the aortic wall, and medial cell proliferation in the bFGF group was significantly greater than that in the control group (Fig 6, a-c). DISCUSSION The purpose of this study was to evaluate whether enhanced aortic cellularity reduces aneurysmal degeneration and progressive increase in diameter in a rat elastaseinduced AAA model. Unlike previous experiments in which aortic cellularity was modified in response to variable flow conditions,9 in this case medial SMC proliferation was stimulated by upregulation of bFGF gene expression. Although used primarily to effect gene transfer in cultured cells, recent work suggests that electroporation efficiently transfers plasmid DNA to tissues and organs in vivo.17,19 Matsumoto et al17 reported that in vivo electroporation enhanced gene expression after plasmid-based gene transfer more than 10-fold in a rabbit carotid artery. We modified these methods to transfect and upregulate bFGF gene expression in rat elastase-treated aorta. Indeed, in the in vivo evidence from the present study, bFGF expression in the pCAcchbFGFcs23-treated aorta was greater than that in the pCAZ3-treated aorta, which might be endogenous bFGF expression in the aorta of this model.14,20 Linking the bFGF gene to the secretory signal sequence of interleukin-2 enabled successful extracellular transport, as confirmed with Western blotting in our cultured rat SMCs (Fig 2). It is presumed that augmenting bFGF expression stimulated cell replication and increased medial SMC density in transfected rats. It is important that no meaningful change was detected in the number of aortic macrophages or endothelial cells in bFGF-treated animals. That bFGF treatment status influenced AAA diameter in these experiments supports and extends our previous observations regarding the role of medial SMC density in maintaining aortic structural integrity.9 Although increased medial cellularity is one of many potential anti-aneurysmal consequences of increased aortic flow,9,11 this experimental design facilitates analysis of SMC proliferation and retention independent of changes in wall shear stress or relative wall strain. Medial SMCs synthesize and assemble collagen and elastin precursors, maintain matrix organization, and mediate the tensile characteristics of the aortic wall. It seems self-evident, therefore, that reduced medial SMC density compromises aortic integrity and facilitates aneurysmal degeneration. Medial SMC apoptosis and cell loss is a morphologic hallmark of AAAs in human beings, possibly induced by inflammatory cytokines and reactive oxygen species released from accumulated inflammatory cells.21,22 The findings of the current study strengthen the association between medial SMC density and aortic structural integrity

Fig 6. Photomicrographs of aneurysm wall from control group (a) and bFGF group (b) at 7 days after treatment (⫻400). Sections were immunostained with anti-bromodeoxyuridine (BrdU) antibody and counterstained with hematoxylin. BrdU-positive cells stain brown. Cell replication in the media was evaluated by counting BrdU-positive cells. Values are shown as mean ⫾ SD. *P ⬍ .05. Cont, control group; bFGF, bFGF group.

in experimental models, and suggest new treatment strategies to limit the progression of clinical AAA disease. Alternative explanations for the observed results include the possibility that increased bFGF directly suppresses aneurysm progression independent of its effect on medial SMC density. Previous in vitro work suggests that bFGF inhibits transendothelial monocyte migration and downregulates expression of proinflammatory endothelial adhesion molecules.23 These effects may limit aortic wall inflammation after bFGF gene transfer, potentially inhibiting aneurysm progression in this manner. No significant difference in medial macrophage accumulation was noted between the bFGF and control groups in this study, however, which suggests that the anti-inflammatory consequences of increased bFGF expression are negligible in this context. These findings support our conclusion that growth factor–mediated SMC proliferation is the most likely explanation for limited aneurysm progression in these experiments. The exact mechanisms by which increased SMC density limits AAA progression remains unknown. Investigations

518 Hoshina et al

into these mechanisms are limited by the complexity of the interactions between various cells types in the extracellular matrix. Matrix homeostasis is a balance of synthesis and degradation. Several matrix molecules contribute to aortic architecture, and degradation is regulated by conflicting systems of proteases and inhibitors generated by infiltrative inflammatory cells as well as SMCs and endothelial cells.24,25 Inasmuch as different cell types with ubiquitous functions are present in or migrate to the aneurysm wall at varyious times during the course of the disease, experimental data using whole specimens of aortic wall have proved to be of limited value in clarifying the cause of the disease process. Indeed, neither real-time reverse transcription polymerase chain reaction nor gelatin zymography analyses using whole aortic samples provided meaningful insight in the present study (data not shown). To determine cellspecific gene expression in the aneurysm wall it might be necessary to use experimental methods that isolate groups of cells within organs, such as laser-capture microdissection.26 We are adopting these methods to facilitate analysis of cell-specific gene expression and further define the influence of increased medial SMC density on AAA progression. In conclusion, we delivered the bFGF gene to enhance medial cell density in the aortic wall of a rat elastase-induced AAA model and assessed the relation between medial cellularity and AAA formation. In vivo electroporation was used to deliver the bFGF gene, and gene-transferred cells released bFGF, because the bFGF gene was fused with the signal sequence. The dominant component of the enhanced cellularity was SMCs, and no increase in macrophages or endothelial cells was observed. At 7 days after elastase treatment and gene transfer of bFGF, aneurysm size was significantly smaller than that of control. These findings suggest that enhancement of medial cell density inhibited aneurysm formation after elastase treatment, which possibly offers a clue for development of a new strategy for AAA treatment. REFERENCES 1. Wassef M, Baxter BT, Chisholm R, Dalman RL, Fillinger MF, Heinecke J, et al. Pathogenesis of abdominal aortic aneurysms: a multidisciplinary research program supported by the National Heart, Lung and Blood Institute. J Vasc Surg 2001;34:730-8. 2. Henderson EL, Geng YJ, Sukhova GK, Whittemore AD, Knox J, Libby P. Death of smooth muscle cells and expression of mediators of apoptosis by T lymphocytes in human abdominal aortic aneurysms. Circulation 1999;99:96-104. 3. Freestone T, Turner RJ, Coady A, Higman DJ, Greenhalgh RM, Powell JT. Inflammation and matrix metalloproteinases in the enlarging abdominal aortic aneurysms. Arterioscler Thromb Vasc Biol 1995;15:1145-51. 4. McMillan WD, Tamarina NA, Cipollone M, Johnson DA, Parker MA, Pearce WH. Size matters: the relationship between MMP-9 expression and aortic diameter. Circulation 1997;96:2228-32. 5. Moore G, Liao S, Curci JA, Starcher BC, Martin RL, Hendricks RT, et al. Suppression of experimental abdominal aortic aneurysms by systemic treatment with a hydrozamate-based matrix metalloproteinase inhibitor (RS 132908). J Vasc Surg 1999;29:522-32. 6. Bigatel DA, Elmore JR, Carey DJ, Cizmeci-Smith G, Franklin DP, Youkey JR. The matrix metalloproteinase inhibitor BB-94 limits expansion of experimental abdominal aortic aneurysms. J Vasc Surg 1999;29:130-9.

JOURNAL OF VASCULAR SURGERY September 2004

7. Sakalihasan N, Delvenne P, Nusgens BV, Limet R, Lapiere CM. Activated forms of MMP2 and MMP9 in abdominal aortic aneurysms. J Vasc Surg 1996;24:127-33. 8. Carrell TWG, Burnand KG, Wells GMA, Clements JM, Smith A. Stromelysin-1 (matrix metalloproteinase-3) and tissue inhibitor of metalloproteinase-3 are overexpressed in the wall of abdominal aortic aneurysms. Circulation 2002;105:477-82. 9. Hoshina K, Sho E, Sho M, Nakahashi TK, Dalman RL. Wall shear stress and strain modulate experimental aneurysm cellularity. J Vasc Surg 2003;37:1067-74. 10. Mendoza AC, Karwowski JK, Dalman RL. Increased flow limits enlargement of experimental aneurysms. Surg Forum 1999;133:321-5. 11. Nakahashi TK, Hoshina K, Tsao PS, Sho E, Sho M, Karwowski JK, et al. Flow loading induces macrophage anti-oxidative gene expression in experimental aneurysms. Arterioscler Thromb Vasc Biol 2002;22: 2017-22. 12. Lindner V, Lappi DA, Baird A, Majack RA, Reidy MA. Role of basic fibroblast growth factor in vascular lesion formation. Circ Res 1991;68: 106-13. 13. 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;88:3739-43. 14. Ohara N, Koyama H, Miyata T, Hamada H, Miyatake SI, Akimoto M, et al. Adenovirus-mediated ex vivo gene transfer of basic fibroblast growth factor promotes collateral development in a rabbit model of hind limb ischemia. Gene Ther 2001;8:837-45. 15. Anidjar S, Salzmann J, Gentric D, Lagneau P, Camilleri J, Michel J. Elastase-induced experimental aneurysms in rats. Circulation 1990;82: 973-81. 16. Anidjar S, Dobrin PB, Eichorst M, Graham GP, Chejfec G. Correlation of inflammatory infiltrate with the enlargement of experimental aortic aneurysms. J Vasc Surg 1992;16:139-47. 17. Matsumoto Y, Komori K, Shoji T, Kuma S, Kume M, Yamaoka T, et al. Successful and optimized in vivo gene transfer to rabbit carotid artery mediated by electronic pulse. Gene Ther 2001;8:1174-9. 18. Koyama H, Reidy MA. Expression of extracellular matrix proteins accompanies lesion growth in a model of intimal reinjury. Circ Res 1998;82:988-95. 19. Aihara H, Miyazaki J. Gene transfer into muscle by electroporation in vivo. Nat Biotechnol 1998;16:867-70. 20. Ninomiya M, Koyama H, Miyata T, Hamada H, Miyatake S, Shigematsu H, et al. Ex vivo gene transfer of basic fibroblast growth factor improves cardiac function and blood flow in a swine chronic myocardial ischemia model. Gene Ther 2003;10:1152-60. 21. Thompson RW, Liao S, Curci JA. Vascular smooth muscle cell apoptosis in abdominal aortic aneurysms. Coronary Artery Dis 1997;8:623-31. 22. Lopez-Candales A, Holmes DR, Liao S, Scott MJ, Wickline SA, Thompson RW. Decreased vascular smooth muscle cell density in medial degeneration of human abdominal aortic aneurysms. Am J Pathol 1997;150:993-1007. 23. Zang H, Issekutz AC. Down-modulation of monocyte transendothelial migration and endothelial adhesion molecule expression by fibroblast growth factor: reversal by the anti-angiogenic agent SU6668. Am J Pathol 2002;160:2219-30. 24. Ocana E, Bohorquez J-C, Perez-Requena J, Brieva JA, Rodriguez C. Characterisation of T and B lymphocytes infiltrating abdominal aortic aneurysms. Atherosclerosis 2003;170:39-48. 25. McMillan WD, Patterson BK, Keen RR, Shively VP, Cipollone M, Pearce WH. In situ localization and quantification of mRNA for 92-kD type IV collagenase and its inhibitor in aneurysmal, occlusive, and normal aorta. Arterioscler Thromb Vasc Biol 1995;15:1139-44. 26. Bonner RF, Emmert-Buck M, Cole K, Pohida T, Chuaqui R, Goldstein S, et al. Laser capture microdissection: molecular analysis of tissue. Science 1997;278:1481-3. Submitted Nov 20, 2003; accepted Jun 9, 2004.