- Email: [email protected]
Vascular endothelial growth factor enhances angiotensin II-induced aneurysm formation in apolipoprotein E-deficient mice
Vascular endothelial growth factor enhances angiotensin II-induced aneurysm formation in apolipoprotein E-deficient mice Edward Choke, PhD, MRCS, Gillian W. Cockerill, PhD, Joseph Dawson, MD, MRCS, Franklyn Howe, PhD, William R. W. Wilson, MD, MRCS, Ian M. Loftus, MD, FRCS, and Matthew M. Thompson, MD, FRCS, London, United Kingdom Objective: Abdominal aortic aneurysm (AAA) development is associated with increased angiogenesis and overexpression of vascular endothelial growth factor (VEGF). Inhibition of angiogenesis results in attenuation of experimental aneurysms. This study investigated the effects of recombinant human (rh)VEGF on experimental aneurysms. Methods: Apolipoprotein E-deficient (apoE–/–) mice were assigned to one of four groups: (1) normal saline infusion (sham), (2) angiotensin-II (AngII) infusion, (3) AngII infusion plus 100 g daily rhVEGF for 14 days (AngIIⴙ14dVEGF), or (4) AngII infusion plus 100 g daily rhVEGF for 21 days (AngIIⴙ21dVEGF). Aortic maximum diameter and cross-sectional area were determined by magnetic resonance imaging and microscopy. All mice were sacrificed at day 28. Results: Aneurysms developed in all mice in the AngIIⴙ14dVEGF and AngIIⴙ21dVEGF groups by day 21 compared with 40% in the AngII group. Treatment with rhVEGF increased maximum aortic diameter (P < .002) and crosssectional area of aneurysms (P < .005) at day 21. This effect was maintained at day 28 (P < .0005). Decreasing rhVEGF treatment from 21 to 14 days did not attenuate aneurysm formation. Treatment with rhVEGF upregulated matrix metalloproteinase 2 gene expression within the aortic wall (P < .0009). Conclusions: Treatment with rhVEGF intensified the formation of AngII-induced aneurysms. Further studies are needed to investigate if antiangiogenic therapy may be a valid medical therapy against aneurysm expansion or rupture. ( J Vasc Surg 2010;52:159-66.) Clinical Relevance: Abdominal aortic aneurysm (AAA) rupture ranks as the 13th leading cause of death in the United States and is responsible for 0.8% of all deaths. No pharmacologic treatment is currently available to inhibit aneurysm expansion or rupture, and the only treatment option is surgical repair by conventional or endovascular means. Previous studies have demonstrated that the AAA rupture site is associated with increased angiogenic response, including increased expression of vascular endothelial growth factor. This study demonstrated that recombinant human vascular endothelial growth factor promoted the severity of experimental aneurysm formation. Angiogenesis may be a valid target for therapeutic intervention against aneurysms.
Abdominal aortic aneurysm (AAA) rupture is the 13th most common cause of death in the Western world,1 and the only effective approach to preventing aneurysm rupture remains surgical repair by conventional or endovascular means. Although collective endeavors in basic science research during the past 2 decades have led to progress in the understanding of aneurysm development and expansion,
From St. George’s Vascular Institute. This work was supported by a research fellowship from the Royal College of Surgeons of England and grants from Peel Medical Research, Society of Academic and Research Surgery, UK, and EU-FP7-HEALTH Fighting Aneurysmal Disease 200467. Competition of interest: none. Reprint requests: Professor Matt Thompson, St. George’s Vascular Institute, St. George’s Hospital, Blackshaw Rd, London SW17 0RE, UK (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest. 0741-5214/$36.00 Copyright © 2010 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2010.02.015
the process is not yet fully understood, and there is no effective medical therapy against aneurysms. The association between angiogenesis and human aneurysms has been recognized by previous investigators2-4 who suggested that angiogenesis might contribute toward the development of aneurysms. The localization of increased angiogenesis and increased expression of proangiogenic cytokines to the site of aneurysm rupture in humans5 implied a further role for angiogenesis in end-stage aneurysm disease and aneurysm rupture. Extracellular proteolysis is an absolute requisite for angiogenesis, primarily contributing toward proteolytic degradation of the vascular basement membrane and remodeling of the extracellular matrix to allow endothelial cells to migrate and invade into the surrounding tissue.6 It is this link between proangiogenic cytokines and induction of extracellular proteolysis that makes these cytokines potentially relevant in aneurysm disease. Recently, Tedesco et al7 demonstrated that the mural expression of vascular endothelial growth factor (VEGF) receptor (VEGFR), as determined by fluorescent imaging and VEGFR-2 immu159
JOURNAL OF VASCULAR SURGERY July 2010
160 Choke et al
Day 1
Day 28
Saline infusion (sham control) 28 day saline infusion
Daily IP saline injection Day 8
28 day AngII infusion
Daily IP saline injection Day 8
Angiotensin II group (positive control)
AngII+14dVEGF short duration group 28 day AngII infusion
Daily IP 100µg VEGF injection Day 15
AngII+21dVEGF long duration group 28 day AngII infusion
Daily IP 100µg VEGF injection Day 8
Fig 1. Schematic diagram of study design. AngII, Angiotensin II; IP, intraperitoneal; VEGF, vascular endothelial growth factor.
nostaining, was increased in a diameter-dependent fashion in experimental aneurysms. These authors also showed that inhibition of angiogenesis resulted in the attenuation of experimental aneurysms. To date, the effects of proangiogenic mediators in experimental models of aneurysm have never been tested, and whether the observed overexpression of proangiogenic cytokines at the site of aneurysm rupture is a cause or effect of end-stage aneurysmal disease remains unknown. The present study investigated the effects of recombinant human (rh) VEGF on angiotensin (Ang)-II–induced aneurysm in apolipoprotein (Apo) E–/– mice. The results indicated that rhVEGF significantly enhanced AAA formation and upregulated gene expression of matrix metalloproteinase (MMP)-2 in the aortic wall. METHODS Experimental animals. The Home Office Science and Research Group (Animals Scientific Procedures Division) approved all animal protocols. Six-month-old apoE–/– male mice (Jackson Laboratories, Bar Harbor, Me) underwent a minimum 1-month acclimatization period. Water and regular mouse chow was available 24 hours daily. Mice were assigned to one of four groups: (1) infusion of normal saline (sham), (2) infusion of AngII (1000 ng/kg/min, Sigma-Aldrich Corp, St. Louis, Mo), (3) infusion of AngII plus 14 days daily intraperitoneal (IP) injections of 100 g rhVEGF (AngII⫹14dVEGF, short-duration group), or (4) infusion of AngII plus 21 days daily IP injections of 100 g rhVEGF (AngII⫹21dVEGF, long-duration group; Fig 1). The short-duration group commenced daily rhVEGF injections (Genentech Inc, South San Francisco, Calif) from day 15, and the long-duration group commenced daily rhVEGF injections from day 8. Mice from the sham group and AngII-only group received daily IP saline injections from day 8. All mice were sacrificed on day 28.
Fig 2. Left, Representative magnetic resonance image coronal section with (Right) photograph of corresponding arterial tree showing suprarenal abdominal aortic aneurysm. In each mouse, nine axial T1-weighted spin-echo cross-sectional images (green lines) were acquired with 2-mm-thick contiguous slices. To ensure reproducible positioning of axial images, the central slice (5th) was aligned at the superior pole of the left kidney.
Implantation of miniosmotic pumps. Alzet model 2004 mini-osmotic pumps (Durect Corp, Cupertino, Calif) were loaded with individual concentrations of AngII (variable according to individual animal weight) for 28 days of continuous delivery of 1000 ng/kg/min of AngII. These were incubated in normal saline at 37°C for 24 hours before subcutaneous implantation at the interscapular area. The miniosmotic pumps implanted in the sham group were loaded with normal saline. Magnetic resonance imaging protocol. Mice underwent magnetic resonance imaging (MRI) on days 21 and 28 (before sacrifice). MRI data were obtained using a Varian Unity INOVA imaging system fitted with a 4.7-T horizontal bore magnet. Coronal scout images were acquired to allow reproducible positioning of axial images aligned with the central slice at the superior end of the left kidney (Fig 2). In each mouse, nine axial T1-weighted spinecho cross-sectional images were acquired with 2-mm-thick contiguous slices. Using the Varian ImageBrowser (Varian Medical Systems, Palo Alto, Calif) software and delineation of regions of interest, an operator (F. H.) who was blinded to the experimental arm of the mice determined the abdominal aorta diameter (mm) and cross-sectional area (mm2) in each cross-section were determined. Details of the experimental protocol are presented in the Supplement (online only).
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
Choke et al 161
Table I. Weight and mortality dataa Variableb Total number Weight, g Pretreatment 4th week Survived to 28 days Spontaneous death Rupture Nonrupture Schedule 1 sacrifice Paraplegia General lethargy Total aneurysmsd
Saline
AngII
AngII⫹14dVEGF
AngII⫹21dVEGF
5
11
9
10
32.5 ⫾ 0.8 35.3 ⫾ 0.9 4 (80)
34.0 ⫾ 0.9 37.1 ⫾ 1.0 5 (45)
34.8 ⫾ 1.1 36.0 ⫾ 1.5 3 (33)
35.0 ⫾ 0.7 35.5 ⫾ 1.2 4 (40)
0 1
2 (18) 2 (18)c
2 (22) 1 (11)
1 (10) 2 (20)
0 0 0 (0)
2 (18)c 0 9 (81)
2 (22)c 1 (11) 8 (89)
2 (20)c 1 (10) 9 (90)
AngII, Angiotensin II; VEGF, vascular endothelial growth factor. a P values for all variables were not significant. b Data are presented as mean ⫾ standard error of the mean, or No. (%). c One mouse in each of these groups did not develop an aneurysm. d Includes mice that died ⬍28 days.
Mice aortic tissue collection. At the end of the 28day treatment period, mice were sacrificed and blood was drawn from the right ventricle. To maximize RNA stability, aortas were perfused with RNALater (Applied Biosystems, Warrington UK) through the heart, and the entire arterial trees were rinsed with RNALater while in situ. A dissection microscope (Olympus, Tokyo, Japan) was used to carefully dissect the arterial tree from fat and connective tissue and was photographed. Ex vivo measurements of diameter were taken with calibrated calipers through the entire length of the aorta to determine the region of maximum dilatation. Samples for molecular analysis were harvested from aneurysmal segments of the aorta and placed in RNALater at –20°C until extraction of total RNA. As previously described,5 total RNA was extracted using RNeasy Fibrous Tissue RNA extraction kit (Qiagen, West Sussex, UK) and checked for integrity and concentration using RNA 6000 Nano LabChip Kit and Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, Calif). Quantitative real-time reverse-transcription polymerase chain reaction. Complimentary DNA (cDNA) was synthesized from total RNA by reverse-transcription using a high-capacity cDNA archive kit (Applied Biosystems). Qualitative reverse-transcription polymerase chain reaction (QRT-PCR) was done with the Mx4000 Multiplex Quantitative PCR System (Stratagene, Edinburgh, UK) using TaqMan Gene Expression Assays (Assays-onDemand Gene Expression Products, Applied Biosystems, UK) and TaqMan Universal PCR Master Mix. Transcript levels were normalized to 18s rRNA (internal control). The expression of genes encoding for the proinflammatory cytokines E-selectin and tumor necrosis factor (TNF), the proteolytic enzymes from the MMP family (membrane type-1-MMP [MT1-MMP] and MMPs ⫺2 and ⫺9) and the plasmin activator system (urokinase plasminogen activator [uPA] and tissue plasminogen activator [tPA]) was determined. Details of the experimental protocol are presented in the Supplement (online only).
Statistics. All data were reported as mean ⫾ standard error of the mean. Multiple comparisons of mean values of aortic aneurysm diameters, cross-sectional areas, and gene expression were performed by one-way analysis of variance, followed by a subsequent Newman-Keuls multiple comparison test. Differences were considered statistically significant at a value of P ⬍ .05. RESULTS Weight and mortality data. No significant differences were observed for weight at the beginning and completion of the study among all experimental groups of mice (Table I). Necropsy was performed ⱕ24 hours on animals that died before completion of study. These animals were not included in AAA quantification analysis but only in the mortality data. The mortality rate was increased in the groups that received AngII infusions compared with the sham group, although this did not reach statistical significance (Table I). Aortic ruptures causing death occurred in AngII (n ⫽ 2), AngII⫹14dVEGF (n ⫽ 2), and AngII⫹21dVEGF (n ⫽ 1) groups. All aortic ruptures occurred between days 6 and 14. Paraplegia or general lethargy necessitated schedule 1 euthanasia before 28 days of infusion had been completed in AngII (n ⫽ 2), AngII⫹14dVEGF (n ⫽ 3), and AngII⫹ 21dVEGF (n ⫽ 3; P ⫽ .48). In the placebo control group, one mouse died on day 18 after insertion of the osmotic pump, and necropsy revealed normal aorta. Including mice that died before 28 days, the incidence of aneurysms was 0% (saline), 81% (AngII), 89% (AngII⫹ 14dVEGF), and 90% (AngII⫹21dVEGF). The aneurysms in mice that died ⬍28 days were excluded from further analysis because their size would have been different had they been allowed to develop to 28 days. Effect of rhVEGF on maximum diameter and crosssectional areas. Treatment with rhVEGF increased the maximum diameter and maximum cross-sectional area of
JOURNAL OF VASCULAR SURGERY July 2010
162 Choke et al
A
Saline
Ang II
Ang II plus 14 days VEGF
Ang II plus 21 days VEGF
Maximal diameter (mm) day 21
Maximal diameter (mm) day 28 #
3.0
§
3.0
2.5
2.5
2.0
2.0
1.5
1.5
1.0
†
††
1.0
Saline
AngII
AngII+14dVEGF
AngII+21dVEGF
Saline
AngII+14dVEGF
AngII+21dVEGF
C
B Maximal cross sectional area (mm2) day 21 7
Maximal cross sectional area (mm2) day 28 ‡
‡‡
7
6
6
5
5
4
4
3
3
2
2
1
1
*
**
0
0
Saline
D
AngII
AngII
AngII+14dVEGF AngII+21dVEGF
Saline
AngII
AngII+14dVEGF AngII+21dVEGF
E
Fig 3. Angiotensin (Ang) II-induced abdominal aortic aneurysm (AAA) size was significantly increased at days 21 and 28 in apolipoprotein E–/– mice treated with vascular endothelial growth factor (VEGF) for 14 (AngII⫹14dVEGF) or 21 (AngII⫹14dVEGF) days. A, Representative macroscopic photographs of the AAA are shown with corresponding cross-sectional magnetic resonance imaging scans at day 28. The maximum diameter of the aorta is outlined by the red line. Box and whisker plots show the AAA maximum diameter at (B) day 21 and (C) day 28 and maximum cross-sectional area at (D) day 21 and (E) day 28 in saline-infused (n ⫽ 4), AngII-infused (n ⫽ 5), Ang II⫹14dVEGF-infused (n ⫽ 3), and AngII⫹21dVEGF-infused (n ⫽ 4) mice. #P ⬍ .05 vs AngII infusion; §P ⬍ .05 vs Ang II infusion; †P ⬍ .05 vs AngII infusion; ††P ⬍ .05 vs Ang II infusion; ‡P ⬍ .01 vs AngII infusion; ‡‡P ⬍ .05 vs Ang II infusion; *P ⬍ .05 vs Ang II infusion; **P ⬍ .05 vs Ang II-infusion. The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively, and the whiskers mark the 90th and 10th percentiles.
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
Choke et al 163
Table II. Summary of gene expressions (arbitrary units normalized to 18S recombinant RNA) Gene
Saline Mean ⫾ SEM
AngII Mean ⫾ SEM
AngII⫹VEGF (pooled) Mean ⫾ SEM
P valuea
MMP-2 MMP-9 MT1-MMP uPA tPA E-selectin TNF
2.20 ⫾ 0.26 4.96 ⫾ 0.16 6.42 ⫾ 0.15 6.00 ⫾ 0.31 5.65 ⫾ 0.16 5.20 ⫾ 0.25 4.71 ⫾ 0.32
1.76 ⫾ 0.22 5.09 ⫾ 0.46 6.79 ⫾ 0.32 5.55 ⫾ 0.27 5.40 ⫾ 0.17 4.75 ⫾ 0.21 4.19 ⫾ 0.26
4.31 ⫾ 0.13 5.14 ⫾ 0.36 7.00 ⫾ 0.06 6.03 ⫾ 0.09 5.60 ⫾ 0.07 5.04 ⫾ 0.07 4.80 ⫾ 0.21
.0009b .94 .40 .46 .52 .39 .36
AngII, Angiotensin II; MMP, matrix metalloproteinase; MT1-MMP, membrane type-1-MMP; SEM, standard error of the mean; TNF, tumor necrosis factor; tPA, tissue-type plasminogen activator; uPA, urokinase-type plasminogen activator; VEGF, vascular endothelial growth factor. a Comparison used one-way analysis of variance. b Significant.
AngII-induced aneurysms (Fig 3). MRI measurements at day 21 demonstrated a 52% increase in maximum aortic diameter (P ⬍ .05) and a 100% increase in maximum cross-sectional area (P ⬍ .05) in the short-duration group (AngII⫹14dVEGF) compared with AngII mice. Similarly, a 38% increase in maximum aortic diameter (P ⬍ .05) and a 68% increase in the maximum cross-sectional area (P ⬍ .05) occurred in the long-duration group (AngII⫹ 21dVEGF). The difference in the maximum aortic diameter or cross-sectional area between short-duration and longduration VEGF groups was not statistically significant. MRI measurements at day 28 demonstrated a 36% increase in maximum aortic diameter (P ⬍ .05) and a 76% increase in maximum cross-sectional area (P ⬍ .05) in the short-duration group (AngII⫹14dVEGF) compared with AngII mice. A 25% increase in maximum aortic diameter (P ⬍ .05) and a 52% increase in maximum cross-sectional area (P ⬍ .05) occurred in the long-duration group (AngII⫹ 21dVEGF). At day 28, no statistically significant difference was evident in maximum diameter or maximum crosssectional area between short- and long-duration VEGF groups. After mice were sacrificed at day 28, ex vivo measurements demonstrated that maximum aortic diameter was increased by 67% (P ⬍ .01) in the short-duration group (AngII⫹14dVEGF) and by 51% (P ⬍ .01) in the longduration group (AngII⫹21dVEGF) compared with AngII mice. A comparison of ex vivo aortic maximum diameter measurements with in vivo measurements by MRI at day 28 (Fig 4, online only) demonstrated that these two measurements were highly correlated with each other (Pearson r2 ⫽ 0.88). Aneurysm formation in rhVEGF-treated mice. Aneurysms formed earlier in rhVEGF-treated mice. Aortic aneurysm was defined as ⬎50% enlargement of maximum abdominal aortic diameter, in line with current clinical definitions.8 The average diameter of the suprarenal aorta in sham mice at 3 weeks was 1.23 mm and a threshold of 1.85 mm was therefore set as the minimum diameter for incidence of aneurysm formation. Aneurysms had already developed at day 21 in all mice in the AngII⫹14dVEGF and AngII⫹21dVEGF groups compared with only two (40%) of the AngII mice.
At day 28, the average diameter of the suprarenal aorta in the control mice was very similar (1.25 mm), and a threshold of 1.88 mm was therefore set as the minimum diameter for incidence of aneurysm formation. At this time point, aneurysm formation was evident in 80% of AngII mice. From days 21 to 28, there was an increase in maximum aortic diameter (14%) and maximum cross-sectional area (25%; P ⬍ .05) in mice from the AngII group. In the rhVEGF-treated groups, no statistically significant changes to diameter or cross-sectional area were evident between days 21 and 28. These data suggested that aneurysms formed more quickly in mice exposed to a proangiogenic stimulus. Effect of VEGF on MMP-2 gene expression in the aortic wall. Gene expression of MMP-2 was significantly increased (P ⫽ .0009) in aneurysmal sections of the aorta in the pooled rhVEGF-treated group compared with aneurysmal sections of the aorta in the AngII group or normal aorta in the saline group (Table II). There was no difference in gene expression of MMP-9, MT1-MMP, uPA, tPA, Eselectin, or TNF. DISCUSSION This study investigated the functional significance of increasing the exposure to the proangiogenic VEGF in experimental aneurysms. VEGF was selected for investigation because it is a potent inducer of angiogenesis and has consistently been shown to be a potentially important participant in aneurysm development3 and rupture.5 Previous work has shown VEGF is increased in aneurysmal tissues.3 The present study demonstrated that exogenous treatment with rhVEGF led to earlier development and increased maximum expansion of AngII-induced aneurysms. This enhancement of aneurysms after rhVEGF treatment was independent of the duration of rhVEGF treatment. In addition to VEGF, an increase in other proangiogenic molecules, such as ␣v-integrin, VE-cadherin, monocyte chemoattractant protein-1, and vimentin, has also been observed at the site of AAA rupture in humans.5,9 However, data from the present study demonstrated that
JOURNAL OF VASCULAR SURGERY July 2010
164 Choke et al
rhVEGF treatment did not lead to an increased rate of aortic rupture within the 28-day experimental period. MRI has previously been validated as a robust noninvasive imaging technique for detecting in vivo phenotypic abnormalities, including the characterization of atherosclerotic plaque,10-14 diagnosis, and measurements of arterial remodeling during aortic lumen enlargement in apoE–/– mice.15,16 In the present study, MRI using a Varian Unity INOVA imaging system fitted with a 4.7-T horizontal bore magnet was used for in vivo monitoring, detecting, and measuring maximum diameter and cross-sectional area of AngII-induced aneurysms in apoE–/– mice. This work endorsed the potential of MRI to obtain accurate vessel dimensions in diseased abdominal aorta, verified against macroscopic measurements, and to monitor disease progression by serial imaging at different time points. No current studies have clearly demonstrated an effect of age on the development of AngII-induced aneurysms. The use of younger mice is limited by the practical constraints of their smaller size and therefore difficulties in implantation of the Alzet miniosmotic pumps. Effective drug delivery required the body weight of mice to be at least 20 grams. We used 6-month-old mice in this study according to the methods described in the first published article on AngII-mediated aneurysms.17 Mice ranging in age from 3 to 11 months at the initiation of AngII infusion have been used, and the described incidence of AngIIinduced aneurysms has been fairly consistent, at 70% to 90%, in hyperlipidemic mice. The incidence of aneurysms in the AngII group (without VEGF) was 80% at 28 days, and this is consistent with that of other published studies. On the other hand, the use of older mice might have resulted in the lower-than-anticipated overall survival to 28 days. The rates of aortic rupture (all ⬍day 14) were 18% (AngII), 22% (AngII⫹14dVEGF), and 10% (AngII⫹ 21dVEGF). These results were consistent with rates of 10%18 to 40%19 published in previous reports. However, the overall survival was low due to other causes such as paraplegia and general lethargy that necessitated schedule I sacrifice. Paraplegia is a recognized complication of AngII infusion.20 Saraff et al18 demonstrated that neovascularization within AngII-induced aneurysms was a late-stage response that occurred during maturation of aneurysmal tissues between days 14 and 28. In this study, rhVEGF IP injection was started at day 15 (short-duration group) to coincide with this response to investigate the effects of proangiogenic cytokine on the angiogenic process. IP rhVEGF injection was also started at day 8 (long-duration group) to investigate the effects of introducing a proangiogenic cytokine before any “naturally occurring” neovascularization process within the developing aneurysms. The results indicated that the effects of rhVEGF were not different, whether it was administered before (day 8) or during (day 15) the neovascularization process. This suggested that there was a dose-saturating effect, in which 14 days of rhVEGF treatment resulted in maximum enhancement of the aneurysm, with no further increases demon-
strable with an earlier additional 7 days of rhVEGF. It is interesting to note that aneurysms in the short-duration group were larger than aneurysms that had longer exposure to rhVEGF, but this result did not reach statistical significance. Tedesco et al7 recently reported that mural VEGFR expression was increased in AngII-induced aneurysms in a diameter-dependent fashion. Because in vivo measurements of VEGFR expression using near-infrared fluorescent imaging is not yet possible due to intrinsic autofluorescence, there are currently no data on the temporal relationship between VEGFR expression and the development of aneurysms. The time course of VEGFR expression may be a possible explanation for the larger (although not statistically significant) size of aneurysms in the short-duration VEGF group compared to the long-duration group. The regulation of MMP and serine protease expression by VEGF is well described. The regulation of MMPs and tissue inhibitor of matrix metalloproteinases (TIMPs) in endothelial cells by VEGF is variable and cell-type dependent:6 ●
●
● ● ●
In human umbilical vein endothelial cells, VEGF increased MMP-1, MMP-2, and TIMP-1, and this change was accompanied by increased activation of MMP-2.21,22 In human dermal microvascular endothelial cells, VEGF increased MMP-2, MMP-9, and MT-1 MMP, had no effect on MMP-1, and decreased TIMP-1 and TIMP-2.23,24 Circulating VEGF induced MMP-9 expression in bone marrow.25 VEGF treatment enhanced production of MMPs-1, -3, and -9 by human smooth muscle cells.26 In bovine microvascular endothelial cells27 and human umbilical vein endothelial cells,28 VEGF increased tPA and uPA.
The messenger RNA (mRNA) levels of MMPs-2 and MMPs-9, MT1-MMP, tPA, and uPA were analyzed in this study to investigate if VEGF treatment had any effect on the expression of these VEGF downstream target genes in the aortic tissue of the mice. In addition, this study analyzed the gene expression of the proinflammatory cytokines selectin-E and TNF. There is evidence that VEGF may act as a proinflammatory cytokine by increasing E-selectin mRNA in cultured endothelial cells.29 Furthermore, VEGF “primed” endothelial cells, sensitizing them to TNF and leading to heightened selective proinflammatory responses, including upregulation of E-selectin.30 This study showed that rhVEGF treatment led to the upregulation of gene expression of MMP-2 in aortic aneurysm tissues. No differences in gene expression of MMP-9, MT1-MMP, uPA, tPA, E-selectin, or TNF were demonstrable in aortic tissues harvested at day 28 after treatment. This suggested that proteolysis by MMP-2 may be one mechanism of enhanced aneurysm formation in this group of mice and that this was independent of an E-selectininduced or TNF-induced inflammatory response, or that
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
the expression of these other genes occurred at different time points. This study had several limitations. First, the sample size of mice that survived to 28 days was small. As such, the results from this study should be cautiously viewed as preliminary. Second, this study investigated an accentuation of function rather than a loss of function. A loss of function experiment might represent a more authoritative method of proving a causal role for VEGF in aneurysm disease. However, experiments using VEGF-knock-out mice are not possible because targeted inactivation of VEGF genes causes embryonic lethality in mice.31 Another limitation was that tissue VEGF concentrations were not assayed due to technical difficulties and therefore there was no definitive proof of VEGF infiltration into aortic tissue. There are currently no studies that provide data on the kinetics of intraperitoneally administered rhVEGF in apolipoprotein-deficient mice. Asahara et al32 demonstrated that 10-g IP injections of rhVEGF in mice daily for a week increased circulating endothelial progenitor cells and augmented corneal neovascularization. Celletti et al33 demonstrated that even single IP injections of low-dose (2 g/kg) rhVEGF had effects on augmenting neovascularization in the thoracic aorta of hyperlipidemic mice. The results from these previous studies suggested the higher dose of daily IP rhVEGF injections (100 g) as used in the present study was likely to have an effect of augmenting the neovascularization process within the mice aortas. Because all the available mouse aortic tissues were used for molecular analysis with QRT-PCR, this study was not able to analyze the histologic appearance of the aortic tissues for the extent of angiogenesis in rhVEGF-treated mice. Further studies are needed to clarify if rhVEGF administration was associated with increased CD32⫹ cells or microcapillaries and monocyte/macrophage infiltration-2 (or similar). The area of aneurysm wall that responded to the rhVEGF should also be investigated. The protein expression of MMP-2 should also ideally be investigated. CONCLUSIONS This study showed that VEGF enhanced the formation of AngII-induced aneurysms in apoE–/– mice and increased gene expression of MMP-2 within the aortic aneurysm wall. These findings suggest an important role for VEGF in the process of experimental aneurysm development. Further work is needed to determine if anti-VEGF treatment has any effect in retarding aneurysm expansion or preventing aneurysm rupture. We are grateful to the Medical Biomics Centre, St George’s, University of London for their technical support. AUTHOR CONTRIBUTIONS Conception and design: EC, GC, IL, MT Analysis and interpretation: EC, GC, FH, MT Data collection: EC, GC, JD, WW, FH
Choke et al 165
Writing the article: EC Critical revision of the article: EC, GC, JD, FH, WW, IL, MT Final approval of the article: EC, GC, JD, FH, WW, IL, MT Statistical analysis: EC Obtained funding: EC, GC, IL, MT Overall responsibility: MT REFERENCES 1. Choke E, Cockerill G, Wilson WR, Sayed S, Dawson J, Loftus I, et al. A review of biological factors implicated in abdominal aortic aneurysm rupture. Eur J Vasc Endovasc Surg 2005;30:227-44. 2. Thompson MM, Jones L, Nasim A, Sayers RD, Bell PR. Angiogenesis in abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 1996;11:464-9. 3. Kobayashi M, Matsubara J, Matsushita M, Nishikimi N, Sakurai T, Nimura Y. Expression of angiogenesis and angiogenic factors in human aortic vascular disease. J Surg Res 2002;106:239-45. 4. Holmes DR, Liao S, Parks WC, Thompson RW. Medial neovascularization in abdominal aortic aneurysms: a histopathologic marker of aneurysmal degeneration with pathophysiologic implications. J Vasc Surg 1995;21:761-71. 5. Choke E, Thompson MM, Dawson J, Wilson WR, Sayed S, Loftus IM, et al. Abdominal aortic aneurysm rupture is associated with increased medial neovascularization and overexpression of proangiogenic cytokines. Arterioscler Thromb Vasc Biol 2006;26:2077-82. 6. Pepper MS. Role of the matrix metalloproteinase and plasminogen activator-plasmin systems in angiogenesis. Arterioscler Thromb Vasc Biol 2001;21:1104-17. 7. Tedesco MM, Terashima M, Blankenberg FG, Levashova Z, Spin JM, Backer MV, et al. Analysis of in situ and ex vivo vascular endothelial growth factor receptor expression during experimental aortic aneurysm progression. Arterioscler Thromb Vasc Biol 2009;29:1452-7. 8. Johnston KW, Rutherford RB, Tilson MD, Shah DM, Hollier L, Stanley JC. Suggested standards for reporting on arterial aneurysms. Subcommittee on Reporting Standards for Arterial Aneurysms, Ad Hoc Committee on Reporting Standards, Society for Vascular Surgery and North American Chapter, International Society for Cardiovascular Surgery. J Vasc Surg 1991;13:452-8. 9. Wilson WR, Anderton M, Schwalbe EC, Jones JL, Furness PN, Bell PR, et al. Matrix metalloproteinase-8 and -9 are increased at the site of abdominal aortic aneurysm rupture. Circulation 2006;113:438-45. 10. Chaabane L, Soulas EC, Contard F, Salah A, Guerrier D, Briguet A, et al. High-resolution magnetic resonance imaging at 2 Tesla: potential for atherosclerotic lesions exploration in the apolipoprotein E knockout mouse. Invest Radiol 2003;38:532-8. 11. Hockings PD, Roberts T, Galloway GJ, Reid DG, Harris DA, VidgeonHart M, et al. Repeated three-dimensional magnetic resonance imaging of atherosclerosis development in innominate arteries of low-density lipoprotein receptor-knockout mice. Circulation 2002;106:1716-21. 12. Choudhury RP, Aguinaldo JG, Rong JX, Kulak JL, Kulak AR, Reis ED, et al. Atherosclerotic lesions in genetically modified mice quantified in vivo by non-invasive high-resolution magnetic resonance microscopy. Atherosclerosis 2002;162:315-21. 13. Chaabane L, Canet E, Serfaty JM, Contard F, Guerrier D, Douek P, et al. Microimaging of atherosclerotic plaque in animal models. MAGMA 2000;11:58-60. 14. Fayad ZA, Fallon JT, Shinnar M, Wehrli S, Dansky HM, Poon M, et al. Noninvasive In vivo high-resolution magnetic resonance imaging of atherosclerotic lesions in genetically engineered mice. Circulation 1998;98:1541-7. 15. Marjamaa J, Tulamo R, Abo-Ramadan U, Hakovirta H, Frosen J, Rahkonen O, et al. Mice with a deletion in the first intron of the Col1a1 gene develop dissection and rupture of aorta in the absence of aneurysms: high-resolution magnetic resonance imaging, at 4.7 T, of the aorta and cerebral arteries. Magn Reson Med 2006;55:592-7. 16. McFadden EP, Chaabane L, Contard F, Guerrier D, Briguet A, Douek P, et al. In vivo magnetic resonance imaging of large spontaneous aortic
166 Choke et al
17.
18.
19.
20.
21.
22.
23.
24.
25.
aneurysms in old apolipoprotein E-deficient mice. Invest Radiol 2004;39:585-90. Daugherty A, Manning MW, Cassis LA. Angiotensin II promotes atherosclerotic lesions and aneurysms in apolipoprotein E-deficient mice. J Clin Invest 2000;105:1605-12. Saraff K, Babamusta F, Cassis LA, Daugherty A. Aortic dissection precedes formation of aneurysms and atherosclerosis in angiotensin II-infused, apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol 2003;23:1621-6. Barisione C, Charnigo R, Howatt DA, Moorleghen JJ, Rateri DL, Daugherty A. Rapid dilation of the abdominal aorta during infusion of angiotensin II detected by noninvasive high-frequency ultrasonography. J Vasc Surg 2006;44:372-6. Wang Y, Cassis L, Daugherty A. Angiotensin II-induced aortic aneurysms. In: Xu Q, editor. A handbook of mouse models of cardiovascular diseases. New York: John Wiley and Sons Ltd; 2006. p. 125-35. Unemori EN, Ferrara N, Bauer EA, Amento EP. Vascular endothelial growth factor induces interstitial collagenase expression in human endothelial cells. J Cell Physiol 1992;153:557-62. Zucker S, Mirza H, Conner CE, Lorenz AF, Drews MH, Bahou WF, et al. Vascular endothelial growth factor induces tissue factor and matrix metalloproteinase production in endothelial cells: conversion of prothrombin to thrombin results in progelatinase A activation and cell proliferation. Int J Cancer 1998;75:780-6. Chan VT, Zhang DN, Nagaravapu U, Hultquist K, Romero LI, Herron GS. Membrane-type matrix metalloproteinases in human dermal microvascular endothelial cells: expression and morphogenetic correlation. J Invest Dermatol 1998;111:1153-9. Lamoreaux WJ, Fitzgerald ME, Reiner A, Hasty KA, Charles ST. Vascular endothelial growth factor increases release of gelatinase A and decreases release of tissue inhibitor of metalloproteinases by microvascular endothelial cells in vitro. Microvasc Res 1998;55:29-42. Heissig B, Rafii S, Akiyama H, Ohki Y, Sato Y, Rafael T, et al. Low-dose irradiation promotes tissue revascularization through VEGF release from mast cells and MMP-9-mediated progenitor cell mobilization. J Exp Med 2005;202:739-50.
JOURNAL OF VASCULAR SURGERY July 2010
26. Wang H,.Keiser JA. Vascular endothelial growth factor upregulates the expression of matrix metalloproteinases in vascular smooth muscle cells: role of flt-1. Circ Res 1998;83:832-40. 27. Pepper MS, Rosnoblet C, Di Sanza C, Kruithof EK. Synergistic induction of t-PA by vascular endothelial growth factor and basic fibroblast growth factor and localization of t-PA to Weibel-Palade bodies in bovine microvascular endothelial cells. Thromb Haemost 2001;86: 702-9. 28. Sahni A, Sahni SK, Simpson-Haidaris PJ, Francis CW. Fibrinogen binding potentiates FGF-2 but not VEGF induced expression of u-PA, u-PAR, and PAI-1 in endothelial cells. J Thromb Haemost 2004;2: 1629-36. 29. Kim I, Moon SO, Kim SH, Kim HJ, Koh YS, Koh GY. Vascular endothelial growth factor expression of intercellular adhesion molecule 1 (ICAM-1), vascular cell adhesion molecule 1 (VCAM-1), and Eselectin through nuclear factor-kappa B activation in endothelial cells. J Biol Chem 2001;276:7614-20. 30. Stannard AK, Khurana R, Evans IM, Sofra V, Holmes DI, Zachary I. Vascular endothelial growth factor synergistically enhances induction of E-selectin by tumor necrosis factor-alpha. Arterioscler Thromb Vasc Biol 2007;27:494-502. 31. Ferrara N, Carver-Moore K, Chen H, Dowd M, Lu L, O’Shea KS, et al. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 1996;380:439-42. 32. Asahara T, Takahashi T, Masuda H, Kalka C, Chen D, Iwaguro H, et al. VEGF contributes to postnatal neovascularization by mobilizing bone marrow-derived endothelial progenitor cells. EMBO J 1999;18: 3964-72. 33. Celletti FL, Waugh JM, Amabile PG, Brendolan A, Hilfiker PR, Dake MD. Vascular endothelial growth factor enhances atherosclerotic plaque progression. Nat Med 2001;7:425-9. Submitted Sep 21, 2009; accepted Feb 6, 2010.
Additional material for this article may be found online at www.jvascsurg.org.
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
METHODS Magnetic resonance imaging protocol. Mice underwent magnetic resonance imaging (MRI) scans at day 21 and day 28 (before euthanasia). MRI data were obtained using a Varian Unity INOVA imaging system fitted with a 4.7-T horizontal bore magnet. Mice were positioned supine within a 63-mm-diameter quadrature imaging coil into which heated air was blown to maintain normal body temperature. Coronal scout images were acquired to allow reproducible positioning of axial images aligned with the central slice at the superior end of the left kidney (Fig 1). Because of the ferromagnetic core of the infusion pump spin echo (rather than gradient echo), images were acquired with strong slice selection and readout gradients to minimize image distortions. Nine axial T1-weighted spinecho cross-sectional images were acquired with 2-mmthick contiguous slices with repetition time of 400 milliseconds, echo time of 8.5 milliseconds, and 16 averages. Outer slice 12-mm-thick saturation bands were used to suppress the high-intensity blood signal of the lumen. In-plane image resolution was 512 ⫻ 256 over a 4-cm field of view and zero-filled to 1024 ⫻ 1024 for region-of-interest (ROI) analysis of abdominal aorta. Using the Varian ImageBrowser software and delineation of the ROI, an operator (F. H.) who was blinded to the experimental arm of the mice determined abdominal aorta diameter (mm) and cross-sectional area (mm2) in each cross-section. Real-time quantitative reverse-transcription polymerase chain reaction. Qualitative reverse-transcription polymerase chain reaction (QRT-PCR) mixtures were es-
Choke et al 166.e1
tablished according to manufacturer’s recommendations (Assays-on-Demand Gene Expression Products and TaqMan Universal PCR Master Mix, Applied Biosystems, Warrington, UK). Each sample was run in triplicates, and for all reactions, negative controls were run with no template present. In addition, total RNA from mice samples was used for sham reverse-transcription reactions with no reverse transcriptase present and then subjected to standard PCR using 18S ribosomal RNA (rRNA) primers to verify that no amplification was produced. QRT-PCR was done using an Mx4000 Multiplex Quantitative PCR System (Stratagene, UK). Each PCR cycle started with an initial 10-minute denaturation step at 95°C, followed by 40 cycles of shuttle heating at 95°C for 15 seconds, and at 60°C for 1 minute. The associated Mx4000 software was used to analyze the data and determine the threshold count (Ct). The specimens were normalized to the endogenous reference control 18S rRNA. For each sample, Cttarget gene and Ct18S rRNA were determined, and ⌬Ct ⫽ Cttarget gene ⫺ Ct18S rRNA. The relative level of target gene normalized to 18S rRNA was determined by calculating: 2–⌬Ct. For the 2–⌬Ct calculation to be valid, preliminary validation experiments were performed to verify that the efficiencies of target gene amplification and the efficiency of 18S rRNA amplification were approximately equal. To achieve this, standard curves of template dilution and Ct were obtained for each target gene and 18S rRNA to confirm comparative rates of transcriptional efficiency for each probe set.
Fig 4, online only. Magnetic resonance imaging was used to compare the ex vivo aortic maximum diameter measurement with the in vivo measurements at day 28. The two measurements were highly correlated with each other (Pearson r2 ⫽ 0.88).
Abdominal aortic aneurysm (AAA) rupture is the 13th most common cause of death in the Western world,1 and the only effective approach to preventing aneurysm rupture remains surgical repair by conventional or endovascular means. Although collective endeavors in basic science research during the past 2 decades have led to progress in the understanding of aneurysm development and expansion,
From St. George’s Vascular Institute. This work was supported by a research fellowship from the Royal College of Surgeons of England and grants from Peel Medical Research, Society of Academic and Research Surgery, UK, and EU-FP7-HEALTH Fighting Aneurysmal Disease 200467. Competition of interest: none. Reprint requests: Professor Matt Thompson, St. George’s Vascular Institute, St. George’s Hospital, Blackshaw Rd, London SW17 0RE, UK (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a competition of interest. 0741-5214/$36.00 Copyright © 2010 by the Society for Vascular Surgery. doi:10.1016/j.jvs.2010.02.015
the process is not yet fully understood, and there is no effective medical therapy against aneurysms. The association between angiogenesis and human aneurysms has been recognized by previous investigators2-4 who suggested that angiogenesis might contribute toward the development of aneurysms. The localization of increased angiogenesis and increased expression of proangiogenic cytokines to the site of aneurysm rupture in humans5 implied a further role for angiogenesis in end-stage aneurysm disease and aneurysm rupture. Extracellular proteolysis is an absolute requisite for angiogenesis, primarily contributing toward proteolytic degradation of the vascular basement membrane and remodeling of the extracellular matrix to allow endothelial cells to migrate and invade into the surrounding tissue.6 It is this link between proangiogenic cytokines and induction of extracellular proteolysis that makes these cytokines potentially relevant in aneurysm disease. Recently, Tedesco et al7 demonstrated that the mural expression of vascular endothelial growth factor (VEGF) receptor (VEGFR), as determined by fluorescent imaging and VEGFR-2 immu159
JOURNAL OF VASCULAR SURGERY July 2010
160 Choke et al
Day 1
Day 28
Saline infusion (sham control) 28 day saline infusion
Daily IP saline injection Day 8
28 day AngII infusion
Daily IP saline injection Day 8
Angiotensin II group (positive control)
AngII+14dVEGF short duration group 28 day AngII infusion
Daily IP 100µg VEGF injection Day 15
AngII+21dVEGF long duration group 28 day AngII infusion
Daily IP 100µg VEGF injection Day 8
Fig 1. Schematic diagram of study design. AngII, Angiotensin II; IP, intraperitoneal; VEGF, vascular endothelial growth factor.
nostaining, was increased in a diameter-dependent fashion in experimental aneurysms. These authors also showed that inhibition of angiogenesis resulted in the attenuation of experimental aneurysms. To date, the effects of proangiogenic mediators in experimental models of aneurysm have never been tested, and whether the observed overexpression of proangiogenic cytokines at the site of aneurysm rupture is a cause or effect of end-stage aneurysmal disease remains unknown. The present study investigated the effects of recombinant human (rh) VEGF on angiotensin (Ang)-II–induced aneurysm in apolipoprotein (Apo) E–/– mice. The results indicated that rhVEGF significantly enhanced AAA formation and upregulated gene expression of matrix metalloproteinase (MMP)-2 in the aortic wall. METHODS Experimental animals. The Home Office Science and Research Group (Animals Scientific Procedures Division) approved all animal protocols. Six-month-old apoE–/– male mice (Jackson Laboratories, Bar Harbor, Me) underwent a minimum 1-month acclimatization period. Water and regular mouse chow was available 24 hours daily. Mice were assigned to one of four groups: (1) infusion of normal saline (sham), (2) infusion of AngII (1000 ng/kg/min, Sigma-Aldrich Corp, St. Louis, Mo), (3) infusion of AngII plus 14 days daily intraperitoneal (IP) injections of 100 g rhVEGF (AngII⫹14dVEGF, short-duration group), or (4) infusion of AngII plus 21 days daily IP injections of 100 g rhVEGF (AngII⫹21dVEGF, long-duration group; Fig 1). The short-duration group commenced daily rhVEGF injections (Genentech Inc, South San Francisco, Calif) from day 15, and the long-duration group commenced daily rhVEGF injections from day 8. Mice from the sham group and AngII-only group received daily IP saline injections from day 8. All mice were sacrificed on day 28.
Fig 2. Left, Representative magnetic resonance image coronal section with (Right) photograph of corresponding arterial tree showing suprarenal abdominal aortic aneurysm. In each mouse, nine axial T1-weighted spin-echo cross-sectional images (green lines) were acquired with 2-mm-thick contiguous slices. To ensure reproducible positioning of axial images, the central slice (5th) was aligned at the superior pole of the left kidney.
Implantation of miniosmotic pumps. Alzet model 2004 mini-osmotic pumps (Durect Corp, Cupertino, Calif) were loaded with individual concentrations of AngII (variable according to individual animal weight) for 28 days of continuous delivery of 1000 ng/kg/min of AngII. These were incubated in normal saline at 37°C for 24 hours before subcutaneous implantation at the interscapular area. The miniosmotic pumps implanted in the sham group were loaded with normal saline. Magnetic resonance imaging protocol. Mice underwent magnetic resonance imaging (MRI) on days 21 and 28 (before sacrifice). MRI data were obtained using a Varian Unity INOVA imaging system fitted with a 4.7-T horizontal bore magnet. Coronal scout images were acquired to allow reproducible positioning of axial images aligned with the central slice at the superior end of the left kidney (Fig 2). In each mouse, nine axial T1-weighted spinecho cross-sectional images were acquired with 2-mm-thick contiguous slices. Using the Varian ImageBrowser (Varian Medical Systems, Palo Alto, Calif) software and delineation of regions of interest, an operator (F. H.) who was blinded to the experimental arm of the mice determined the abdominal aorta diameter (mm) and cross-sectional area (mm2) in each cross-section were determined. Details of the experimental protocol are presented in the Supplement (online only).
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
Choke et al 161
Table I. Weight and mortality dataa Variableb Total number Weight, g Pretreatment 4th week Survived to 28 days Spontaneous death Rupture Nonrupture Schedule 1 sacrifice Paraplegia General lethargy Total aneurysmsd
Saline
AngII
AngII⫹14dVEGF
AngII⫹21dVEGF
5
11
9
10
32.5 ⫾ 0.8 35.3 ⫾ 0.9 4 (80)
34.0 ⫾ 0.9 37.1 ⫾ 1.0 5 (45)
34.8 ⫾ 1.1 36.0 ⫾ 1.5 3 (33)
35.0 ⫾ 0.7 35.5 ⫾ 1.2 4 (40)
0 1
2 (18) 2 (18)c
2 (22) 1 (11)
1 (10) 2 (20)
0 0 0 (0)
2 (18)c 0 9 (81)
2 (22)c 1 (11) 8 (89)
2 (20)c 1 (10) 9 (90)
AngII, Angiotensin II; VEGF, vascular endothelial growth factor. a P values for all variables were not significant. b Data are presented as mean ⫾ standard error of the mean, or No. (%). c One mouse in each of these groups did not develop an aneurysm. d Includes mice that died ⬍28 days.
Mice aortic tissue collection. At the end of the 28day treatment period, mice were sacrificed and blood was drawn from the right ventricle. To maximize RNA stability, aortas were perfused with RNALater (Applied Biosystems, Warrington UK) through the heart, and the entire arterial trees were rinsed with RNALater while in situ. A dissection microscope (Olympus, Tokyo, Japan) was used to carefully dissect the arterial tree from fat and connective tissue and was photographed. Ex vivo measurements of diameter were taken with calibrated calipers through the entire length of the aorta to determine the region of maximum dilatation. Samples for molecular analysis were harvested from aneurysmal segments of the aorta and placed in RNALater at –20°C until extraction of total RNA. As previously described,5 total RNA was extracted using RNeasy Fibrous Tissue RNA extraction kit (Qiagen, West Sussex, UK) and checked for integrity and concentration using RNA 6000 Nano LabChip Kit and Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, Calif). Quantitative real-time reverse-transcription polymerase chain reaction. Complimentary DNA (cDNA) was synthesized from total RNA by reverse-transcription using a high-capacity cDNA archive kit (Applied Biosystems). Qualitative reverse-transcription polymerase chain reaction (QRT-PCR) was done with the Mx4000 Multiplex Quantitative PCR System (Stratagene, Edinburgh, UK) using TaqMan Gene Expression Assays (Assays-onDemand Gene Expression Products, Applied Biosystems, UK) and TaqMan Universal PCR Master Mix. Transcript levels were normalized to 18s rRNA (internal control). The expression of genes encoding for the proinflammatory cytokines E-selectin and tumor necrosis factor (TNF), the proteolytic enzymes from the MMP family (membrane type-1-MMP [MT1-MMP] and MMPs ⫺2 and ⫺9) and the plasmin activator system (urokinase plasminogen activator [uPA] and tissue plasminogen activator [tPA]) was determined. Details of the experimental protocol are presented in the Supplement (online only).
Statistics. All data were reported as mean ⫾ standard error of the mean. Multiple comparisons of mean values of aortic aneurysm diameters, cross-sectional areas, and gene expression were performed by one-way analysis of variance, followed by a subsequent Newman-Keuls multiple comparison test. Differences were considered statistically significant at a value of P ⬍ .05. RESULTS Weight and mortality data. No significant differences were observed for weight at the beginning and completion of the study among all experimental groups of mice (Table I). Necropsy was performed ⱕ24 hours on animals that died before completion of study. These animals were not included in AAA quantification analysis but only in the mortality data. The mortality rate was increased in the groups that received AngII infusions compared with the sham group, although this did not reach statistical significance (Table I). Aortic ruptures causing death occurred in AngII (n ⫽ 2), AngII⫹14dVEGF (n ⫽ 2), and AngII⫹21dVEGF (n ⫽ 1) groups. All aortic ruptures occurred between days 6 and 14. Paraplegia or general lethargy necessitated schedule 1 euthanasia before 28 days of infusion had been completed in AngII (n ⫽ 2), AngII⫹14dVEGF (n ⫽ 3), and AngII⫹ 21dVEGF (n ⫽ 3; P ⫽ .48). In the placebo control group, one mouse died on day 18 after insertion of the osmotic pump, and necropsy revealed normal aorta. Including mice that died before 28 days, the incidence of aneurysms was 0% (saline), 81% (AngII), 89% (AngII⫹ 14dVEGF), and 90% (AngII⫹21dVEGF). The aneurysms in mice that died ⬍28 days were excluded from further analysis because their size would have been different had they been allowed to develop to 28 days. Effect of rhVEGF on maximum diameter and crosssectional areas. Treatment with rhVEGF increased the maximum diameter and maximum cross-sectional area of
JOURNAL OF VASCULAR SURGERY July 2010
162 Choke et al
A
Saline
Ang II
Ang II plus 14 days VEGF
Ang II plus 21 days VEGF
Maximal diameter (mm) day 21
Maximal diameter (mm) day 28 #
3.0
§
3.0
2.5
2.5
2.0
2.0
1.5
1.5
1.0
†
††
1.0
Saline
AngII
AngII+14dVEGF
AngII+21dVEGF
Saline
AngII+14dVEGF
AngII+21dVEGF
C
B Maximal cross sectional area (mm2) day 21 7
Maximal cross sectional area (mm2) day 28 ‡
‡‡
7
6
6
5
5
4
4
3
3
2
2
1
1
*
**
0
0
Saline
D
AngII
AngII
AngII+14dVEGF AngII+21dVEGF
Saline
AngII
AngII+14dVEGF AngII+21dVEGF
E
Fig 3. Angiotensin (Ang) II-induced abdominal aortic aneurysm (AAA) size was significantly increased at days 21 and 28 in apolipoprotein E–/– mice treated with vascular endothelial growth factor (VEGF) for 14 (AngII⫹14dVEGF) or 21 (AngII⫹14dVEGF) days. A, Representative macroscopic photographs of the AAA are shown with corresponding cross-sectional magnetic resonance imaging scans at day 28. The maximum diameter of the aorta is outlined by the red line. Box and whisker plots show the AAA maximum diameter at (B) day 21 and (C) day 28 and maximum cross-sectional area at (D) day 21 and (E) day 28 in saline-infused (n ⫽ 4), AngII-infused (n ⫽ 5), Ang II⫹14dVEGF-infused (n ⫽ 3), and AngII⫹21dVEGF-infused (n ⫽ 4) mice. #P ⬍ .05 vs AngII infusion; §P ⬍ .05 vs Ang II infusion; †P ⬍ .05 vs AngII infusion; ††P ⬍ .05 vs Ang II infusion; ‡P ⬍ .01 vs AngII infusion; ‡‡P ⬍ .05 vs Ang II infusion; *P ⬍ .05 vs Ang II infusion; **P ⬍ .05 vs Ang II-infusion. The horizontal line in the middle of each box indicates the median; the top and bottom borders of the box mark the 75th and 25th percentiles, respectively, and the whiskers mark the 90th and 10th percentiles.
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
Choke et al 163
Table II. Summary of gene expressions (arbitrary units normalized to 18S recombinant RNA) Gene
Saline Mean ⫾ SEM
AngII Mean ⫾ SEM
AngII⫹VEGF (pooled) Mean ⫾ SEM
P valuea
MMP-2 MMP-9 MT1-MMP uPA tPA E-selectin TNF
2.20 ⫾ 0.26 4.96 ⫾ 0.16 6.42 ⫾ 0.15 6.00 ⫾ 0.31 5.65 ⫾ 0.16 5.20 ⫾ 0.25 4.71 ⫾ 0.32
1.76 ⫾ 0.22 5.09 ⫾ 0.46 6.79 ⫾ 0.32 5.55 ⫾ 0.27 5.40 ⫾ 0.17 4.75 ⫾ 0.21 4.19 ⫾ 0.26
4.31 ⫾ 0.13 5.14 ⫾ 0.36 7.00 ⫾ 0.06 6.03 ⫾ 0.09 5.60 ⫾ 0.07 5.04 ⫾ 0.07 4.80 ⫾ 0.21
.0009b .94 .40 .46 .52 .39 .36
AngII, Angiotensin II; MMP, matrix metalloproteinase; MT1-MMP, membrane type-1-MMP; SEM, standard error of the mean; TNF, tumor necrosis factor; tPA, tissue-type plasminogen activator; uPA, urokinase-type plasminogen activator; VEGF, vascular endothelial growth factor. a Comparison used one-way analysis of variance. b Significant.
AngII-induced aneurysms (Fig 3). MRI measurements at day 21 demonstrated a 52% increase in maximum aortic diameter (P ⬍ .05) and a 100% increase in maximum cross-sectional area (P ⬍ .05) in the short-duration group (AngII⫹14dVEGF) compared with AngII mice. Similarly, a 38% increase in maximum aortic diameter (P ⬍ .05) and a 68% increase in the maximum cross-sectional area (P ⬍ .05) occurred in the long-duration group (AngII⫹ 21dVEGF). The difference in the maximum aortic diameter or cross-sectional area between short-duration and longduration VEGF groups was not statistically significant. MRI measurements at day 28 demonstrated a 36% increase in maximum aortic diameter (P ⬍ .05) and a 76% increase in maximum cross-sectional area (P ⬍ .05) in the short-duration group (AngII⫹14dVEGF) compared with AngII mice. A 25% increase in maximum aortic diameter (P ⬍ .05) and a 52% increase in maximum cross-sectional area (P ⬍ .05) occurred in the long-duration group (AngII⫹ 21dVEGF). At day 28, no statistically significant difference was evident in maximum diameter or maximum crosssectional area between short- and long-duration VEGF groups. After mice were sacrificed at day 28, ex vivo measurements demonstrated that maximum aortic diameter was increased by 67% (P ⬍ .01) in the short-duration group (AngII⫹14dVEGF) and by 51% (P ⬍ .01) in the longduration group (AngII⫹21dVEGF) compared with AngII mice. A comparison of ex vivo aortic maximum diameter measurements with in vivo measurements by MRI at day 28 (Fig 4, online only) demonstrated that these two measurements were highly correlated with each other (Pearson r2 ⫽ 0.88). Aneurysm formation in rhVEGF-treated mice. Aneurysms formed earlier in rhVEGF-treated mice. Aortic aneurysm was defined as ⬎50% enlargement of maximum abdominal aortic diameter, in line with current clinical definitions.8 The average diameter of the suprarenal aorta in sham mice at 3 weeks was 1.23 mm and a threshold of 1.85 mm was therefore set as the minimum diameter for incidence of aneurysm formation. Aneurysms had already developed at day 21 in all mice in the AngII⫹14dVEGF and AngII⫹21dVEGF groups compared with only two (40%) of the AngII mice.
At day 28, the average diameter of the suprarenal aorta in the control mice was very similar (1.25 mm), and a threshold of 1.88 mm was therefore set as the minimum diameter for incidence of aneurysm formation. At this time point, aneurysm formation was evident in 80% of AngII mice. From days 21 to 28, there was an increase in maximum aortic diameter (14%) and maximum cross-sectional area (25%; P ⬍ .05) in mice from the AngII group. In the rhVEGF-treated groups, no statistically significant changes to diameter or cross-sectional area were evident between days 21 and 28. These data suggested that aneurysms formed more quickly in mice exposed to a proangiogenic stimulus. Effect of VEGF on MMP-2 gene expression in the aortic wall. Gene expression of MMP-2 was significantly increased (P ⫽ .0009) in aneurysmal sections of the aorta in the pooled rhVEGF-treated group compared with aneurysmal sections of the aorta in the AngII group or normal aorta in the saline group (Table II). There was no difference in gene expression of MMP-9, MT1-MMP, uPA, tPA, Eselectin, or TNF. DISCUSSION This study investigated the functional significance of increasing the exposure to the proangiogenic VEGF in experimental aneurysms. VEGF was selected for investigation because it is a potent inducer of angiogenesis and has consistently been shown to be a potentially important participant in aneurysm development3 and rupture.5 Previous work has shown VEGF is increased in aneurysmal tissues.3 The present study demonstrated that exogenous treatment with rhVEGF led to earlier development and increased maximum expansion of AngII-induced aneurysms. This enhancement of aneurysms after rhVEGF treatment was independent of the duration of rhVEGF treatment. In addition to VEGF, an increase in other proangiogenic molecules, such as ␣v-integrin, VE-cadherin, monocyte chemoattractant protein-1, and vimentin, has also been observed at the site of AAA rupture in humans.5,9 However, data from the present study demonstrated that
JOURNAL OF VASCULAR SURGERY July 2010
164 Choke et al
rhVEGF treatment did not lead to an increased rate of aortic rupture within the 28-day experimental period. MRI has previously been validated as a robust noninvasive imaging technique for detecting in vivo phenotypic abnormalities, including the characterization of atherosclerotic plaque,10-14 diagnosis, and measurements of arterial remodeling during aortic lumen enlargement in apoE–/– mice.15,16 In the present study, MRI using a Varian Unity INOVA imaging system fitted with a 4.7-T horizontal bore magnet was used for in vivo monitoring, detecting, and measuring maximum diameter and cross-sectional area of AngII-induced aneurysms in apoE–/– mice. This work endorsed the potential of MRI to obtain accurate vessel dimensions in diseased abdominal aorta, verified against macroscopic measurements, and to monitor disease progression by serial imaging at different time points. No current studies have clearly demonstrated an effect of age on the development of AngII-induced aneurysms. The use of younger mice is limited by the practical constraints of their smaller size and therefore difficulties in implantation of the Alzet miniosmotic pumps. Effective drug delivery required the body weight of mice to be at least 20 grams. We used 6-month-old mice in this study according to the methods described in the first published article on AngII-mediated aneurysms.17 Mice ranging in age from 3 to 11 months at the initiation of AngII infusion have been used, and the described incidence of AngIIinduced aneurysms has been fairly consistent, at 70% to 90%, in hyperlipidemic mice. The incidence of aneurysms in the AngII group (without VEGF) was 80% at 28 days, and this is consistent with that of other published studies. On the other hand, the use of older mice might have resulted in the lower-than-anticipated overall survival to 28 days. The rates of aortic rupture (all ⬍day 14) were 18% (AngII), 22% (AngII⫹14dVEGF), and 10% (AngII⫹ 21dVEGF). These results were consistent with rates of 10%18 to 40%19 published in previous reports. However, the overall survival was low due to other causes such as paraplegia and general lethargy that necessitated schedule I sacrifice. Paraplegia is a recognized complication of AngII infusion.20 Saraff et al18 demonstrated that neovascularization within AngII-induced aneurysms was a late-stage response that occurred during maturation of aneurysmal tissues between days 14 and 28. In this study, rhVEGF IP injection was started at day 15 (short-duration group) to coincide with this response to investigate the effects of proangiogenic cytokine on the angiogenic process. IP rhVEGF injection was also started at day 8 (long-duration group) to investigate the effects of introducing a proangiogenic cytokine before any “naturally occurring” neovascularization process within the developing aneurysms. The results indicated that the effects of rhVEGF were not different, whether it was administered before (day 8) or during (day 15) the neovascularization process. This suggested that there was a dose-saturating effect, in which 14 days of rhVEGF treatment resulted in maximum enhancement of the aneurysm, with no further increases demon-
strable with an earlier additional 7 days of rhVEGF. It is interesting to note that aneurysms in the short-duration group were larger than aneurysms that had longer exposure to rhVEGF, but this result did not reach statistical significance. Tedesco et al7 recently reported that mural VEGFR expression was increased in AngII-induced aneurysms in a diameter-dependent fashion. Because in vivo measurements of VEGFR expression using near-infrared fluorescent imaging is not yet possible due to intrinsic autofluorescence, there are currently no data on the temporal relationship between VEGFR expression and the development of aneurysms. The time course of VEGFR expression may be a possible explanation for the larger (although not statistically significant) size of aneurysms in the short-duration VEGF group compared to the long-duration group. The regulation of MMP and serine protease expression by VEGF is well described. The regulation of MMPs and tissue inhibitor of matrix metalloproteinases (TIMPs) in endothelial cells by VEGF is variable and cell-type dependent:6 ●
●
● ● ●
In human umbilical vein endothelial cells, VEGF increased MMP-1, MMP-2, and TIMP-1, and this change was accompanied by increased activation of MMP-2.21,22 In human dermal microvascular endothelial cells, VEGF increased MMP-2, MMP-9, and MT-1 MMP, had no effect on MMP-1, and decreased TIMP-1 and TIMP-2.23,24 Circulating VEGF induced MMP-9 expression in bone marrow.25 VEGF treatment enhanced production of MMPs-1, -3, and -9 by human smooth muscle cells.26 In bovine microvascular endothelial cells27 and human umbilical vein endothelial cells,28 VEGF increased tPA and uPA.
The messenger RNA (mRNA) levels of MMPs-2 and MMPs-9, MT1-MMP, tPA, and uPA were analyzed in this study to investigate if VEGF treatment had any effect on the expression of these VEGF downstream target genes in the aortic tissue of the mice. In addition, this study analyzed the gene expression of the proinflammatory cytokines selectin-E and TNF. There is evidence that VEGF may act as a proinflammatory cytokine by increasing E-selectin mRNA in cultured endothelial cells.29 Furthermore, VEGF “primed” endothelial cells, sensitizing them to TNF and leading to heightened selective proinflammatory responses, including upregulation of E-selectin.30 This study showed that rhVEGF treatment led to the upregulation of gene expression of MMP-2 in aortic aneurysm tissues. No differences in gene expression of MMP-9, MT1-MMP, uPA, tPA, E-selectin, or TNF were demonstrable in aortic tissues harvested at day 28 after treatment. This suggested that proteolysis by MMP-2 may be one mechanism of enhanced aneurysm formation in this group of mice and that this was independent of an E-selectininduced or TNF-induced inflammatory response, or that
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
the expression of these other genes occurred at different time points. This study had several limitations. First, the sample size of mice that survived to 28 days was small. As such, the results from this study should be cautiously viewed as preliminary. Second, this study investigated an accentuation of function rather than a loss of function. A loss of function experiment might represent a more authoritative method of proving a causal role for VEGF in aneurysm disease. However, experiments using VEGF-knock-out mice are not possible because targeted inactivation of VEGF genes causes embryonic lethality in mice.31 Another limitation was that tissue VEGF concentrations were not assayed due to technical difficulties and therefore there was no definitive proof of VEGF infiltration into aortic tissue. There are currently no studies that provide data on the kinetics of intraperitoneally administered rhVEGF in apolipoprotein-deficient mice. Asahara et al32 demonstrated that 10-g IP injections of rhVEGF in mice daily for a week increased circulating endothelial progenitor cells and augmented corneal neovascularization. Celletti et al33 demonstrated that even single IP injections of low-dose (2 g/kg) rhVEGF had effects on augmenting neovascularization in the thoracic aorta of hyperlipidemic mice. The results from these previous studies suggested the higher dose of daily IP rhVEGF injections (100 g) as used in the present study was likely to have an effect of augmenting the neovascularization process within the mice aortas. Because all the available mouse aortic tissues were used for molecular analysis with QRT-PCR, this study was not able to analyze the histologic appearance of the aortic tissues for the extent of angiogenesis in rhVEGF-treated mice. Further studies are needed to clarify if rhVEGF administration was associated with increased CD32⫹ cells or microcapillaries and monocyte/macrophage infiltration-2 (or similar). The area of aneurysm wall that responded to the rhVEGF should also be investigated. The protein expression of MMP-2 should also ideally be investigated. CONCLUSIONS This study showed that VEGF enhanced the formation of AngII-induced aneurysms in apoE–/– mice and increased gene expression of MMP-2 within the aortic aneurysm wall. These findings suggest an important role for VEGF in the process of experimental aneurysm development. Further work is needed to determine if anti-VEGF treatment has any effect in retarding aneurysm expansion or preventing aneurysm rupture. We are grateful to the Medical Biomics Centre, St George’s, University of London for their technical support. AUTHOR CONTRIBUTIONS Conception and design: EC, GC, IL, MT Analysis and interpretation: EC, GC, FH, MT Data collection: EC, GC, JD, WW, FH
Choke et al 165
Writing the article: EC Critical revision of the article: EC, GC, JD, FH, WW, IL, MT Final approval of the article: EC, GC, JD, FH, WW, IL, MT Statistical analysis: EC Obtained funding: EC, GC, IL, MT Overall responsibility: MT REFERENCES 1. Choke E, Cockerill G, Wilson WR, Sayed S, Dawson J, Loftus I, et al. A review of biological factors implicated in abdominal aortic aneurysm rupture. Eur J Vasc Endovasc Surg 2005;30:227-44. 2. Thompson MM, Jones L, Nasim A, Sayers RD, Bell PR. Angiogenesis in abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 1996;11:464-9. 3. Kobayashi M, Matsubara J, Matsushita M, Nishikimi N, Sakurai T, Nimura Y. Expression of angiogenesis and angiogenic factors in human aortic vascular disease. J Surg Res 2002;106:239-45. 4. Holmes DR, Liao S, Parks WC, Thompson RW. Medial neovascularization in abdominal aortic aneurysms: a histopathologic marker of aneurysmal degeneration with pathophysiologic implications. J Vasc Surg 1995;21:761-71. 5. Choke E, Thompson MM, Dawson J, Wilson WR, Sayed S, Loftus IM, et al. Abdominal aortic aneurysm rupture is associated with increased medial neovascularization and overexpression of proangiogenic cytokines. Arterioscler Thromb Vasc Biol 2006;26:2077-82. 6. Pepper MS. Role of the matrix metalloproteinase and plasminogen activator-plasmin systems in angiogenesis. Arterioscler Thromb Vasc Biol 2001;21:1104-17. 7. Tedesco MM, Terashima M, Blankenberg FG, Levashova Z, Spin JM, Backer MV, et al. Analysis of in situ and ex vivo vascular endothelial growth factor receptor expression during experimental aortic aneurysm progression. Arterioscler Thromb Vasc Biol 2009;29:1452-7. 8. Johnston KW, Rutherford RB, Tilson MD, Shah DM, Hollier L, Stanley JC. Suggested standards for reporting on arterial aneurysms. Subcommittee on Reporting Standards for Arterial Aneurysms, Ad Hoc Committee on Reporting Standards, Society for Vascular Surgery and North American Chapter, International Society for Cardiovascular Surgery. J Vasc Surg 1991;13:452-8. 9. Wilson WR, Anderton M, Schwalbe EC, Jones JL, Furness PN, Bell PR, et al. Matrix metalloproteinase-8 and -9 are increased at the site of abdominal aortic aneurysm rupture. Circulation 2006;113:438-45. 10. Chaabane L, Soulas EC, Contard F, Salah A, Guerrier D, Briguet A, et al. High-resolution magnetic resonance imaging at 2 Tesla: potential for atherosclerotic lesions exploration in the apolipoprotein E knockout mouse. Invest Radiol 2003;38:532-8. 11. Hockings PD, Roberts T, Galloway GJ, Reid DG, Harris DA, VidgeonHart M, et al. Repeated three-dimensional magnetic resonance imaging of atherosclerosis development in innominate arteries of low-density lipoprotein receptor-knockout mice. Circulation 2002;106:1716-21. 12. Choudhury RP, Aguinaldo JG, Rong JX, Kulak JL, Kulak AR, Reis ED, et al. Atherosclerotic lesions in genetically modified mice quantified in vivo by non-invasive high-resolution magnetic resonance microscopy. Atherosclerosis 2002;162:315-21. 13. Chaabane L, Canet E, Serfaty JM, Contard F, Guerrier D, Douek P, et al. Microimaging of atherosclerotic plaque in animal models. MAGMA 2000;11:58-60. 14. Fayad ZA, Fallon JT, Shinnar M, Wehrli S, Dansky HM, Poon M, et al. Noninvasive In vivo high-resolution magnetic resonance imaging of atherosclerotic lesions in genetically engineered mice. Circulation 1998;98:1541-7. 15. Marjamaa J, Tulamo R, Abo-Ramadan U, Hakovirta H, Frosen J, Rahkonen O, et al. Mice with a deletion in the first intron of the Col1a1 gene develop dissection and rupture of aorta in the absence of aneurysms: high-resolution magnetic resonance imaging, at 4.7 T, of the aorta and cerebral arteries. Magn Reson Med 2006;55:592-7. 16. McFadden EP, Chaabane L, Contard F, Guerrier D, Briguet A, Douek P, et al. In vivo magnetic resonance imaging of large spontaneous aortic
166 Choke et al
17.
18.
19.
20.
21.
22.
23.
24.
25.
aneurysms in old apolipoprotein E-deficient mice. Invest Radiol 2004;39:585-90. Daugherty A, Manning MW, Cassis LA. Angiotensin II promotes atherosclerotic lesions and aneurysms in apolipoprotein E-deficient mice. J Clin Invest 2000;105:1605-12. Saraff K, Babamusta F, Cassis LA, Daugherty A. Aortic dissection precedes formation of aneurysms and atherosclerosis in angiotensin II-infused, apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol 2003;23:1621-6. Barisione C, Charnigo R, Howatt DA, Moorleghen JJ, Rateri DL, Daugherty A. Rapid dilation of the abdominal aorta during infusion of angiotensin II detected by noninvasive high-frequency ultrasonography. J Vasc Surg 2006;44:372-6. Wang Y, Cassis L, Daugherty A. Angiotensin II-induced aortic aneurysms. In: Xu Q, editor. A handbook of mouse models of cardiovascular diseases. New York: John Wiley and Sons Ltd; 2006. p. 125-35. Unemori EN, Ferrara N, Bauer EA, Amento EP. Vascular endothelial growth factor induces interstitial collagenase expression in human endothelial cells. J Cell Physiol 1992;153:557-62. Zucker S, Mirza H, Conner CE, Lorenz AF, Drews MH, Bahou WF, et al. Vascular endothelial growth factor induces tissue factor and matrix metalloproteinase production in endothelial cells: conversion of prothrombin to thrombin results in progelatinase A activation and cell proliferation. Int J Cancer 1998;75:780-6. Chan VT, Zhang DN, Nagaravapu U, Hultquist K, Romero LI, Herron GS. Membrane-type matrix metalloproteinases in human dermal microvascular endothelial cells: expression and morphogenetic correlation. J Invest Dermatol 1998;111:1153-9. Lamoreaux WJ, Fitzgerald ME, Reiner A, Hasty KA, Charles ST. Vascular endothelial growth factor increases release of gelatinase A and decreases release of tissue inhibitor of metalloproteinases by microvascular endothelial cells in vitro. Microvasc Res 1998;55:29-42. Heissig B, Rafii S, Akiyama H, Ohki Y, Sato Y, Rafael T, et al. Low-dose irradiation promotes tissue revascularization through VEGF release from mast cells and MMP-9-mediated progenitor cell mobilization. J Exp Med 2005;202:739-50.
JOURNAL OF VASCULAR SURGERY July 2010
26. Wang H,.Keiser JA. Vascular endothelial growth factor upregulates the expression of matrix metalloproteinases in vascular smooth muscle cells: role of flt-1. Circ Res 1998;83:832-40. 27. Pepper MS, Rosnoblet C, Di Sanza C, Kruithof EK. Synergistic induction of t-PA by vascular endothelial growth factor and basic fibroblast growth factor and localization of t-PA to Weibel-Palade bodies in bovine microvascular endothelial cells. Thromb Haemost 2001;86: 702-9. 28. Sahni A, Sahni SK, Simpson-Haidaris PJ, Francis CW. Fibrinogen binding potentiates FGF-2 but not VEGF induced expression of u-PA, u-PAR, and PAI-1 in endothelial cells. J Thromb Haemost 2004;2: 1629-36. 29. Kim I, Moon SO, Kim SH, Kim HJ, Koh YS, Koh GY. Vascular endothelial growth factor expression of intercellular adhesion molecule 1 (ICAM-1), vascular cell adhesion molecule 1 (VCAM-1), and Eselectin through nuclear factor-kappa B activation in endothelial cells. J Biol Chem 2001;276:7614-20. 30. Stannard AK, Khurana R, Evans IM, Sofra V, Holmes DI, Zachary I. Vascular endothelial growth factor synergistically enhances induction of E-selectin by tumor necrosis factor-alpha. Arterioscler Thromb Vasc Biol 2007;27:494-502. 31. Ferrara N, Carver-Moore K, Chen H, Dowd M, Lu L, O’Shea KS, et al. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 1996;380:439-42. 32. Asahara T, Takahashi T, Masuda H, Kalka C, Chen D, Iwaguro H, et al. VEGF contributes to postnatal neovascularization by mobilizing bone marrow-derived endothelial progenitor cells. EMBO J 1999;18: 3964-72. 33. Celletti FL, Waugh JM, Amabile PG, Brendolan A, Hilfiker PR, Dake MD. Vascular endothelial growth factor enhances atherosclerotic plaque progression. Nat Med 2001;7:425-9. Submitted Sep 21, 2009; accepted Feb 6, 2010.
Additional material for this article may be found online at www.jvascsurg.org.
JOURNAL OF VASCULAR SURGERY Volume 52, Number 1
METHODS Magnetic resonance imaging protocol. Mice underwent magnetic resonance imaging (MRI) scans at day 21 and day 28 (before euthanasia). MRI data were obtained using a Varian Unity INOVA imaging system fitted with a 4.7-T horizontal bore magnet. Mice were positioned supine within a 63-mm-diameter quadrature imaging coil into which heated air was blown to maintain normal body temperature. Coronal scout images were acquired to allow reproducible positioning of axial images aligned with the central slice at the superior end of the left kidney (Fig 1). Because of the ferromagnetic core of the infusion pump spin echo (rather than gradient echo), images were acquired with strong slice selection and readout gradients to minimize image distortions. Nine axial T1-weighted spinecho cross-sectional images were acquired with 2-mmthick contiguous slices with repetition time of 400 milliseconds, echo time of 8.5 milliseconds, and 16 averages. Outer slice 12-mm-thick saturation bands were used to suppress the high-intensity blood signal of the lumen. In-plane image resolution was 512 ⫻ 256 over a 4-cm field of view and zero-filled to 1024 ⫻ 1024 for region-of-interest (ROI) analysis of abdominal aorta. Using the Varian ImageBrowser software and delineation of the ROI, an operator (F. H.) who was blinded to the experimental arm of the mice determined abdominal aorta diameter (mm) and cross-sectional area (mm2) in each cross-section. Real-time quantitative reverse-transcription polymerase chain reaction. Qualitative reverse-transcription polymerase chain reaction (QRT-PCR) mixtures were es-
Choke et al 166.e1
tablished according to manufacturer’s recommendations (Assays-on-Demand Gene Expression Products and TaqMan Universal PCR Master Mix, Applied Biosystems, Warrington, UK). Each sample was run in triplicates, and for all reactions, negative controls were run with no template present. In addition, total RNA from mice samples was used for sham reverse-transcription reactions with no reverse transcriptase present and then subjected to standard PCR using 18S ribosomal RNA (rRNA) primers to verify that no amplification was produced. QRT-PCR was done using an Mx4000 Multiplex Quantitative PCR System (Stratagene, UK). Each PCR cycle started with an initial 10-minute denaturation step at 95°C, followed by 40 cycles of shuttle heating at 95°C for 15 seconds, and at 60°C for 1 minute. The associated Mx4000 software was used to analyze the data and determine the threshold count (Ct). The specimens were normalized to the endogenous reference control 18S rRNA. For each sample, Cttarget gene and Ct18S rRNA were determined, and ⌬Ct ⫽ Cttarget gene ⫺ Ct18S rRNA. The relative level of target gene normalized to 18S rRNA was determined by calculating: 2–⌬Ct. For the 2–⌬Ct calculation to be valid, preliminary validation experiments were performed to verify that the efficiencies of target gene amplification and the efficiency of 18S rRNA amplification were approximately equal. To achieve this, standard curves of template dilution and Ct were obtained for each target gene and 18S rRNA to confirm comparative rates of transcriptional efficiency for each probe set.
Fig 4, online only. Magnetic resonance imaging was used to compare the ex vivo aortic maximum diameter measurement with the in vivo measurements at day 28. The two measurements were highly correlated with each other (Pearson r2 ⫽ 0.88).