Intramural coronary lipid injection induces atheromatous lesions expressing proinflammatory chemokines: implications for the development of a porcine model of atherosclerosis

Cardiovascular Revascularization Medicine 12 (2011) 304 – 311

Biology

Intramural coronary lipid injection induces atheromatous lesions expressing proinflammatory chemokines: implications for the development of a porcine model of atherosclerosis Armando Tellez a,⁎, David S. Schuster a , Carlos Alviar a , Gabriel López-Berenstein b , Angela Sanguino b , Christie Ballantyne c , Xiao-Yuan Dai Perrard c , Daryl G. Schulz d , Serge Rousselle e , Greg L. Kaluza a , Juan F. Granada a a Skirball Center for Cardiovascular Research, Cardiovascular Research Foundation, Orangeburg, NY, USA Department of Experimental Therapeutics, University of Texas, MD Anderson Cancer Center, Houston TX, USA c Center for Cardiovascular Disease Prevention, Methodist DeBakey Heart and Vascular Center, Houston, TX, USA d Research Institute, Methodist Hospital, Houston, TX, USA e Alizée Pathology, Thurmont, MD, USA b

Received 3 January 2011; received in revised form 21 March 2011; accepted 25 March 2011

Abstract

Background: Intramural delivery of lipids into the coronaries of pigs fed high-cholesterol diet results in the formation of localized atherosclerotic-like lesions within 12 weeks. These lesions are located in positively remodeled vessels and are associated to the development of abundant adventitial vasa vasorum and mononuclear cell infiltrate. In this study, we aimed to analyze the degree of expression of various inflammatory chemokines within the developed lesions compared with control segments injected with saline. Methods: Balloon injury was performed in 15 coronary arteries of pigs fed high-cholesterol diet for 12 weeks. Two weeks after procedure, 60 coronary segments were randomized to either intramural injections of complex lipids (n=30) or normal saline (n=30). Neovessel density in the lesions was analyzed by lectin stain. Segments were processed for RNA expression of inflammatory chemokines such as monocyte chemoattractant protein-1 and vascular endothelial growth factor. Results: At 12 weeks, the percentage area of stenosis seen in histological sections was modest in both groups (lipids: 17.3±15 vs. saline: 32.4±22.8, P=.017). The lipid group showed higher vasa vasorum (VV) quantity (saline: 18.2±14.9 VV/section vs. lipids: 30.6±21.6 VV/section, Pb.05) and vasa vasorum density (saline: 7.3±4.6 VV/mm2 vs. lipids: 16.5±9 VV/mm2, Pb.001). In addition, monocyte chemoattractant protein-1 expression was higher in the lipid group (1.5±1.12) compared with saline control group (0.83±0.34, Pb.01). Vascular endothelial growth factor expression was also higher in the lipid group (1.36±0.9) compared with saline group (0.87±0.33, Pb.05). Conclusion: The intramural injection of complex lipids into the coronary arteries of pigs maintained in a high-cholesterol diet results in focal lesions located in positively remodeled vessels that have a high neovessel count and express proinflammatory chemokines. © 2011 Elsevier Inc. All rights reserved.

Keywords:

Coronary atherosclerosis; Animal model of disease; Porcine model; MCP-1; VEGF

⁎ Corresponding author. Skirball Center for Cardiovascular Research, Cardiovascular Research foundation, 8 Corporate Dr., Orangeburg, NY, 10965, USA. Tel.: +1 845 580 3099; fax: +1 845 359 5084. E-mail address: [email protected] (A. Tellez). 1553-8389/11/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.carrev.2011.03.007

1. Introduction One of the primary limitations in atherosclerosis research is the lack of a large animal model that accurately represents

A. Tellez et al. / Cardiovascular Revascularization Medicine 12 (2011) 304–311

the disease state seen in humans. In the domestic swine, the introduction of a high-cholesterol diet for a prolonged period of time induces atherosclerotic lesions [1,2] that are variably located in the arterial vasculature and lack the human pathological characteristics of complex atheromas [1,3,4]. Porcine animal models developing naturally occurring disease usually display complex lesions that take a long time to develop and usually are randomly located and at different stages of development [5]. Recently, several injurybased models have been proposed for the study of atherosclerosis imaging and interventions [6,7]. Although some of the pathological features of injury-based models are different to what is seen in human disease, these models may offer a cost-effective alternative for the validation of imaging and therapeutic technologies in the atherosclerosis field. We have previously published preliminary data in regard to the development of a particular injury-based model of atherosclerosis [4,5,8]. In this model, complex lipids are injected into the coronary arteries several weeks following initial balloon denudation. Within 12 weeks, complex atherosclerotic lesions are formed and consist of abundant adventitial neovascularization, mononuclear cell infiltrates and positive vascular remodeling. In this study, we aimed to further characterize this model by analyzing the expression profile of proinflammatory chemokines within the developed lesions injected with complex lipids compared with control segments injected with saline.

2. Methods 2.1. Experimental design The study was approved by the Institutional Animal Care and Use Committee. All animals received standard care according to the study protocol and following the act of animal welfare and the “Principles of Care of Laboratory Animals” formulated by the Institute of Laboratory Animal Resources (National Research Council, National Institutes of Health Publication No. 85-23, revised 1996). A total of five female or castrated Domestic Yorkshire swine (mean body weight, 30 kg) were included in this study. Animals were pretreated with aspirin (650 mg) and clopidogrel (300 mg) 1 day prior to the procedure and aspirin (325 mg) and clopidogrel (150 mg) on the day of the procedure. Baseline quantitative coronary angiography (QCA) and intravascular ultrasound (IVUS) were performed. Coronary segments from a total of 15 coronaries were selected according to accessibility and appropriate lumen caliber. A dilatation balloon catheter was used to achieve ~30% overstretch of baseline reference arterial diameter, according to IVUS, to produce vascular injury. The inflation was repeated twice. Two weeks after the initial balloon injury, coronary segments were randomized to two treatment groups: intramural delivery of a liposome-based formulation of complex lipids or control saline injection group. Each coronary segment

305

received four microinjections of 250 μl for a total of 1 ml. Each coronary received only one type of solution (lipid solution injection or control saline injection) to avoid mixture between therapies. All animals were maintained under a high-cholesterol diet (2% cholesterol, 20% lard and 1.5% sodium cholate) until the termination of the study. 2.2. IVUS imaging IVUS pullback images were obtained and analyzed using a coronary ultrasound catheter (Atlantis SR Pro 40 MHz Coronary Imaging Catheter; Boston Scientific, Natick, MA, USA) and a commercially available measurement analytic system (iLab Boston Scientific). Using fluoroscopy, the IVUS catheter was placed distal to the vascular segment selected for intervention by angiography and an automated pullback performed at a speed of 0.5 mm/s covering at least 10 mm proximal and distal to selected vascular segment. The starting position of the IVUS catheter was determined by fluoroscopy and situated by anatomical landmarks in a live image during the pullback. The morphometric analysis in each vessel was performed using standard definitions [9]. 2.3. Lipidic solution injectate One part cholesteryl linoleate powder (Sigma Aldrich, St. Louis, MO, USA) was added to two parts olive oil and vortex mixed. In a modification of the previously described method, human oxidized low-density lipoprotein (LDL) isoform-5 was added to the cholesterol ester injectate. The resulting turbid and viscous solution was loaded into an injection catheter and immersed in sterile hot saline, thereby increasing the temperature and decreasing the viscosity of the solution immediately prior to injection. LDL used for the injectate was isolated from homozygotic familial hypercholesterolemic (human) and separated according to charge using a LCC-500 programmer controlling two P-500 pumps on an UnoQ12 column, an anion exchange column (BioRad, Hercules, CA, USA) preequilibrated with buffer A (0.02 M Tris-HCl, pH 8.0, 0.5 mM EDTA) at 4°C. LDL protein in buffer A was loaded onto the UnoQ12 column and eluted with a multistep gradient of buffer B (1 M NaCl in buffer A): 0%, 10 min; 0–15%, 10 min; 15%–20%, 30 min; isocratic 20%, 10 min; 20%–40%, 25 min; 40%–100%, 10 min; 100%, 15 min; 100%–0%, 5 min and 0%, 25 min. LDL fractions were pooled according to NaCl concentration into five subfractions, L1 through L5 (0.08–0.17, 0.17–0.18, 0.18–0.20, 0.20, and 0.20–0.38 M, respectively). The isolates were concentrated, sterilized and stored at 4 °C. 2.4. Intramural vascular injection Two weeks following balloon injury, endovascular intramural injection was performed using a needle injection catheter (Mercator MedSystems, San Leandro, CA, USA) customized for local drug delivery. The system contains a

306

A. Tellez et al. / Cardiovascular Revascularization Medicine 12 (2011) 304–311

needle in its distal end designed to penetrate the arterial wall and allow solution infusion during balloon deployment in a symmetric fashion using a handheld control mechanism. The needle injection catheter was advanced and located within the previously balloon injured segment. Assessment by fluoroscopy was performed to have the needle expanded to a diameter assuring solution delivery. Following coronary treatment randomization, the infusion catheter system delivered approximately 250 μl of saline control or lipid infusion in each segment per injection. The procedure was repeated circumferentially for a total of four times within the target coronary segment to achieve a total of 1 cc of solution injected. Each animal receive one treatment in at least two of three coronaries. Each coronary received only one treatment. At 12 weeks following injection, terminal angiography and IVUS were obtained, and animals were euthanized while under general anesthesia by intravenous injection of a commercially available euthanasia solution. Following euthanasia, the heart was removed, and coronaries were flushed with 1 L of saline and excised from the myocardium. The coronary arteries were further harvested in 2- to 3-mm vascular rings. The vascular segments were selected in a sequential fashion to be prepared for either histology or quantitative messenger RNA (mRNA) gene expression analysis. Thus, for each vascular segment assigned to gene expression analysis, there was an adjacent vascular segment assigned to histological analysis. 2.5. Histology protocol Arterial segments of the injected site destined for histology were retrieved from the freezer, rinse off the Tissue-Tek® (Sakura Finetek, Torrance, CA) embedding medium for frozen tissue in saline and immersed in 10% normal buffered formalin for 12 h for complete fixation. Arterial segments were processed under standard histology protocols. Arterial segments were embedded in paraffin and sliced to produce 5-μmthick sections. Slides were stained with hematoxylin and eosin, elastin trichrome and Movat's pentachrome. In addition, endothelial cell detection was performed in each arterial sample using Biotinylated Griffonia simplicifolia Lectin I (Vector Laboratories, Burlingame, CA, USA) under general immunohistochemistry protocols. Digital images of the vessels were captured using Spot Advanced v 4.1.1 Software and histomorphometry was performed using BIOQUANT NOVA PRIME v6.70.10 software. Lumen (L) area (mm2), internal elastic lamina (IEL) and external elastic lamina (EEL) area (mm2) were measured. Neointimal area was calculated as IEL − L. Percentage area of stenosis was calculated as [(IEL−L)/ IEL]⁎100. Following slide quadrants, a semiqualitative score was used to determine vasa vasorum (VV) quantification. The VV was quantified in the neointima, media and adventitial vascular area. The VV quantification was performed in each single histological sample. VV density was assessed by calculating VV quantification divided by the neointimal area

(EEL area−L area) with the purpose of determining VV density according to the neointima formation in each arterial segment. 2.6. RNA isolation, reverse transcription polymerase chain reaction Arterial segments of the injected site destined for gene expression were immersed in Tissue-Tek® embedding medium for frozen tissue specimens and snap frozen in liquid nitrogen. Then, a two-step reverse transcription polymerase chain reaction (RT-PCR) methodology was used to determine relative quantities of various RNA transcripts in the arterial segments destined for gene expression. Total cellular RNA was isolated from the vascular samples with TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. A polytron homogenizer was used to homogenize the tissue samples in TRIzol prior to RNA isolation. Following isolation, the RNA samples were treated with DNase to remove residual genomic DNA. First-strand complement DNA (cDNA) was synthesized from 5.0 μg RNA in 100 μl RT reactions using the High Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. The RT method used random hexamers to prime cDNA synthesis. TaqMan real-time QPCR was used to determine the relative cDNA quantities of several target genes. In addition, the relative cDNA quantity of 18S (ribosomal) was determined for use as a normalizer. All QPCR assays for target genes were designed in-house. The 18S QPCR assay was ordered predesigned (Applied Biosystems). QPCR was conducted using the ABI PRISM 7000 Sequence Detection System (Applied Biosystems) in 50-μl singleplex reactions according to the manufacturer's default assay conditions. Serial dilutions of a representative cDNA sample were used as a standard curve for relative cDNA quantification. This methodology was used to quantify the mRNA expression of monocyte chemoattractant protein 1 (MCP-1) and vascular endothelial growth factor (VEGF) in the intercalated arterial samples destined for gene expression. 2.7. Statistical analysis The results of each analytic method were presented as mean±S.D. The statistical analysis was performed using SigmaStat 3.11 software (2004; Sustat Inc., San Jose, California, USA). Unpaired Student's t tests were used to test for differences between groups. A P valueb.05 was considered statistically significant.

3. Results 3.1. Procedural characteristics A total of 60 coronary arterial segments were injected with either a human LDL-rich complex lipid mixture (n=30)

A. Tellez et al. / Cardiovascular Revascularization Medicine 12 (2011) 304–311

307

Table 1 Morphological features and VV quantification of the lipid injection group compared with the saline injection group

2

Lumen area (mm ) Internal elastic lamina (mm2) Vessel size (EEL, mm2) Percent area stenosis VV count VV density (no./mm2)

Lipid Injection (n=30)

Saline Injection (n=30)

P

2.11±1.15 2.55±1.06 4.18±1.62 17.3±15.0 30.6±21.6 16.5±9.0

1.57±0.9 2.22±0.94 4.17±1.59 32.4±22.8 18.2±14.9 7.3±4.6

.09, NS .27, NS .9, NS .017, S .04, S b.001, S

A P value of ≤0.05 was considered statistical significant. S, statistically significant; NS, not statistically significant.

or normal saline (n=30). On QCA, minimal lumen diameter at baseline showed no statistical difference between saline (2.7±0.4 mm) or lipid injection (2.8±0.3 mm, P .5). At termination, a reduction in minimal lumen diameter was observed in both groups (saline: 2.3±0.3 mm, 15% reduction vs. lipid: 2.3±0.4 mm, 18% reduction). Despite these differences, the resulting percentage of Domestic Yorkshire swine observed by QCA was not statistically different between groups (saline: 14.7±5.7% vs. lipid: 19.5±16%, P .5). At termination, IVUS analysis of all injected segments showed a mean lesion length of 16.6±6.9 mm and a percent area of stenosis of 41.9%±7.9%.

3.2. Histomorphometric analysis The mean vessel areas of the treated vascular segments were not different between any of the treated groups (lipids: 4.18±1.62 mm2 vs. saline: 4.17±1.59 mm2, P=.997, nonsignificant). Percentage area of stenosis was higher in saline group compared with lipid group (saline: 32.4±22.8 vs. lipids: 17.3±15.0, P=17; Table 1). There was a statistically significant increase in VV values in the lipid group (30.6 ±21.6 VV/slide) compared with saline control (18.2±14.9 VV /slide, P=.04). VV density (VV count per neointimal area) in the lipid group was found to be statistically higher than that found in the saline group (saline group: 7.3±4.6 VV/mm2 vs. lipid group: 16.5±9.0 VV/mm2, Pb.001; Fig. 1). 3.3. Histological description The degree of plaque formation was variable in sections of vessels selected for microscopy and was observed in both control and lipid-injected vessels. The lesions were characterized by locally extensive infiltration of the neointima by foamy macrophages (Fig. 2), frequently extending into the media. Lipid injected vessels showed a slightly greater incidence of adventitial inflammation (21% of lipid vascular segments vs. 5% of control). The inflammation was generally focal and lymphohistiocytic. Occasional microgranulomas were also observed in the lipid-injected group (Fig. 2). Other microscopic features were typical of vascular healing following balloon injury and were characterized by variable degrees of mural injury with formation of fibromuscular neointima. These changes were observed at comparable degrees in both the control and lipid-injected groups. The presence of foamy macrophages was evident in 50% of the vascular segments in both groups. However, a highest degree of foamy macrophage infiltration was observed in lipid (10.5%) when compared with saline control (5%; Table 2). 3.4. mRNA gene expression analysis

Fig. 1. Histology neovascularization analysis. (A) The VV amount in relation to neointimal area showed that the statistical difference of VV density in lipid injection was more evident when compared with the VV density in saline control group. (B) The amount of VV was significantly higher in lipid injection group when analyzing the number of VV per slide compared with saline control.

Compared with a standardized curve for cDNA quantification, there was a higher expression of MCP-1 in the lipid group compared with the saline group (lipids: 1.55±1.12, vs. saline: 0.83±0.34, Pb.01). VEGF showed similar results with a statistically significant increase in expression in the

308

A. Tellez et al. / Cardiovascular Revascularization Medicine 12 (2011) 304–311

Fig. 2. Plaque formation was inconsistent in both control and lipid-injected vessels. The lipid injection group (A, B, D, E, G, 2× magnification 1 mm reference; C, 20× magnification 100µm reference; F, 20× magnification 200µm reference) shows a neointimal formation less stenotic that the one observed in the saline injection group (G, H, 2× magnification 1 mm reference). A significant neointimal formation was observed as the effect of balloon injury and saline injection on hematoxylin and eosin (H) and Masson's trichrome (G). Other microscopic features were typical of vascular healing following balloon injury and were characterized by variable degrees of mural injury with formation of fibromuscular neointima. However, in the Masson's trichrome, it can be observed that the collagen/fibrotic content (blue) in the saline group (G) is higher than the collagen/fibrotic content of the lipid group (A), suggesting that the saline group produced a more stable, firm plaque compared with the lipid group. The neointimal formation in the lipid injection group observed in the hematoxylin and eosin stain (B) showed a locally extensive infiltration of the neointima by foamy macrophages (C, magnified image of neointima) and proteoglycan infiltration (D). Lipid injected vessels showed a slightly greater incidence of lymphohistiocytic adventitial inflammation. To assess the VV, biotinylated G. simplicifolia Lectin I was used to detect endothelial cells (E), and VV was mainly identified in neointima-media region sometimes extending into the media of the vascular segments (F, magnified image of VV).

lipid group compared with the saline group (lipid group: 1.36±0.92 vs. saline group: 0.87±0.33, P=.013: Fig. 3). The biodistribution of chemokines throughout the treated vascular area was also determined according to the levels of RNA gene expression. In general, the distribution of expression was homogeneous, and there was no statistical difference neither between the groups nor between the analyzed chemokines in regard to vascular expression. However, it was interesting to observe that the highest expression of MCP-1 occurred in at the shoulders of the lesion (57% proximal, 28% distal) compared with the center. Conversely, the VEGF expression occurred more frequently in the distal part of the vascular lesions (43%),

and this distribution correlated with a higher presence of adventitial neovascularization. In contrast, vessels injected with saline have a highest expression of VEGF in the center of the lesion and also correlated with a higher presence of adventitial neovascularization in this region.

4. Discussion The domestic swine is the most commonly used model for preclinical cardiovascular research studies as their anatomy closely resembles that of humans [1]. When maintained on a high-cholesterol diet for a prolonged period of time, this

A. Tellez et al. / Cardiovascular Revascularization Medicine 12 (2011) 304–311

309

Table 2 Histological analysis

Cellular plaque (foamy macrophages) Grade 1 Grade 2 Grade 3 Vascular healing Grade 1 Grade 2 Grade 3 Adventitial inflammation Lymphohistiocytic aggregates Grade 1 Grade 2 Granuloma Grade 1 Grade 2

Lipid injection (%)

Saline injection (%)

52.63 36.84 5.26 10.53 31.58 26.32 5.26 0.00 31.58

52.63 31.58 15.79 5.26 52.63 36.84 10.53 5.26 5.26

15.79 5.26

5.26 0.00

5.26 5.26

0.00 0.00

The same percent of intervened vascular segments demonstrated the presence of foamy macrophages; however, the degree of dense infiltrates widely distributed throughout the neointima presented in higher percentage in the lipid injection group. The neointimal proliferation was observed more prominent in the saline control with the fibromuscular neointima. The presence of granulomas in different degrees and the increased presence of lymphohistiocytic infiltrates indicate an ongoing inflammatory process, which is absent in the saline injection control. 0=absent; 1=slight or minimal feature that barely departs from normal; 2=mild feature that is morphologically distinct and may be locally pronounced but is not widely distributed; 3=moderate feature forming variably dense infiltrates widely distributed in the tissue or may be substenosing when applied to vascular elements but is not regarded as significantly compromising function under normal conditions. 4=marked or severe, overwhelming feature effacing preexisting tissue components or feature that is occlusive or subocclusive (neointima). This highest score would typically produce a compromise tissue function.

model develops atherosclerotic-like lesions. These early lesions, however, are more fibroelastic in nature and do not display some of the biological components seen in humans [1,2], namely, the formation of complex atheromas. Numerous investigators have aimed to develop complex atherosclerotic lesions using the swine model, either by diet supplementation, inducing vascular injury or both. However,

Fig. 3. RNA genetic expression of chemokines. A statistical increase was observed in the lipid injection group in both MCP-1 (A) and VEGF (B) when compared with saline injection control group.

although still effective in inducing lesions, all these models are limited by the development time, costs and composition of the resulting lesion. Gerrity et al. [1] accelerated the development process (20 weeks) by maintaining streptozotocin-induced diabetic pigs on a similar diet [1]. In addition, other investigators have aimed to accelerate the process by inducing direct vascular injury. Shi et al. [3] recently described a model of atherosclerosis in the carotid artery closely resembling human atherosclerosis induced by maintaining pigs on a high fat and cholesterol and by partially ligating the target arteries [3]. We previously proposed a method for inducing atherosclerotic lesions by combining balloon injury and the adventitial delivery of lipidic substances. We have previously shown that by 2 weeks, these lesions are located in positively remodeled vessels and are easily identifiable using IVUS. These early lesions, containing abundant inflammatory cells and adventitial neovascularization, resemble the early stages of human atherosclerosis development [4,5]. Autopsy studies have shown that human atherosclerosis is strongly associated with the presence of macrophages [10] and neovascularization [10–14]. The chemokine molecular system is known to activate and recruit inflammatory cells in different conditions and promote neovascularization in atherosclerosis [11]. It has also been described that neointimal thickening is associated with an increase in VV density [13]. In an attempt to provide oxygen to the cellular components of the pathological medial thickening, immature VV develops leading to red blood cell and LDL deposition in the atherosclerotic plaque early in the process of development. The accumulation of these biological components in

310

A. Tellez et al. / Cardiovascular Revascularization Medicine 12 (2011) 304–311

the medial vascular wall are controlled by chemotactic cytokines (chemokines) that promote inflammatory cell recruitment and activation, resulting in the inflammatory vascular infiltration observed in complex atherosclerotic plaques [13,15]. Two of the main chemokines involved in inflammatory cell chemotaxis and vasculogenesis are MCP1 and VEGF. MCP-1 role in angiogenesis and atherosclerosis has been extensively studied [16–20]. The presence of MCP-1 protein has been correlated with neointimal proliferation [18] and foam cell presence [11,16–19,21]. It has also been detected in human atherosclerotic plaques and macrophage-rich areas of atherosclerotic lesions [22,23]. Also, MCP-1 is thought to play a role in the transition from a stable lesion to a more complex state of disease [11,24]. MCP-1−/− knockout mice usually develop stable lesions with high-collagen content and minimal necrotic core formation [24]. VEGF has been shown to induce neovascularization in physiological and pathological processes [13,19,25–27]. The administration of VEGF-A protein in apolipoprotein Edeficient mice promotes angiogenesis, while the administration of anti-VEGFR-1 antibody in apolipoprotein E-deficient mice reduces plaque development and increases plaque stability [28,29]. In order to determine the similarity of our lipid injection model of accelerated atherosclerosis to the development of early human atherosclerosis [13], we aimed to study the relationship of VV density with MCP-1 and VEGF expression. In our study, the degree of vascular injury induced by balloon overexpansion and the physical introduction of the needle were controlled variables present for both groups. Angiographic and IVUS analysis demonstrated that the resulting lesions were eccentric, nonocclusive and located in positively remodeled segments. Histomorphometric analysis showed that the lesions were nonocclusive in both intervened groups. The nature of the developed lesions in this model relates to previously published data that suggest that the majority of coronary lesions are nonocclusive and appear to remain stable over time [30]. It has been described that the degree of angiographic stenosis may not be the predominant feature leading to thrombosis and that a significant number of events occur among patients having angiograms that were regarded as normal or mildly. Current knowledge in atherosclerosis research suggests that necrotic core size and composition may be more important than obstruction severity in regard to plaque vulnerability [31,32]. In our study, we found an increase of 44% in VV density in lipid injection group when compared with saline injection. This result together with the 55% and 63% increase of MCP1 and VEGF expression, respectively, within the same injected segments suggested the promotion of angiogenic and inflammatory processes resembling the characteristics observed in human atherosclerosis. In general, the resulting lesions were composed of fatty infiltrates that were occasionally significant, forming dense aggregates effacing the neointima. The media and adventitia were generally unaffected, although increased inflammatory in-

filtrates were recorded in the adventitia of lipid-injected vessels. More advanced plaque features such as necrotic core or calcification were not observed. Interestingly, although the “overall” histological appearance of the resulting plaques did not reveal appreciable differences between saline and lipid-injected vessels (Fig. 2), significant differences were observed when the degree of chemokine expression was compared between the groups. The biggest limitation of our study is the short follow-up time that was chosen for this study. Therefore, although the resulting lesions were evaluated for the presence of neovascularization and inflammation, other aspects of complex atherosclerosis, including the timing required to develop a necrotic core and their ability to spontaneously rupture, could not be tested. Further research is necessary to more accurately determine the model's relevance to human disease and whether it could be used in preclinical trials. Nonetheless, the model herein described produced early atherosclerotic lesions that bear striking microanatomical, morphological and biological similarities to human lesions. Hence, it appears as a cost-effective method that induces atherosclerotic lesions in swine that bear some resemblance with human lesions with respect to neovascularization and chemokines expression profile. Due to its controlled and expedited induction, it could prove useful for the evaluation of emerging endovascular imaging modalities and possibly complement other models requiring months to develop mature atherosclerotic lesions. We believe the described model to be promising and a viable research or screening tool for selective technologies in the atherosclerosis research field. References [1] Gerrity RG, Natarajan R, Nadler JL, Kimsey T. Diabetes-induced accelerated atherosclerosis in swine. Diabetes 2001;50:1654–65. [2] Gerrity RG, Naito HK, Richardson M, Schwartz CJ. Dietary induced atherogenesis in swine. Morphology of the intima in prelesion stages. Am J Pathol 1979;95:775–92. [3] Shi ZS, Feng L, He X, Ishii A, Goldstine J, Vinters HV, Vinuela F. Vulnerable plaque in a Swine model of carotid atherosclerosis. AJNR Am J Neuroradiol 2009;30:469–72. [4] Granada JF, Moreno PR, Burke AP, Schulz DG, Raizner AE, Kaluza GL. Endovascular needle injection of cholesteryl linoleate into the arterial wall produces complex vascular lesions identifiable by intravascular ultrasound: early development in a porcine model of vulnerable plaque. Coron Artery Dis 2005;16:217–24. [5] Granada JF, Kaluza GL, Wilensky RL, Biedermann BC, Schwartz RS, Falk E. Porcine models of coronary atherosclerosis and vulnerable plaque for imaging and interventional research. EuroIntervention 2009; 5:140–8. [6] Busnelli M, Froio A, Bacci ML, Giunti M, Cerrito MG, Giovannoni R, Forni M, Gentilini F, Scagliarini A, Deleo G, Benatti C, Leone BE, Biasi GM, Lavitrano M. Pathogenetic role of hypercholesterolemia in a novel preclinical model of vascular injury in pigs. Atherosclerosis 2009;207:384–90. [7] Ferns GA, Avades TY. The mechanisms of coronary restenosis: insights from experimental models. Int J Exp Pathol 2000;81:63–88. [8] Alviar CL, Tellez A, Wallace-Bradley D, Lopez-Berestein G, Sanguino A, Schulz DG, Builes A, Ballantyne CM, Yang CY, Kaluza

A. Tellez et al. / Cardiovascular Revascularization Medicine 12 (2011) 304–311

[9]

[10] [11]

[12]

[13]

[14] [15]

[16]

[17]

[18] [19]

GL, Granada JF. Impact of adventitial neovascularisation on atherosclerotic plaque composition and vascular remodelling in a porcine model of coronary atherosclerosis. EuroIntervention;5:981-8. Mintz GS, Nissen SE, Anderson WD, Bailey SR, Erbel R, Fitzgerald PJ, Pinto FJ, Rosenfield K, Siegel RJ, Tuzcu EM, Yock PG. American College of Cardiology Clinical Expert Consensus Document on Standards for Acquisition, Measurement and Reporting of Intravascular Ultrasound Studies (IVUS). A report of the American College of Cardiology Task Force on Clinical Expert Consensus Documents. J Am Coll Cardiol 2001;37:1478–92. Fan J, Watanabe T. Inflammatory reactions in the pathogenesis of atherosclerosis. J Atheroscler Thromb 2003;10:63–71. Frangogiannis NG. The prognostic value of monocyte chemoattractant protein-1/CCL2 in acute coronary syndromes. J Am Coll Cardiol 2007; 50:2125–7. Langheinrich AC, Michniewicz A, Sedding DG, Walker G, Beighley PE, Rau WS, Bohle RM, Ritman EL. Correlation of vasa vasorum neovascularization and plaque progression in aortas of apolipoprotein E(−/−)/low-density lipoprotein(−/−) double knockout mice. Arterioscler Thromb Vasc Biol 2006;26:347–52. Langheinrich AC, Kampschulte M, Buch T, Bohle RM. Vasa vasorum and atherosclerosis — Quid novi? Thromb Haemost 2007; 97:873–9. Libby P. Inflammation in atherosclerosis. Nature 2002;420:868–74. Moreno PR, Purushothaman KR, Zias E, Sanz J, Fuster V. Neovascularization in human atherosclerosis. Curr Mol Med 2006;6: 457–77. Heil M, Ziegelhoeffer T, Wagner S, Fernandez B, Helisch A, Martin S, Tribulova S, Kuziel WA, Bachmann G, Schaper W. Collateral artery growth (arteriogenesis) after experimental arterial occlusion is impaired in mice lacking CC-chemokine receptor-2. Circ Res 2004; 94:671–7. Ito WD, Arras M, Winkler B, Scholz D, Schaper J, Schaper W. Monocyte chemotactic protein-1 increases collateral and peripheral conductance after femoral artery occlusion. Circ Res 1997;80:829–37. Charo IF, Taubman MB. Chemokines in the pathogenesis of vascular disease. Circ Res 2004;95:858–66. Fujii T, Yonemitsu Y, Onimaru M, Tanii M, Nakano T, Egashira K, Takehara T, Inoue M, Hasegawa M, Kuwano H, Sueishi K. Nonendothelial mesenchymal cell-derived MCP-1 is required for FGF-2mediated therapeutic neovascularization: critical role of the inflammatory/arteriogenic pathway. Arterioscler Thromb Vasc Biol 2006;26: 2483–9.

311

[20] Hoefer IE, van Royen N, Buschmann IR, Piek JJ, Schaper W. Time course of arteriogenesis following femoral artery occlusion in the rabbit. Cardiovasc Res 2001;49:609–17. [21] Hartl D, Griese M, Nicolai T, Zissel G, Prell C, Reinhardt D, Schendel DJ, Krauss-Etschmann S. A role for MCP-1/CCR2 in interstitial lung disease in children. Respir Res 2005;6:93. [22] Takeya M, Yoshimura T, Leonard EJ, Takahashi K. Detection of monocyte chemoattractant protein-1 in human atherosclerotic lesions by an anti-monocyte chemoattractant protein-1 monoclonal antibody. Hum Pathol 1993;24:534–9. [23] Yla-Herttuala S, Lipton BA, Rosenfeld ME, Sarkioja T, Yoshimura T, Leonard EJ, Witztum JL, Steinberg D. Expression of monocyte chemoattractant protein 1 in macrophage-rich areas of human and rabbit atherosclerotic lesions. Proc Natl Acad Sci USA 1991;88: 5252–6. [24] Inoue S, Egashira K, Ni W, Kitamoto S, Usui M, Otani K, Ishibashi M, Hiasa K, Nishida K, Takeshita A. Anti-monocyte chemoattractant protein-1 gene therapy limits progression and destabilization of established atherosclerosis in apolipoprotein E-knockout mice. Circulation 2002;106:2700–6. [25] Moulton KS, Vakili K, Zurakowski D, Soliman M, Butterfield C, Sylvin E, Lo KM, Gillies S, Javaherian K, Folkman J. Inhibition of plaque neovascularization reduces macrophage accumulation and progression of advanced atherosclerosis. Proc Natl Acad Sci USA 2003;100:4736–41. [26] Kolodgie FD, Narula J, Yuan C, Burke AP, Finn AV, Virmani R. Elimination of neoangiogenesis for plaque stabilization: is there a role for local drug therapy? J Am Coll Cardiol 2007;49:2093–101. [27] von Hundelshausen P, Petersen F, Brandt E. Platelet-derived chemokines in vascular biology. Thromb Haemost 2007;97:704–13. [28] Matsumoto T, Mugishima H. Signal transduction via vascular endothelial growth factor (VEGF) receptors and their roles in atherogenesis. J Atheroscler Thromb 2006;13:130–5. [29] Ferrara N, Gerber HP, LeCouter J. The biology of VEGF and its receptors. Nat Med 2003;9:669–76. [30] Haft JI, Haik BJ, Goldstein JE, Brodyn NE. Development of significant coronary artery lesions in areas of minimal disease. A common mechanism for coronary disease progression. Chest 1988;94:731–6. [31] Falk E, Shah PK, Fuster V. Coronary plaque disruption. Circulation 1995;92:657–71. [32] Mann JM, Davies MJ. Vulnerable plaque. Relation of characteristics to degree of stenosis in human coronary arteries. Circulation 1996;94: 928–31.