Induction of Atherosclerotic Plaque Rupture in Apolipoprotein E−/− Mice After Adenovirus-Mediated Transfer of p53
Background— The presence of the tumor-suppressor gene p53 in advanced atherosclerotic plaques and the sensitivity to p53-induced cell death of smooth muscle cells isolated from these plaques have fueled speculation about the role of p53 in lesion destabilization and plaque rupture. In this study, we describe a strategy to promote (thrombotic) rupture of preexisting atherosclerotic lesions using p53-induced lesion remodeling.
Methods and Results— Carotid atherogenesis was initiated in apolipoprotein E knockout mice by placement of a perivascular silastic collar. The resulting plaques were incubated transluminally with recombinant adenovirus carrying either a p53 or β-galactosidase (lacZ) transgene. p53 transfection was restricted to the smooth muscle cell-rich cap of the plaque and led to an increase in cap cell apoptosis 1 day after transfer. p53 overexpression resulted in a marked decrease in the cellular and extracellular content of the cap, reflected by a markedly reduced cap/intima ratio (0.21±0.04 versus 0.46±0.03, P<0.001). The latter is a characteristic feature of plaque vulnerability to rupture, and whereas spontaneous rupture of p53-treated lesions was rare, it was found in 40% of cases after treatment with the vasopressor compound phenylephrine (P=0.003).
Conclusions— We have demonstrated a potential role of p53-induced remodeling in atherosclerotic plaque destabilization. Being the first example of inducible rupture at a predefined location, this model offers a unique opportunity to delineate the processes that precede rupture and to evaluate plaque-stabilizing therapies.
Received December 4, 2001; revision received February 12, 2002; accepted February 14, 2002.
The turnover of the cellular compartment,1,2⇓ including macrophages,3 smooth muscle cells,4 and possibly endothelial cells,5 is thought to play an important role in atherogenesis. This is supported by the presence of proapoptotic/antiproliferative gene products in the advanced atherosclerotic plaque, including bax, Fas, and p53.6–11⇓⇓⇓⇓⇓ Whether the overall effect of these factors in the plaque is beneficial or deleterious has hitherto remained subject to controversy.12
See p 2010
The tumor-suppressor protein p53 is involved in both cell proliferation and apoptosis and upregulated by various inducers of cellular stress known to be present in an atheromatous setting, including DNA damage, nitrogen oxide, hypoxia, oxidative stress, and oxidized lipoproteins.13 p53 expression is increased in human atherosclerotic lesions, both in lipid-laden macrophages and vascular smooth muscle cells.9,10⇓ Interestingly, vascular smooth muscle cells (VSMCs) isolated from atherosclerotic plaques are more susceptible to p53-mediated apoptosis than normal VSMCs.14,15⇓
Guevara et al16 have demonstrated that atherosclerosis is aggravated in p53/apolipoprotein E (ApoE) double-knockout (p53−/−/ApoE−/−) mice through an increase in p53-controlled proliferation. Moreover, we have shown that ApoE*3-Leiden transgenic mice suffer from accelerated atherosclerosis after bone marrow reconstitution with p53−/− bone marrow, which is believed to result from enhanced intimal accumulation of macrophages.17 To study the effect of p53-mediated apoptosis in preexisting atherosclerosis, we transfected collar-induced18 carotid lesions in ApoE−/− mice with adenovirus carrying a p53 transgene. One day after transfection, we analyzed these plaques for transgene expression and markers of apoptosis and proliferation. After 2 weeks, we assessed lesion stability by analyzing plaque composition and by inducing plaque stress in vivo through systemic administration of phenylephrine.
To generate the replication-defective adenovirus vector Ad5-CMV.p53, the coding region of the human p53 cDNA was polymerase chain reaction amplified from plasmid pCMV-p5319 and inserted in the adapter plasmid pCA13 (Microbix Biosystems Inc). The clone pCA13/p53 was cotransfected with plasmid pJM1720 (a gift from Dr Frank Graham, Hamilton, Canada) into 911 helper cells. Homologous recombination generates the E1-deleted adenovirus vector Ad5-CMV.p53. The vector was plaque purified, propagated on PER.C6 cells, and quantified according to our standard protocols.21 All batches were tested for the presence of replication-competent adenoviruses with polymerase chain reaction.22 An analogous method was used to generate a replication-defective adenovirus vector Ad5-CMV.lacZ expressing β-galactosidase.
Carotid Collar Placement and Adenovirus Injection
Carotid atherosclerotic lesions were induced by perivascular collar placement in female ApoE−/− mice aged 10 to 12 weeks, as described previously.18 Three weeks after surgery, the collars were removed and 10 μL of adenoviral suspension at 1.5×1010 pfu/mL was instilled bilaterally (1 side with Ad5-CMV.p53, the contralateral side with Ad5-CMV.lacZ) into the common carotid artery via the external carotid (Figure 1A). The suspension was left in situ for 10 minutes after temporary ligation of the proximal common carotid artery and the internal carotid artery and subsequently drawn off before ligation of the external carotid and closure of the skin wound with silk sutures. In total, 54 animals were used in this study, of which 5 did not recover from the anesthetic; the remainder were sacrificed 1 (n=7), 4 (n=6), or 14 days (n=16) after transfection. In a separate experiment, phenylephrine (8 μg/kg IV, Sigma Diagnostics) was administered to 20 animals 14 days after bilateral intravascular adenovirus incubation, and the effects on plaque integrity were assessed after 24 hours.
Tissue Harvesting and Preparation for Histological Analysis
Transverse 5-μm cryosections were prepared from carotid artery specimens, as described previously,18 and routinely stained with hematoxylin (Sigma Diagnostics) and eosin (Merck Diagnostica), Masson’s trichrome (Accustain kit, Sigma), and picrosirius red (Direct red 80, Sigma). 32 β-Galactosidase was demonstrated by incubation with x-gal (1 mg/mL, Eurogentec) at 37°C for 4 hours. Slides were stained with antibodies against recombinant human p53 (FL-393, rabbit polyclonal IgG, Santa Cruz Biotechnology, Santa Cruz, CA), a macrophage-specific antigen (MOMA-2, polyclonal rat IgG2b, diluted 1:10; Research Diagnostics Inc, Flanders, NJ), α-smooth muscle cell actin (clone 1A4, dilution 1:500; Sigma), intracellular adhesion molecule-1 (clone BSA2, dilution 1:200; R&D Systems), vascular cellular adhesion molecule-1 (clone 429, dilution 1:100; Pharmingen), and proliferating cell nuclear antigen (PCNA) (clone 19F4, dilution 1:40; Roche). Goat anti-mouse IgG peroxidase conjugate (dilution 1:100 and 1:500; Nordic, Tilburg, the Netherlands) and goat anti-rat IgG alkaline phosphatase conjugate (dilution 1:100; Sigma, St Louis, Mo) were used as secondary antibodies, with 3,3′-diamino-benzidine, nitro blue tetrazolium, and 5-bromo-4-chloro-3-indolyl phosphate as enzyme substrates (all Sigma).
Quantification of Apoptosis
Apoptosis was assessed by terminal deoxynucleotidyl transferase end-labeling (TUNEL).6,23⇓ Only TUNEL-positive nuclei were included that displayed morphological features of apoptosis, including cell shrinkage, aggregation of chromatin into dense masses, and nuclear fragmentation.
The point of maximal stenosis of each vessel was determined by analysis of sections at 50-μm intervals. At this point (on average ≈0.5 mm proximal to the collar), morphometry was performed using LeicaQwin software (Leica Imaging Systems). The intimal surface area was calculated by subtracting the free lumen area from the area circumscribed by the internal elastic lamina. The intima was subdivided into a fibrocellular cap and a necrotic core on the basis of extracellular matrix staining by HE, which was confirmed by staining for collagen by picrosirius red and Masson’s trichrome. Actin-positive and MOMA-2-positive areas were determined by computer-assisted color-gated measurement and related to the total intimal surface area (LeicaQwin). Intimal cell numbers were assessed by counting the nuclei in a 5-μm section after H&E staining.
Values are expressed as mean±SEM. A 2-tailed Student’s t test was used in the comparison of continuous data, which was used in a nonpaired form when comparing different animals and paired when comparing contralateral values in the same animal. Frequency data analysis was carried out by means of the Fisher’s exact test. A level of P<0.05 was considered significant.
Adenoviral Expression Pattern
Adenoviral activity was confirmed by intravenous injection of both constructs (2×109 pfu/mouse) in control animals, which resulted in the expression of lacZ and p53 in ≈50% of hepatocytes 5 days after injection (data not shown). To establish the efficiency and distribution of vascular transduction, initial studies were performed with Ad5-CMV.lacZ. The pattern of β-galactosidase activity 4 days after transfection differed markedly, depending on the site examined. The area opposite the external carotid artery, in which the endothelium had been denuded by the instillation catheter, showed extensive transduction of the media (Figure 1B). Efficient transduction of the endothelium and a superficial layer of α-smooth muscle actin-positive cells was observed in the fibrous cap of the collar-induced plaque (Figures 1C, 1D, and 2A⇓). The percentage of positive plaque cells increased from 3.6±2.8% (n=6) 1 day after transfection to 18.4±10.6% (n=3) 4 days after transfection. After Ad5-CMV.p53 transfection, immunohistochemical staining for p53 was minimal and only present in the lesion core by 4 days (n=6), but a substantial number of p53-positive cells were identified in the intima 1 day after transfection (1.3±0.6%, n=7), of which 60% were located in the fibrous cap (Figure 2C). No expression of p53 was seen at either time point in Ad5-CMV.lacZ-treated plaques, and although the antibody used cross-reacts with murine p53, this clear difference in staining intensity implies expression of the transgene rather than aspecific effects of Ad5-CMV.p53 transduction. Intracarotid adenovirus instillation did not affect normal functioning of the animals and did not result in a change in weight (20.9±1.9 versus 21.3±0.9 g in controls) or serum cholesterol (31.4±0.6 versus 33.1±1.1 mmol/L) at the time of harvest. Administration of adenovirus at the titer used in this study was not found to result in local cytopathic effects.
Proliferation and Apoptosis
PCNA staining was reduced from 12.7±2.9% (n=3) of cap cells in Ad5-CMV.lacZ to 7.1±2.1% (n=3) in Ad5-CMV.p53-treated animals 1 day after transfection (P=0.19, Figure 2E). In the Ad5-CMV.lacZ-treated vessels, 0.7±0.3% (n=7) of plaque cells were TUNEL-positive 1 day after transfection compared with 1.9±1.0% (n=7) in the Ad5-CMV.p53-incubated plaques (P=0.20, Figures 2B, 2D, and 2F). A total of 48.8±10.0% of TUNEL-positive cells in the Ad5-CMV.p53-treated plaques, but only 14.3±7.2% in Ad5-CMV.lacZ-treated vessels, were located in the fibrous cap (P<0.05). Fourteen days after transfection, plaque TUNEL staining amounted to 0.2±0.1% (n=7) in Ad5-CMV.lacZ and 0.9±0.7% (n=8) in Ad5-CMV.p53-treated animals. PCNA and TUNEL staining were minimal in the endothelium, with only occasional positive cells observed in both groups at either time point (data not shown).
Total cross-sectional plaque area did not differ between treatment groups (P=0.63, Figure 3A), being 57.3±7.7×103 μm2 in the Ad5-CMV.lacZ-treated vessels and 68.9±21.3×103 μm2 in Ad5-CMV.p53-treated plaques. The plaques of both groups consisted of a distinctive fibrous cap overlying a hypocellular core (Figures 4A, 4B, 4G, and 4H). Whereas the core area in Ad5-CMV.p53-treated animals was increased compared with the Ad5-CMV.lacZ control group (54.1±16.6×103 μm2 versus 32.0±5.5×103 μm2, P=0.25), the cap area was decreased by Ad5-CMV.p53 treatment (14.7±5.4×103 μm2 versus 25.3±2.3×103 μm2, P=0.11). This effect was seen to translate to a highly significant decrease in the cap/intima ratio in Ad5-CMV.p53-treated arteries (0.21±0.04 versus 0.46±0.03, P<0.001; Figure 3B), which was mirrored by a decrease in the cap/core ratio (0.28±0.13 versus 0.86±0.24, P<0.01). The decrease in fibrocellular cap surface detected with H&E staining was confirmed as a decrease in collagen-rich matrix by Masson’s trichrome (Figures 4A, 4D, 4G, and 4H) and picrosirius red (Figures 4C and 4F). The average number of nuclei contained within each 5 μm section of the cap was decreased from 208±28 to 103±26 by Ad5-CMV.p53 treatment (Figure 3C; P=0.03), but the number of nuclei per cap surface area remained unchanged (Figure 3D). None of the morphometric parameters differed between the Ad5-CMV.lacZ-treated vessels and untreated controls (eg, cap/intima ratio: 0.46±0.03 versus 0.41±0.13, respectively; P=0.53).
Intimal MOMA-2 staining was unaffected by Ad5-CMV.p53 (16.0±7.2% versus 5.8±1.9%, P=0.30). Staining for α-smooth muscle actin was confined to the media and the fibrous cap in both groups (Figure 4). A paradoxical increase in actin staining was seen in the Ad5-CMV.p53-treated cap (Figures 4B and 4E), which led to a nonsignificantly increased (P=0.36) α-smooth muscle actin/cap ratio (6.8±2.6% versus 1.4±0.5%). One day after transfection, the endothelium overlying the plaque was positive for intracellular adhesion molecule-1 and vascular cellular adhesion molecule-1 in both groups, confirming the integrity of the endothelial lining, but staining intensity was not seen to differ (data not shown).
Induction of Plaque Rupture
The vulnerability of Ad5-CMV.p53-treated plaques to spontaneous rupture was suggested by the occurrence of cap breaks in three of a total of 16 animals in three separate experiments 14 days after transfection, which was accompanied by partial extrusion of the core contents in all cases. Extensive thrombosis, intraplaque hemorrhage, and phagocytosis of erythrocytes by macrophages accompanied cap rupture in one animal (Figure 4I). Plaque integrity was maintained in all of the Ad5-CMV.lacZ-treated vessels. To study the liability of plaques to rupture after a hemodynamic challenge, we systemically administered the vasopressor compound phenylephrine (8 μg/kg IV) to 20 animals 14 days after transfection, the collar-induced plaques of which had been treated on one side with Ad5-CMV.p53 and on the contralateral side with Ad5-CMV.lacZ. This was found to result in a moderate and transient rise in systolic blood pressure (15.0±1.4 mm Hg, as recorded by tail-cuff manometry), which was sustained for at least 15 minutes. To permit the development of possible rupture-related thrombosis, euthanasia was delayed for 24 hours after phenylephrine injection. Intraoperative examination of the plaques before PBS perfusion (ie, with an intact circulation) revealed evidence of intra-plaque hemorrhage in 5, thrombosis in 1, and cap break without intra-plaque hemorrhage in 1 of the p53-treated vessels. No such events were found in the lacZ control vessels. These findings were confirmed by histological analysis of the collar-induced plaques, which revealed evidence of plaque rupture in these 7 animals, and in one additional plaque out of 20 Ad5-CMV.p53-treated vessels. This was accompanied by intracarotid thrombosis in 1 case (Figure 4J) and by intraplaque hemorrhage in 7 vessels (Table). Because we did not find evidence of plaque rupture or its sequelae in the Ad5-CMV.lacZ-treated control plaques (n=20), the number of events in the Ad5-CMV.p53-treated group was highly significant (Fisher’s exact test, P=0.003).
In this study, Ad5-CMV.p53-treated plaques were seen to acquire a phenotype that has been associated with increased vulnerability to rupture.24 Spontaneous rupture of these predisposed lesions was rare, but administration of a vasopressor compound resulted in evidence of plaque disruption in 40% of the Ad5-CMV.p53-treated vessels.
We consider the overexpression of p53 in the cap to be instrumental in inducing this plaque destabilization. The number of p53-expressing cells was indeed raised in the cap after Ad5-CMV.p53 transfection, but expression levels of this transgene were lower than those found with β-galactosidase. This difference may be related to the discrepancy in intracellular half-life of β-galactosidase and wild-type p53, which have been estimated to be at least 1 to 2 days25 and only ≈10 minutes,26 respectively. Intravenous injection of the vectors used in this study, however, revealed equipotent hepatocyte transduction rates. Untransformed hepatocytes are known to be relatively resistant to p53-induced apoptosis,27 in stark contrast to plaque-derived smooth muscle cells, which display increased rates of spontaneous apoptosis14 and high susceptibility to p53-mediated apoptosis.14,15⇓ In the vessel wall, high levels of p53 expression may therefore lead to rapid clearance of transduced cells and consequently to a possible underestimation of transduction efficiency. Apoptosis of these cells and their subsequent phagocytosis by surrounding macrophages and smooth muscle cells are both known to be extremely rapid processes, lasting ≈2 to 4 hours.3,14,15,28,29⇓⇓⇓⇓ This is supported by the fact that staining for p53 was seen in plaques 1 day after Ad5-CMV.p53 transfection, whereas staining was minimal on day 4. Although p53-mediated apoptosis is considered to contribute to the observed 50% reduction in plaque cell number, an important role may also be attributed to a concomitant decrease in cellular proliferation.
The reduction in cell number is believed to be primarily responsible for the development of the destabilized cap phenotype through a proportional decrease in collagen production by cap VSMCs. In addition, p53 may exert a destabilizing effect by selectively eliminating synthetic VSMCs or promoting smooth muscle cell differentiation10,30⇓ and thus inducing the transition of cap smooth muscle cells from a synthetic to a contractile phenotype. Both processes may cause the observed tendency toward a relative increase in α-smooth muscle cell actin staining after Ad5-CMV.p53 treatment, because this suggests a selective elimination of α-smooth muscle cell actin-negative VSMCs or a net increase in (re)differentiated VSMCs.
The findings of this study may be considered to contradict previous work, in which p53 deficiency was seen to lead to an acceleration of atherosclerosis.16,17⇓ Apoptosis has indeed been implicated in atherosclerotic plaque remodeling, and the induction of macrophage apoptosis may constitute an intriguing approach to atherosclerosis prevention or regression.12 The apparent lack of such a modulation of lesion size and macrophage content in this study does not refute this hypothesis but is likely to be attributable to the restriction of p53 expression to the superficial, smooth muscle cell-rich areas of the plaque and the relative lack of macrophage transduction in our model.
In summary, we have provided evidence for a potential role of p53-mediated effects in the destabilization and rupture of atherosclerotic plaques. Plaque rupture and subsequent thrombotic vascular occlusion are the most deleterious sequelae of atherosclerosis, underlying most fatal myocardial infarctions and thromboembolic strokes. A satisfactory animal model for this disease entity has not been available to date. Spontaneous plaque rupture has previously been observed in atherosclerosis-prone mouse strains.31,32⇓ In our model, however, rupture occurs more frequently and in a controlled fashion, allowing the delineation of the molecular pathways involved and the evaluation of therapies aimed at plaque stabilization and the prevention of acute cardiovascular events.
This study was supported by grants M93.001 (Dr von der Thüsen) and 2000.051 (Dr van Vlijmen) from the Netherlands Heart Foundation and by NWO grant 902-26-207 (Dr van Vlijmen).
↵*These authors contributed equally to this study.
- ↵Kockx MM, Knaapen MW. The role of apoptosis in vascular disease. J Pathol. 2000; 190: 267–280.
- ↵Walsh K, Isner JM. Apoptosis in inflammatory-fibroproliferative disorders of the vessel wall. Cardiovasc Res. 2000; 45: 756–765.
- ↵Bjorkerud S, Bjorkerud B. Apoptosis is abundant in human atherosclerotic lesions, especially in inflammatory cells (macrophages and T cells), and may contribute to the accumulation of gruel and plaque instability. Am J Pathol. 1996; 149: 367–380.
- ↵Bennett MR, Boyle JJ. Apoptosis of vascular smooth muscle cells in atherosclerosis. Atherosclerosis. 1998; 138: 3–9.
- ↵Dimmeler S, Hermann C, Zeiher AM. Apoptosis of endothelial cells: contribution to the pathophysiology of atherosclerosis? Eur Cytokine Netw. 1998; 9: 697–698.
- ↵Kockx MM. Apoptosis in the atherosclerotic plaque: quantitative and qualitative aspects. Arterioscler Thromb Vasc Biol. 1998; 18: 1519–1522.
- ↵Isner JM, Kearney M, Bortman S, et al. Apoptosis in human atherosclerosis and restenosis. Circulation. 1998; 91: 2703–2711.
- ↵Mallat Z, Ohan J, Leseche G, et al. Colocalization of CPP-32 with apoptotic cells in human atherosclerotic plaques. Circulation. 1997; 96: 424–428.
- ↵Ihling C, Menzel G, Wellens E, et al. Topographical association between the cyclin-dependent kinases inhibitor P21, p53 accumulation, and cellular proliferation in human atherosclerotic tissue. Arterioscler Thromb Vasc Biol. 1997; 17: 2218–2224.
- ↵Ihling C, Haendeler J, Menzel G, et al. Co-expression of p53 and MDM2 in human atherosclerosis: implications for the regulation of cellularity of atherosclerotic lesions. J Pathol. 1998; 185: 303–312.
- ↵Geng YJ, Henderson LE, Levesque EB, et al. Fas is expressed in human atherosclerotic intima and promotes apoptosis of cytokine-primed human vascular smooth muscle cells. Arterioscler Thromb Vasc Biol. 1997; 17: 2200–2208.
- ↵Kockx MM, Herman AG. Apoptosis in atherosclerosis: beneficial or detrimental? Cardiovasc Res. 1997; 45: 736–746.
- ↵Ko LJ, Prives C. p53: puzzle and paradigm. Genes Dev. 1996; 10: 1054–1072.
- ↵Bennett MR, Evan GI, Schwartz SM. Apoptosis of human vascular smooth muscle cells derived from normal vessels and coronary atherosclerotic plaques. J Clin Invest. 1995; 95: 2266–2274.
- ↵Bennett MR, Littlewood TD, Schwartz SM, et al. Increased sensitivity of human vascular smooth muscle cells from atherosclerotic plaques to p53-mediated apoptosis. Circ Res. 1997; 81: 591–599.
- ↵Guevara NV, Kim HS, Antonova EI, et al. The absence of p53 accelerates atherosclerosis by increasing cell proliferation in vivo. Nat Med. 1999; 5: 335–339.
- ↵van Vlijmen BJM, Gerritsen G, Franken AL, et al. Macrophage p53 deficiency leads to enhanced atherosclerosis in APOE*3-Leiden transgenic mice. Circ Res. 2001; 88: 780–786.
- ↵von der Thüsen JH, van Berkel ThJC, Biessen EAL. Induction of rapid atherogenesis by perivascular carotid collar placement in ApoE−/− and LDLr−/− mice. Circulation. 2001; 103: 1164–1170.
- ↵Steegenga WT, Shvarts A, Riteco N, et al. Distinct regulation of p53 and p73 activity by adenovirus E1A, E1B, and E4orf6 proteins. Mol Cell Biol. 1999; 19: 3885–3894.
- ↵McGrory WJ, Bautista DS, Graham FL. A simple technique for the rescue of early region I mutations into infectious human adenovirus type 5. Virology. 1988; 163: 614–617.
- ↵Fallaux FJ, Bout A, van der Velde I, et al. New helper cells and matched early region 1-deleted adenovirus vectors prevent generation of replication-competent adenoviruses. Hum Gene Ther. 1998; 9: 1909–1917.
- ↵Pietersen AM, van der Eb MM, Rademaker HJ, et al. Specific tumor-cell killing with adenovirus vectors containing the apoptin gene. Gene Ther. 1999; 6: 882–892.
- ↵Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol. 1992; 119: 493–501.
- ↵Kolodgie FD, Burke AP, Farb A, et al. The thin-cap fibroatheroma: a type of vulnerable plaque. The major precursor lesion to acute coronary syndromes. Curr Opin Cardiol. 2001; 16: 285–292.
- ↵Smith RL, Geller AI, Escudero KW, et al. Long-term expression in sensory neurons in tissue culture from herpes simplex virus type 1 (HSV-1) promoters in an HSV-1-derived vector. J Virol. 1995; 69: 4593–4599.
- ↵Soussi T. The p53 tumor suppressor gene: from molecular biology to clinical investigation. Ann N Y Acad Sci. 2000; 910: 121–137.
- ↵Drazan KE, Shen XD, Csete ME, et al. In vivo adenoviral-mediated human p53 tumor suppressor gene transfer and expression in rat liver after resection. Surgery. 1994; 116: 197–203.
- ↵Bennett MR, Evan GI, Newby AC. Deregulated expression of the c-myc oncogene abolishes inhibition of proliferation of rat vascular smooth muscle cells by serum reduction, interferon-γ, heparin, and cyclic nucleotide analogues and induces apoptosis. Circ Res. 1994; 74: 525–536.
- ↵Bennett MR., Gibson DF, Schwartz SM, et al. Binding and phagocytosis of apoptotic vascular smooth muscle cells is mediated in part by exposure of phosphatidylserine. Circ Res. 1995; 77: 1136–1142.
- ↵Li ZR, Hromchak R, Mudipalli A, et al. Tumor suppressor proteins as regulators of cell differentiation. Cancer Res. 1998; 58: 4282–4287.
- ↵Johnson JL, Jackson CL. Atherosclerotic plaque rupture in the apolipoprotein E knockout mouse. Atherosclerosis. 2001; 154: 399–406.
- ↵Calara F, Silvestre M, Casanada F, et al. Spontaneous plaque rupture and secondary thrombosis in apolipoprotein E-deficient and LDL receptor-deficient mice. J Pathol. 2001; 195: 257–263.