Plaque Rupture After Short Periods of Fat Feeding in the Apolipoprotein E–Knockout Mouse
http://www.100md.com
循环学杂志 2005年第3期
the Bristol Heart Institute, University of Bristol, Bristol, United Kingdom.
Abstract
Background— These studies examined the early time course of plaque development and destabilization in the brachiocephalic artery of the apolipoprotein E–knockout mouse, the effects of pravastatin thereon, and the effects of pravastatin on established unstable plaques.
Methods and Results— Male apolipoprotein E–knockout mice were fed a high-fat, cholesterol-enriched diet from the age of 8 weeks. Animals were euthanized at 1-week intervals between 4 and 9 weeks of fat feeding. Acutely ruptured plaques were observed in the brachiocephalic arteries of 3% of animals up to and including 7 weeks of fat feeding but in 62% of animals after 8 weeks, which suggests that there is a sharp increase in the number of plaque ruptures at 8 weeks. These acute plaque ruptures then appear to heal and form buried fibrous caps; after 9 weeks of fat feeding, mice had 1.05±0.15 buried fibrous caps at a single site in the brachiocephalic artery. Pravastatin (40 mg/kg of body weight per day for 9 weeks; resultant plasma concentration 16±4 nmol/L) had no effect on plasma cholesterol concentration in fat-fed apolipoprotein E–knockout mice but reduced the number of buried fibrous caps by 43% (P<0.0001). In longer-term experiments, the delay of pravastatin treatment until unstable plaques had developed reduced the incidence of acute plaque rupture by 36% (P<0.0001).
Conclusions— Plaque rupture occurs at high frequency in the brachiocephalic arteries of male apolipoprotein E–knockout mice after 8 weeks of fat feeding. Pravastatin treatment inhibits early plaque rupture and is also effective when begun after unstable plaques have developed.
Key Words: atherosclerosis ; plaque ; pathology ; drugs
Introduction
Over the years, many animal models of atherosclerosis have been developed, but virtually all of them suffer the disadvantage that they do not progress to plaque rupture accompanied by mural thrombosis. Because this is probably the single most important event in the life history of a human atherosclerotic plaque, the animal models have been of limited value for investigating plaque rupture; therefore, they have in the main been used to study atherogenesis. Recently, we and others1–4 have described features of advanced atherosclerotic lesions in older apolipoprotein E (apoE)–knockout mice that suggest that these lesions can indeed become unstable to the point of rupture, as evidenced by breaches of the fibrous cap with thrombi extending from the lumen to the core.
In those studies,1–4 the mice were fed either normal chow or fat-enriched diet for periods up to 14 months. This extended time period makes intervention studies, and even the study of the pathophysiology of plaque destabilization, slow and difficult; a model in which plaques rupture after a reasonably short period would be much more useful. In the first part of the present report, we describe the development of unstable plaques in the brachiocephalic arteries of male apoE-knockout mice after 8 weeks of fat feeding.
Probably the greatest utility of an animal model of plaque rupture would be the screening and development of potential plaque-stabilizing agents. This requires the animal model reliably to predict the activity of the agent in humans. One way of assessing the likelihood of this happening is to test an agent known to be effective in humans in the animal model in question. For plaque-stabilizing agents, this is difficult, because there is no unequivocal therapy. The best candidates are the statins, which reduce the incidence of clinical cardiovascular events to a greater extent than predicted from the degree of lipid-lowering produced.5,6 It has been suggested that this is a consequence of a direct effect on plaque stability, a hypothesis supported by beneficial changes in the composition of atherosclerotic plaques in patients treated with pravastatin.7 Intravascular ultrasound imaging of human coronary plaques has shown that pravastatin does not induce plaque regression even at doses that do reduce the incidence of clinical events,8 which is also suggestive of a direct stabilizing action. We have therefore tested the effect of pravastatin on plaque stability in the fat-fed apoE-knockout mice in the second part of this report.
Methods
Animals
Homozygous C57BL/6,129 male apoE-knockout mice (strain background 71% C57BL/6 and 29% 129, by microsatellite analysis) were bred within the Animal Unit of the University of Bristol. The housing and care of the animals and all the procedures used in these studies were performed in accordance with the guidelines and regulations of the United Kingdom Home Office.
Husbandry
Starting at 8 weeks of age, mice were switched from normal rodent diet to a high-fat diet that contained 21% fat from lard and was supplemented with 0.15% (wt/wt) cholesterol (Special Diets Services). Animals were inspected at least once every 24 hours.
Drug Treatment
Pravastatin sodium salt was obtained from LKT Laboratories. Analysis by nuclear magnetic resonance, high-performance liquid chromatography, gas chromatography, and mass spectrometry confirmed the material to be 98.3% pure pravastatin sodium. It was administered as a solution in the drinking water (ordinary tap water) at a gravimetric dose of 40 mg/kg of body weight per day, with twice-weekly monitoring of water consumption and adjustment of drug concentration as required. Controls received tap water without pravastatin.
Pravastatin Study Design
Three treatment schedules were used. In the first, animals started to receive pravastatin at the same time as commencement of fat feeding and then were euthanized 9 weeks later. This experiment was originally scheduled to be terminated at 8 weeks but because of staff sickness had to be extended by a further week. The second study was similar to the first, except that scheduled termination was after 40 weeks of treatment. In the third study, pravastatin treatment was not started until 16 weeks of fat feeding had elapsed, and scheduled termination was 24 weeks after that. There was a single control group for the second and third studies. A satellite group of control animals were euthanized after 16 weeks of fat feeding.
Euthanasia
Animals were surgically anesthetized by intraperitoneal injection of sodium pentobarbitone before thorough terminal exsanguination by arterial perfusion via the abdominal aorta with PBS at a constant pressure of 100 mm Hg, with outflow through the severed jugular veins. This was followed by constant pressure perfusion in situ with 10% formalin. Brachiocephalic arteries were removed with a piece of the aortic arch and the stump of the right subclavian artery still attached to aid orientation during histological processing. In some animals, the heart was also removed to study the aortic sinus.
Histology
Brachiocephalic arteries were embedded in paraffin or optimum cutting temperature compound (OCT; BDH Laboratory Supplies). Hearts were embedded in paraffin. Sections were cut at 3 μm for paraffin-embedded sections and 6 μm for OCT-embedded sections. Sections were cut every 30 μm along the brachiocephalic artery (starting from the proximal end) or until the aortic valve cusps were visible for the aortic sinus. Sections were stained with hematoxylin and eosin, Miller’s elastin/van Gieson, Masson’s trichrome, or oil red O.
Plaque Morphology
Sections stained for elastin were inspected for the presence or absence of disruption of the cap of the plaque. Acute plaque rupture was accepted when a visible defect in the cap was accompanied by intrusion of erythrocytes into the plaque below it. In many animals, 1 or more buried fibrous caps (smooth muscle cell–rich layers, invested with elastin, and usually overlain with foam cells) were seen within the body of the plaque. These were also counted.
Plaque Morphometry
All morphometric analyses were made on elastin-stained sections, and 1 vessel cross section was quantified per mouse. Any vessels that were not sectioned exactly perpendicular to the long axis, as revealed by shearing distortion of the elastic laminae, were re-embedded, and fresh sections were cut. Morphometry was performed with a computerized image-analysis program (Image Pro Plus, Media Cybernetics). The lengths of the internal and external elasticae were recorded. These were used to derive the media area, by assuming them to be the circumferences of perfect circles. The plaque area was measured directly and was subtracted from the area enclosed by the internal elastic lamina to derive the true lumen area.
Plaque Lipid Content
Color images of delipidated paraffin sections were digitally processed into monochrome with Image Pro Plus software. This causes any tissue to be rendered as black and delipidated areas to appear as white. The total white area in the plaque was expressed as a percentage of the total plaque area to give the fractional lipid content. The validity of this method was assessed by taking 15 cryostat sections of lesioned brachiocephalic arteries and staining them with oil red O. The area stained red was determined by computerized morphometry, then the same sections were delipidated and stained for elastin before measurement of the lipid content from the monochrome image as described above. The mean values for the lipid content, 46.4±3.2% and 47.1±2.9% for oil red O and elastin-stained sections, respectively, were not significantly different but were highly significantly correlated (Pearson’s r2=0.76; P<0.0001), which shows that this is a valid method for estimating plaque lipid content.
Determination of Plasma Lipid Concentration
Blood samples were drawn from the left ventricle of a cohort of male apoE-knockout mice that had received pravastatin and high-fat diet for 9 weeks. Total cholesterol and phosphotungstic acid/magnesium chloride–precipitated HDL cholesterol were determined enzymatically in heparinized plasma, and VLDL/LDL cholesterol was determined indirectly by Friedewald’s formula.
Determination of Plasma Pravastatin Concentration
Blood samples were drawn from the left ventricle of a cohort of male apoE-knockout mice that had received pravastatin and high-fat diet for 9 weeks. Pravastatin concentrations were determined by high-performance liquid chromatography/mass spectrometry with protein-precipitated plasma samples; a calibration curve was obtained by spiking known amounts of pravastatin into blank mouse plasma.
Immunocytochemistry
Serial 3-μm paraffin sections were dewaxed and rehydrated. Endogenous peroxidase activity was inhibited by incubation with 3% hydrogen peroxide. After sections were blocked with 20% (vol/vol) goat serum in PBS, they were incubated overnight at 4°C either with a rabbit anti-human fibrin antibody (a kind gift from Dr Douglas Thompson, Department of Pathology, University of Aberdeen, United Kingdom) or a mouse monoclonal antibody against -smooth muscle actin (Sigma) in 1% (wt/vol) BSA in PBS. Sections were then incubated with the appropriate secondary antibodies. For the rabbit anti-human fibrin antibody, this was Fluor 488-conjugated (Molecular Probes), which was diluted 1:200 in 1% (wt/vol) BSA in PBS, and sections were mounted with Vectashield aqueous mountant (Vector Laboratories) that contained 4‘,6-diamidino-2-phenylindole to reveal nuclei. For -smooth muscle actin, the Mouse-on-Mouse kit was used (Vector Laboratories). Negative controls had the primary antibodies replaced with rabbit or mouse IgG at the same dilution and were uniformly devoid of staining (data not shown).
Statistical Analysis
Values are expressed as mean±SEM. Treatment group values were compared with their controls with the computer programs InStat and Prism (both GraphPad Software). For the comparison of group means, a check was first made for similar variances; if this was passed, then an unpaired, 2-sample, 2-tailed Student t test was performed, or 1-way ANOVA if there were more than 2 groups. If the variances were significantly different, then an unpaired, 2-sample, 2-tailed t test with Welch’s correction was used for 2 groups and a Kruskal-Wallis test for more than 2 groups. Survival data were compared with the log-rank test. Contingency data (presence or absence of plaque disruption, presence or absence of fibrin staining) were analyzed by Fisher exact test for 2-group comparisons and by the 2 test for 3-group comparisons. Discontinuous data (incidence of buried fibrous layers) were analyzed by the Mann-Whitney test. In all cases, statistical significance was concluded if the 2-tailed probability was <0.05.
Results
Fat Feeding Rapidly Provokes Advanced Atherosclerosis in the Brachiocephalic Artery
The intima of the brachiocephalic artery of the male apoE-knockout mouse was histologically normal for the first 3 weeks of fat feeding, although some cytoplasmic lipid droplets were seen in medial smooth muscle cells. At the fourth week, the first intimal changes became apparent, with the development of small fatty streaks (Figure 1A). These nascent lesions were in the proximal part of the vessel. From the sixth week of fat feeding onward, fibrofatty atherosclerotic lesions with well-defined fibrous caps were present in the brachiocephalic arteries of all animals (Figure 1B). By 8 weeks of fat feeding, the lesions were complex and had a multiply layered appearance (Figure 1C).
Acute Plaque Rupture Is an Early and Frequent Occurrence in the Brachiocephalic Artery
Acute plaque rupture, defined, as in a previous study,4 as a visible breach in the cap with intraplaque hemorrhage intruding into the lesion at the same site, was first seen in the brachiocephalic artery after 5 weeks of fat feeding, although this was in only 1 animal, and no other acute plaque ruptures were seen in 33 other animals that were fed a high-fat diet for periods up to and including 7 weeks (8 or 10 animals at each time point). After 8 weeks of fat feeding, 107 (62%) of 173 animals exhibited acute plaque rupture in the brachiocephalic artery. One week later, this proportion had fallen to 30% (11 of 37 animals), which suggests that there had been a wave of plaque ruptures with a peak around the 8-week time point. All of the ruptured plaques were in the proximal 150 μm of this vessel. Examples of acute plaque ruptures are shown in Figures 1C and 2, and the data are summarized in Figure 3B. No disrupted plaques were seen in aortic sinus sections.
Presence of Healed Plaque Ruptures
Structures were observed within the plaque that closely resembled buried fibrous caps. They were rich in -smooth muscle actin–positive cells (Figure 4A) and elastin. Use of a specific anti-fibrin antibody revealed accumulation of fibrin at the same site (Figure 4B); no such accumulation occurred in plaques without these structures (Figure 4C). After 8 weeks of fat feeding, the number of buried fibrous caps averaged 0.29±0.05 per animal (n=173). During the next 7 days, this rose to 1.05±0.15 per animal (n=37), with some mice showing 2 buried fibrous caps, one above the other, at this site. These data are summarized in Figure 3B.
To test whether the presence of buried fibrous caps and the occurrence of acute plaque rupture are associated events, we examined aortic sinuses in 28 apoE-knockout mice that had been fat fed for up to 40 weeks and had already been found to have an acute plaque rupture in the brachiocephalic artery. These animals were a separate group whose data are not reported elsewhere in this report. The sinus lesions were stained for elastin and inspected for the presence of buried fibrous caps and acute plaque disruption. There were no acute ruptures and no buried caps at this site in these 28 mice, but there were a total of 35 buried fibrous caps in the brachiocephalic arteries (P<0.0001, Fisher exact test), which shows that there is a statistically significant association between acute plaque rupture and the development of buried fibrous caps.
We also examined proximal brachiocephalic artery plaques, taken from a series of 50 male apoE-knockout mice that had been fat fed for 8 weeks, for the presence of fibrin as assessed by bright red staining with Masson’s trichrome. These mice formed a subset of the total of 173 mice that were fat fed for 8 weeks. The mice were selected for the absence of acute plaque rupture. Of these mice, 23 had buried fibrous caps at this site, and 19 of these 23 plaques were positive for fibrin. Among the 27 mice with no buried fibrous caps, 4 stained positively for fibrin (P=0.000001, Fisher exact test), which shows that there is a statistically significant association between buried fibrous cap formation and intraplaque fibrin accumulation.
Pravastatin Has No Effect on Plasma Lipid Levels
Pravastatin treatment for 9 weeks at 40 mg/kg of body weight per day had no effect on total plasma cholesterol concentration (control [n=18] 41.7±3.3 mmol/L, pravastatin [n=13] 41.5±2.8 mmol/L). Pravastatin also had no effect on LDL cholesterol, VLDL/LDL cholesterol, or triglyceride concentration (Table 1).
Pravastatin Has a Weak Inhibitory Effect on Acute Plaque Rupture
The concentration of pravastatin in the plasmas of a group of 15 male apoE-knockout mice fed a high-fat diet for 9 weeks and receiving pravastatin in the drinking water at a dose of 40 mg/kg of body weight per day throughout, was 16±4 nmol/L. This treatment had no effect on the incidence of acute plaque rupture after 9 weeks of fat feeding (Table 2). When animals were kept on a high-fat diet for 40 weeks, the incidence of acute plaque disruption differed significantly between controls and pravastatin-treated mice (control, 24 in 65 [37%]; pravastatin throughout, 4 in 76 [5%]; pravastatin from 16 weeks, 12 in 74 [16%]; P<0.0001). These data are summarized in Tables 2 and 3 .
Pravastatin Reduces the Occurrence of Buried Fibrous Caps
In the present study, after 9 weeks of fat feeding, control male apoE-knockout mice exhibited 1.21±0.13 buried fibrous layers per animal. This was significantly reduced by 43% to 0.69±0.13 in pravastatin-treated animals (P<0.0001). After 40 weeks of fat feeding, controls had 2.72±0.15 buried fibrous layers. Treatment throughout this period with pravastatin reduced this by 54% to 1.26±0.10 (P<0.0001), and delayed treatment from 16 weeks onward reduced it by 36% to 1.73±0.12 (P<0.0001). These data are summarized in Tables 2 and 3.
There is a statistically significant linear correlation between the number of buried fibrous caps and plaque size, which suggests that plaque rupture can drive plaque growth. When this correlation is compared between control and pravastatin-treated animals, the slopes of the lines are similar (47.5 and 42.1x103 μm2/buried cap, respectively), but the intercept points are significantly different (25.6 and 3.8x103 μm2, respectively; P=0.004). These data support the idea that pravastatin can retard plaque initiation and early plaque growth (as well as reduce the likelihood of a plaque rupture occurring), but once a rupture does occur, the amount of plaque expansion that follows is unaffected by pravastatin treatment.
Pravastatin Inhibits Atherogenesis
As shown in Table 2, in the 9-week study, pravastatin reduced plaque lipid content by 42% (P<0.0001) and plaque cross-sectional area by 61% (P<0.0001). Lumen area was increased by 21% (P=0.029). It was not possible to measure fibrous cap thickness in the pravastatin-treated mice because the lesions had not yet progressed to the stage at which a fibrous cap could be defined. When animals were treated continuously for up to 40 weeks, plaque lipid content was reduced by 44% (P<0.001) and plaque cross-sectional area by 19% (P<0.05). Lumen area was unchanged, and fibrous cap thickness was increased 5-fold (P<0.001).
Delayed Pravastatin Treatment Alters the Phenotype but Not the Size of Mature Plaques
As shown in Table 3, pravastatin, given after a 16-week delay, caused a 263% increase in fibrous cap thickness (P<0.001) and a 36% reduction in plaque lipid content (P<0.001). There was no change in plaque or lumen cross-sectional area, which suggests that pravastatin does not induce plaque regression.
Pravastatin Reduces the Incidence of Sudden Death
Because of the possibility that combining animals that survived to scheduled termination with those that suffered sudden death could bias the observations, we compared the data from these 2 categories of animals. In control mice, there were 2.65±0.21 and 2.75±0.18 buried fibrous caps per mouse in those that reached scheduled termination and those that died of sudden death, respectively (P=NS); for pravastatin-treated mice, the values were 1.33±0.13 and 1.14±0.17 buried fibrous caps per mouse, respectively (P=NS). The 49.8% and 58.5% reductions in the number of buried fibrous caps in pravastatin-treated animals compared with controls were both statistically significant.
Discussion
We have shown that plaques in the proximal brachiocephalic arteries of 62% of male apoE-knockout mice suddenly become unstable and rupture after 8 weeks of fat feeding and that this is followed by the formation of buried fibrous layers within the plaque that may represent previous healed plaque ruptures. Treatment with pravastatin reduces the number of the buried fibrous layers. Because inhibition of plaque progression by pravastatin could indirectly reduce the number of plaque ruptures that occur, we also commenced treatment in animals that had already been fat fed for 16 weeks. This reduced the incidence of acute plaque rupture, the number of buried fibrous layers, and the incidence of sudden death. These data suggest that the fat-fed apoE-knockout mouse is a useful model of plaque rupture.
Evidence for Plaque Rupture in ApoE-Knockout Mice
Despite reports from 3 independent laboratories,1–4 the suggestion that apoE-knockout mice suffer plaque rupture has proved contentious. It has been suggested that what has been described as plaque rupture in these mice is really intraplaque hemorrhage,9 presumably occurring either through the fibrous cap or from plaque microvessels. Plaque neovascularization has been reported to occur in aortic sinus lesions in apoE-knockout mice,10 but this has been contradicted by other reports1,11,12 and remains controversial. We have not observed intimal neovascularization in the brachiocephalic arteries of nearly 700 apoE-knockout mice, which suggests that any intraplaque hemorrhage must occur through the fibrous cap; Figures 1C and 2 show examples of precisely this phenomenon. In these brachiocephalic arteries, the formed blood elements within the plaque are continuous with a mural thrombus. This indicates that contact between the blood and the interior of the plaque has provoked thrombosis; the alternative explanation is that postmortem coagulation has somehow eroded the fibrous cap and forced erythrocytes into the lesion, a process for which there is no known mechanism. This suggests that either plaque erosion or plaque rupture was the initiating event. Plaque erosion is defined as loss of endothelium, leading to thrombus formation, without any associated fissure or rupture.13 In our studies, we identify acute plaque rupture by the presence of formed blood elements within the plaque accompanied by a visible defect in the fibrous cap, and therefore we have not included superficial plaque erosions. We conclude that acute plaque ruptures do occur in the brachiocephalic arteries of fat-fed apoE-knockout mice.
Relation Between Buried Fibrous Caps and Previous, Now Healed, Plaque Rupture
We believe that, as in human coronary arteries, acute plaque ruptures in brachiocephalic arteries can heal over and form buried fibrous caps within the plaque14,15; however, it has also been suggested that it may not be necessary for plaque rupture to occur in order for buried fibrous caps to form.16 They could, for example, be the consequence of episodic growth of the lesion, with waves of smooth muscle cell migration from the media occurring at intervals and leading to the layered appearance of plaques like those in Figures 1C and 4B. Alternatively, the migration and/or accumulation of foam cells over an existing cap may be part of normal plaque development in the fat-fed apoE-knockout mouse, again requiring no plaque disruption for it to occur. For such hypotheses to be true, acute plaque rupture and the development of buried fibrous caps would have to be independent events, but this appears not to be the case. In 28 apoE-knockout mice that had been fat fed for up to 40 weeks and had already been found to have an acute plaque rupture in the brachiocephalic artery, aortic sinus lesions contained no acute ruptures and no buried caps (P<0.0001). The absence of buried fibrous caps at a site where acute plaque rupture is rare or absent, and their presence at a site where rupture is relatively frequent, is supportive of the idea that acute plaque rupture is followed by incorporation of the disrupted cap into the growing plaque.
It could be the case that buried fibrous cap formation is a special feature of the brachiocephalic artery that does not occur at sites such as the aortic sinus; however, the significant association between the presence of buried fibrous caps in the brachiocephalic artery and another marker of plaque rupture, fibrin deposition, supports our working hypothesis that buried fibrous caps equate to healed plaque ruptures.
Experimental Assessment of Plaque Stability
The establishment of a link between acute plaque rupture and buried fibrous caps is important because acute plaque rupture is a variable parameter, especially at early time points. In our early time series (Figure 3), there appeared to be a wave of acute plaque ruptures after 8 weeks of fat feeding; the number of acute plaque ruptures at 9 weeks was lower than at 8 weeks. It is easy to see how small differences in the timing of this wave could result in some studies having rather low incidences of acute plaque rupture at the chosen time point, and our 9-week study with pravastatin is a case in point. The incidence of acute plaque rupture in control animals at 9 weeks was 13%, which provides little signal for pravastatin to inhibit. The use of buried fibrous caps as an indicator of the number of previous ruptures was much more robust, with control animals averaging 1.2 ruptures each, a figure significantly reduced by pravastatin treatment (Table 2).
The longer-term studies do not suffer from this problem so much. Presumably this is because with time, individual differences in the timing of repeated episodes of acute plaque rupture mean that what starts out as a relatively coherent wave, with a well-defined peak, becomes a more-constant average signal. In the 40-week study, the incidence of acute plaque rupture in control animals was 37%, and this was significantly reduced by both continuous and delayed pravastatin treatment (Table 3). Therefore, it appears sensible to use buried fibrous cap counts to assess plaque stability at early time points, whereas for long-term studies, acute plaque rupture incidence and buried fibrous cap counts are both useful.
The present data show that for intervention studies, it is important to use both short-term and delayed-treatment protocols. If an intervention inhibits plaque initiation and/or progression, it may appear to have an effect on plaque stability in short-term studies simply because the plaques have not developed to the point at which they can rupture. Delaying the commencement of treatment until the plaques are already fully established and unstable circumvents this problem.
Pharmacodynamic Validation of the Model
In these studies, pravastatin reduced the incidence of plaque rupture in fat-fed apoE-knockout mice at a plasma concentration almost identical to that seen in humans.17 Although it has not yet been established that statins do indeed stabilize vulnerable human plaques, the beneficial effect of pravastatin on plaque disruption in the fat-fed apoE-knockout mouse suggests that the murine events are congruent with human plaque rupture (otherwise it would be necessary to posit that buried fibrous caps are unrelated to plaque instability, that pravastatin serendipitously influences their formation, and that the effect of pravastatin on acute plaque rupture is a statistical fluke). The clinical relevance of the model is further supported by the reduction of plaque lipid content in pravastatin-treated mice (Tables 2 and 3 ), an effect also observed in carotid endarterectomy plaques from patients treated with pravastatin.7
We found that pravastatin reduced the incidence of sudden death in the long-term studies. We have previously reported that plaque rupture in the brachiocephalic artery and sudden death are unrelated events in fat-fed apoE-knockout mice4; however, data from the present long-term studies contradict this. Acute plaque rupture was observed in 50% of control animals that suffered sudden death but in none of the animals that survived to scheduled euthanasia (P<0.0001). (There were no differences in morphometric parameters between survivors and decedents.) Although we do not know why there should be this difference between studies, it means that it is possible that plaque rupture in the brachiocephalic artery does contribute to morbidity and mortality in fat-fed apoE-knockout mice. This is consistent with the reduction by pravastatin of both the incidence of plaque rupture and the incidence of sudden death; however, we cannot be certain of this in the absence of definitive cause-of-death data. We therefore conclude that the pravastatin data support the use of the fat-fed apoE-knockout mouse as a test bed for potential plaque-stabilizing therapies.
Do These Studies Provide Evidence for a Pleiotropic Effect of Pravastatin on Plaque Stability;
Pravastatin had no effect on plasma cholesterol concentration in apoE-knockout mice (Table 1), as has already been shown for simvastatin by Bea et al.18 In the study by Bea et al,18 which used older chow-fed apoE knockouts, the effect of simvastatin on brachiocephalic plaque instability, as manifested by the presence of intraplaque hemorrhage, was rather variable; indeed, treatment periods longer than 6 weeks produced no significant benefits. The robust effect of pravastatin on plaque rupture in the present studies may derive from some difference in plaque development between chow-fed and fat-fed apoE-knockout mice, with the latter being more sensitive to statin treatment. Because there were no changes in plasma lipids, it is clear that the improvements in plaque stability must have resulted from some other action of pravastatin, an effect usually described as pleiotropism.
Conclusions
Plaque ruptures frequently occur in the brachiocephalic arteries of apoE-knockout mice after 8 weeks of fat feeding. We suggest that these early acute plaque ruptures heal over and become incorporated into the lesion as buried fibrous caps, which can be counted to give another estimate of plaque stability. Pravastatin, at a dose that produced a clinically relevant plasma concentration, had no effect on plasma lipid levels. When pravastatin treatment commenced at the same time as fat feeding, it impaired atherogenesis; when treatment was delayed until plaques had already become fully established and unstable, it had no effect on plaque size but reduced plaque lipid content, increased fibrous cap thickness, and decreased the incidence of plaque rupture. These data support the use of the fat-fed apoE-knockout mouse as a model of plaque rupture and indicate that pravastatin has a direct beneficial effect on plaque stability that is unrelated to lipid lowering.
Acknowledgments
This work was supported by the British Heart Foundation. The rabbit anti-human fibrin antibody was a kind gift from Dr Douglas Thompson (Department of Pathology, University of Aberdeen, Scotland). The authors gratefully thank Drs Jonathan Bennett and Laura Johnston, both of Organon Laboratories Ltd, Newhouse, Scotland, for chemical analysis of pravastatin and measurement of pravastatin levels in mouse plasma, respectively.
Footnotes
J. Johnson and K. Carson contributed equally to this work.
References
Johnson JL, Jackson CL. The apolipoprotein E knockout mouse: an animal model of atherosclerotic plaque rupture. Atherosclerosis. 2001; 154: 399–406.
Rosenfeld ME, Polinsky P, Virmani R, Kauser K, Rubanyi G, Schwartz SM. Advanced atherosclerotic lesions in the innominate artery of the apoE knockout mouse. Arterioscler Thromb Vasc Biol. 2000; 20: 2587–2592.
Calara F, Silvestre M, Casanada F, Yuan N, Napoli C, Palinski W. Spontaneous plaque rupture and secondary thrombosis in apolipoprotein E–deficient and LDL receptor–deficient mice. J Pathol. 2001; 195: 257–263.
Williams H, Johnson JL, Carson KGS, Jackson CL. Characteristics of intact and ruptured atherosclerotic plaques in the brachiocephalic arteries of apolipoprotein E knockout mice. Arterioscler Thromb Vasc Biol. 2002; 22: 788–792.
Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994; 344: 1383–1389.
Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW, McKillop JH, Packard CJ. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia: West of Scotland Coronary Prevention Study Group. N Engl J Med. 1995; 333: 1301–1307.
Crisby M, Nordin-Fredriksson G, Shah PK, Yano J, Zhu J, Nilsson J. Pravastatin treatment increases collagen content and decreases lipid content, inflammation, metalloproteinases, and cell death in human carotid plaques: implications for plaque stabilization. Circulation. 2001; 103: 926–933.
Nissen SE, Tuzcu EM, Schoenhagen P, Brown BG, Ganz P, Vogel RA, Crowe T, Howard G, Cooper CJ, Brodie B, Grines CL, DeMaria AN, for the REVERSAL Investigators. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis: a randomized controlled trial. JAMA. 2004; 291: 1071–1080.
Lutgens E, van Suylen RJ, Faber BC, Gijbels MJ, Eurlings PM, Bijnens AP, Cleutjens KB, Heeneman S, Daemen MJ. Atherosclerotic plaque rupture: local or systemic process; Arterioscler Thromb Vasc Biol. 2003; 23: 2123–2130.
Moulton KS, Heller E, Konerding MA, Flynn E, Palinski W, Folkman J. Angiogenesis inhibitors endostatin or TNP-470 reduce intimal neovascularization and plaque growth in apolipoprotein E–deficient mice. Circulation. 1999; 99: 1726–1732.
Silvestre JS, Gojova A, Brun V, Potteaux S, Esposito B, Duriez M, Clergue M, Le Ricousse-Roussanne S, Barateau V, Merval R, Groux H, Tobelem G, Levy B, Tedgui A, Mallat Z. Transplantation of bone marrow-derived mononuclear cells in ischemic apolipoprotein E–knockout mice accelerates atherosclerosis without altering plaque composition. Circulation. 2003; 108: 2839–2842.
Nakata Y, Maeda N. Vulnerable atherosclerotic plaque morphology in apolipoprotein E–deficient mice unable to make ascorbic acid. Circulation. 2002; 105: 1485–1490.
van der Wal AC, Becker AE, van der Loos CM, Das PK. Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of the dominant plaque morphology. Circulation. 1994; 89: 36–44.
Mann J, Davies MJ. Mechanisms of progression in native coronary artery disease: role of healed plaque disruption. Heart. 1999; 82: 265–268.
Burke AP, Kolodgie FD, Farb A, Weber DK, Malcom GT, Smialek J, Virmani R. Healed plaque ruptures and sudden coronary death: evidence that subclinical rupture has a role in plaque progression. Circulation. 2001; 103: 934–940.
Bennett MR. Breaking the plaque: evidence for plaque rupture in animal models of atherosclerosis. Arterioscler Thromb Vasc Biol. 2002; 22: 713–714.
Nishizato Y, Ieiri I, Suzuki H, Kimura M, Kawabata K, Hirota T, Takane H, Irie S, Kusuhara H, Urasaki Y, Urae A, Higuchi S, Otsubo K, Sugiyama Y. Polymorphisms of OATP-C (SLC21A6) and OAT3 (SLC22A8) genes: consequences for pravastatin pharmacokinetics. Clin Pharmacol Ther. 2003; 73: 554–565.
Bea F, Blessing E, Bennett B, Levitz M, Wallace EP, Rosenfeld ME. Simvastatin promotes atherosclerotic plaque stability in apoE-deficient mice independently of lipid lowering. Arterioscler Thromb Vasc Biol. 2002; 22: 1832–1837.(Jason Johnson, MSc; Kevin)
Abstract
Background— These studies examined the early time course of plaque development and destabilization in the brachiocephalic artery of the apolipoprotein E–knockout mouse, the effects of pravastatin thereon, and the effects of pravastatin on established unstable plaques.
Methods and Results— Male apolipoprotein E–knockout mice were fed a high-fat, cholesterol-enriched diet from the age of 8 weeks. Animals were euthanized at 1-week intervals between 4 and 9 weeks of fat feeding. Acutely ruptured plaques were observed in the brachiocephalic arteries of 3% of animals up to and including 7 weeks of fat feeding but in 62% of animals after 8 weeks, which suggests that there is a sharp increase in the number of plaque ruptures at 8 weeks. These acute plaque ruptures then appear to heal and form buried fibrous caps; after 9 weeks of fat feeding, mice had 1.05±0.15 buried fibrous caps at a single site in the brachiocephalic artery. Pravastatin (40 mg/kg of body weight per day for 9 weeks; resultant plasma concentration 16±4 nmol/L) had no effect on plasma cholesterol concentration in fat-fed apolipoprotein E–knockout mice but reduced the number of buried fibrous caps by 43% (P<0.0001). In longer-term experiments, the delay of pravastatin treatment until unstable plaques had developed reduced the incidence of acute plaque rupture by 36% (P<0.0001).
Conclusions— Plaque rupture occurs at high frequency in the brachiocephalic arteries of male apolipoprotein E–knockout mice after 8 weeks of fat feeding. Pravastatin treatment inhibits early plaque rupture and is also effective when begun after unstable plaques have developed.
Key Words: atherosclerosis ; plaque ; pathology ; drugs
Introduction
Over the years, many animal models of atherosclerosis have been developed, but virtually all of them suffer the disadvantage that they do not progress to plaque rupture accompanied by mural thrombosis. Because this is probably the single most important event in the life history of a human atherosclerotic plaque, the animal models have been of limited value for investigating plaque rupture; therefore, they have in the main been used to study atherogenesis. Recently, we and others1–4 have described features of advanced atherosclerotic lesions in older apolipoprotein E (apoE)–knockout mice that suggest that these lesions can indeed become unstable to the point of rupture, as evidenced by breaches of the fibrous cap with thrombi extending from the lumen to the core.
In those studies,1–4 the mice were fed either normal chow or fat-enriched diet for periods up to 14 months. This extended time period makes intervention studies, and even the study of the pathophysiology of plaque destabilization, slow and difficult; a model in which plaques rupture after a reasonably short period would be much more useful. In the first part of the present report, we describe the development of unstable plaques in the brachiocephalic arteries of male apoE-knockout mice after 8 weeks of fat feeding.
Probably the greatest utility of an animal model of plaque rupture would be the screening and development of potential plaque-stabilizing agents. This requires the animal model reliably to predict the activity of the agent in humans. One way of assessing the likelihood of this happening is to test an agent known to be effective in humans in the animal model in question. For plaque-stabilizing agents, this is difficult, because there is no unequivocal therapy. The best candidates are the statins, which reduce the incidence of clinical cardiovascular events to a greater extent than predicted from the degree of lipid-lowering produced.5,6 It has been suggested that this is a consequence of a direct effect on plaque stability, a hypothesis supported by beneficial changes in the composition of atherosclerotic plaques in patients treated with pravastatin.7 Intravascular ultrasound imaging of human coronary plaques has shown that pravastatin does not induce plaque regression even at doses that do reduce the incidence of clinical events,8 which is also suggestive of a direct stabilizing action. We have therefore tested the effect of pravastatin on plaque stability in the fat-fed apoE-knockout mice in the second part of this report.
Methods
Animals
Homozygous C57BL/6,129 male apoE-knockout mice (strain background 71% C57BL/6 and 29% 129, by microsatellite analysis) were bred within the Animal Unit of the University of Bristol. The housing and care of the animals and all the procedures used in these studies were performed in accordance with the guidelines and regulations of the United Kingdom Home Office.
Husbandry
Starting at 8 weeks of age, mice were switched from normal rodent diet to a high-fat diet that contained 21% fat from lard and was supplemented with 0.15% (wt/wt) cholesterol (Special Diets Services). Animals were inspected at least once every 24 hours.
Drug Treatment
Pravastatin sodium salt was obtained from LKT Laboratories. Analysis by nuclear magnetic resonance, high-performance liquid chromatography, gas chromatography, and mass spectrometry confirmed the material to be 98.3% pure pravastatin sodium. It was administered as a solution in the drinking water (ordinary tap water) at a gravimetric dose of 40 mg/kg of body weight per day, with twice-weekly monitoring of water consumption and adjustment of drug concentration as required. Controls received tap water without pravastatin.
Pravastatin Study Design
Three treatment schedules were used. In the first, animals started to receive pravastatin at the same time as commencement of fat feeding and then were euthanized 9 weeks later. This experiment was originally scheduled to be terminated at 8 weeks but because of staff sickness had to be extended by a further week. The second study was similar to the first, except that scheduled termination was after 40 weeks of treatment. In the third study, pravastatin treatment was not started until 16 weeks of fat feeding had elapsed, and scheduled termination was 24 weeks after that. There was a single control group for the second and third studies. A satellite group of control animals were euthanized after 16 weeks of fat feeding.
Euthanasia
Animals were surgically anesthetized by intraperitoneal injection of sodium pentobarbitone before thorough terminal exsanguination by arterial perfusion via the abdominal aorta with PBS at a constant pressure of 100 mm Hg, with outflow through the severed jugular veins. This was followed by constant pressure perfusion in situ with 10% formalin. Brachiocephalic arteries were removed with a piece of the aortic arch and the stump of the right subclavian artery still attached to aid orientation during histological processing. In some animals, the heart was also removed to study the aortic sinus.
Histology
Brachiocephalic arteries were embedded in paraffin or optimum cutting temperature compound (OCT; BDH Laboratory Supplies). Hearts were embedded in paraffin. Sections were cut at 3 μm for paraffin-embedded sections and 6 μm for OCT-embedded sections. Sections were cut every 30 μm along the brachiocephalic artery (starting from the proximal end) or until the aortic valve cusps were visible for the aortic sinus. Sections were stained with hematoxylin and eosin, Miller’s elastin/van Gieson, Masson’s trichrome, or oil red O.
Plaque Morphology
Sections stained for elastin were inspected for the presence or absence of disruption of the cap of the plaque. Acute plaque rupture was accepted when a visible defect in the cap was accompanied by intrusion of erythrocytes into the plaque below it. In many animals, 1 or more buried fibrous caps (smooth muscle cell–rich layers, invested with elastin, and usually overlain with foam cells) were seen within the body of the plaque. These were also counted.
Plaque Morphometry
All morphometric analyses were made on elastin-stained sections, and 1 vessel cross section was quantified per mouse. Any vessels that were not sectioned exactly perpendicular to the long axis, as revealed by shearing distortion of the elastic laminae, were re-embedded, and fresh sections were cut. Morphometry was performed with a computerized image-analysis program (Image Pro Plus, Media Cybernetics). The lengths of the internal and external elasticae were recorded. These were used to derive the media area, by assuming them to be the circumferences of perfect circles. The plaque area was measured directly and was subtracted from the area enclosed by the internal elastic lamina to derive the true lumen area.
Plaque Lipid Content
Color images of delipidated paraffin sections were digitally processed into monochrome with Image Pro Plus software. This causes any tissue to be rendered as black and delipidated areas to appear as white. The total white area in the plaque was expressed as a percentage of the total plaque area to give the fractional lipid content. The validity of this method was assessed by taking 15 cryostat sections of lesioned brachiocephalic arteries and staining them with oil red O. The area stained red was determined by computerized morphometry, then the same sections were delipidated and stained for elastin before measurement of the lipid content from the monochrome image as described above. The mean values for the lipid content, 46.4±3.2% and 47.1±2.9% for oil red O and elastin-stained sections, respectively, were not significantly different but were highly significantly correlated (Pearson’s r2=0.76; P<0.0001), which shows that this is a valid method for estimating plaque lipid content.
Determination of Plasma Lipid Concentration
Blood samples were drawn from the left ventricle of a cohort of male apoE-knockout mice that had received pravastatin and high-fat diet for 9 weeks. Total cholesterol and phosphotungstic acid/magnesium chloride–precipitated HDL cholesterol were determined enzymatically in heparinized plasma, and VLDL/LDL cholesterol was determined indirectly by Friedewald’s formula.
Determination of Plasma Pravastatin Concentration
Blood samples were drawn from the left ventricle of a cohort of male apoE-knockout mice that had received pravastatin and high-fat diet for 9 weeks. Pravastatin concentrations were determined by high-performance liquid chromatography/mass spectrometry with protein-precipitated plasma samples; a calibration curve was obtained by spiking known amounts of pravastatin into blank mouse plasma.
Immunocytochemistry
Serial 3-μm paraffin sections were dewaxed and rehydrated. Endogenous peroxidase activity was inhibited by incubation with 3% hydrogen peroxide. After sections were blocked with 20% (vol/vol) goat serum in PBS, they were incubated overnight at 4°C either with a rabbit anti-human fibrin antibody (a kind gift from Dr Douglas Thompson, Department of Pathology, University of Aberdeen, United Kingdom) or a mouse monoclonal antibody against -smooth muscle actin (Sigma) in 1% (wt/vol) BSA in PBS. Sections were then incubated with the appropriate secondary antibodies. For the rabbit anti-human fibrin antibody, this was Fluor 488-conjugated (Molecular Probes), which was diluted 1:200 in 1% (wt/vol) BSA in PBS, and sections were mounted with Vectashield aqueous mountant (Vector Laboratories) that contained 4‘,6-diamidino-2-phenylindole to reveal nuclei. For -smooth muscle actin, the Mouse-on-Mouse kit was used (Vector Laboratories). Negative controls had the primary antibodies replaced with rabbit or mouse IgG at the same dilution and were uniformly devoid of staining (data not shown).
Statistical Analysis
Values are expressed as mean±SEM. Treatment group values were compared with their controls with the computer programs InStat and Prism (both GraphPad Software). For the comparison of group means, a check was first made for similar variances; if this was passed, then an unpaired, 2-sample, 2-tailed Student t test was performed, or 1-way ANOVA if there were more than 2 groups. If the variances were significantly different, then an unpaired, 2-sample, 2-tailed t test with Welch’s correction was used for 2 groups and a Kruskal-Wallis test for more than 2 groups. Survival data were compared with the log-rank test. Contingency data (presence or absence of plaque disruption, presence or absence of fibrin staining) were analyzed by Fisher exact test for 2-group comparisons and by the 2 test for 3-group comparisons. Discontinuous data (incidence of buried fibrous layers) were analyzed by the Mann-Whitney test. In all cases, statistical significance was concluded if the 2-tailed probability was <0.05.
Results
Fat Feeding Rapidly Provokes Advanced Atherosclerosis in the Brachiocephalic Artery
The intima of the brachiocephalic artery of the male apoE-knockout mouse was histologically normal for the first 3 weeks of fat feeding, although some cytoplasmic lipid droplets were seen in medial smooth muscle cells. At the fourth week, the first intimal changes became apparent, with the development of small fatty streaks (Figure 1A). These nascent lesions were in the proximal part of the vessel. From the sixth week of fat feeding onward, fibrofatty atherosclerotic lesions with well-defined fibrous caps were present in the brachiocephalic arteries of all animals (Figure 1B). By 8 weeks of fat feeding, the lesions were complex and had a multiply layered appearance (Figure 1C).
Acute Plaque Rupture Is an Early and Frequent Occurrence in the Brachiocephalic Artery
Acute plaque rupture, defined, as in a previous study,4 as a visible breach in the cap with intraplaque hemorrhage intruding into the lesion at the same site, was first seen in the brachiocephalic artery after 5 weeks of fat feeding, although this was in only 1 animal, and no other acute plaque ruptures were seen in 33 other animals that were fed a high-fat diet for periods up to and including 7 weeks (8 or 10 animals at each time point). After 8 weeks of fat feeding, 107 (62%) of 173 animals exhibited acute plaque rupture in the brachiocephalic artery. One week later, this proportion had fallen to 30% (11 of 37 animals), which suggests that there had been a wave of plaque ruptures with a peak around the 8-week time point. All of the ruptured plaques were in the proximal 150 μm of this vessel. Examples of acute plaque ruptures are shown in Figures 1C and 2, and the data are summarized in Figure 3B. No disrupted plaques were seen in aortic sinus sections.
Presence of Healed Plaque Ruptures
Structures were observed within the plaque that closely resembled buried fibrous caps. They were rich in -smooth muscle actin–positive cells (Figure 4A) and elastin. Use of a specific anti-fibrin antibody revealed accumulation of fibrin at the same site (Figure 4B); no such accumulation occurred in plaques without these structures (Figure 4C). After 8 weeks of fat feeding, the number of buried fibrous caps averaged 0.29±0.05 per animal (n=173). During the next 7 days, this rose to 1.05±0.15 per animal (n=37), with some mice showing 2 buried fibrous caps, one above the other, at this site. These data are summarized in Figure 3B.
To test whether the presence of buried fibrous caps and the occurrence of acute plaque rupture are associated events, we examined aortic sinuses in 28 apoE-knockout mice that had been fat fed for up to 40 weeks and had already been found to have an acute plaque rupture in the brachiocephalic artery. These animals were a separate group whose data are not reported elsewhere in this report. The sinus lesions were stained for elastin and inspected for the presence of buried fibrous caps and acute plaque disruption. There were no acute ruptures and no buried caps at this site in these 28 mice, but there were a total of 35 buried fibrous caps in the brachiocephalic arteries (P<0.0001, Fisher exact test), which shows that there is a statistically significant association between acute plaque rupture and the development of buried fibrous caps.
We also examined proximal brachiocephalic artery plaques, taken from a series of 50 male apoE-knockout mice that had been fat fed for 8 weeks, for the presence of fibrin as assessed by bright red staining with Masson’s trichrome. These mice formed a subset of the total of 173 mice that were fat fed for 8 weeks. The mice were selected for the absence of acute plaque rupture. Of these mice, 23 had buried fibrous caps at this site, and 19 of these 23 plaques were positive for fibrin. Among the 27 mice with no buried fibrous caps, 4 stained positively for fibrin (P=0.000001, Fisher exact test), which shows that there is a statistically significant association between buried fibrous cap formation and intraplaque fibrin accumulation.
Pravastatin Has No Effect on Plasma Lipid Levels
Pravastatin treatment for 9 weeks at 40 mg/kg of body weight per day had no effect on total plasma cholesterol concentration (control [n=18] 41.7±3.3 mmol/L, pravastatin [n=13] 41.5±2.8 mmol/L). Pravastatin also had no effect on LDL cholesterol, VLDL/LDL cholesterol, or triglyceride concentration (Table 1).
Pravastatin Has a Weak Inhibitory Effect on Acute Plaque Rupture
The concentration of pravastatin in the plasmas of a group of 15 male apoE-knockout mice fed a high-fat diet for 9 weeks and receiving pravastatin in the drinking water at a dose of 40 mg/kg of body weight per day throughout, was 16±4 nmol/L. This treatment had no effect on the incidence of acute plaque rupture after 9 weeks of fat feeding (Table 2). When animals were kept on a high-fat diet for 40 weeks, the incidence of acute plaque disruption differed significantly between controls and pravastatin-treated mice (control, 24 in 65 [37%]; pravastatin throughout, 4 in 76 [5%]; pravastatin from 16 weeks, 12 in 74 [16%]; P<0.0001). These data are summarized in Tables 2 and 3 .
Pravastatin Reduces the Occurrence of Buried Fibrous Caps
In the present study, after 9 weeks of fat feeding, control male apoE-knockout mice exhibited 1.21±0.13 buried fibrous layers per animal. This was significantly reduced by 43% to 0.69±0.13 in pravastatin-treated animals (P<0.0001). After 40 weeks of fat feeding, controls had 2.72±0.15 buried fibrous layers. Treatment throughout this period with pravastatin reduced this by 54% to 1.26±0.10 (P<0.0001), and delayed treatment from 16 weeks onward reduced it by 36% to 1.73±0.12 (P<0.0001). These data are summarized in Tables 2 and 3.
There is a statistically significant linear correlation between the number of buried fibrous caps and plaque size, which suggests that plaque rupture can drive plaque growth. When this correlation is compared between control and pravastatin-treated animals, the slopes of the lines are similar (47.5 and 42.1x103 μm2/buried cap, respectively), but the intercept points are significantly different (25.6 and 3.8x103 μm2, respectively; P=0.004). These data support the idea that pravastatin can retard plaque initiation and early plaque growth (as well as reduce the likelihood of a plaque rupture occurring), but once a rupture does occur, the amount of plaque expansion that follows is unaffected by pravastatin treatment.
Pravastatin Inhibits Atherogenesis
As shown in Table 2, in the 9-week study, pravastatin reduced plaque lipid content by 42% (P<0.0001) and plaque cross-sectional area by 61% (P<0.0001). Lumen area was increased by 21% (P=0.029). It was not possible to measure fibrous cap thickness in the pravastatin-treated mice because the lesions had not yet progressed to the stage at which a fibrous cap could be defined. When animals were treated continuously for up to 40 weeks, plaque lipid content was reduced by 44% (P<0.001) and plaque cross-sectional area by 19% (P<0.05). Lumen area was unchanged, and fibrous cap thickness was increased 5-fold (P<0.001).
Delayed Pravastatin Treatment Alters the Phenotype but Not the Size of Mature Plaques
As shown in Table 3, pravastatin, given after a 16-week delay, caused a 263% increase in fibrous cap thickness (P<0.001) and a 36% reduction in plaque lipid content (P<0.001). There was no change in plaque or lumen cross-sectional area, which suggests that pravastatin does not induce plaque regression.
Pravastatin Reduces the Incidence of Sudden Death
Because of the possibility that combining animals that survived to scheduled termination with those that suffered sudden death could bias the observations, we compared the data from these 2 categories of animals. In control mice, there were 2.65±0.21 and 2.75±0.18 buried fibrous caps per mouse in those that reached scheduled termination and those that died of sudden death, respectively (P=NS); for pravastatin-treated mice, the values were 1.33±0.13 and 1.14±0.17 buried fibrous caps per mouse, respectively (P=NS). The 49.8% and 58.5% reductions in the number of buried fibrous caps in pravastatin-treated animals compared with controls were both statistically significant.
Discussion
We have shown that plaques in the proximal brachiocephalic arteries of 62% of male apoE-knockout mice suddenly become unstable and rupture after 8 weeks of fat feeding and that this is followed by the formation of buried fibrous layers within the plaque that may represent previous healed plaque ruptures. Treatment with pravastatin reduces the number of the buried fibrous layers. Because inhibition of plaque progression by pravastatin could indirectly reduce the number of plaque ruptures that occur, we also commenced treatment in animals that had already been fat fed for 16 weeks. This reduced the incidence of acute plaque rupture, the number of buried fibrous layers, and the incidence of sudden death. These data suggest that the fat-fed apoE-knockout mouse is a useful model of plaque rupture.
Evidence for Plaque Rupture in ApoE-Knockout Mice
Despite reports from 3 independent laboratories,1–4 the suggestion that apoE-knockout mice suffer plaque rupture has proved contentious. It has been suggested that what has been described as plaque rupture in these mice is really intraplaque hemorrhage,9 presumably occurring either through the fibrous cap or from plaque microvessels. Plaque neovascularization has been reported to occur in aortic sinus lesions in apoE-knockout mice,10 but this has been contradicted by other reports1,11,12 and remains controversial. We have not observed intimal neovascularization in the brachiocephalic arteries of nearly 700 apoE-knockout mice, which suggests that any intraplaque hemorrhage must occur through the fibrous cap; Figures 1C and 2 show examples of precisely this phenomenon. In these brachiocephalic arteries, the formed blood elements within the plaque are continuous with a mural thrombus. This indicates that contact between the blood and the interior of the plaque has provoked thrombosis; the alternative explanation is that postmortem coagulation has somehow eroded the fibrous cap and forced erythrocytes into the lesion, a process for which there is no known mechanism. This suggests that either plaque erosion or plaque rupture was the initiating event. Plaque erosion is defined as loss of endothelium, leading to thrombus formation, without any associated fissure or rupture.13 In our studies, we identify acute plaque rupture by the presence of formed blood elements within the plaque accompanied by a visible defect in the fibrous cap, and therefore we have not included superficial plaque erosions. We conclude that acute plaque ruptures do occur in the brachiocephalic arteries of fat-fed apoE-knockout mice.
Relation Between Buried Fibrous Caps and Previous, Now Healed, Plaque Rupture
We believe that, as in human coronary arteries, acute plaque ruptures in brachiocephalic arteries can heal over and form buried fibrous caps within the plaque14,15; however, it has also been suggested that it may not be necessary for plaque rupture to occur in order for buried fibrous caps to form.16 They could, for example, be the consequence of episodic growth of the lesion, with waves of smooth muscle cell migration from the media occurring at intervals and leading to the layered appearance of plaques like those in Figures 1C and 4B. Alternatively, the migration and/or accumulation of foam cells over an existing cap may be part of normal plaque development in the fat-fed apoE-knockout mouse, again requiring no plaque disruption for it to occur. For such hypotheses to be true, acute plaque rupture and the development of buried fibrous caps would have to be independent events, but this appears not to be the case. In 28 apoE-knockout mice that had been fat fed for up to 40 weeks and had already been found to have an acute plaque rupture in the brachiocephalic artery, aortic sinus lesions contained no acute ruptures and no buried caps (P<0.0001). The absence of buried fibrous caps at a site where acute plaque rupture is rare or absent, and their presence at a site where rupture is relatively frequent, is supportive of the idea that acute plaque rupture is followed by incorporation of the disrupted cap into the growing plaque.
It could be the case that buried fibrous cap formation is a special feature of the brachiocephalic artery that does not occur at sites such as the aortic sinus; however, the significant association between the presence of buried fibrous caps in the brachiocephalic artery and another marker of plaque rupture, fibrin deposition, supports our working hypothesis that buried fibrous caps equate to healed plaque ruptures.
Experimental Assessment of Plaque Stability
The establishment of a link between acute plaque rupture and buried fibrous caps is important because acute plaque rupture is a variable parameter, especially at early time points. In our early time series (Figure 3), there appeared to be a wave of acute plaque ruptures after 8 weeks of fat feeding; the number of acute plaque ruptures at 9 weeks was lower than at 8 weeks. It is easy to see how small differences in the timing of this wave could result in some studies having rather low incidences of acute plaque rupture at the chosen time point, and our 9-week study with pravastatin is a case in point. The incidence of acute plaque rupture in control animals at 9 weeks was 13%, which provides little signal for pravastatin to inhibit. The use of buried fibrous caps as an indicator of the number of previous ruptures was much more robust, with control animals averaging 1.2 ruptures each, a figure significantly reduced by pravastatin treatment (Table 2).
The longer-term studies do not suffer from this problem so much. Presumably this is because with time, individual differences in the timing of repeated episodes of acute plaque rupture mean that what starts out as a relatively coherent wave, with a well-defined peak, becomes a more-constant average signal. In the 40-week study, the incidence of acute plaque rupture in control animals was 37%, and this was significantly reduced by both continuous and delayed pravastatin treatment (Table 3). Therefore, it appears sensible to use buried fibrous cap counts to assess plaque stability at early time points, whereas for long-term studies, acute plaque rupture incidence and buried fibrous cap counts are both useful.
The present data show that for intervention studies, it is important to use both short-term and delayed-treatment protocols. If an intervention inhibits plaque initiation and/or progression, it may appear to have an effect on plaque stability in short-term studies simply because the plaques have not developed to the point at which they can rupture. Delaying the commencement of treatment until the plaques are already fully established and unstable circumvents this problem.
Pharmacodynamic Validation of the Model
In these studies, pravastatin reduced the incidence of plaque rupture in fat-fed apoE-knockout mice at a plasma concentration almost identical to that seen in humans.17 Although it has not yet been established that statins do indeed stabilize vulnerable human plaques, the beneficial effect of pravastatin on plaque disruption in the fat-fed apoE-knockout mouse suggests that the murine events are congruent with human plaque rupture (otherwise it would be necessary to posit that buried fibrous caps are unrelated to plaque instability, that pravastatin serendipitously influences their formation, and that the effect of pravastatin on acute plaque rupture is a statistical fluke). The clinical relevance of the model is further supported by the reduction of plaque lipid content in pravastatin-treated mice (Tables 2 and 3 ), an effect also observed in carotid endarterectomy plaques from patients treated with pravastatin.7
We found that pravastatin reduced the incidence of sudden death in the long-term studies. We have previously reported that plaque rupture in the brachiocephalic artery and sudden death are unrelated events in fat-fed apoE-knockout mice4; however, data from the present long-term studies contradict this. Acute plaque rupture was observed in 50% of control animals that suffered sudden death but in none of the animals that survived to scheduled euthanasia (P<0.0001). (There were no differences in morphometric parameters between survivors and decedents.) Although we do not know why there should be this difference between studies, it means that it is possible that plaque rupture in the brachiocephalic artery does contribute to morbidity and mortality in fat-fed apoE-knockout mice. This is consistent with the reduction by pravastatin of both the incidence of plaque rupture and the incidence of sudden death; however, we cannot be certain of this in the absence of definitive cause-of-death data. We therefore conclude that the pravastatin data support the use of the fat-fed apoE-knockout mouse as a test bed for potential plaque-stabilizing therapies.
Do These Studies Provide Evidence for a Pleiotropic Effect of Pravastatin on Plaque Stability;
Pravastatin had no effect on plasma cholesterol concentration in apoE-knockout mice (Table 1), as has already been shown for simvastatin by Bea et al.18 In the study by Bea et al,18 which used older chow-fed apoE knockouts, the effect of simvastatin on brachiocephalic plaque instability, as manifested by the presence of intraplaque hemorrhage, was rather variable; indeed, treatment periods longer than 6 weeks produced no significant benefits. The robust effect of pravastatin on plaque rupture in the present studies may derive from some difference in plaque development between chow-fed and fat-fed apoE-knockout mice, with the latter being more sensitive to statin treatment. Because there were no changes in plasma lipids, it is clear that the improvements in plaque stability must have resulted from some other action of pravastatin, an effect usually described as pleiotropism.
Conclusions
Plaque ruptures frequently occur in the brachiocephalic arteries of apoE-knockout mice after 8 weeks of fat feeding. We suggest that these early acute plaque ruptures heal over and become incorporated into the lesion as buried fibrous caps, which can be counted to give another estimate of plaque stability. Pravastatin, at a dose that produced a clinically relevant plasma concentration, had no effect on plasma lipid levels. When pravastatin treatment commenced at the same time as fat feeding, it impaired atherogenesis; when treatment was delayed until plaques had already become fully established and unstable, it had no effect on plaque size but reduced plaque lipid content, increased fibrous cap thickness, and decreased the incidence of plaque rupture. These data support the use of the fat-fed apoE-knockout mouse as a model of plaque rupture and indicate that pravastatin has a direct beneficial effect on plaque stability that is unrelated to lipid lowering.
Acknowledgments
This work was supported by the British Heart Foundation. The rabbit anti-human fibrin antibody was a kind gift from Dr Douglas Thompson (Department of Pathology, University of Aberdeen, Scotland). The authors gratefully thank Drs Jonathan Bennett and Laura Johnston, both of Organon Laboratories Ltd, Newhouse, Scotland, for chemical analysis of pravastatin and measurement of pravastatin levels in mouse plasma, respectively.
Footnotes
J. Johnson and K. Carson contributed equally to this work.
References
Johnson JL, Jackson CL. The apolipoprotein E knockout mouse: an animal model of atherosclerotic plaque rupture. Atherosclerosis. 2001; 154: 399–406.
Rosenfeld ME, Polinsky P, Virmani R, Kauser K, Rubanyi G, Schwartz SM. Advanced atherosclerotic lesions in the innominate artery of the apoE knockout mouse. Arterioscler Thromb Vasc Biol. 2000; 20: 2587–2592.
Calara F, Silvestre M, Casanada F, Yuan N, Napoli C, Palinski W. Spontaneous plaque rupture and secondary thrombosis in apolipoprotein E–deficient and LDL receptor–deficient mice. J Pathol. 2001; 195: 257–263.
Williams H, Johnson JL, Carson KGS, Jackson CL. Characteristics of intact and ruptured atherosclerotic plaques in the brachiocephalic arteries of apolipoprotein E knockout mice. Arterioscler Thromb Vasc Biol. 2002; 22: 788–792.
Scandinavian Simvastatin Survival Study Group. Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet. 1994; 344: 1383–1389.
Shepherd J, Cobbe SM, Ford I, Isles CG, Lorimer AR, MacFarlane PW, McKillop JH, Packard CJ. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia: West of Scotland Coronary Prevention Study Group. N Engl J Med. 1995; 333: 1301–1307.
Crisby M, Nordin-Fredriksson G, Shah PK, Yano J, Zhu J, Nilsson J. Pravastatin treatment increases collagen content and decreases lipid content, inflammation, metalloproteinases, and cell death in human carotid plaques: implications for plaque stabilization. Circulation. 2001; 103: 926–933.
Nissen SE, Tuzcu EM, Schoenhagen P, Brown BG, Ganz P, Vogel RA, Crowe T, Howard G, Cooper CJ, Brodie B, Grines CL, DeMaria AN, for the REVERSAL Investigators. Effect of intensive compared with moderate lipid-lowering therapy on progression of coronary atherosclerosis: a randomized controlled trial. JAMA. 2004; 291: 1071–1080.
Lutgens E, van Suylen RJ, Faber BC, Gijbels MJ, Eurlings PM, Bijnens AP, Cleutjens KB, Heeneman S, Daemen MJ. Atherosclerotic plaque rupture: local or systemic process; Arterioscler Thromb Vasc Biol. 2003; 23: 2123–2130.
Moulton KS, Heller E, Konerding MA, Flynn E, Palinski W, Folkman J. Angiogenesis inhibitors endostatin or TNP-470 reduce intimal neovascularization and plaque growth in apolipoprotein E–deficient mice. Circulation. 1999; 99: 1726–1732.
Silvestre JS, Gojova A, Brun V, Potteaux S, Esposito B, Duriez M, Clergue M, Le Ricousse-Roussanne S, Barateau V, Merval R, Groux H, Tobelem G, Levy B, Tedgui A, Mallat Z. Transplantation of bone marrow-derived mononuclear cells in ischemic apolipoprotein E–knockout mice accelerates atherosclerosis without altering plaque composition. Circulation. 2003; 108: 2839–2842.
Nakata Y, Maeda N. Vulnerable atherosclerotic plaque morphology in apolipoprotein E–deficient mice unable to make ascorbic acid. Circulation. 2002; 105: 1485–1490.
van der Wal AC, Becker AE, van der Loos CM, Das PK. Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of the dominant plaque morphology. Circulation. 1994; 89: 36–44.
Mann J, Davies MJ. Mechanisms of progression in native coronary artery disease: role of healed plaque disruption. Heart. 1999; 82: 265–268.
Burke AP, Kolodgie FD, Farb A, Weber DK, Malcom GT, Smialek J, Virmani R. Healed plaque ruptures and sudden coronary death: evidence that subclinical rupture has a role in plaque progression. Circulation. 2001; 103: 934–940.
Bennett MR. Breaking the plaque: evidence for plaque rupture in animal models of atherosclerosis. Arterioscler Thromb Vasc Biol. 2002; 22: 713–714.
Nishizato Y, Ieiri I, Suzuki H, Kimura M, Kawabata K, Hirota T, Takane H, Irie S, Kusuhara H, Urasaki Y, Urae A, Higuchi S, Otsubo K, Sugiyama Y. Polymorphisms of OATP-C (SLC21A6) and OAT3 (SLC22A8) genes: consequences for pravastatin pharmacokinetics. Clin Pharmacol Ther. 2003; 73: 554–565.
Bea F, Blessing E, Bennett B, Levitz M, Wallace EP, Rosenfeld ME. Simvastatin promotes atherosclerotic plaque stability in apoE-deficient mice independently of lipid lowering. Arterioscler Thromb Vasc Biol. 2002; 22: 1832–1837.(Jason Johnson, MSc; Kevin)