Author + information
- Received July 3, 2007
- Revision received September 10, 2007
- Accepted September 17, 2007
- Published online March 18, 2008.
- Borja Ibanez, MD⁎,
- Gemma Vilahur, DVM, PhD†,
- Giovanni Cimmino, MD⁎,
- Walter S. Speidl, MD⁎,
- Antonio Pinero, MD⁎,
- Brian G. Choi, MD⁎,
- M. Urooj Zafar, MD⁎,
- Carlos G. Santos-Gallego, MD⁎,
- Brian Krause, PhD‡,
- Lina Badimon, PhD† and
- Valentin Fuster, MD, PhD, FACC⁎
- Juan J. Badimon, PhD, FACC⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Juan J. Badimon, Mount Sinai School of Medicine, One Gustave L. Levy Place, Box 1030, New York, New York 10029.
Objectives This study sought to assess the effect of short-term apolipoprotein (apo) A-IMilano administration on plaque size and on suspected markers of plaque vulnerability.
Background Long-term lipid-lowering interventions can regress and stabilize atherosclerotic plaques. However, the majority of recurrent events occur early after the first episode. Interventions able to acutely induce plaque regression and stabilization are lacking. Regression of human coronary lesions after 5 weeks of treatment with apoA-IMilano administration has been shown. However, there are no data regarding its effect on plaque vulnerability.
Methods Advanced aortic lesions were induced in New Zealand White rabbits (n = 40). Plaque size was assessed by magnetic resonance imaging (MRI) at the end of atherosclerosis induction. Animals were randomized to placebo or apoA-IMilano phospholipids (ETC-216), 2 infusions 4 days apart. After the last dose, another MRI study was performed and aortas were processed for cellular composition and gene protein expression of markers associated with plaque instability.
Results Pre-treatment MRI showed similar plaque size in both groups, whereas post-treatment MRI showed 6% smaller plaques in apoA-IMilano–treated animals compared with placebo (p = 0.026). The apoA-IMilano treatment induced a 5% plaque regression (p = 0.003 vs. pre-treatment), whereas the placebo showed no significant effect. Plaque regression by apoA-IMilano was associated with a reduction in plaque macrophage density and a significant down-regulation in gene and protein expression of tissue factor, monocyte chemoattractant protein-1, and cyclooxygenase-2, as well as marked decrease in gelatinolytic activity. Conversely, cyclooxygenase-1 was significantly up-regulated.
Conclusions Acute plaque regression observed after short-term apoA-IMilano administration was associated with a significant reduction in suspected makers of plaque vulnerability in an experimental model of atherosclerosis.
Lipid-lowering interventions have shown that atherosclerosis is not necessarily a constant cumulative process, as it is feasible to stop or even regress it (1–3). Initial studies showed that atherosclerosis progression can be halted by low-density-lipoprotein cholesterol lowering (2). More recent evidence has suggested that increasing high-density-lipoprotein (HDL) cholesterol can further regress atherosclerotic lesions (3) The latter observation, added to other pre-clinical (4,5) and clinical work (6), set up the basis for the HDL-raising approaches in the treatment of atherosclerotic disease.
Acute complications of atherothrombosis are mostly secondary to plaque disruption with superimposed thrombus formation. Several post-mortem studies have strongly suggested the importance of plaque composition, rather than its stenotic severity, in the clinical manifestations of atherosclerosis. An increased macrophage density combined with the presence of high gelatinolytic activity is among the pathological features associated with plaque vulnerability. In addition, it is known that certain components are clearly associated with high-risk plaques, especially because on plaque rupture the local environment is highly pro-coagulant.
Statins have been shown to reduce cardiovascular events and even to induce plaque stabilization (7), but these plaque-stabilizing effects were always seen after long-term treatments. However, the majority of recurrent cardiovascular events take place within the first weeks after the initial episode (8).
Apolipoprotein (apo) A-IMilano is a mutant form of apoA-I associated with low incidence of cardiovascular disease. Recombinant-ApoA-IMilano (rApoA-IM) has been tested in numerous pre-clinical studies showing a reduced plaque size and lower lipid and macrophage plaque content compared with control subjects (9,10). It has also been tested in humans, showing a plaque reduction after 5 weekly injections (6). What remains to be determined is whether this rapid plaque shrinkage is associated with plaque stabilization.
In this study we have aimed to study the effects of rApoA-IM not only on plaque size but also on its composition and activity. In a model of advanced human-like atherosclerotic lesions, short-term rApoA-IM administration induced an acute and significant regression of the lesions. Plaque regression was associated with cellular and molecular changes, suggesting a plaque stabilizing effect.
See the Online Appendix for an expanded methods section.
The atherosclerotic plaques were induced in the abdominal aortas of New Zealand White rabbits (n = 40). At the end of the atherosclerosis induction, all animals underwent a pre-treatment magnetic resonance imaging (MRI) study for plaque size quantification. Animals were then randomized to receive 2 intravenous injections (75 mg/kg), 4 days apart, of rApoA-IM phospholipids (ETC-216, n = 22) (Pfizer, Groton, Connecticut) or equal volume of placebo (n = 18). Four days after the last dose, a second MRI study was performed. Subsequently, rabbits were euthanized and aortas were processed for further analyses.
The study protocol was approved by an institutional research committee, and animals received humane care in compliance with the Guide for the Care and Use of Laboratory Animals.
On MRI, sequential transverse images (3-mm thickness) of the 5 cm of abdominal aorta immediately distal to the celiac trunk were obtained. The initial and final images were matched for anatomical position as previously described (11). Cross-sectional areas of the lumen and vessel wall were determined by a validated semiautomatic quantification method (12).
The mean values for each rabbit were considered for statistical analysis. A secondary analysis included the effect of treatments in the most diseased lesion, which was defined as the largest plaque in 3 consecutive segments on initial MRI study.
Histological sections were stained with Masson trichrome-elastin stain, rabbit alveolar macrophage (RAM)-11 (macrophages), and alpha-actin (vascular smooth muscle cells). Tissue factor (TF), monocyte chemoattractant protein (MCP)-1, cyclooxygenase (COX)-1, and COX-2 antigen expression were analyzed by Western blot analysis. Polymerase chain reaction was used to assess messenger ribonucleic acid expression of the same markers. Gelatinase activity of matrix metalloproteinase (MMP)-2 was assessed in protein extracts from atherosclerotic plaques by zymography.
Continuous variables are expressed as mean ± standard deviation. Statistical comparisons of means were made by Student paired and unpaired t tests for normally distributed variables. For nonnormally distributed variables, Wilcoxon and Mann-Whitney U tests were applied appropriately. A value of p < 0.05 (2-tailed) was considered statistically significant. The investigators had full access to the data and take responsibility for its integrity. All investigators have read and agree to the article as written.
Circulating levels of rApoA-IM peaked 30 min after ETC-216 infusion. Thereafter there was a slow reduction in circulating rApoA-IM levels. The increase in rApoA-IM levels was still significant up to 48 h post-administration (Online Appendix).
Effect of rApoA-IMilano on vessel wall
Results of the effect of the treatments on vessel wall measurements assessed by MRI are presented in Table 1. Before treatments, there were no differences in plaque size between the 2 groups. Post-treatment MRI showed that plaque size was 6.2% smaller in animals receiving rApoA-IM compared with placebo (p = 0.026). The administration of rApoA-IM resulted in a 5% plaque regression (p = 0.003 vs. initial MRI), whereas a nonsignificant effect was observed in the placebo group. Figure 1 shows an example of plaque regression after rApoA-IM administration. A similar effect was observed on the most diseased lesion: a statistically significant 6% regression was observed in the rApoA-IM group, and no significant effect in placebo. Despite the significant changes observed in plaque size, there were no significant differences in the luminal size when comparing the pre-treatment and post-treatment values in either group.
Plaque cellular composition and gelatinolytic activity
There was a significantly reduced macrophage density in the lesions of the rApoA-IM group compared with placebo. A 53% reduction in RAM-11+ cells was observed in the rApoA-IM group compared with placebo (0.7 ± 0.2% vs. 1.5 ± 0.5%, p = 0.008). The smooth muscle cell-to-macrophage ratio was 2.1-fold higher in the aortas of the rApoA-IM–treated animals (p < 0.05). On zymography, MMP-2 activity was significantly lower in rApoA-IM–treated animals compared with placebo (p < 0.001) (Fig. 2).
Molecular footprint of rApoA-IMilano administration
At the gene level, rApoA-IM significantly down-regulated the expression of TF, MCP-1, and COX-2 in the atherosclerotic lesions (p = 0.01, p = 0.04, and p = 0.03 vs. placebo, respectively). Concordant observations were rendered at the antigen level. The rApoA-IM down-regulated the protein expression of these parameters (p = 0.03, p = 0.029, and p = 0.02 vs. placebo, respectively) (Fig. 3).
Conversely, we found an up-regulation in the gene and protein expression of markers associated with normalization of endothelial activity. We detected increased gene and antigen COX-1 expression levels in the aortic lesions of the rApoA-IM group (p < 0.05 vs. placebo for both comparisons). The COX-1 and -2 gene expression is shown in Figure 4.
See the Online Appendix for an expanded discussion section.
We report the acute and significant regression of advanced atherosclerotic lesions by short-term administration of rApoA-IM by MRI. Furthermore, the shrinkage of the lesions was combined with a profound change in the molecular footprint of the atherosclerotic lesions, suggesting a change into a more stable phenotype. To date, this is the first study correlating the in vivo plaque regression with changes in markers of plaque instability (both at the gene expression level and at the protein level) after rApoA-IM treatment.
The effects of apoA-IMilano on plaque size have been previously reported by other investigators using different models of atherosclerosis (6,9,10). We report similar observations on plaque size, but in a model of more advanced atherosclerotic lesions more similar to human pathology (5).
We found a significant plaque regression with no effect on lumen size. The advent of novel noninvasive imaging modalities, such as MRI, enables the study of the real impact of different approaches on the actual plaque size. We previously described the use of MRI to monitor plaque progression and regression by inducing atherosclerotic lesions, as described in the present work, and a subsequent withdrawal of the lipid overfeeding (13).
The thrombogenic potential of disrupted atherosclerotic plaques seems to be highly modulated by their TF content (14). The significant reduction of TF expression at the vascular levels observed in the animals receiving rApoA-IM strongly suggests a reduction in the thrombogenic potential of the residual lesions in the case of an eventual disruption.
The reduction in MCP-1 expression seen in our work might have critical implications because MCP-1 is implicated in the genesis and perpetuation of the atherosclerotic process (via progressive monocyte recruitment and inflammation environment generation). Its reduction may stop the vicious cycle of atherosclerosis perpetuation. In addition, MCP-1 down-regulation has been associated with the prevention of in-stent restenosis (15). In this context, Kaul et al. (16) reported an inhibitory effect on neointimal formation post-stenting by the administration of apoA-IMilano, something that could have been secondary to the MCP-1 down-regulation by rApoA-IM administration.
The results found in this work on COX expression suggest a dual beneficial effect of rApoA-IM administration: on one hand, anti-inflammatory (down-regulation of COX-2), and on the other hand, a normalization of endothelial dysfunction (COX-1 overexpression).
Degradation of the matrix components by certain proteases (MMP) is believed to be a major player in plaque instability. Atherosclerotic plaques with high MMP activity are considered vulnerable. In the present work, rApoA-IM administration resulted in a dramatic reduction in plaque gelatinolytic activity, as evidenced by lower MMP-2 activity. It was previously reported that HDL infusion was associated with lower MMP antigen levels in rabbit atherosclerotic plaques (17). The present work extends this finding, providing changes in MMP activity after treatment with rApoA-IM.
Currently there are several interventions that have been shown to induce plaque stabilization; however, the time needed to achieve this was relatively long (months). It is important to remember that the majority of recurrent events take place in the early phase after the first ischemic event (8). In this regard, statins have been tested early after an acute coronary syndrome, showing a reduction in recurrent events in some studies (18). Our results in an animal model of atherosclerosis suggest that an acute loading intravenous regimen of rApoA-IM might induce a rapid regression and stabilization of high-risk atherosclerotic plaques.
The recent failure of a cholesteryl ester transfer protein (CETP) inhibitor shed doubt on the impact of increasing HDL for the treatment of atherosclerosis (19). Despite the fact that both approaches (CETP inhibition and apoA-I infusion) might be considered HDL-raising therapies, the mechanisms of action are different. The reconstituted lipid-poor HDL used in the current study (rApoA-IMilano/phospholipids) is a highly active molecule in removing cholesterol from extrahepatic tissues, whereas the fate and activity of the large, cholesterol-rich HDL particle generated by the CETP inhibition is not completely understood (20).
The present work does not address whether rApoA-IM is more effective than native apoA-I in regressing and stabilizing atherosclerotic lesions. Different lines of clinical evidence suggest that not all apoA-I/HDL compounds are equally efficient, even at equivalent doses (6,21). The strongest evidence came from Shah’s group (22), showing that macrophage-specific expression of the apoA-IM gene was more effective than wild-type apoA-I in reducing atherosclerosis and plaque inflammation despite comparable circulating levels of the transgene and lipid profile.
The authors thank Noemi Escalera for the proper conduct of the experimental work.
For the expanded versions of the Methods and Discussion sections as well as supplemental tables, please see the online version of this article.
This work has been partially funded by Pfizer Research & Development. Dr. Ibanez has received a grant from the Working Group on Ischemic Heart Disease of the Spanish Society of Cardiology. Dr. Vilahur has received a grant from the Science-Education Spanish Ministry. Dr. Badimon has served as a consultant for Pfizer.
- Abbreviations and Acronyms
- cholesteryl ester transfer protein
- recombinant apolipoprotein A-IMilano and 1-palmitoyl-2-oleoyl phosphatidylcholine complexes
- high-density lipoprotein
- monocyte chemoattractant protein
- matrix metalloproteinase
- magnetic resonance imaging
- recombinant apolipoprotein A-IMilano
- tissue factor
- Received July 3, 2007.
- Revision received September 10, 2007.
- Accepted September 17, 2007.
- American College of Cardiology Foundation
- Corti R.,
- Fuster V.,
- Fayad Z.A.,
- et al.
- Shah P.K.,
- Yano J.,
- Reyes O.,
- et al.
- Chiesa G.,
- Monteggia E.,
- Marchesi M.,
- et al.
- Corti R.,
- Osende J.I.,
- Fallon J.T.,
- et al.
- Helft G.,
- Worthley S.G.,
- Fuster V.,
- et al.
- Badimon J.J.,
- Lettino M.,
- Toschi V.,
- et al.
- Joner M.,
- Farb A.,
- Cheng Q.,
- et al.
- Kaul S.,
- Rukshin V.,
- Santos R.,
- et al.
- Nicholls S.J.,
- Cutri B.,
- Worthley S.G.,
- et al.
- Wang L.,
- Sharifi B.G.,
- Pan T.,
- Song L.,
- Yukht A.,
- Shah P.K.