Recapitulating the complex pathology of atherosclerosis; Which model to use?

Atherosclerosis, defined as the formation of fibrofatty lesions within the artery, is a leading cause of morbidity and mortality worldwide and a major contributing factor to myocardial infarction, stroke and peripheral arterial disease [1]. Heart disease, which is often attributed to atherosclerotic disease of the coronary arteries, is the leading cause of death in the United States, with more than 35% of Americans over the age of 20 having some form of cardiovascular illness, underscoring the urgent need for feasible therapeutic strategies for the treatment of atherosclerosis [2].

In the current issue of Circulation Research, Wei et al [3] have sought to target the pathogenic phenotype switching of vascular smooth muscle cells (vSMCs) to attenuate high cholesterol diet induced atherosclerosis. Rabbits, which better mimic the unique lipoprotein metabolism of humans have been utilized in the aforementioned study instead of a murine model, to better address the question of cholesterol driven atherosclerosis. While mice continue to be the model of choice to decipher underlying genetic and epigenetic induced mechanisms in the progression of atherosclerosis, the authors have used the transgenic (Tg) rabbit model in order to better recapitulate metabolic and lipid changes within atherosclerosis. Previously described aberrant expression of peroxisome proliferator activated receptor gamma coactivator 1 alpha (PGC-1α), has been suggested to contribute to atherosclerosis progression through control of mitochondrial biogenesis, which regulates numerous cellular functions [4]. While primarily investigated for its therapeutic potential in obesity and skeletal muscle differentiation [5], the multifaceted nature of PGC-1α in both endothelial and smooth muscle regulation makes it an interesting target in atherosclerosis, in which inflammation acts as a key driver [5]. Wei and co-authors demonstrate via the utilization of a SM22α-promotor driven PGC-1α smooth muscle conditional overexpression Tg rabbits, that PGC-1α expression can ameliorate atherosclerosis progression via the inhibition of vSMC phenotypic switching, thus promoting an atheroprotective phenotype. Tg rabbits on a high cholesterol diet exhibited a more contractile phenotype within the aortic smooth muscle; characterised both histologically and functionally. The ability for vSMCs to retain a contractile phenotype resulted in decreased macrophage infiltration, less inflammation, reduced ROS production and a decrease in both senescence and proliferation within the intima, contributing to a reduced atherosclerotic burden. More specifically, the authors elucidate an ERK1/2 dependent ELK1 phosphorylation mechanism, which allows vSMCs to maintain a contractile phenotype in both human and rabbit vSMCs in vitro. In addition, Wei et al found that moderate PGC-1α overexpression did not disrupt plasma lipid levels, or the characteristics of arteries in Tg rabbits fed a normal diet, suggesting local targeting of pro-atherosclerotic pathways may indeed be translatable. Furthermore, while both non-Tg and Tg rabbits exhibited hypercholesteremia whilst on high-cholesterol diet, the plaque burden in the Tg rabbits was significantly reduced. In human atherosclerotic plaques examined by the authors, PGC-1α was substantially lower in the medial layer of atherosclerotic vessels compared to non-diseased specimens. In addition, genes promoting the vSMC phenotypic switch to contractile vSMCs were negatively correlated with the presence of PGC-1α, again suggesting a possible translational role of PGC-1α in both rabbit atherosclerosis and humans. Most interestingly, perhaps is the comparison between human and rabbit plaques. Rabbits on a diet containing >0.5% cholesterol for four months had plaques containing predominantly monocyte/macrophage enriched foam cells in the arterial intima, which are comparable to early human lesions. However, rabbits fed a diet containing <0.5% cholesterol for eight months exhibit plaques similar to that of stable and more advanced human atherosclerotic plaques, consisting of large fibrous cap [6]. To mimic an early lesion within the rabbits, the authors chose to focus on attenuation of atherosclerosis development, rather than regression, which begs the question whether PGC-1α will have a similar impact in humans, who do not receive treatment prior to a clinical atherosclerosis diagnosis. Furthermore, with the ubiquity of SMCs throughout the body, further thought should be given to how PGC-1α overexpression may impact other organs and non-diseased vascular beds, and whether targeting the smooth muscle cell is in fact viable in such a complex organism as humans [7].

Although this manuscript thoroughly elucidates the role of PGC-1α in high cholesterol driven atherosclerosis development, it raises some thought-provoking questions for the wider atherosclerosis field: 1) Has translational research outgrown the traditional murine model of atherosclerosis? 2) How central is recapitulating the lipid profile to targeting atherosclerotic development?

To translate basic research to the clinic, human samples are the obvious choice to work on to garner insights to both the mechanistic processes and pharmacological approaches pertaining to treatment of atherosclerosis. However, the chronic nature of atherosclerosis and the period of time taken for clinical manifestations to become apparent inhibits sample collection prior to end-stage disease, and thus pre-clinical models must be utilized to better address these questions (Figure 1).

FIGURE: Comparison of rabbit, mouse and pig models of atherosclerosis.

Small animal (rabbit and mouse) and large animal (pig) models are used in atherosclerosis. All three models have unique benefits and limitations for modeling atherosclerosis; however, all offer the ability to track atherosclerotic development in both early-stage and late-stage disease, which in inaccessible when working with human tissue.

HDL: High density lipoprotein; LDL: low density lipoprotein; CETP: cholesteryl ester transfer protein.

Currently, the majority of contemporary studies are performed in mice. Murine models, due to their affordable cost of care and husbandry, and their ability to develop atherosclerosis quickly have been particularly useful for elucidating the impact of individual genes roles in atherosclerosis, often seen as a proof of principle model. Notably, mouse models recently provided the first insight into the role of proprotein convertase subtilisin/kexin type 9 (PCSK9) in increasing plasma low-density lipoprotein (LDL) concentration through reduction of expression of LDL receptors, suggesting the potential of targeting PCKS9 via monoclonal antibody as a therapeutic approach [8, 9]. Through this approach, the efficacy of PCSK9 antibodies in lowering LDL cholesterol has now been demonstrated in a multitude on clinical trials and is currently used within the clinic [10]. Nevertheless, the difference between mouse and human atherosclerosis is vast and sometimes difficult to translate. Mouse models exhibit significantly more inflammation within the vasculature compared to that found in human disease and thus although targeting inflammation and the related pathways have anti-atherogenic therapeutic benefits in mice, these do not translate well to humans [11]. Moreover, murine atherosclerotic plaque lacks a thick fibrous cap, and atherosclerosis is rarely assessed within the coronary arteries as it is with humans, but rather in larger more accessible vessels, further limiting the translational benefits. In terms of calcification, mice exhibit relatively low calcification, whereas calcification is present in 80% of ruptured human plaques, a pathologic complication of atherosclerosis which is not easily recapitulated within the mouse and thus are often seen as proof of principle rather than translational models for humans.

Rabbits, on the other hand, may better represent the pathobiology of human atherosclerosis. Seen here and elsewhere [3, 12], rabbits, when fed a cholesterol-enriched diet, exhibit a similar lipid profile to humans [13]. With dyslipidaemia a well-recognized epidemic across the American population [2], recapitulating a lipid profile for development of novel atherosclerosis treatments should be a high priority moving forwards. The size of rabbits, and their ability to recapitulate not only early but late-stage atherosclerosis, smaller, more focused studies may be undertaken to better understand the entirety of atherosclerosis, with one rabbit able to provide sufficient tissue for multiple analyses, thus allowing a reduction in the number of animals used for research. While large animal models recapitulate atherosclerosis better than mice, rabbits may present a compromise between high-cost large animal models such as ovine or swine and the less costly and easier to handle mice. Cost, coupled with the advancement and availability of Tg rabbits may provide a unique tool to assess the effects of individual lipoproteins and atherosclerotic genes successfully [14]. Similarly, non-invasive imaging, while more expensive in the rabbit compared to the mice is easier [15], as is catheterization of the rabbit coronary arteries, a technique which is problematic in murine models.

Overall, the current findings of Wei et al have clearly demonstrated the efficacy and value of rabbit models within atherosclerotic studies, and in the times of lipid-driven research, rabbits may contribute more to the future translational atherosclerosis research.