Cholesterol crystal induced arterial inflammation and destabilization of atherosclerotic plaque

Janoudi A, Shamoun, FE, Kalavakunta JK et al.
Eur Heart J (2016) 37, 1959–1967. DOI: http://dx.doi.org/10.1093/eurheartj/ehv653

Morphological descriptions of vulnerable plaque and plaque rupture do not take into account the dynamic events that led to rupture. When cholesterol crystallises from a liquid to a solid state, its volume expands. Sharp-tipped cholesterol crystals (CC) within the plaque can perforate the fibrous cap. In addition, CC appear to trigger local and systemic inflammation, which is the topic of this review.
 Cholesterol enters the arterial wall via LDL particles. When LDL is oxidised, it is taken up by macrophages to become foam cells. Initially, macrophages exert an anti-atherogenic effect by producing nascent HDL.
Esterified LDL particles convert into free cholesterol (FRC). Accumulation of FRC in the arterial wall can occur due to an imbalance of esterified cholesterol (ESC) and FRC or changes in HDL functionality, and predisposes to CC formation. CC build up in the plaque’s necrotic core causes mechanical injury.

Equilibrium between esterified and free cholesterol

Esterification to FRC is done by cholesterol ester hydrolase (CEH), while acyl-coenzyme A cholesterol acyltransferase 1 (ACAT-1) de-esterifies FRC back to ESC. Disruption of this equilibrium can cause accumulation of either ESC or FRC.
Inhibition of ACAT has been shown to result in cholesterol domains in the plasma membrane bilayer of macrophages, which can serve as nucleation sites for formation of CC that are released into the extracellular space. In a mouse model, overexpression of CEH led to reduced necrotic core size and a higher number of viable foam cells in the presence of normal HDL function. No accumulation of FRC was seen due to increased reverse cholesterol transport via HDL.
ACAT-1 inhibitors have been developed to potentially prevent conversion of macrophages to foam cells, possibly to inhibit plaque growth. While animal study results evaluating these inhibitors were hopeful, trials in humans were not (CAPTIVATE).
FRC may also derive from dying foam cells and red blood cells with rich cholesterol membranes from injured vasa vasorum. Also a change of pH, due to release of cellular content of dying cells can enhance cholesterol crystallisation. This may lead to multiple plaque ruptures in various arterial beds, thus ‘vulnerable patient’ may be more applicable than ‘vulnerable plaque’.

High-density lipoprotein functionality and free cholesterol accumulation

Before HDL reverse cholesterol transport out of arterial wall cells can take place by means of membrane-bound ABCA-1 and ABCG-1 transporters, ESC needs to be converted to FRC. When HDL is dysfunctional, for instance as a result of inflammation, FRC accumulates, promoting CC. Atherosclerosis building up in the arterial wall increases local and systemic inflammation. Thus, these processes strengthen each other towards atherosclerosis build up and further worsening of HDL functioning.
HDL has been shown to dissolve CC, which is being investigated further.

Inflammation and plaque destabilisation

Oxidised LDL causes endothelial dysfunction and results in expression of proinflammatory cytokines, chemokines and adhesion molecules. Oxidised LDL also activates macrophages, and causes them to release cytokines that attract other macrophages and metalloproteinases with autolytic properties. Trough induction of CD14 and TLR-4, the NLRP3 inflammasome and pro-IL-beta protein expression is induced. Other inflammatory markers are also increased during atherogenesis, promoting foam cell formation and apoptosis, impairing endothelial function and enhancing smooth muscle cell migration and proliferation.
CC form early during atherogenesis, and activate NLRP3 to secrete IL-1beta, subsequently leading to IL-6 secretion and an acute-phase response in the liver. Clinical studies have identified C-reactive protein produced during an acute-phase response as a marker of risk of developing acute CV events.
If CC grow and eventually perforate the intima overlying the necrotic core, this will trigger a systemic response. In an atherosclerotic rabbit model, lowering serum cholesterol levels with ezetimibe reduced arterial wall CC content, plaque rupture and thrombosis.

Local conditions that enhance plaque rupture

It is hypothesised that the volume expansion that accompanies crystallisation of liquid FRC may lead to plaque rupture. Evidence supporting this hypothesis includes:

• Post-mortem, patients who died from acute myocardial infarction (MI) showed CC perforating the intimal surface and fibrous cap, while those who died of other causes did not show arterial plaques with perforating CC.
• Autopsy of patients who died from MI showed CC perforating the intima overlying ruptured plaques in both the culprit artery and other coronary arteries.
• In atherosclerotic rabbits, greater plaque burden was associated with higher arterial wall cholesterol content, CC density, serum inflammation, macrophage infiltration in plaque and greater events of intimal disruption with thrombosis. Lowering total serum cholesterol diminished the crystal content and associated thrombosis.
• CC has been detected in human coronary arteries by means of optical coherence tomography (OCT) during percutaneous procedures, suggesting a link between CC and potentially vulnerable plaques.

Several physical conditions, including increased cholesterol saturation, 1-2 degrees Celcius drop in local temperature, an alkaline pH and formation of cholesterol monohydrate, can enhance cholesterol crystallisation.
A causal link between CC and plaque rupture has not been established in human, but various lines of evidence suggest that plaque rupture involves penetration of sharp-tipped CC into the fibrous cap. And expansion of crystals may stretch the fibrous cap to create thin-cap fibroatheroma, which is likely vulnerable to CC.

Mechanisms of plaque rupture and clinical implications

It has been shown that during a heart attack, several arteries are involved with plaque rupture, but not all develop occlusive thrombosis. It is postulated that plaques with large necrotic cores will release greater amounts of CC into the circulation causing more injury and arterial thrombosis, thus underscoring the importance of plaque burden.

Imaging of vulnerable plaques

CC perforating the intima over plaque can be made visible with scanning electron microscopy, when tissue samples are processed by air or vacuum dehydration (not with standard approach with ethanol) and by confocal microscopy of fresh human plaque samples.
Protruding CC can be stained with Bodipy, a fluorescence dye and the intima counterstained with acyl-LDL, without any tissue fixation or dehydration.
While standard tissue preparation methods dissolve cholesterol, light microscopy can reveal CC penetrating the fibrous cap and communicating as empty channels or clefts with the intimal surface.
In addition, CC perforating plaque caps have been visualised using micro-OCT of human carotid plaques, coherent anti-Stokes Raman scattering (CARS) and by means of MRI using beta-cyclodextrin conjugated with super paramagnetic iron oxide nanoparticles in rabbit aortas.

Therapeutic implications

Considering the proposed role of CC, compounds that have the potential to dissolve CC or inhibit their formation may provide an effective therapeutic approach to reduce the incidence of acute CV events. Recently, statins have been shown to dissolve CC and interfere with crystal formation. PCKS9 inhibition may also impact CC formation through its LDL-lowering effect.
The ongoing CANTOS trial that evaluates the efficacy of the selective inhibitor of IL1beta canakinumab will provide interesting information on the effect of reducing plaque inflammation. The same is true for the CIRT trial that evaluates low-dose methotrexate in patients with high CV risk.

Future directions

Various stages of atherosclerotic plaque development, the evolution to vulnerable plaque and eventual rupture need to be studied to understand more of the role of CC and their effect on clinical outcomes.

Find this article online at Eur Heart J