Keywords
atherosclerosis - endothelial dysfunction - inflammation - matrix metalloproteinases
- platelets
Beyond their physiological function in hemostasis, platelets play a pivotal role in
the pathogenesis of various disorders, including thrombosis, cancer, allergy, acute
lung injury, autoimmune diseases, and sepsis.[1] Over the last few decades an important role of platelets has emerged also in the
initiation and progression of atherosclerosis.[2]
[3]
[4] Circulating activated platelets are associated with greater atherosclerotic plaque
burden, increased plaque vulnerability, and enhanced risk of arterial thrombosis.
Given that atherosclerosis is the cause of myocardial infarction, stroke, and peripheral
arterial disease, which represent the leading global causes of mortality and morbidity,
it is of great importance to deeply understand the role of platelets and of their
interplay with other cells in atherogenesis in order to identify new targets for the
prevention of these widespread disorders. In this review, we recapitulate the historical
steps that have led to the characterization of the role of platelets in the development
and progression of atherosclerosis.
The Pathogenic Theories of Atherosclerosis
The Pathogenic Theories of Atherosclerosis
The word “atherosclerosis” was introduced in 1904 by the German pathologist Felix
Marchand to emphasize the presence of lipid-rich lesions in the arterial wall, leading
to vascular stiffening and obstruction (Greek: atheré = hulled grain; sklerosis = sclerosis).
In the mid-19th century, two opposing theories on atherogenesis were formulated: on
one hand Rudolf Virchow proposed that inflammation had a primary role in atherogenesis,[5] on the other hand, Carl von Rokitansky claimed that inflammation was a mere consequence
of lipid deposition.[6]
During the 20th century, intense research on lipoprotein metabolism identified low-density
lipoprotein (LDL)-cholesterol as the leading trigger of atherosclerosis. In 1910,
Adolf Windaus, Nobel Prize in Chemistry in 1928, showed that atheromatous lesions
contain 25-fold more cholesterol than the normal arterial wall. A few years later,
Russian researchers elicited atherosclerotic lesions in rabbits by feeding animals
with pure cholesterol and first characterized the cell types infiltrating atheromas,
including macrophages, lymphocytes, smooth muscle cells (SMCs), and macrophage-derived
foam cells.[7] Numerous subsequent epidemiological and clinical studies confirmed the central role
of cholesterol in atherogenesis, showing that its circulating levels correlate with
cardiovascular events and that its reduction prevents them. The 20th century brought
many other advancements in the understanding of atherosclerosis, including the discovery
in 1973 of the LDL receptor, and soon after the identification of the first cholesterol
lowering agent, a β-hydroxy β-methylglutaryl-CoA (HMG-CoA) reductase inhibitor.[8] More recently, the discovery of proprotein convertase subtilisin/kexin type 9 (PCSK9)
brought new insight in the regulation of cholesterol levels and led to the introduction
of a new class of lipid-lowering agents, the PCSK9 inhibitors, either anti-PCSK9 monoclonal
antibodies (alirocumab and evolocumab) or the gene-silencing small-interfering RNA
inclisiran.[9]
[10]
Atherosclerosis
The arterial wall is formed by three layers: the outermost layer called adventitia,
formed by connective tissue with elastic and collagen fibers and vasa vasorum, i.e.,
small blood vessels feeding the arterial wall; the middle layer called media, formed
by SMCs and extracellular matrix proteins, which provides a scaffold for the vessel
and regulates blood flow and blood pressure modifying vessel diameter; and the innermost
layer called intima, formed by a basement membrane made of elastic fibers lined by
the endothelium, a single layer of tightly connected cells which regulates the exchanges
between the bloodstream and surrounding tissues ([Fig. 1]).[11] Atherosclerosis begins with a functional impairment of the endothelium, resulting
in disordered intercellular junctions with consequent increased permeability leading
to the penetration of circulating ApoB-containing lipoproteins, such as LDLs, very
low density lipoproteins (VLDLs), and apolipoprotein E (ApoE) remnants, in the subendothelial
space.[12] Subendothelium-retained lipoproteins are then modified by oxidation or glycation,
becoming atherogenic and further promoting the dysfunction of endothelial cells. Altered
endothelial cells lose their antiadhesive properties and express adhesion molecules,
including P-selectin, vascular cell adhesion protein 1 (VCAM-1), intercellular adhesion
molecule 1 (ICAM-1), proinflammatory receptors, like toll like receptor-2 (TLR2),
and cytokines, like monocyte chemoattractant protein-1 (MCP-1) and interleukin-8 (IL-8),
thus promoting the recruitment of immune cells in the vessel wall, such as monocytes,
dendritic cells, mast cells, regulatory T cells, and T-helper subset 1 (Th1) cells.
Monocytes penetrated within the vessel wall differentiate into macrophages that can
then be polarized either to a pro-inflammatory phenotype (M1), prone to differentiation
in foam cells, or to an anti-inflammatory one (M2) with scavenging activity. The switch
to one or the other phenotype depends on the microenvironment. In particular, oxidized
LDL (ox-LDL) skew the phenotypic distribution towards the pro-inflammatory M1 subset
while IL-4 skews it towards the anti-inflammatory M2 subset. Although both M1 and
M2 macrophages can be found in atherosclerotic lesions, the M1 subtype dominates in
the rupture-prone shoulder region of unstable plaques, while the M2 subtype in the
adventitia of stable plaques.[13] These macrophage subsets are present also in murine atherosclerotic plaques, and
a phenotypic shift from the M2 to the M1 subtype correlates with plaque progression,
while a shift towards M2 with reduced atherosclerosis.[14]
[15]
[16]
Fig. 1 Schematic representation of the vascular wall. (A) Normal artery, (B) artery with early atherosclerotic features, and (C) artery with advanced atherosclerotic features.
The penetration of minute cholesterol crystals coincides with the first appearance
of inflammatory cells in the vessel wall and acts by stimulating the caspase-1-activating
NOD-like receptor protein 3 (NLRP3) inflammasome, leading to the cleavage and secretion
of proinflammatory IL-1.[17] Once macrophages phagocytose ApoB-containing lipoproteins enriched of cholesterol,
they turn into foam cells, which later undergo regulated apoptosis, generating a necrotic
core that promotes further inflammation.[18]
[19] In response, SMCs migrate from the media to the intima and proliferate, leading
to intimal hyperplasia.[20] As this process progresses, SMCs deposit extracellular matrix, forming a protruding
fibrous cap which progressively leads to a hemodynamically significant vessel stenosis.
Ultimately, the rupture of the fibrous cap, due to the gradual loss of SMCs and to
the erosion of the collagen-rich cap, leads to the exposure of thrombogenic materials
which trigger the thrombotic occlusion of the artery with resultant organ ischemia.[21] Recent data suggest that some inhaled or ingested micro- and nanoplastics, mainly
polyethylene and polyvinyl chloride, accumulate at sites of atherosclerosis promoting
oxidative stress, inflammation, and apoptosis in vascular cells and associating with
an enhanced risk of acute cardiovascular events.[22]
Moreover, clonal hematopoiesis of indeterminate potential, a condition associated
with somatic mutations in genes that lead to the constitutive activation of the Janus-associated
kinase signaling pathway and to an abnormal function of circulating clones of mutated
leukocytes and platelets, has recently been found to be associated with accelerated
atherosclerosis.[23]
The Crucial Role of Inflammation in Atherosclerosis
The Crucial Role of Inflammation in Atherosclerosis
Virchow's theory laid the foundations for the famous response-to-injury hypothesis
of Russell Ross, which maintained that atherosclerosis is caused by an initial endothelial
injury resulting in arterial wall denudation with subsequent activation of platelets
that release factors stimulating the migration and proliferation of SMCs. Subsequent
research demonstrated that the initial endothelial injury can be mild, insufficient
to denude the vessel wall, but strong enough to generate endothelial dysfunction.
The accomplished hypothesis was finally published by Ross in 1999 in a seminal paper
in the New England Journal of Medicine[2] purporting that atherosclerosis is an inflammatory disorder in which LDL represents
a primary cause of endothelial injury that in turn starts a vicious circle of inflammation,
lipoprotein oxidation, and further inflammation. In particular, the binding of ox-LDL
to the scavenger receptor class A (SR-A) LOX-1 on endothelial cells causes an increase
in leukocyte adhesion molecules, activates apoptotic pathways, increases ROS generation,
and causes endothelial dysfunction, thus promoting the interaction with circulating
monocytes and platelets, the secretion of proinflammatory chemokines, and the impairment
of nitric oxide (NO) generation.[24]
Currently atherosclerosis is considered a chronic, multifactorial inflammatory disorder
of large and medium-size arteries, the pathogenesis of which involves plasma components,
circulating blood cells, vascular wall cells, the extracellular matrix, and their
reciprocal interactions.[25]
[26] Innate immune cells, such as monocytes and macrophages, develop a long-lasting proinflammatory
phenotype after the exposure to microbiological or atherogenic stimuli through metabolic
and epigenetic reprogramming, a phenomenon called trained immunity.[27] The permanent hyperactivation of the innate immune system together with a persistent
low-grade vascular inflammation are key events in the development of atherosclerosis.[28] In recent years, an increasing body of evidence suggests that endothelial cells
participate in the clearance of apoptotic cells, cell debris and pathogens and in
the recruitment of immune cells to vulnerable sites in the vasculature in order to
limit pathogen dissemination.[29]
[30] A large number of DAMPs, such as high-mobility group box 1 (HMGB1), S100 proteins,
ox-LDL, and heat shock proteins, have been implicated in the pathogenesis and progression
of atherosclerosis. Damage-associated molecular patterns (DAMPs) initiate an inflammatory
response in the absence of pathogens by activating pattern recognition receptors,
such as TLRs on vascular cells. Upon contact with DAMPs, SMCs adopt a pro-inflammatory
phenotype and transdifferentiate into macrophage-like cells that enhance atherogenesis.[31]
[32]
A role in atherosclerosis in also played by the shear forces generated by blood flow
on the vessel wall. While laminar shear stress maintains the antiadhesive properties
of the endothelium by inducing the release of nitric oxide (NO), turbulent flow activates
endothelial cells favoring lesion formation, explaining why plaques generally localize
at arterial branch points and curvatures.[20]
Platelets as Inflammatory Cells
Platelets as Inflammatory Cells
Platelets were first clearly identified as the third morphological element of blood
involved in clot formation by Giulio Bizzozero in 1882. Since then, their central
role in hemostasis and thrombosis has become increasingly clear.[33]
Over the last few decades it has emerged, however, that platelets are also involved
in inflammatory processes and in the innate immune responses against microorganisms.
Although the adhesion of platelets to bacteria was first described already in 1946,[34] only in the last few years the ability of platelets to migrate and to scavenge bacteria
bridging innate and adaptive immunity has been fully characterized.[35] Platelets express a plethora of receptors that allow them to detect, bind, and elicit
a host response to intravascular and opsonized pathogens and also to kill microorganisms,
thus contributing to the defense against infection.[34]
[36] Platelet α-granules store a variety of peptides with microbicidal activity, such
as connective tissue-activating peptide (CTAP)-3, thymosin β-4, fibrinopeptide A and
B, and thrombocidins, able to prevent the growth of bacteria, such as Bacillus subtilis, Escherichia coli, Staphylococcus aureus, and fungi, such as Cryptococcus neoformans.[37]
[38]
Platelets also participate in the inflammatory reactions to noxious stimuli, either
directly or by interacting with other cells in the circulation.[33]
[35] Inflammation of the vessel wall is a central feature of atherogenesis and most of
the activities of platelets in inflammation have been implicated in the initiation
and progression of atherosclerosis. Crucial initial evidence for the contribution
of activated platelets to atherosclerosis include the observation in 1984 that platelets
adhere to fatty streaks[39] and in 1994 that platelets are activated by atherogenic triggers, including ox-LDL.[40] Another important step in the development of the concept that the participation
of platelets in inflammation and immunity implies also their involvement in atherogenesis
was the discovery in 2004 that human platelets possess functional TLRs, pathogen-associated
molecular pattern-, and DAMP-recognition receptors.[41] TLRs activate platelets to release pro-inflammatory cytokines, such as IL-1, IL-6,
and IL-8, which are involved in the early phases of atherogenesis.[42]
[43]
[44]
Platelets contribute to inflammation by releasing cytokines and chemokines, stored
in their granules or de novo synthesized, which recruit leukocytes to sites of inflammation
or injury.[45] Among them, platelet factor 4 (PF4 or CXCL4), the most abundant protein secreted
by activated platelets, functions as a chemoattractant for monocytes and promotes
the retention of lipoproteins in the vascular wall.[46] Platelet-derived IL-1α mediates inflammation in vivo[47] and induces the expression of ICAM-1 and VCAM-1 by endothelial cells, accelerating
the transendothelial migration of neutrophils.[46] Platelets are a major source of IL-1β which is not stored in granules but is produced
upon platelet activation when IL-1β pre-mRNA is spliced, IL-1β mRNA translated into
pro-IL-1β, and the latter is then caspase-1-processed resulting in the release of
functional IL-1β.[48] Platelet-derived IL-1β causes the up-regulation of endothelial adhesion receptors
and the release of proinflammatory IL-6 and IL-8 from endothelial cells. IL-1β is
also responsible for the activation of NF-κB in endothelial cells, which is required
for the transcription of the inflammatory genes MCP-1 and ICAM-1.[46] Platelet-derived growth factor is a chemoattractant for monocytes and eosinophils.
The chemokine RANTES (Regulated on Activation, Normal T Cell Expressed and Secreted)
monocytes to the inflamed endothelium.[49] Among other platelet-derived chemokines, macrophage inflammatory protein (MIP)-1α,
a chemoattractant, is involved in the transendothelial migration of neutrophils and
monocytes at sites of inflammation.[50] Platelets also store and release large amounts of the pro-inflammatory trimeric
transmembrane protein CD40 ligand (CD40L). The interaction of CD40L with CD40 on endothelial
cells and macrophages causes the release of IL-8 and MCP-1, which attract neutrophils
and monocytes and induces adhesion receptor expression by the endothelium.[51] Platelets also contain and release polyphosphates (polyP) which activate the contact
coagulation pathway resulting in the generation of bradykinin, an inflammatory mediator
that increases vascular permeability by binding to its endothelial receptor BR2 and
thus favors the local accumulation of neutrophils.[52]
Contribution of Animal Models to the Understanding of Atherosclerosis
Contribution of Animal Models to the Understanding of Atherosclerosis
Animal models have greatly contributed to the understanding of the molecular mechanisms
leading to atherosclerosis, plaque rupture, and thrombus formation ([Table 1]). Moreover, animal models have been essential for the development of drugs preventing
atherosclerosis. Much of the data showing the crucial role of platelets in atherosclerosis
come from studies in these models (see the next section; [Fig. 2]).
Fig. 2 Timeline of the discoveries of the platelet involvement in atherosclerosis.
Table 1
Principal mouse models of atherosclerosis
|
Model
|
ApoE-/-
|
LDLR-/-
|
ApoE*Leiden
|
ApoE-/-/LDLR-/-
|
Apobec1−/−
|
LDLR−/− Leiden
|
PCSK9-D374Y-AAV
|
|
Publication date
|
1992a
|
1993b
|
1993 e
|
1994c
|
1996f
|
2019g
|
2004d
|
|
Lipid profile on SD
LDL CHOL
HDL CHOL
TOT CHOL
TG
|
421 mg/dLc
41 mg/dL
∼400mg/dL Greatly ↑
|
108 mg/dLc
105 mg/dL
275mg/dL
62 mg/dL
|
Modestly ↑e
Decreased
165 mg/dL
182 mg/dL
|
447 mg/dLc1
35 mg/dL
620 mg/dL
Greatly ↑
|
268.3 mg/dLf1
48.7 mg/dL
343 mg/dL
249 mg/dL
|
35 mg/dLg1
21 mg/dL
1216 mg/dL
421 mg/dL
|
Modestly ↑d
Similar to WT
202 mg/dL
Doubled ↑
|
|
Lipid profile on HFD
LDL CHOL
HDL CHOL
TOT CHOL
TG
|
11 w HFDa1:
107 mg/dL
37 mg/dL
1821 mg/dL
45 mg/dL
|
19 w HFDb1:
731 mg/dL
155 mg/dL
998 mg/dL
558 mg/dL
|
6 w HFDe1:
271-414 mg/dL
244-313 mg/dL
1014-1544 mg/dL
310 mg/dL
|
8 w HFDc1:
581 mg/dL
194 mg/dL
1591 mg/dL
221 mg/dL
|
3mo HFDf1:
-
-
1439 mg/dL
292 mg/dL
|
24w HFD:
146 mg/dLg1
30 mg/dL
1769 mg/dL
531 mg/dL
|
16w HFDd1:
600 mg/dL
181 mg/dL
722 mg/dL
266 mg/dL
|
|
SPONTANEOUS PLAQUE
DEVELOPMENT
|
3 mo:
fatty streaks in the proximal aorta
8mo:
aortic plaques
|
No or very mild lesions
|
No or only very mild lesionse1
|
Similar to ApoE-/-
|
Medium plaque formation in aortic root and abdominal aorta at 24 and 36 mo old micef2
|
Virtually absentg2
|
Noned1
|
|
PLAQUE DEVELOPMENT
UNDER HFD
|
Fast development in the first month
|
Medium plaques after 4 mo after HFD
|
Severe plaques after 6 w HFDe1
|
Advanced lesions 8 mo after HFH
|
Severe plaque after 3mo HFDf3
|
Severe plaque after 4 w HFDg2
|
Medium plaque after 16 w HFDd1
|
HFD: high fat diet; SD: standard diet; LDL CHOL: Low Density Lipoprotein cholesterol;
HDL CHOL: High Density Lipoprotein colesterol; TG: tryglycerides;
a -Piedrahita JA et al., Proc Natl Acad Sci U S A 2004; 89(6):4471–4475.
a1- Plump AS et al., Cell 1992;71(2):343–353.
b -Ishibashi S et al., J Clin Invest 1993;92(2):883-93.
b1 - Momi S et al., Cardiovasc Res 2012;94(3):428-38.
c - Ishibashi C et al., Proc Natl Acad Sci U S A 1994;91(10):4431-5
c1 - Franczyk-Zarów M et al., Br J Nutr 2008;99(1):49-58
d -Maxwell KN & Breslow JL. Proc Natl Acad Sci U S A 2004;101(18):7100-5
d1 - Keeter WC et al., Eur Heart J Open 2022;2(3):oeac028.
e - van den Maagdenberg AM et al., J Biol Chem 1993;268(14):10540-5.
e1 -van Vlijmen BJ et al., J Clin Invest 1994;93(4):1403-10.
f - Hirano K et al. J Biol Chem 1996;271(17):9887-90.
f1 - Nakamuta M et al., Arterioscler Thromb Vasc Biol 1998;18(5):747-55.
f2 -Dutta R et al., Atherosclerosis 2003;169(1):51-62.
f3 - Takayuki Y et al., Blood. 2006;107(10):3883-91.
g - Inia JA et al., FASEB J 2024;38(20):e70087.
g1 - Salic K et al., PLoS One. 2019;14(6):e0218459.
g2 - van den Hoek Am et al., Cells 2020;9(9):2014
One of the first animal models was developed in the Macaca nemestrina by removing the endothelium of the iliac artery through an intra-arterial balloon,
showing that de-endothelization is followed by the formation of a platelet layer covering
the denuded area and by the migration of SMCs from the media to the intima and to
their proliferation, reproducing the pre-atherosclerotic intimal hyperplasia observed
in humans.[53]
A strong boost to the understanding of the mechanisms of atherogenesis came from the
generation of a particular rabbit strain, the Watanabe heritable hyperlipidemic (WHHL)
rabbit, obtained by inbreeding a mutant discovered in 1973.[54] These rabbits show abnormally high serum cholesterol and triglyceride levels and
progressively develop stacks of strongly birefringent lipids in the aortic intima
and media, generating fatty streaks, raised foam cell lesions, and atheromas with
close similarities with human lesions.[55]
[56] The main factors initiating atherosclerosis in WHHL rabbits are adhesion of leukocytes
and platelets to endothelial cells and the accumulation of lipids in the aortic wall.[56]
In the last few decades, mice have been the most widely used animal model in atherosclerosis
research. Their advantages, compared with other species, include the availability
of numerous inbred strains and the ease to generate transgenic and knock-out breeds.
Mice do not express the cholesterol ester transfer protein, responsible for the exchange
of cholesteryl esters and triglycerides between lipoproteins,[57] and thus they have very low LDL levels compared to humans and do not develop arterial
lesions even when fed a high-cholesterol diet.[58] However, the addition of cholic acid (which inhibits bile acid biosynthesis, thus
increasing hepatic cholesterol levels) to high-fat diets resulted in aortic lesions
similar to the early fatty streaks of humans.[59]
A widely used model of atherosclerosis, the LDL receptor knockout mice (LDLR−/−), was generated in 1993 by Ishibashi et al in the laboratory of Joseph Goldstein,
who was awarded the 1985 Nobel Prize for discoveries on cholesterol metabolism, obtaining
a mouse that develops widespread atherosclerosis when fed a cholesterol-rich diet.[60] Unlike wild-type mice, LDLR−/− mice respond to moderate amounts of dietary cholesterol (0.2% cholesterol, 10% fat)
with a significant increase in circulating intermediate-density and LDL cholesterol
particles and form atheromatous plaques covering one-third of the aortic surface;
instead, when kept on a normal chow they develop only minimal lesions.[61] Given that the lipoprotein profile of LDLR−/− mice is close to the human profile, this mouse strain is considered a suitable model
for the study of cholesterol and lipoprotein metabolism.[62]
ApoE knockout mice (ApoE−/−) lacking ApoE, a glycoprotein (GP) synthesized mainly by the liver and the brain
that mediates the uptake of chylomicrons and VLDLs by hepatic receptors (LDLR and
LDL receptor-related protein 1), is another widely used model. ApoE−/− mice show delayed lipoprotein clearance and develop severe hypercholesterolemia and
atherosclerotic lesions.[63] In particular, normal diet-fed ApoE−/− mice develop moderate hypercholesterolemia and spontaneously mature fatty streaks
in the proximal aorta,[64] while high-cholesterol-fed animals develop severe hypercholesterolemia with widespread
and accelerated atherosclerotic lesions.[65] In fact, aortic foam cell lesions develop at 8 weeks, fibrous lesions at 15 weeks,
and at 60 weeks an entire spectrum of arterial lesions similar to human atherosclerosis,
which are widely distributed but do not involve the coronary arteries, which instead
are one of the primary sites of atherosclerosis in humans.[66]
[67]
These two classic models have served as the platform for the generation of numerous
other atherosclerosis-susceptible or -resistant mouse strains, such as the ApoE−/−/LDLR−/−, SRBI−/−/LDLR−/−, SRBI−/−/ApoE−/−, Apobec1−/−, LDLR/Apobec1−/−, ApoE*3 Leiden, and others.[67]
[68] ApoE−/−/LDLR−/− mice display significantly higher levels of circulating activated platelets compared
to wild-type mice, as shown by increased TxB2 production, P-selectin expression, and GPIIb/IIIa activation.[69] Also, Apobec1−/− mice display enhanced in vivo platelet activation, in particular increased GPIIIa
(CD61) and P-selectin (CD62P) on circulating platelets and higher soluble plasma P-selectin.[70] In both SRBI/LDLR and LDLR/Apobec1 double knockout mice, a significant accumulation
of platelets in atherosclerotic plaques was observed.[70]
[71]
In the 1990s, the human APOE*3-Leiden CETP mouse strain, a transgenic strain expressing
human cholesteryl ester transfer protein (CETP), was generated.[72] These mice display high plasma cholesterol levels, exhibit increased VLDL and LDL,
and decreased high-density lipoprotein levels, similar to humans with the metabolic
syndrome, and develop severe atherosclerosis when fed a high-fat diet.[73] However, in ApoE*3-Leiden mice, differently from humans, spontaneous plaque rupture,
intraplaque hemorrhage, and thrombus formation are not observed.[74]
Although mouse models have provided important informations on the mechanisms underlying
atherosclerosis, a gap exists between the results obtained in this species and translation
to human disease. Mice and humans diverged about 80 million years ago and have been
subjected to very different evolutionistic pressures. Nevertheless, their genomes
remain rather similar, sharing about 95% of their protein-coding genes, although noncoding
genes and regulatory regions are considerably less conserved. Moreover, inbreeding
of mice yields a high genetic homogeneity which makes a great difference with humans
who exhibit an enormous genetic diversity. Also, the strictly controlled laboratory
environment may have an impact on the development and evolution of atherosclerosis
by altering cellular behavior. Additional reasons highlighting problems on the translatability
of animal experiments include the lack of comorbidities, the prevailing use of young
mice while human atherosclerosis prevails in advanced age, and the quite extreme experimental
conditions required to speed up atherosclerosis, reducing the congruence with human
cardiovascular disease.[75] Even considering these limitations, animal studies have provided crucial insight
into the pathophysiology of atherosclerosis and have contributed to the development
of treatments of proven benefit to patients. The strictness, reproducibility, and
practicability of mouse experiments provide compelling reasons for their continued
use in cardiovascular investigation, using critical judgement in the extrapolation
of results to the clinic.
The Role of Platelets in Atherosclerosis
The Role of Platelets in Atherosclerosis
Platelets in atherosclerotic cardiovascular disease have long been studied for their
central role in the final thrombotic event triggered by plaque rupture.[76] In the last few decades, however, it started to emerge that platelets are also involved
in the initiation of atherosclerosis thanks to their special ability to bind to the
damaged arterial wall and to interact with leukocytes.[3]
[4]
[77] The adhesion of platelets to atherosclerosis predilection sites before the appearance
of manifest lesions to the endothelium has been shown by scanning electron microscopy
in hypercholesterolemic New Zealand white rabbits and in ApoE−/− mice.[78] Indeed, activated platelets contribute to induce endothelial dysfunction by secreting
inflammatory mediators, such as chemokines, tumor necrosis factor α (TNF-α), RANTES,
and PF4.[79] Risk factors for atherosclerosis, including hypercholesterolemia, hypertension,
cigarette smoking, and diabetes, trigger platelet activation in the circulation[80] and all these conditions are characterized by endothelial dysfunction, which in
turn facilitates platelet activation.[81] The primary manifestation of endothelial dysfunction is impaired release of NO.
NO has antiatherosclerotic properties, like the inhibition of vascular SMC proliferation,[82] of platelet activation,[83]
[84] and of inflammatory cell adhesion to and transmigration through the endothelium.[85]
[86] Thus, an impairment of NO biosynthesis may increase the progression of atherosclerosis.
Besides the endothelium, platelets also release NO[87]
[88]
[89]
[90] and this represents a mechanism of self-regulation of platelet activation but also
an antiatherogenic system. Indeed, impaired platelet NO release is associated with
increased circulating platelet/monocyte aggregates,[91] which in turn promote atherosclerotic lesion formation in ApoE−/− mice by supporting the adhesion of monocytes to the endothelium.[2]
[92] Platelet–leukocyte aggregates are present in blood of mice and patients with atherosclerosis,
and they accelerate atherosclerosis development.[93]
Platelet–endothelium interactions regulate leukocyte attachment and transendothelial
migration. Once adherent to the vascular wall, platelets provide a sticky surface
that recruits leukocytes which form initially loose, transient tethers with P-selectin
on the activated platelets and PSGL-1 on leukocytes, and subsequently stable adhesion
mediated by the binding of leukocyte β2 integrins, including Mac-1, to platelet GPIbα,
and of leukocyte junctional adhesion molecule-3 and ICAM-2 to high-molecular-weight
kininogen bound to GPIbα and to fibrinogen bound to platelet αIIbβ3.[94]
[95]
[96]
[97]
[98] Activated platelets modulate the chemotactic and adhesive properties of endothelial
cells also through the synthesis and release of IL-1β that triggers endothelial cells
to secrete chemokines and to express adhesion molecules that favor neutrophil and
monocyte attachment.[46]
[51]
[99] Finally, platelets recruit circulating CD34+ endothelial progenitor cells and bind
them to the altered vascular wall, promoting their differentiation either into endothelial
cells or into macrophages and foam cells, thus directing the switch for vessel regeneration
versus disease progression.[100]
Platelet and neutrophil activation are closely intertwined and synergize in response
to infection, but also in inducing inflammation and atherosclerosis ([Fig. 3]). Upon activation, neutrophils release decondensed chromatin, decorated with histones
and proteins, forming extracellular networks that entrap and kill bacteria, called
neutrophil extracellular traps (NETs). Activated platelets secrete HMGB-1 that induces
neutrophils to release NETs. In turn, elastase, proteases, and cathepsin-G embedded
in NETs activate platelets in a mutual positive feed-back process.[101] Cholesterol crystals of atherosclerotic plaques stimulate NET formation, and the
latter in turn stimulate macrophages to produce cytokines, including IL-1β, which
recruit additional neutrophils and activate TH17 cells which amplify immune cell recruitment.[102] Indeed, NETs have been detected in sections of human and murine atherosclerotic
lesions.[103]
Fig. 3 Platelet interactions with neutrophils, monocytes, macrophages, and endothelial cells
contribute to inflammation leading to atherosclerosis.
Platelets migrate in tissue in response to inflammatory stimuli driving the subsequent
influx of leukocytes, as shown in vivo in models of allergic asthma and of microvascular
inflammation.[104]
[105] Platelets are equipped with all the machinery required to migrate in response to
cytokines, sensitizing allergens, and other stimuli.[34]
[106]
[107]
[108] Platelets also actively migrate through the endothelium toward the plaque core.[109] Indeed, immunohistochemistry studies in ApoE−/− mice showed platelets within atherosclerotic plaques, in the intima or in close proximity
to tissue macrophages in the subendothelial space, and in the lipidic accumulation
sites.[110] The specific sites in which platelets were visualized made it possible to speculate
that they may have a role in foam cell formation, reinforcing previous observations
showing that platelets can accumulate lipids in a hypercholesterolemic environment.[111]
On the other hand, high-speed intravital microscopy in ApoE−/− mice suggested that the recruitment of activated platelets to atherosclerotic plaques
is driven by neutrophils previously penetrated in the vascular wall.[111] Monocytes too were found to favor platelet transendothelial migration in co-culture
experiments in vitro. In this model, monocytes and platelets were stimulated by TNF-α
and ox-LDL before testing transendothelial migration. Interestingly, when pretreated
with conditioned medium derived from monocyte cultures, platelets migrated as well,
suggesting that factors secreted by monocytes stimulate platelet migration.[109]
Platelet/leukocyte interactions give also rise to the generation of reactive oxygen
species (ROS), tissue factor expression, and thrombin prodution.[112] ROS are highly reactive molecules which promote protein, lipid and DNA damage, endothelial
dysfunction, and cell apoptosis contributing to the development of atherosclerosis.
Platelets contain mitochondria, on average 5–8/cell, mostly located in the vicinity
of the plasma membrane.[113] Platelet mitochondria, by increasing ROS production, have been implicated in platelet
activation and thrombosis.[114] Activated platelets release respiratory-competent mitochondria, either as free organelles
or encapsulated within microparticles. The hydrolysis of the mitochondrial membranes
by sPLA2-IIA then releases inflammatory mediators (i.e., lysophospholipids, fatty
acids, and mitochondrial DNA) that promote leukocyte activation, and possibly contribute
to atherogenesis.[113]
Activated platelets release and express on their surface also CD40L, which triggers
the liberation of IL-8 and MCP-1 and increases endothelial tissue factor expression
by interacting with endothelial CD40, thus generating signals for the recruitment
and extravasation of leukocytes.[51] CD40L exposed on the platelet membrane is then cleaved and shed as soluble CD40L,
which also produces transactivation of the endothelium and of leukocytes exacerbating
atherosclerosis.[115] The body of evidence implicating the CD40–CD40L interaction in atherogenesis is
compelling: the genetic disruption of CD40/CD40L in hyperlipidemic mice downregulated
plaque formation[115]
[116]
[117]; an antibody against mouse CD40L prevented aortic atherosclerotic lesion formation
in LDLR−/− mice[118]; platelets of healthy human volunteers activated at a localized site of vascular
damage express CD40L in amounts sufficient to induce VCAM-1 expression and soluble
MCP-1 and IL-8 secretion by endothelial cells[119]; the expression of CD40L on circulating platelets from patients with type 2 diabetes
mellitus correlated with the progression of atherosclerotic carotid artery disease.[120]
The binding of CD40L to ECs, smooth muscle cells, and macrophages triggers also the
expression of matrix metalloproteinases (MMPs), a family of zinc-dependent endoproteases
which degrade subendothelial matrix proteins and contribute to the remodeling of the
vessel wall associated with plaque formation.[121] Platelets contain and release several MMPs, including MMP-2 which participates in
thrombogenesis.[122]
[123]
[124] Platelet-expressed MMP-2 cleaves endothelial protease-activated receptor 1 (PAR-1)
triggering p38MAPK signaling, which leads to the expression of surface adhesion molecules.[124]
[125]
[126] LDLR/MMP-2 double knockout mice kept on a high-fat diet showed reduced atherosclerotic
plaque formation as compared with LDLR−/− mice. Similar results were obtained with LDLR−/− mice transplanted with bone marrow from MMP-2−/− mice or with LDLR−/− mice made thrombocytopenic and subsequently reinfused with activated platelets from
mice deficient in MMP-2, showing that the main source of MMP-2 involved in plaque
promotion is platelets. Interestingly, monocytes/macrophages infiltrated in aortic
plaques showed a shift from M1 to M2 in LDLR−/−/MMP-2−/− mice compared with LDLR−/− mice, suggesting that platelet MMP-2 is involved also in the pro-inflammatory polarization
of macrophages.[16] Finally, MMP-2 was overexpressed on circulating platelets from patients with coronary
artery disease and patients with chronic human immunodeficiency virus infection and
significantly correlated with the degree of carotid artery atherosclerotic stenosis.[16]
Platelet-released P-selectin binds endothelial cell PSGL-1.[94] In a study investigating the role of platelet- versus endothelium-released P-selectin
in generating atherosclerotic lesions in ApoE-deficient mice, the authors found that
although endothelial P-selectin played a crucial role, also platelet P-selectin contributed
to plaque development, in fact plaques of wild-type mice receiving bone marrow transplants
from P-selectin-deficient mice were 30% smaller than those of mice receiving bone
marrow from wild-type mice.[127] Moreover, repeated transfusions of activated wild-type platelets, but not of activated
platelets from P-selectin−/− mice, in hypercholesterolemic ApoE−/− mice increased aortic plaque formation.[4]
P-selectin expressed on activated platelets mediates the formation of rosettes with
monocytes and neutrophils and the rolling of monocytes on activated endothelium.[4]
[16] In P-selectin knockout mice, a marked inhibition of leukocyte rolling and a delayed
recruitment of monocytes to sites of inflammation were observed.[128] While the above studies were performed in vitro or in animal models, a study in
humans with risk factors for atherosclerosis showed that high levels of P-selectin
expression on circulating platelets significantly and positively correlated with carotid
intima-media thickness and arterial stiffness,[129] two recognized surrogate markers of atherosclerosis progression.[130]
[131]
Activated platelets interacting with the vessel wall also release the chemokines RANTES,
which triggers monocyte arrest, and PF4, which attracts monocytes and promotes their differentiation into macrophages.
The deposition of PF4 on the vascular endothelium by activated platelets induces the
formation of heterodimers of PF4 with RANTES promoting monocyte adherence and transmigration
through the endothelium. Monocytes exposed to PF4 differentiate then towards a specific
type of macrophage associated with vascular inflammation and plaque instability.[132] Indeed, platelet-derived chemokines have been detected in murine and human atherosclerotic
plaques.[49]
[133] PF4 also enhances the uptake of ox-LDL by macrophages, thus promoting the formation
of foam cells contributing to the development of the atherosclerotic lipid core.[134] PF4 also stimulates the proliferation, migration, cytokine release, and calcification
of vascular SMCs.[135] Moreover, PF4 binds to LDLR and disrupts the cellular endocytic machinery, resulting
in the permanence of LDLs on the cell surface, which may facilitate proatherogenic
modifications. Indeed, the knocking out of the PF4 gene leads to a significant reduction
of atherosclerotic lesion formation in ApoE−/− mice.[136]
Potential Role of Antiplatelet Agents in Atherosclerosis
Potential Role of Antiplatelet Agents in Atherosclerosis
Given the important role of activated platelets in atherogenesis, it seems logical
to assess whether antiplatelet therapy could potentially prevent or delay atherosclerosis.
However, while randomized controlled trials have provided unequivocal evidence that
low-dose aspirin is effective in the secondary prevention of cardiovascular disease,
no clinical study has convincingly shown that chronic aspirin intake may lessen atherosclerotic
lesions. Notwithstanding, a role of platelet COX-1 in the development of atherosclerosis
has been suggested, showing that the selective inhibition of COX-1, the only isoform
present in platelets, with SC-560 (5-(4-chlorophenyl)-1-(4-methoxyphenyl)-3-trifluoromethylpyrazole)
prevented lesion formation in ApoE−/− mice.[137] Similar findings were obtained with chronic treatment with low-dose aspirin in LDL-R−/− mice. However, the authors recognized that it is not possible to tell whether the
antiatherosclerotic effect was due to the antiplatelet or to the anti-inflammatory
action of the drug.[138] On the other hand, the selective genetic ablation of COX-1 in platelets and megakaryocytes,
which would be expected to recapitulate the effects of low-dose aspirin, when obtained
by two different recombination methods (PF4-ΔCre vs Gp1bα-ΔCre) generated conflicting
results: restrained atherogenesis with one and accelerated with the other. These divergent
effects were ascribed to a different impact of the two recombination methods on COX-1-dependent
prostanoid formation by the vasculature and leave unsettled the question of the potential
antiatherogenic effect of platelet COX-1 inhibition.[139]
The effect of P2Y12 receptor antagonism on atherogenesis has been explored in mice with a double ApoE
and P2Y12 gene deletion showing a reduced atherosclerotic lesion area, an increased fibrous
content of the plaque, and a decreased monocyte/macrophage infiltration.[140]
[141] Moreover, the P2Y12 antagonist clopidogrel reduced macrophage and T-cell infiltration in atherosclerotic
lesions and delayed the development of atherosclerosis in some preclinical models,[141]
[142]
[143]
[144] but not in others.[145] Similarly, ticagrelor reduced atherosclerosis in ApoE−/− mice in one study[146] but not in another.[147] Discrepancies between the positive and negative studies may be related to drug dose,
timing, and duration of treatment,[147] but leave quite open the question of whether P2Y12 antagonists may reduce atherosclerosis. The combination of aspirin and a P2Y12 antagonist, largely used in patients who suffered an acute coronary event, did not
seem either to prevent atherosclerosis in hypercholesterolemic mice.[145]
Vorapaxar, the first in class PAR-1 antagonist approved for clinical use, has also
been explored for its effect on the development of atherosclerosis in the ApoE−/− mouse model, showing a reduction of aortic lesion size associated with a reduced
expression of vascular adhesion molecules and a shift of infiltrating monocytes toward
the M2 phenotype.[148] Indeed, in clinical studies vorapaxar prevented the progression of atherosclerosis
of the lower limb arteries[149]; however, here too it is difficult to ascribe these actions only to the antiplatelet
action of vorapaxar given that PAR-1 is also present on the endothelial surface.
Nitric oxide has strong antiplatelet properties[150] but also ideal features as an antiatherosclerotic agent. Drugs favoring the release
of NO or increasing its bioavailability, or indirectly enhancing the biologic effects
of endogenously released NO, have shown efficacy against atherosclerosis. In particular,
the coupling of nonsteroidal anti-inflammatory drugs (NSAIDs) to appropriate chemical
spacers bearing a NO-donating moiety has led to a new class of drugs called NO-NSAIDs
or CINODs (cyclooxygenase-inhibiting nitric oxide donors). The NO-donating aspirin
derivative, NCX 4016, has been one of the leading compounds of this new class.[151] In LDLR−/− and ApoE−/− mice kept on a high-fat diet, NCX 4016 reduced balloon injury-induced carotid artery
intimal hyperplasia significantly more than aspirin.[152] Moreover, chronic administration of NCX4016 to mice strikingly reduced arterial
intimal thickening induced by photochemical injury, an effect potentiated by the concomitant
administration of clopidogrel. The observation that the drug combination producing
the most striking inhibition of platelet aggregation gave the most effective inhibition
of intimal proliferation, suggests that a very profound suppression of platelet activation
may be required to prevent atherosclerosis.[145] In a prospective, randomized, placebo-controlled, double blind clinical trial in
patients with peripheral arterial disease, the administration of NCX-4016 for 6 months
significantly reduced carotid intima-media thickness,[153] an index of atherosclerosis progression.[154] Similarly, the NO-releasing statins NCX 6560 and NCX 6550, which combine atorvastatin
and pravastatin with a NO moiety, reduced atherosclerosis, the generation of ROS,
circulating inflammatory cytokines, and endothelial dysfunction in strongly hyperlipidemic
mice.[155]
[156]
GPVI is a key platelet receptor for collagen involved in platelet-mediated atheroprogression.
The soluble dimeric GPVI receptor fusion protein Revacept, an antagonist of collagen-mediated
platelet activation, significantly improved endothelial dysfunction and decreased
vessel wall thickening in cholesterol-fed rabbit.[157]
Platelet adhesion is initiated by the binding of von Willebrand factor (VWF) linked
to subendothelial collagen to platelet GPIb which slows down the movement of platelets
along the vessel surface. Prolonged blockade of platelet GPIbα with a blocking monoclonal
antibody profoundly reduced leukocyte accumulation in the arterial intima and attenuated
atherosclerotic lesion formation in hyperlipidemic mice.[3] However, no drug inhibiting the GPIb/VWF interaction is currently available for
clinical use.
Firm adhesion of platelets to activated endothelial cells depends also on the binding
of platelet αIIbβ3 to endothelial ICAM-1.[158] The knocking out of the gene for GPIIb (integrin αIIb) reduced lesion formation
in ApoE−/− mice, showing that platelet GPIIb mediates firm platelet adhesion at damaged arterial
sites contributing to the development of atherosclerosis.[159] A small mechanistic study in patients with acute coronary syndromes caused by plaque
erosion serially studied by optical coherence tomography showed that in patients treated
with tirofiban, a GPIIb/IIIa antagonist, the increase in coronary stenosis 1 year
after the acute event was significantly smaller than in those not receiving tirofiban.[160] However, currently available GPIIb/IIIa antagonists are short-lived and only used
intravenously, and it seems unlikely that a short-lasting drug administration may
prevent atherosclerosis development on the long term.
Conclusion
While it is well established that platelets are the crucial actors in arterial thrombosis,
their role in atherogenesis is still being unraveled. The inflammatory nature of atherosclerosis
and the now clarified characterization of platelets as inflammatory cells explain
the participation of platelets in atherosclerosis. Platelets contribute to atherosclerosis
by participating in all the phases of the atherogenic process, from the very initial
functional damage to the endothelium, favoring LDL and inflammatory cell penetration,
through the formation of the lipidic necrotic core, to plaque unstabilization. The
interaction of activated platelets with the endothelium at lesion-prone sites is crucial
in this sequence of events and is a multistep process involving several platelet-exposed
or -released proteins, the relative contribution of which probably changes depending
on rheological and systemic factors. While current antiplatelet agents seem unable
to prevent the contribution of platelets to atherogenesis, in line with the observation
that there is a dichotomy in the signaling pathways regulating the platelet hemostatic
versus platelet inflammatory functions,[108] the inhibition of platelet secretion, of the release of MMPs, and of some specific
pathways of platelet adhesion to the vessel wall, may represent promising future strategies
for the prevention of atheroprogression. Further research to completely unravel the
mechanisms regulating the participation of platelets to atherosclerosis in humans
and to identify new platelet targets is highly warranted in order to develop novel
therapeutic approaches, further improving the antiatherosclerotic effects of the current,
highly active hypolipemic agents.
