Vascular smooth muscle cells in atherosclerosis: time for a re-assessment

4.1.1 Contractile vs. synthetic VSMCs

The switch from contractile to synthetic phenotype is defined by loss of contractile markers (e.g. ACTA2, MYH11, TAGLN and calponin) with gain of synthetic organelles, migratory and proliferative properties, and plays a critical role in both neointimal hyperplasia and atherogenesis⁴² (Figure 3).²⁶ Historically, this phenotypic switch was viewed as the hallmark of vascular repair, a reversible process wherein VSMCs return to their contractile state after completing vessel repair.^5,44 For example, neointimal VSMCs re-acquire their contractile markers after injury in a rat balloon injury model.⁴⁵ In contrast, VSMCs retain their non-contractile status during atherosclerosis due to the continuous exposure to phenotypic switching-inducing stimuli, although they may maintain their ability to revert back to their contractile phenotype.³⁸

4.1.2 Foam cells

Historically, foam cells were viewed as bone marrow-derived macrophages in fatty deposits in artery walls, where they ingest low-density lipoproteins (LDL), giving them a foamy appearance. While macrophages do contribute to foam cells within atherosclerotic plaques, VSMCs also transform into foam cells when exposed to aggregated or oxidised LDL in the intima and media of hyperlipidaemic mice and humans,^16,46 and the majority of foam cells in atherosclerotic plaques appear derived from VSMCs.^16,46 Lipid uptake traditionally occurs through specialized scavenger receptors (e.g. SR-A and CD36), but may also involve alternative pathways, such as phagocytosis and autophagy.^27,47 While clearance of lipid species may be beneficial at first, lipid loaded-VSMCs in culture secrete a variety of pro-inflammatory mediators⁴⁸ and undergo apoptosis from free cholesterol overload,⁴⁹ likely compromising plaque stability (Figure 3). The cholesterol excess in VSMCs is likely the result of inadequate cholesterol efflux due to reduced expression of the ABCA1 efflux transporter^16,46 and/or impaired autophagy in advanced plaques.⁵⁰

4.1.3 Macrophage-like

Macrophage-like VSMCs occur in both human^11,16 and mouse^17,27 atherosclerotic plaques, and also in the media of atherosclerotic mice.²⁶ The switch to a macrophage-like cell is likely driven by lipid accumulation, as cholesterol loading transdifferentiates VSMCs into macrophage-like cells in culture,⁵¹ and can be reversed by stimulating cholesterol efflux via apolipoprotein A1 and high-density lipoprotein (HDL).¹⁷ Macrophage-like VSMCs up-regulate ‘macrophage markers’, such as LGALS3, CD68,⁵¹ and pro-inflammatory cytokines [e.g. monocyte chemoattractant protein 1 (MCP1)],¹¹ yet not all lipid-laden VSMCs adopt a macrophage-like phenotype in vivo³⁶ and not all plaque VSMCs expressing LGALS3 adopt a macrophage-like state.³⁷ Instead, LGALS3 is expressed in >60% of mouse plaque VSMCs during a transition state to a non-macrophage but pro-inflammatory and osteochondrogenic (see Section 4.1.5) VSMC phenotype, and may be the first cell type to invest the plaque.³⁷ Hence, the question arises whether macrophage-like VSMCs are as harmful for plaque stability as their myeloid counterparts. Indeed, macrophage-like VSMCs are transcriptionally and functionally different from myeloid-derived macrophages in culture^17,52 and emerge later in atherogenesis than myeloid-derived macrophages that enter via the circulation.³⁸ Overall, the presence of macrophage markers in VSMCs does not imply full functionality as macrophages nor cell specificity, making it difficult to predict their impact on atherosclerotic plaque growth and stability. Nevertheless, based on their pro-inflammatory profile, macrophage-like VSMCs are likely detrimental for plaque stability (Figure 3), although their reversibility might offer the opportunity to therapeutically inhibit phenotypic switching.

4.1.4 Adipocyte-like

Lipid accumulation in VSMCs may result from both uptake (and impaired degradation or efflux) of extracellular lipoproteins and de novo lipogenesis of fatty acids via peroxisome proliferator-activated receptors gamma and the liver X receptor pathway.⁵³ For example, human VSMCs cultured in an adipogenic differentiation medium transdifferentiate into adipocyte-like cells expressing adipsin and leptin. Although VSMCs do not fully develop into mature adipocytes, lipid-containing VSMCs express elevated adipsin and lipogenesis genes in human atherosclerotic plaques.⁵³ Exposure of mouse VSMCs to a similar adipogenic cocktail promotes VSMC proliferation and activation of pro-inflammatory and pro-fibrotic pathways,⁵⁴ although whether this is associated with a switch to adipocyte-like VSMCs is unclear. These studies emphasize again the high level of plasticity of VSMCs, but the role and abundance of adipocyte-like VSMCs in atherogenesis is unclear, illustrated by a postulated neutral-to-negative position on the ‘plaque-stability scale’ (Figure 3).

4.1.5 Osteochondrogenic

Vascular calcification is characterised by hydroxyapatite deposits in the vessel wall, and is associated with increased CVD risk.⁵⁵ Vascular calcification is a tightly regulated process, primarily driven by VSMCs developing an osteochondrogenic phenotype in response to high levels of calcium and phosphate,^56,57 and lineage-tracing studies in mice show that 98% of all osteochondrogenic cells in plaques are VSMC-derived.⁵⁸ Osteochondrogenic VSMCs up-regulate osteochondrogenic markers [e.g. runt-related transcription factor 2 (RUNX2), SRY-related HMG-box (SOX9), and osteopontin)], deposit calcification-prone matrix (e.g. collagen type II and X), and secrete calcifying vesicles.⁵⁷ Simultaneously, cells down-regulate expression of molecules that inhibit calcification (e.g. vit K-dependent matrix Gla-protein and fetuin A). However, it is unclear whether contractile VSMCs can directly transform into an osteochondrogenic phenotype⁵⁷ or first have to undergo a switch to a synthetic phenotype,^59,60 or an intermediary LGALS3⁺ cell state.³⁷ Vascular calcification occurs in the intima of atherosclerotic plaques but also in the media where it promotes arterial stiffness.^61,62 Moreover, mouse and human coronary VSMCs with an osteogenic gene signature are associated with plaque destabilisation.³⁷ In particular, micro-calcified deposits (<50 µm) are associated with increased inflammation and mostly observed in the fibrous cap of human lesions, and thus considered detrimental for plaque stability. In contrast, macrocalcifications (>200 µm) often accumulate in the deep intima or necrotic core in organized structures and may promote plaque stability.⁶³ Interestingly, cultured calcifying VSMCs are still functionally different from mature, bone-forming osteoblasts, and more susceptible to apoptosis.⁶⁴ The opposite roles of micro- and macro-calcification in atherosclerosis limits its therapeutic potential and is illustrated by a neutral position on the ‘plaque stability scale’ (Figure 3).

4.1.6 Myofibroblast-like

A myofibroblast-like or ‘fibromyocyte’³⁶ refers to differentiation of a contractile VSMC into a fibroblast-like cell. Myofibroblast-like VSMCs exhibit a fibroblast-like transcriptional signature characterised by increased expression of fibronectin 1, osteoprotegerin, collagen type 1α1, and various small proteoglycans, including up-regulation of fibroblast-specific pathways. These myofibroblast-like cells have been identified in mouse plaques, particularly accumulating in the fibrous cap, and in a scRNA-seq dataset from human coronary plaques.³⁶ Differentiation into fibromyocytes is regulated, at least partially, by the CAD-associated transcription factor TCF21. Reduced TCF21 expression is associated with increased CVD risk, and loss of Tcf21 inhibits phenotypic modulation and attenuates fibromyocyte number in the fibrous cap, suggesting an athero-protective role (Figure 3).³⁶ This study also showed that the majority of VSMCs undergo phenotypic modulation into one main cell type rather multiple different phenotypes,^27,37,38 that phenotype modulation can occur in the media, and that genes that are causally associated with CVD can fundamentally alter VSMC phenotype switching in vivo.

4.1.7 EC-like

A limited number of studies report that contractile VSMCs can transdifferentiate into an EC-like phenotype. For example, mouse VSMCs cultured under laminar shear stress show increased expression of EC markers (CD31, von Willebrand factor, and VE-cadherin).⁶⁵ More recent work shows that human VSMCs can convert into ECs in culture, albeit indirectly, via an intermediary progenitor state elicited by stem cell reprogramming factors,⁶⁶ with further differentiation into ECs regulated via Notch signalling. Interestingly, these VSMC-derived ECs exhibit fully fledged EC functions and express CD31 in vitro and in vivo but could potentially promote intraplaque neovascularisation and haemorrhage associated with human plaque rupture⁶⁷(Figure 3). Direct evidence of endothelial-like VSMCs in human atherosclerosis is currently lacking, although multipotent VSMC-derived intermediary cells expressing endothelial markers were found in a scRNA-seq dataset from mouse and human plaques.³⁸

4.1.8 MSC-like

A rare population of MSC-like Sca1⁺ cells has been identified in medial differentiated VSMCs, which is increased upon vessel injury and exposure to stimuli known to induce phenotypic switching.³⁵ These medial Sca1+ VSMCs might represent a ‘first responder’ subpopulation that could undergo or regulate clonal expansion in atherosclerosis in response to environmental cues. Differentiated medial VSMCs can also give rise to adventitial Sca1⁺ progenitor cells that contribute to adventitial remodelling.⁶⁸ The VSMCs undergo a ‘reprogramming process’ controlled by Krüppel-like factor 4 (KLF4) and can transdifferentiate into different cell types, including macrophage-like, EC-like, chondrocyte-like, and adipocyte-like under defined in vitro conditions. Although these studies identify VSMC migration in opposite directions, they highlight the ability of Sca1⁺ VSMC-derived cells to promote VSMC transdifferentiation and underline the extremely high degree of plasticity of VSMCs. The highly plastic nature of mesenchymal-like VSMCs offers the opportunity to therapeutically push these cells into a plaque-stabilising phenotype (Figure 3).

4.2.1 Transcriptional

The expression of contractile genes is controlled by the serum response factor (SRF) and its co-activator myocardin (MYOCD), the former binding as a homodimer to the DNA consensus sequence CC(A/T)6GG known as a CArG box within the promoters of contraction-related genes.⁷² SRF is ubiquitously expressed and cannot activate transcription of contractile genes alone; binding to MYOCD—which is selectively expressed in VSMCs and cardiac muscle—is required to promote contractile gene transcription.⁷³ Many factors that control VSMC phenotypic switching alter MYOCD expression or activity either by direct binding to the Myocd promoter or interfering with its binding to SRF (Figures 1 and 3). MYOCD is the chief regulator of VSMC differentiation, although other TCFs have been identified to interact with SRF, including co-activators and co-repressors, such as KLF4 and ETS-like-1 protein (ELK-1).⁶⁹

The pluripotency factor KLF4 is a major regulator of VSMC phenotype modulation.¹¹ KLF4 represses expression of contractile genes via different mechanisms,⁷⁴ including: (i) binding to the G/C repressor element of contractile gene promoters, (ii) recruitment of histone deacetylases (HDAC2/5), causing transcriptional repression of contractile genes, (iii) binding to SRF, thereby antagonizing SRF binding to CArG-boxes, and (iv) complexing with phosphorylated ELK1 and SRF, thereby antagonizing MYOCD binding to SRF. KLF4 induces VSMC transdifferentiation into macrophage-like and MSC-like phenotypes and promotes atherosclerosis.¹¹ Moreover, KLF4 contributes to VSMC switching into LGALS3⁺ chondrocyte-like cells,³⁷ and osteogenic cells by binding to and cooperating with RUNX2.⁷⁵ The relative abundance of Runx2 and Sox9 regulates differentiation into osteogenic or chondrogenic lineages, respectively.⁵⁷ Octamer binding transcription factor (OCT4) is another pluripotency factor that regulates phenotypic switching, but has opposite effects to KLF4 (Figure 3).⁷⁶ Indeed, plaques of VSMC Klf4 vs. Oct4 knockout mice have virtually opposite gene expression patterns and their putative target genes play an important role in regulating VSMC phenotypic changes.³⁷ In contrast to Klf4, Oct4 reduces atherogenesis and promotes fibrous cap VSMC investment, potentially via increased migration.⁷⁶ Oct4 activation in VSMCs is associated with hydroxy-methylation of the promoter and transcriptional regulation via hypoxia-inducible factor 1 alpha and KLF4.⁷⁶

Other TCFs regulate VSMC differentiation through MYOCD (Figures 1 and 3). For example, TCF21 binds to the Myocd promoter, reducing its expression and interfering with the formation of the MYOCD-SRF complex.⁷⁷ The nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB) p65 subunit binds to the Myocd promoter, decreasing its expression; repression of Myocd by NFκB acts in concert with KLF4,⁷⁸ while forkhead box O3a (FOXO3a) binds to the Myocd promoter and negatively regulates Myocd and contractile marker gene expression.⁷⁹

4.2.2 MicroRNAs

miRNA143 and miRNA145 are the most well-documented microRNAs regulating the contractile phenotype of VSMCs (Figures 1 and 3).⁸⁰ miRNA143/145 block VSMC de-differentiation and proliferation by inhibiting Klf4 and Elk1 through binding with their mRNA 3′ UTR region.⁸¹ miRNA143/145 expression is positively controlled by SRF and MYOCD, but reduced in injured or atherosclerotic aortas.⁸¹ Moreover, cholesterol loading promotes VSMC transdifferentiation into macrophage-like cells via the miRNA143/145-MYOCD axis.⁵² Hence, down-regulation of miRNA143/145 may promote atherogenesis through VSMC de-differentiation, while the MYOCD-induced miRNA1 can target Klf4 and promote VSMC differentiation.⁸²

MicroRNAs implicated in VSMC phenotypic switching (Figure 1) also include miRNA221 and miRNA222.⁸³ miRNA221 stimulates proliferation by down-regulating the cell-cycle inhibitor p27^Kip1, while miRNA221 targets the proto-oncogene c-kit, which reduces MYOCD expression followed by reduced contractile gene expression.⁸⁴ miRNA221/222 also promote VSMC switching into osteochondrogenic like-cells and calcification.⁸⁵ miRNA21 and miRNA124 inhibit VSMC differentiation by down-regulating specificity protein 1;^86,87 the latter promotes a contractile VSMC phenotype by binding to the G/C repressor element in the promoter of contractile genes.

In addition to microRNAs, some long non-coding RNAs (LncRNAs) regulate VSMC proliferation and phenotype. LncRNAs regulate gene expression via different mechanisms, including through ‘sponging’ miRNAs. For example, the lncRNA SMILR promotes VSMC proliferation through direct regulation of mitotic processes and its expression is increased in unstable vs. stable human atherosclerotic plaques.⁸⁸ The lncRNA MALAT1 stimulates VSMC proliferation and migration⁸⁹ and promotes aortic stiffness.⁹⁰

4.2.3 Epigenetic

Epigenetic modifications of amino acid residues on histones (e.g. acetylation and methylation) regulate contractile gene expression during VSMC phenotypic switching by altering chromatin structure and changing accessibility for TCFs⁷⁰ (Figure 3). For example, Acta2 and Myh11 are activated through histone acetylation during VSMC differentiation allowing SRF binding.⁹¹ The transcriptional activity of MYOCD is enhanced by histone acetylation via p300 histone acetyltransferase (HAT) or inhibited by class II HDACs.⁹² In contrast, KLF4-mediated recruitment of HDAC2/5 to the G/C repressor element represses contractile gene expression.⁹³ Collectively, increased histone acetylation of CarG box chromatin facilitates contractile gene expression by improving chromatin accessibility for MYOCD/SRF complexes through stearic hindering. HATs and HDACs also regulate many other VSMC processes including proliferation, migration, and matrix deposition that may influence VSMC phenotype and behaviour in atherosclerosis.⁹⁴

MYOCD/SRF binding to CArG box chromatin is also associated with histone methylation. The histone modification H3K4me2 is enriched at the promoter of contractile genes during VSMC differentiation and is associated with increased tethering of MYOCD.⁹⁵ Interestingly, H3K4me2 of the Myh11 locus is restricted to VSMCs (Figure 1) in humans and mice, and persists in phenotypically modulated VSMCs that have lost their contractile markers, permitting its use to identify VSMCs in human atherosclerosis.¹² The retention of an epigenetic signature in phenotypically modulated VSMCs as a reminder remnant of their contractile state might explain the reversible nature of phenotype switching.⁴³

4.2.4 Environmental stimuli

The platelet-derived growth factor (PDGF-BB) is up-regulated upon injury and promotes VSMC proliferation and de-differentiation via multiple mechanisms⁶⁹ (Figures 1 and 3). Through the PDGFRβ receptor, PDGF-BB stimulates ELK1 phosphorylation, which competes with MYOCD for the same docking site on SRF, suppressing contractile gene expression.⁹⁶ In addition, the effects of PDGF-BB on phenotypic switching are partially mediated via KLF4⁹⁷ and miRNA221.⁸⁴ The Wnt/β-catenin signalling pathway promotes VSMC proliferation and intimal thickening⁹⁸ and promotes VSMCs osteogenic transdifferentiation through activation of the Runx2 promoter.⁹⁹

TGFβ is one of few growth factors that promotes VSMC differentiation (Figures 1 and 3). Contractile gene expression is regulated via the TCFs SMAD2 or SMAD3 that either bind to SMAD binding sites within contractile gene promoters or interact with SRF in a CArG-dependent manner.¹⁰⁰ Moreover, TGFβ blocks VSMC phenotypic switching by down-regulating Klf4 through miRNA143/145.¹⁰¹ The latter are transcriptionally activated by TGFβ through MYOCD and CArG box binding. TGFβ is a multifunctional growth factor that regulates VSMC proliferation and hypertrophy, and has both matrix synthesising (e.g. stimulates production of collagens) and proteolytic (e.g. activates matrix metalloproteinases (MMPs)) properties, and can thus have secondary effects on VSMC phenotype.⁶⁹

Recently, integrin β3 has been shown to regulate VSMC phenotype and clonality through cell-autonomous and paracrine effects.¹⁹ Integrin β3 is produced by VSMCs and macrophages, and inhibits VSMC transdifferentiation into macrophage-like cells and limits VSMC proliferation and migration (Figure 1). In atherosclerosis, integrin β3 prevents clonal expansion of VSMC-derived cells in the fibrous cap and lesion, suggesting that macrophages may regulate the recruitment of medial VSMCs into the developing lesion via secretion of integrin β3.¹⁹

5.1.1 VSMC migration

VSMC migration from media to intima is an established part of most theories of atherogenesis, but there are multiple unanswered questions about its precise role and regulation. For example: (i) what is the contribution of migration to proliferation in human (or indeed mouse) atherosclerosis, (ii) does migration depend upon contractile-synthetic de-differentiation, and (iii) does migration of VSMCs occur before proliferation, or can both processes occur simultaneously. In general, VSMCs have been described to migrate into the intima followed by proliferation and formation of the fibrous cap;^42,44,102 migration of medial VSMCs may be preceded by proliferation¹⁰³ or both dividing and non-dividing VSMCs can migrate and contribute to lesion formation.²³ However, lineage-tracing studies showing VSMC monochromatic clonal patches in plaques instead of the medial mosaic suggests that VSMC migration occurs independent of proliferation, and is not a major contributor to lesion formation.^13,19,27 Similarly, the observation that VSMC-derived neointimal patches are connected to medial patches expressing the same reporter colour suggests that VSMC proliferation already occurs in the media and thus precedes migration.²⁷ Although suggestive, these theories are based on indirect methods of exclusion of proliferative clonally expanded VSMCs; we therefore need new methods to directly monitor VSMC migration to reveal its true contribution to atherosclerosis.

5.1.2 VSMC proliferation

Historically, VSMCs were thought to accumulate in atherosclerotic lesions by ongoing proliferation in response to injury or inflammatory cues. However, VSMCs in or derived from advanced atherosclerotic plaques exhibit low proliferative indices,^104,105 and recent lineage-tracing studies might explain this proliferation paradox. For example, lineage-tracing indicates that VSMC proliferation initiates in the media and VSMC progeny then migrate into the intima, where they continue to divide in an oligoclonal fashion (Figure 2).²⁷ These few VSMC clones are highly proliferative and form the fibrous cap and invade the plaque core,¹⁹ possibly reflecting migration of ACTA2⁺ fibrous cap cells. Multiple rounds of cell division to form large oligoclonal plaques potentially explain the cellular senescence (and therefore low proliferative indices) observed in advanced atherosclerotic plaques.

5.1.3 VSMC senescence

Cellular senescence is defined as a state of irreversible growth arrest and categorized as replicative senescence, due to replicative exhaustion of cell lifespan, or stress-induced premature senescence induced by many external stimuli, such as DNA-damaging agents or oxidative stress.¹⁰⁶ Both types of senescence activate a DNA damage response to initiate growth arrest and a senescent secretory phenotype (SASP) that creates a pro-inflammatory environment (see Section 5.2.3).¹⁰⁶ Human plaque VSMCs show reduced proliferation, early senescence, and up-regulation of cell-cycle arrest markers including p16^INK4a and p21, associated with hypophosphorylation of the tumour suppressor retinoblastoma protein.¹⁰⁷ Moreover, human plaque VSMCs have much shorter telomeres than medial VSMCs.¹⁰⁷ VSMC senescence promotes atherosclerosis and features of plaque instability in mice when genes promoting senescence are expressed from the Tagln promoter,^108,109 potentially driven via an IL1α-dependent SASP,¹¹⁰ but it is unclear whether different VSMC phenotypes show similar levels of cell senescence and how they are regulated. Nevertheless, there is strong evidence linking VSMC senescence and aging with medial calcification.⁵⁷ For example, replicative senescence¹¹¹ or IL1β-induced senescence¹¹² of VSMCs enhances calcification through initiating the transition to an osteoblastic phenotype in vitro.

5.2.1 ECM proteins

VSMCs produce ECM as a major structural component of the vessel wall.¹¹³ The mechano-elastic properties of arteries are largely maintained by three ECM types: elastic fibres, fibrillar collagen, and large aggregating proteoglycans, but the latter can also promote atherosclerosis by retaining apolipoprotein B-lipoproteins.¹¹³ The interaction between VSMCs and ECM is a dynamic, bi-directional process, and ECM content is determined by the balance between production and degradation. In early plaque development, MMPs are essential for VSMC migration by degrading the cage of connective tissue surrounding VSMCs.¹¹⁴ However, both the type and amount of ECM synthesised varies between VSMC phenotypes, and cultured synthetic VSMCs, macrophage-like VSMCs and VSMC-derived foam cells secrete proteolytic MMPs.

Recent advances in proteomics have enabled significant progress in our understanding of the role of ECM in atherosclerosis. For example, a 4-protein signature comprising MMP9, S100A8/S100A9, cathepsin D, and galectin-3-binding protein can differentiate symptomatic vs. asymptomatic carotid plaques, and could be used as biomarkers for risk stratification of CVD patients.¹¹⁵ Similarly, a set of fibrous cap-related proteins including vitronectin, TGFβ-induced protein, complement factor 7, and apolipoprotein B have been identified in the plasma of CAD patients.¹¹⁶ Mouse atherosclerotic aortas show an osteoclast-specific signature among known collagenases and MMPs,¹¹⁷ and a few novel ECM factors, such as matrilin-2 and peroxidasin, although their exact role in atherosclerosis and relation to VSMCs is unclear. Proteomic profiling of secreted proteins from an atherosclerotic swine model revealed differential expression of many proteins related to VSMC activation or migration.¹¹⁸ Chitinase-3-like-protein 1 in particular was up-regulated in plaque VSMCs in relation to disease stage, and in plasma of fat-fed animals vs. controls. The advances in proteomics and emerging discoveries regarding the role of ECM in the atherosclerosis pave the way for the development of novel clinical disease biomarkers.

5.2.2 Pro-inflammatory mediators

Synthetic VSMCs play a major role in forming and maintaining the fibrous cap by secreting ECM, yet their overall role in plaque stability could be detrimental. Synthetic VSMCs secrete a wide variety of pro-inflammatory molecules and matrix-degrading enzymes that can cause cell death of neighbouring cells.^42,114 Pro-inflammatory cytokines, such as IL1β, IL6, and MCP1 promote atherogenesis by stimulating monocyte recruitment and cell death¹¹⁹ and synthetic VSMCs express a range of adhesion molecules and receptors (e.g. Toll-like-receptors) that promote monocyte recruitment and regulate intracellular inflammatory signalling, respectively. Moreover, synthetic VSMCs secrete extracellular vesicles (EVs) that can drive vascular calcification (see Section 5.2.4).⁵⁹ Hence, synthetic VSMCs are beneficial for fibrous cap formation, but depending on the local environment and the disease stage, may promote inflammation, calcification, cell senescence, and plaque instability.

5.2.3 SASP

Senescent cells secrete a spectrum of cytokines, chemokines, and growth factors known as SASP, that drives many age-related diseases including atherosclerosis by sustaining a state of chronic low-grade inflammation, called ‘inflammaging’.¹²⁰ Clearance of senescence cells by immune cells can be impaired in disease, propagating their accumulation.¹²¹ The SASP can self-amplify its secretory network in an autocrine interleukin-dependent manner,¹²² and induce paracrine senescence mediated via TGFβ¹²³ or via EVs (see Section 5.2.4).¹²⁴ Through their SASP, senescent cells attract immune cells and induce proliferation, apoptosis, or alter the differentiation status of neighbouring healthy cells,¹⁰⁶ including phenotypic transition of neighbouring VSMCs. For example, IL6 released by senescent human VSMCs favours their transition to an osteogenic phenotype¹²⁵ but may also promote reprogramming of neighbouring cells into a stem cell-like phenotype,¹²⁶ implying that a small number of senescent cells can have marked effects on disease development.

5.2.4 VSMC-derived vesicles

VSMCs shed different types of EVs, which can be categorized based on their size: exosomes<micro-vesicles<apoptotic bodies. EVs can mediate cell–cell communications and play important roles in both physiological and pathophysiological processes.¹²⁷ For example, EV contents include bioactive molecules, such as lipids and proteins, but also DNA, mRNA, microRNA, and lncRNA.¹²⁸ In addition to transferring cargo, EVs interact and exchange crucial information with the target cell, to change their phenotype or even reprogramme the recipient cell.¹²⁸

Synthetic VSMCs in particular secrete exosomes and may drive calcification when the EV contains phosphatidylserine (PS), Annexin A6, and low concentrations of calcification inhibitors.¹²⁹ By exposing PS on the surface, and possibly tissue factor, exosomes may also promote vascular coagulation.¹²⁹ Matrix vesicles containing remnants of apoptotic VSMCs can act as a nucleation site for calcification in plaques,¹³⁰ and osteochondrocyte-like VSMCs secrete calcifying vesicles that further propagate calcification.⁶⁰ In addition, senescent cells secrete increased EVs compared to non-senescent cells,^124,131 which promote paracrine senescence,¹²⁴ cell proliferation, inflammation, wound healing repair, and DNA damage response pathways.^124,131,132

5.3.1 Apoptosis

Apoptotic indices are low in early plaques but increase as lesions become more advanced¹³⁵ and are higher in symptomatic plaques compared to stable lesions.¹³⁶ Apoptotic debris accumulates particularly in the core region, but apoptosis markers are transient, such that apoptotic indices do not measure apoptosis rates or efficiency of apoptotic cell clearance, and can label necrotic cells.¹³⁷ Given these limitations, the effects and incidence of apoptosis of different VSMC phenotypes are mostly unknown. VSMC apoptosis driven by expression of a suicide gene from the Tagln promoter accelerates atherogenesis, and promotes plaque vulnerability, calcification, and medial degeneration,¹³⁸ suggesting that apoptosis of medial contractile VSMCs or myofibroblast-like VSMCs in the intima is detrimental. However, further delineation of the effects of apoptosis of different VSMCs requires identification of phenotype-specific TCFs for selective cell killing.

5.3.2 Necrosis

Necrosis has long been considered an accidental and uncontrolled form of cell death. However, recent studies demonstrate that necrosis is a regulated cell death mechanism, with subtypes, such as necroptosis regulated by receptor interacting protein kinases 1 and 3 and ferroptosis characterised by accumulation of lipid-peroxidation products and lethal amounts of reactive-oxygen species derived from iron metabolism.¹³³ Necrosis in atherosclerosis is triggered by various stimuli, such as high levels of reactive-oxygen species, highly oxidised LDL, depletion of cellular energy, increased intracellular calcium levels, and mitochondrial Ca²⁺-overload.¹³⁴ Inefficient efferocytosis of apoptotic VSMCs also leads to secondary necrosis and IL1-driven inflammation.¹³⁹

Plaque cell necrosis contributes to plaque vulnerability by releasing pro-inflammatory mediators (e.g. IL1α), MMPs, and thrombogenic factors (e.g. tissue factor) whereas inhibition of necrosis improves features of plaque stability in mice.¹³⁴ However, lineage-tracing studies suggest that we need to re-evaluate the role of necrosis in atherosclerosis. Plaque cell necrosis is usually quantified by measuring the necrotic core, and necrotic cells have been reported to be predominantly (foam cell) macrophages. However, the necrotic core also contains lipid material (membrane remnants, cholesterol esters, and crystals) and lineage-traced VSMC-derived cells.¹³ The precise incidence and role of necrosis in different VSMC phenotypes will require studies using lineage-tracing coupled to specific markers of necrosis in vivo.