nature reviews cardiology https://doi.org/10.

1038/s41569-024-01115-w

Review article Check for updates

Atherosclerosis in diabetes mellitus: novel

mechanisms and mechanism-based
therapeutic approaches
Abdul Waheed Khan 1
& Karin A. M. Jandeleit-Dahm 1,2

Abstract Sections

Atherosclerosis is a disease of large and medium arteries that can lead Introduction

to life-threatening cardiovascular and cerebrovascular consequences, Emerging mechanisms and

such as myocardial infarction and stroke. Moreover, atherosclerosis is therapeutic targets

a major contributor to cardiovascular-related mortality in individuals Current therapies and

strategies in clinical
with diabetes mellitus. Diabetes aggravates the pathobiological development
mechanisms that underlie the development of atherosclerosis.
Emerging therapeutic
Currently available anti-atherosclerotic drugs or strategies solely strategies
focus on optimal control of systemic risk factors, including Conclusions
hyperglycaemia and dyslipidaemia, but do not adequately target the
diabetes-exacerbated mechanisms of atherosclerotic cardiovascular
disease, highlighting the need for targeted, mechanism-based therapies.
This Review focuses on emerging pathological mechanisms and related
novel therapeutic targets in atherosclerotic cardiovascular disease in
patients with diabetes.

Department of Diabetes, Central Clinical School, Monash University, Melbourne, Victoria, Australia. 2German
1

Diabetes Centre, Leibniz Centre for Diabetes Research at the Heinrich Heine University, Dusseldorf, Germany.
e-mail: [email protected]

Nature Reviews Cardiology

Review article

Key points disease and stroke. Understanding the mechanisms and strategies
for managing atherosclerotic CVD in patients with diabetes, as well
as the potential cardiovascular risks and benefits of glucose-lowering
• Diabetes mellitus is associated with the cellular processes and drugs, is important for the management of CVD in these individuals.
pathways that promote atherosclerosis development, and atherosclerotic Multiple cellular and molecular mechanisms have been hypothesized
cardiovascular disease remains the major cause of the increased mortality to participate in diabetes-accelerated atherosclerosis development
observed in patients with diabetes compared with the general population. (Fig. 1). The development of atherosclerosis in individuals with dia-
betes follows the same disease stages as in patients without diabetes.
• Current clinical practices in the general population, including in Shared processes that contribute to atherosclerotic plaque devel-
patients with diabetes, focus on reducing disease-promoting risk opment include endothelial dysfunction, blood flow-induced shear
factors, such as hyperglycaemia, hyperlipidaemia and hypertension, stress, endothelial-to-mesenchymal transition (EndMT), monocyte
but do not target the diabetes-induced mechanisms of atherosclerosis cross-endothelial migration into the arterial wall and differentiation
development and progression. into macrophages, foam cell formation and vascular smooth muscle cell
(VSMC) proliferation. All these processes are accelerated in diabetes
• Several pathobiological processes that participate in the development (Fig. 1), with some of these effects being first identified decades ago.
of diabetes-accelerated atherosclerosis have been identified, including However, with advances in research technologies and imaging modali-
turbulent blood flow, endothelial-to-mesenchymal transition, inflammation ties, newly identified mechanisms are emerging as potential novel
and oxidative stress. therapeutic avenues for atherosclerotic CVD.

• Although these processes were identified decades ago, advances Blood flow-mediated shear stress
in technologies have now helped to identify promising therapeutic Atherosclerosis occurs at certain sites in blood vessels, determined by
molecular targets such as epigenetic enzymes and transcription the nature of haemodynamic forces4 such as shear stress (the frictional
factors linked to these processes. force of blood flow against the endothelial layer of the vascular wall). Ath-
erosclerotic plaques develop preferentially at branches, bends and bifur-
• New potential mechanism-based approaches, such as anti- cations in the arterial tree, where the vascular endothelium is exposed
inflammatory drugs and cell-specific nanomedicine, are emerging to to turbulent blood flow with oscillatory low shear stress5 (Fig. 2). By con-
combat atherosclerotic cardiovascular diseases, including in patients trast, straight parts of arteries are exposed to sustained laminar blood
with diabetes. flow with high shear stress and are generally spared from atherosclerosis
development4 (Fig. 2). The endothelial cells are a vulnerable vascular
cell population in the context of diabetes because they are continuously
Introduction exposed to haemodynamic forces and metabolic factors. Transmural
The prevalence of diabetes mellitus is exceptionally high worldwide, and circumferential stress in the arterial wall resulting from blood pres-
and the number of individuals with diabetes is expected to rise to sure mainly affects and regulates VSMCs in the vascular media, whereas
700 million by 2045 (ref. 1). Atherosclerotic cardiovascular disease blood flow-mediated frictional shear stress affects predominantly the
(CVD) is the major cause of increased morbidity and premature mor- endothelium, causing structural, functional, transcriptional and epig-
tality in individuals with diabetes2, and is associated with a high eco- enomic changes in the endothelial cells6. Mechanosensory complexes
nomic burden. Approximately two-thirds of deaths in individuals with present on the endothelial cell membrane serve as sensors for the differ-
diabetes are imputable to CVD, with almost 50% being from ischaemic ent blood flow patterns and transduce flow signals into mechanotrans-
heart disease or stroke, with atherosclerosis as the underlying cause3. duction pathways7. Activation of these pathways regulates the activity
Unfortunately, with the ongoing worldwide diabetes epidemic and and/or expression of transcription factors and epigenetic modifiers
despite the improved management of various cardiovascular risk (including epigenetic enzymes and non-coding RNAs), thereby generat-
factors, including hypertension, dyslipidaemia and hyperglycaemia, ing cellular responses to regulate endothelial function4,8. Although these
CVD remains a major burden on our health-care systems. Research blood flow-associated haemodynamic forces are not responsible for the
into atherosclerosis has led to many compelling hypotheses about the initiation of atherosclerotic plaque formation, they condition the vascu-
underlying pathophysiology of atherosclerotic plaque development lar endothelial cells at specific sites to respond distinctly to systemic risk
and the progression and clinical manifestations of atherosclerosis, such factors, including hypercholesterolaemia and hyperglycaemia9 (Fig. 2).
as myocardial infarction and stroke. However, despite these advances, Endothelial cell reprogramming at sites of turbulent, low shear stress
we still lack definitive evidence to identify appropriate therapeutic and oscillatory blood flow induces endothelial inflammation and dys-
strategies that can directly target the intrinsic pathobiology of athero- function, oxidative stress and EndMT10 (Fig. 2). These pro-atherogenic
sclerotic plaque development and progression. In this Review, we first pathways are accelerated in diabetes, driven by the hyperglycaemic
discuss the latest advances in our understanding of the pathobiological pro-inflammatory milieu11–13. The cardiovascular protective effects of
mechanisms of atherosclerosis development and progression in the the glucose-lowering sodium–glucose cotransporter 2 (SGLT2) inhibi-
context of diabetes and the novel therapies that are currently in clinical tors, such as empagliflozin, are not entirely attributable to improved
development. We then highlight emerging therapeutic approaches glycaemic control14. A non-randomized, open-label, prospective cohort
and novel targets that have been identified thanks to these advances. study demonstrated a direct effect of empagliflozin on increasing blood
viscosity and wall shear stress compared with incretin-based therapy in
Emerging mechanisms and therapeutic targets patients with type 2 diabetes15. In addition, intima–media thickness was
Important clinical manifestations of atherosclerotic CVD in the set- markedly decreased early after initiation of empagliflozin treatment
ting of diabetes include myocardial infarction, peripheral artery compared with incretin-based therapy15.

Nature Reviews Cardiology

Review article

Diabetes mellitus

Endothelial dysfunction Inflammation Oxidative stress EndMT Epigenetic changes

Leukocyte Monocyte Macrophage

↓ Antioxidant
mechanisms: ↑ Pro-oxidant ↓ 5mC: ↑ expression of ↑ H3K9ac ↑ HDAC3
↓ SOD1, mechanisms: VEFFB, PLGF, PLCB1, FATP4 ↑ H3K14ac ↑ Expression
Turbulent
SOD2 and ↑ NOX1 ↓ TET2: ↓ 5hmC content ↑ H4K5ac of TNF and
blood flow
Lumen SOD3 ↓ NOX4 ↑ HDAC9 levels ↑ H4K12ac COX2
↑ NOX5
Monocyte
Endothelial LDL Collagen
cell

Intima ICAM1/ IL-1β, IL-6,

VCAM1 TNF
ROS
Cellular debris
oxLDL Foam cell Apoptotic
cell Cholesterol
Macrophage Necrotic core crystals
Vascular smooth muscle cells

Media

Fig. 1 | Atherosclerosis development in diabetes. Diabetes mellitus is associated differentiation into macrophages, foam cell formation, and vascular smooth
with pathogenic processes that are also involved in atherosclerosis development. muscle cell proliferation and migration towards the plaque fibrous cap. 5hmC,
Turbulent blood flow primes endothelial cells at specific sites to respond distinctly 5-hydroxymethyl cytosine; 5mC, 5-methylcytosine; COX2, cyclooxygenase 2;
to systemic risk factors including hypercholesterolaemia and hyperglycaemia. H3K9ac, histone 3 lysine 9 acetylation; H3K14ac, histone 3 lysine 14 acetylation;
The stages of atherosclerotic plaque development are similar in individuals H4K5ac, histone 4 lysine 5 acetylation; H4K12ac, histone 4 lysine
with and without diabetes. However, atherogenesis is markedly accelerated in 12 acetylation; HDAC, histone deacetylase; ICAM1, intercellular adhesion
the context of diabetes. Common processes that contribute to atherosclerotic molecule 1; NOX, NADPH oxidase; oxLDL, oxidized LDL; ROS, reactive oxygen
plaque development include endothelial dysfunction, endothelial-to- species; SOD, superoxide dismutase; TET2, methylcytosine dioxygenase TET2;
mesenchymal transition (EndMT), monocyte cross-endothelial migration and TNF, tumour necrosis factor; VCAM1, vascular cell adhesion protein 1.

Blood flow-associated molecular mechanisms, including flow- Many of the pro-atherogenic genes listed above have been
sensitive transcription factors and epigenetic pathways, are poten- reported to be upregulated in diabetes-associated atherosclerosis
tial therapeutic targets for atherosclerosis. Advances in integrated in humans and in animal models, including ICAM1, VCAM1 and NOX1
cell-specific multiomics technologies and the development of (refs. 19,20). Our group has shown that NADPH oxidase 1 (NOX1) is a
in vitro models of fluid shear stress (including microfluidic devices, pro-atherogenic pro-oxidant whereas NOX4 is atheroprotective in
the parallel-plate chamber and the cone-and-plate viscometer) have streptozotocin (STZ)-induced diabetic Apoe−/− mice, an established
helped to identify novel therapeutic targets for atherosclerosis related mouse model of diabetes-associated atherosclerosis19,21. We have also
to processes in endothelial cells, such as oxidative stress, EndMT and shown in the same mouse model that atheroprone aortic regions
inflammation9. These studies have identified a long list of flow-sensitive exposed to turbulent blood flow have increased AP1 activity compared
gene targets, which have been summarized in multiple reviews8,9. Of with aortic regions exposed to laminar flow22. In addition, in an in vitro
note, the most important molecules with therapeutic potential, tak- microfluidics-based model of pro-atherogenic low shear stress in
ing into consideration the reproducibility of the findings in in vivo human aortic endothelial cells (HAECs), AP1 activity and expression
animal models and in the human context, include genes linked to were increased in the presence of high glucose concentrations com-
anti-atherogenic effects (KLF2, KLF4, NOS3 and NOX4), genes pared with conditions of high shear stress without high glucose22. AP1
encoding antioxidant proteins (SOD2 and SOD3) and genes linked levels were also increased in carotid endarterectomy specimens from
to pro-atherogenic effects (activator protein 1 (AP1) complex, CCL2, patients with diabetes compared with samples from individuals with-
ICAM1, VCAM1, NFKB, NOX1, NOX2, MMPs, MAPK1, MAPK3 and HIF1A)9,16. out diabetes22. Flow-sensitive epigenetic therapeutic targets include
In a mouse model of type 2 diabetes, Klf2 expression induced by physi- microRNAs (such as miR-10a, miR-19a, miR-23b, miR-205 and miR-21),
cal exercise was shown to activate endothelial nitric oxide synthase long non-coding RNAs (such as MALAT1, STEEL, LASSIE and LISPR1),
eNOS (also known as NOS3) to improve vasodilatation17. A study using DNA methyltransferases (DNMT1 and DNMT3), histone deacetylases
RNA sequencing and NanoString technology to investigate the gene (HDAC1–HDAC7 and SIRT1) and ­histone methyltransferases (such as
expression profile of atherosclerotic plaques from pigs and humans EZH2)8,9,16,23–26. On the basis of the cellular response to variable patterns
with or without diabetes identified KFL4 as a gene with a distinct of blood flow, two obvious therapeutic approaches would be to either
expression pattern that reflects the progression of atherosclerosis stimulate stable flow-induced atheroprotective molecules or inhibit
in diabetes18. turbulent flow-mediated pro-atherogenic molecules. These goals can

Nature Reviews Cardiology

Review article

Ageing Hypertension Diabetes mellitus Hyperlipidaemia Smoking

Laminar blood flow Turbulent blood flow

Epigenomic changes Epigenomic changes

• miR-10a, miR-19a, miR-23b • miR-92a, miR-205, miR-21
• HOTAIR, LASSIE, LISPR1, MANTIS • MALAT1, STEEL
• P300, HAT, SIRT1 • DNMT1, DNMT3, HDAC1-7, EZH2
Transcriptomic changes Transcriptomic changes
KLF2, KLF4, NOS3, NOX4, NRF2, Nucleus CCL2, HIF1A, ICAM1, MAPK1, MAPK3,
SOD2, SOD3 MMPs, NFKB, NOX1, NOX2, VCAM1

Endothelial cell

↓ Inflammation ↑ Inflammation
↓ EndMT ↑ EndMT
↓ Oxidative stress ↑ Oxidative stress
↑ Endothelial function ↓ Endothelial function

Fig. 2 | Shear stress-mediated vascular endothelial cell reprogramming. shear stress and oscillatory blood flow includes activation of a cell-specific
Low shear stress induced by turbulent blood flow primes vascular endothelial transcriptomic and epigenomic programme that induces endothelial
cells to respond distinctly to systemic risk factors, including hyperlipidaemia inflammation, oxidative stress, endothelial-to-mesenchymal transition (EndMT)
and hyperglycaemia. Endothelial cell reprogramming at sites of turbulent, low and endothelial dysfunction. miR, microRNA.

be achieved using small-molecule inhibitors or recombinant proteins, Cre–lox gene deletion strategies30–32. A study using an endothelial-specific
but ­preferentially using a site-specific drug delivery approach. lineage tracking system showed that EndMT-derived, fibroblast-like cells
are common in atherosclerotic plaques in mice31. Furthermore, in both
Endothelial-to-mesenchymal transition human and mouse atherosclerotic plaques, single-cell RNA sequencing
EndMT is a complex dynamic process that involves the transdifferentia- studies have also identified vascular endothelial cell subpopulations that
tion of endothelial cells into mesenchymal-like cells. During this process express mesenchymal markers, including in the context of diabetes, which
of transition, endothelial cells undergo a series of molecular changes suggests that EndMT is involved in atherosclerosis development33–36.
that lead to the phenotypic switch towards mesenchymal cells, such as These single-cell RNA sequencing studies have also identified endothelial
myofibroblasts and VSMCs (Fig. 3). Although no consensus exists on subpopulations that co-express different mesenchymal markers, which
the criteria to define EndMT, this process is mostly demonstrable by is indicative of intermediate EndMT (that is, endothelial cells under-
changes in morphology and in cellular markers of either endothelial going a transitioning state towards a mesenchymal phenotype)33–36.
or mesenchymal cells. Phenotypically and morphologically, endothe- The concept of a ‘partial’ or ‘intermediate’ phase in the EndMT process
lial cells lose their endothelial-specific properties, including loss of is gaining increased recognition37–39. The EndMT process is now generally
apical-to-basal membrane polarity and cell-to-cell adhesion27,28. Instead, accepted to be a continuum, in which an endothelial cell undergoing
the cells attain mesenchymal-specific properties, including a migratory EndMT transforms through multiple intermediate phases and can remain
phenotype and a spindle morphology27,28. Molecularly, endothelial cells in an intermediate phase37–39 (Fig. 3). Indeed, transitioning endothelial
undergoing EndMT show reduced expression of endothelial cell mark- cells have also been detected in human atherosclerotic plaques31. In the
ers, such as vascular endothelial cadherin (also known as cadherin 5) partial EndMT state, both endothelial and mesenchymal features can
and von Willebrand factor, and elevated expression of mesenchymal cell be present in the same cell, and this state is generally considered to be a
markers, including transgelin (TAGLN; also known as SM22α), ferropto- reversible phenomenon40.
sis suppressor protein 1, calponin 1 and α-smooth muscle actin (αSMA; To prevent or revert EndMT progression, certain EndMT signalling
also known as ACTA2)29. pathways, including transforming growth factor-β (TGFβ) signalling, have
EndMT is indispensable in the embryonic processes of cardiogenesis been targeted in preclinical studies in atherosclerosis as a potential thera-
and vasculogenesis, but the EndMT process that is aberrantly induced peutic strategy41,42. However, approaches targeting TGFβ systemically
by environmental factors, such as hyperglycaemia, can contribute to are associated with complications, including loss of protection against
diabetes-accelerated atherosclerosis13. Other EndMT-triggering factors vascular inflammation43. Endothelial-specific inhibition of TGFβ sig-
include low shear stress caused by turbulent blood flow, oxidized LDL nalling using 7C1 lipid nanoparticles carrying small interfering RNA
(oxLDL), hypoxia, oxidative stress and sustained inflammation, which are (siRNA) targeting Tgfbr1 and Tgfbr2 has shown promising results by
also common features of atherosclerosis development in diabetes (Fig. 3). suppressing EndMT and reversing atherosclerosis in atheroprone mice41.
Additional evidence of the involvement of EndMT in atherosclerosis has Epigenetic pathways have also been implicated in EndMT induction,
been provided by studies using endothelial-specific lineage systems or and studies have identified DNA methylation and histone modifications

Nature Reviews Cardiology

Review article

associated with EndMT44. Endothelial-specific genetic deletion of Hdac9 EndMT via multiple transcription factors, including SMAD, angioten-
has been shown to block EndMT and reduce atherosclerotic lesion size sin II, endothelin 1, poly(ADP-ribose) polymerase 1 and extracellular
and increase plaque stability in mice45. These studies have revealed the signal-regulated kinase51,52.
therapeutic potential of EndMT manipulation in a cell-specific manner,
but more studies are required to further explore the clinical t­ ranslatability Oxidative stress
of the findings. Oxidative stress, defined as an imbalance between excessive genera-
Although EndMT is commonly observed in diabetes-related compli- tion of reactive oxygen species (ROS) and reduced antioxidant defence
cations, including aortic stiffening36,46,47, only a few studies have explored mechanisms, has a key role in diabetes-associated atherosclerosis53–55.
the modulation of EndMT in diabetes-induced atherosclerosis. In a rat ROS can be generated by various systems in the vascular wall, includ-
model of type 2 diabetes, the SGLT2 inhibitor dapagliflozin was shown ing the mitochondrial electron transport chain, NOX, xanthin oxidase
to have anti-EndMT activity in the setting of cardiac fibrosis48. EndMT and eNOS. ROS can be beneficial, for example in neutrophils for the
has also been observed in aortic sections from a mouse model of type 2 respiratory burst to kill bacteria, but are deleterious if produced in
diabetes (db/db mice) as well as from patients with diabetes with the excessive amounts.
use of immunofluorescent staining for CD31 and αSMA, and real-time In endothelial cells, exposure to high glucose levels leads to
PCR for genes encoding transcription factors such as SLUG, TWIST and increased mitochondrial ROS formation56,57. The high glucose flux
SNAIL47. Furthermore, the transcription factor Krüppel-like factor 7 through the mitochondria in endothelial cells leads to increased hex-
(KLF7) was identified in this study as a target of the microRNA miR-132-3p, osamine and polyol flux, increased formation of intracellular advanced
and was significantly elevated at the mRNA and protein levels in aortic glycation end product (AGE) precursors (such as methylglyoxal) and
sections from db/db mice and from patients with diabetes47. In in vitro protein kinase C activation58. Overexpression of mitochondrial brown
experiments in human umbilical vein endothelial cells (HUVECs) grown in fat uncoupling protein 1 (which disrupts mitochondria) or superox-
conditions of high glucose concentration, overexpression of miR-132-3p ide dismutase 2, mitochondrial (an antioxidant enzyme) prevented
and downregulation of KLF7 expression ameliorated EndMT47. Multiple mitochondria-derived superoxide formation in endothelial cells59. This
transcription factors, including SMAD2, SMAD3, RHO-associated finding explains the observed association between hyperglycaemia
kinase 1, SNAIL and serum response factor, have been shown to regu- and microvascular disease60. By contrast, although the association
late the expression of mesenchymal cell markers, thereby contributing between hyperglycaemia and macrovascular disease is evident from
to hyperglycaemia-induced EndMT49,50. In human endothelial cells epidemiological studies61, the underlying mechanisms are less clear
(HUVECs and HAECs), conditions of high glucose concentration induced than for microvascular disease. Nevertheless, the constellation of

• Turbulent blood flow

• Hyperglycaemia
• Oxidative stress
• oxLDL

EndMT

Loss of endothelial cell phenotype

Gain of mesenchymal cell phenotype

Endothelial phenotype Mesenchymal cell

Endothelial cell
markers:
markers: Reversible partial EndMT Mesenchymal phenotype
• αSMA
• PECAM1
• Transgelin
• VE-cadherin
• Calponin
• von Willebrand factor
• FAP

Endothelial cell-specific
HDAC9 genetic deletion

Anti-TGFB Endothelial cell-specific

siRNA TGFβ inhibition (7C1)
7C1 lipid
nanoparticle Atherosclerosis

Fig. 3 | Endothelial-to-mesenchymal transition. Endothelial-to-mesenchymal EndMT can remain in an intermediate, reversible phase. EndMT modulation
transition (EndMT) is a dynamic process involving the transdifferentiation of at this stage could be a promising therapeutic approach for the treatment and
endothelial cells into mesenchymal-like cells. The aberrant induction of EndMT prevention of atherosclerosis. αSMA, α-smooth muscle actin; FAP, fibroblast
by environmental factors, including turbulent blood flow, hyperglycaemia, activation protein; PECAM1, platelet endothelial cell adhesion molecule; siRNA,
oxidative stress and oxidized LDL (oxLDL), can contribute to atherosclerosis small interfering RNA; TGFβ, transforming growth factor-β; VE-cadherin,
development, including in the context of diabetes mellitus. EndMT is a vascular endothelial cadherin.
continuous process in which an endothelial cell that is transitioning through

Nature Reviews Cardiology

Review article

risk factors associated with metabolic syndrome, including insulin stroke, in patients with atherosclerotic CVD and diabetes compared
resistance and increased fatty acid oxidation, can also contribute to with patients with atherosclerotic CVD but without diabetes74. A mouse
increased production of ROS by mitochondria and other sources in model of unstable atherosclerotic plaques has been developed using
the development of diabetes-associated atherosclerosis. the tandem stenosis plaque model, in which two ligations are placed
The diabetic milieu, which is characterized by high blood glucose on the left carotid artery and atherosclerotic lesions with characteristics
levels, toxic glucose-derived metabolites, increased systemic and of unstable plaques form in the area of the tandem stenosis, closely resem-
local concentrations of AGEs, and activation of the AGE receptor, the bling the disease in human73. Under diabetic conditions (STZ-induced
renin–angiotensin–aldosterone system and growth factors, is closely diabetes), these mice showed increased plaque ­instability, charac-
linked to excessive ROS production by the NOX enzymes53–55. In par- terized by larger necrotic cores, increased macrophage infiltration,
ticular, NOX1 has a key role in diabetes-associated atherosclerosis. In intraplaque haemorrhage, reduced fibrous cap-to-plaque size ratio,
Apoe−/− mice with STZ-induced diabetes, genetic deletion of Nox1 or increased collagen formation and reduced NOX4 expression73. In these
pharmacological inhibition of NOX1 with GKT137831 (also known as mice, these changes were attenuated with treatment with the SGLT2
setanaxib, a dual NOX1–NOX4 inhibitor) reduced atherosclerosis62, inhibitor dapagliflozin 5 weeks after diabetes induction and 3 days
as shown by a reduction in plaque area and improvement in plaque mor- after tandem stenosis as an early intervention study73. These s­ tudies
phology, compared with mice expressing Nox1 or with untreated mice, provided the first proof-of-concept that SGLT2 inhibitors might have
respectively19. However, global genetic deletion of another NOX isoform, direct vasculoprotective and anti-atherosclerotic effects that are in part
Nox4, in mice was associated with increased plaque formation and the independent of the glucose-lowering effects of the drug. The vascu-
development of more complex plaques characterized by increased loprotective and anti-atherosclerotic effects of SGLT2 inhibitors were
newly formed collagen and altered extracellular matrix deposition21. confirmed in clinical studies75.
NOX4 has since been found to be vasculoprotective63, an effect that Additional mechanisms involved in the formation of vulnerable
could be attributable to its role in generating H2O2, which has vasodila- atherosclerotic plaques in diabetes have been reported. Antisense
tory effects and can be either vasculoprotective or pro-atherogenic oligonucleotides against Apoc3 reduced the necrotic core area in a
depending on its localization and concentration63. In subsequent stud- mouse model of type 1 diabetes76. Inflammasome activation also seems
ies, we and others have shown that NOX1 and NOX4 have distinct roles to have an important role. In atherosclerosis-prone mice, a deficiency
in regulating immune cell composition and VSMC morphology and in NLRP3, absent in melanoma 2 (AIM2) or gasdermin D in haematopoi-
phenotype in atherosclerotic plaques in diabetic mice64,65. etic cells was associated with reduced atherosclerotic lesion size, but
Other investigators have shown that the antioxidant defence did not prevent necrotic core formation, suggesting that the effect of
capacity is reduced in diabetes-associated atherosclerosis and that NLRP3, AIM2 or gasdermin D deficiency in reducing plaque size is inde-
restoring the antioxidant defence capacity contributed to reduced pendent of macrophage pyroptosis77. Pharmacological targeting of
plaque formation in mice66. However, targeting some of the antioxidant NLRP3 with the specific NLRP3 inhibitor MCC950 has also been shown
pathways (for example, nuclear factor erythroid 2-related factor 2) to reduce atherosclerotic plaque area and necrotic core formation in
pharmacologically in vivo has been difficult, with either poor efficacy Apoe−/− mice with STZ-induced diabetes78.
or substantial off-target effects, at least at the higher doses67. Therefore,
the more promising approach seems to be the direct inhibition of the Epigenetic mechanisms
most important source of ROS, the NOX enzymes. Gene transcription is regulated by multiple pathways, including epige-
In the past 15 years, the research focus has shifted to the NOX5 iso- netic mechanisms. In response to environmental stimuli, changes in the
form, which is expressed in humans but is absent in mice and rats. In a epigenetic state at regulatory regions of a gene can modify its expres-
mouse model of diabetic kidney disease, mice expressing human NOX5 sion. Aberrant gene expression can lead to tissue injury and contrib-
in either endothelial cells (using the promoter for vascular endothelial ute to disease initiation and progression, such as in atherosclerosis79.
cadherin) or VSMCs (using the transgelin promoter) had accelerated The epigenetic state and DNA–histone interactions have a central role
renal injury compared with control mice68,69. However, the effect of in establishing and maintaining chromatin structure. This packaging
transgenic expression of NOX5 in mouse models of atherosclerosis of genetic material into chromatin is essential for nuclear functions,
was not as clear70,71. In these studies, which used a different promoter such as DNA replication and repair and the regulation of gene expres-
(the promoter for TIE2) to overexpress NOX5 in endothelial cells, NOX5 sion by promoting or restricting gene accessibility to the transcriptional
expression was associated with increased aneurysm formation in mice machinery80. The expression of genes located in condensed chromatin
with diabetes70 and the development of hypertension in mice without regions is suppressed, whereas an open chromatin facilitates gene tran-
diabetes71, but atherosclerotic plaque area did not increase compared scription. The chromatin structure is maintained by enzymes involved in
with that in control mice70. In human coronary artery plaques, NOX5 is epigenetic processes, namely DNA methylation, histone modifications
highly expressed in endothelial cells in the early stages of atheroscle- and non-coding RNA-regulated pathways. These enzymes and epige-
rosis and in both endothelial cells and VSMCs in the more advanced netic mechanisms are emerging as important players in CVD, including
stages of coronary artery disease72. in diabetes-associated CVD81.

Unstable atherosclerotic plaques DNA methylation. DNA methylation is an enzymatic process that
Diabetes is not only associated with increased atherosclerotic plaque involves the addition of a methyl group to a cytosine at CpG sites82.
area but also with the formation of unstable atherosclerotic plaques73, At regulatory elements, DNA methylation can mediate gene silencing
which are prone to rupture. Unstable plaques are characterized by through direct inhibition of the binding of methylation-dependent
large necrotic cores, intraplaque haemorrhage and thin fibrous caps73. transcriptional activators83, or indirectly by altering the affinity of
Clinically, this increase in unstable plaques in diabetes is manifested proteins involved in chromatin remodelling84–86. DNA methylation has
by a higher rate of clinical events, such as myocardial infarction or been implicated in inflammation, endothelial dysfunction, foam cell

Nature Reviews Cardiology

Review article

formation and VSMC proliferation, thus showing the association phosphorylation, sumoylation and ubiquitylation108. These histone
between DNA methylation and atherosclerosis development87,88. How- modifications can regulate gene expression by changing the chromatin
ever, little information is available for the role of DNA methylation in structure into an open or closed conformation80. Histone modification
diabetes-induced atherosclerosis87,89,90. A DNA methylation profil- is an enzyme-mediated process and, therefore, can be modulated by
ing study showed that hypomethylation of multiple genes, including controlling enzyme activity80. Histone acetylation and methylation
VEGFB, PLGF, PLCB1 and FATP4, is associated with CVD in individuals are the most studied post-translational modifications of histones.
with diabetes91. Blood samples from individuals with or without diabe- Specific inhibitors of acetylases, deacetylases, methyltransferases
tes, as well as from control individuals with no history of diabetes or and demethylases have been developed, and many of these inhibi-
CVD, were screened for differentially methylated genes using a methyl- tors have been tested in preclinical studies of atherosclerotic CVD109,
ated DNA immunoprecipitation chip (MeDIP-chip) microarray (n = 3 with positive results suggesting potential therapeutic opportunities.
per group). The findings were validated using methylation-specific However, none of the strategies have yet progressed to clinical develop-
PCR and gene expression analysis with quantitative PCR in a larger ment. Studies have also assessed the effects of both isoform-specific
number of samples (40 in the control group, 57 in the diabetes group and cell-specific HDAC inhibition in atherosclerosis. For example, inhi-
and 48 in the CVD group)91. bition of HDAC3 or HDAC9 reduced atherosclerosis in mouse models
Several DNMT inhibitors are under clinical development for the of atherosclerosis45,110. HDACs have also been targeted specifically in
treatment of cancer. Two DNMT inhibitors have been approved by macrophages and monocytes using ESM-HDAC528, a pan-HDAC inhibi-
the FDA for the treatment of myelodysplasia, including 5-azacytidine tor that contains an esterase-sensitive chemical motif (ESM) that allows
and 5-aza-2-deoxycytidine92. Although few preclinical studies have specific targeting of cells that express carboxylesterase 1, such as mono-
shown promising atheroprotective effects with DNMT inhibitors93,94, nuclear myeloid cells111. Carboxylesterase 1 is preferentially expressed
these inhibitors have not been tested in clinical trials in atherosclerotic in monocytes and macrophages, and drugs with an ESM are mostly
CVD. A plausible reason for the lack of clinical studies using DNMT inhib- hydrolysed in these cells, thereby causing the accumulation of the drug
itors could be that DNA methylation is a ubiquitous gene-regulatory specifically in monocytes and macrophages112. However, administration
mechanism across cell types and tissues and, therefore, targeting this of ESM-HDAC528 in mice showed limited effects on atherosclerosis111.
process is often associated with systemic off-target effects. A targeted Although atherosclerotic plaque size was not reduced, the lesions
therapy using nanocarriers to deliver DNMT inhibitors to specific cell showed a substantial shift towards a less severe phenotype, suggesting
types at the injury site could be an alternative approach to overcome relevant atheroprotective benefits of this drug111.
this limitation. In individuals with atherosclerotic CVD and type 2 diabetes, altered
The epigenetic regulation of gene expression is also mediated by epigenome-wide histone acetylation has been noted in peripheral
enzymes that demethylate DNA. Methylcytosine dioxygenase TET2 blood mononuclear cells compared with patients with atherosclerotic
(TET2) mediates DNA demethylation by catalysing the oxidation of CVD without diabetes113. Acetylation of histone 3 at lysine 9 (H3K9ac)
5-methylcytosine to 5-hydroxymethyl cytosine (5hmC). TET2 has been was increased at 118 locus and reduced at 62 locus which implicates
implicated in the pathobiology of atherosclerosis, including in inflam- histone acetylation in atherosclerotic CVD in diabetes113. However,
matory pathways and endothelial dysfunction95–97. TET2 has also been only a small number of studies tested inhibitors of histone modifiers
implicated in diabetic microvascular complications including diabetic in diabetes-associated atherosclerosis. The class I HDAC inhibitor
retinopathy98. The underlying mechanisms are mostly linked to the valproate has been shown to attenuate atherosclerosis in diabetic
demethylation or methylation status of TET2 target genes, including Apoe−/− mice114. In another study, macrophages in atherosclerotic
KFL4 and autophagy-related genes97,99,100. In addition, TET2 is also one plaques in aortic arch specimens from patients with diabetes showed
of the most well-studied genes related to clonal haematopoiesis of elevated levels of HDAC3 compared with patients with atherosclerosis
indeterminate potential (CHIP) in atherosclerosis101. CHIP is defined without diabetes115. HDAC3 levels in plaque macrophages correlated
as a process of clonal expansion of haematopoietic cells as a result of with plasma LDL, triglyceride and HDL levels115. However, only limited
a single acquired somatic mutation in these cells. CHIP has emerged information on clinical parameters was provided in this study. Hdac3
as a major risk factor for atherosclerotic CVD independent of standard deletion in macrophages using the lysozyme 2 promoter-controlled
risk factors102. Interestingly, a significant reduction in 5hmc levels in Cre–flox system was shown to significantly reduce atherosclero-
circulating leukocytes has been observed in individuals with diabetes sis in Apoe−/− mice fed a high-fat diet115. However, the mouse model
compared with healthy individuals, which was linked to the loss of TET2 of high-fat diet is not a classic mouse model of diabetes, and the
activity in individuals with diabetes103. In addition, preclinical stud- lysozyme-controlled Cre system is not macrophage-specific but rather
ies have shown that diabetes can induce myelopoiesis and increased myeloid cell-specific. Therefore, further genetic and pharmacological
production of atherogenic monocytes, which results in accelerated studies are required to investigate the cell-specific role of HDAC3 in
atherosclerosis development104. Identifying potential therapeutic diabetes-associated atherosclerosis.
approaches for targeting CHIP-related genes is a major challenge.
Some approaches targeting TET2 activity have shown promising results Trained immunity. Trained immunity is a term coined for a phenom-
in preclinical studies105,106. Vitamin C has been demonstrated to pro- enon that refers to the capacity of the innate immune system to remem-
mote TET2 function resulting in restoration of haematopoiesis in ber previous invading agents or sterile triggers through epigenetic
Tet2−/− mice106. In addition, targeting sirtuin 1 has also been shown to mechanisms and respond with increased strength to a secondary similar
influence TET2 activity in haematopoietic stem cells in mice107. or unrelated stimulus116. Trained immunity is non-specific and sustains
the activated innate immune phenotype for an extended period. Various
Histone modifications. Histones are the protein component of the stimuli can induce this specialized type of immunity, including oxLDL
chromatin. The amino-terminal tails of histones are subjected to cova- and hyperglycaemia117–120. Trained immunity has been hypothesized
lent post-translational modifications, such as acetylation, methylation, to be a potential mechanism underlying the chronic inflammation

Nature Reviews Cardiology

Review article

Atherosclerotic Atherosclerotic atherosclerotic plaques and peripheral blood leukocytes, suggesting

plaque progression plaque regression a potential role of RUNX1 in humans118. Furthermore, pharmacological
and instability and stability
inhibition of RUNX1 blunted the trained immunity response in vitro
in bone marrow-derived macrophages, supporting the therapeutic
potential of targeting this transcription factor in diabetes-induced
trained immunity118. Further discovery of new targets for pharma-
cological modulation of the trained innate immunity response in
­diabetes-associated atherosclerosis is a ­promising therapeutic avenue.
Current risk factor-based therapies
• Lipid-lowering drugs Current therapies and strategies in
• Glucose-lowering drugs
• Antihypertensive drugs clinical development
Novel mechanism-based therapeutic targets In addition to traditional CVD risk factors, such as dyslipidaemia and
• Blood flow-mediated shear stress hypertension, hyperglycaemia accompanied by oxidative stress
• EndMT
• Oxidative stress
and inflammation has an important role in accelerated atheroscle-
• Inflammation rosis development in the context of diabetes. Current clinical prac-
• Epigenetic mechanisms tices in the general population focus on reducing atherosclerotic
• Trained immunity
• Transcription factors CVD-promoting risk factors including hypercholesterolaemia,
• Cell-specific targets hyperglycaemia and hypertension (Fig. 4). Current therapies for the
Fig. 4 | New mechanism-based therapeutic approaches in diabetes-associated prevention and treatment of CVD in patients with diabetes include
atherosclerosis. Current available therapies for atherosclerotic cardiovascular anticoagulant and antiplatelet therapies, lipid-lowering treatments,
disease solely focus on traditional risk factors, including hyperglycaemia, anti-glycaemic therapies (such as SGLT2 inhibitors and glucagon-like
hyperlipidaemia and hypertension, but do not adequately target the specific peptide 1 (GLP1) receptor agonists), anti-hypertensive drugs (such as
mechanisms of atherosclerosis that are exacerbated in diabetes mellitus. Novel β-blockers, angiotensin-converting enzyme inhibitors and angiotensin
promising mechanism-based approaches are emerging to prevent and treat II ­receptor blockers) and anti-inflammatory therapies.
atherosclerosis, including in diabetes, such as targeting shear stress, endothelial-to-
mesenchymal transition (EndMT), oxidative stress and inflammation. Antithrombotic therapies
Antithrombotic therapies prevent blood clot formation and are used
to prevent complications of atherosclerotic CVD, including myocar-
observed in the accelerated atherosclerosis that occurs in diabetes121. dial infarction and stroke. Antiplatelet agents include the most widely
Both metabolic and epigenetic reprogramming are important to estab- used aspirin, and drugs such as clopidogrel and prasugrel 129,130. The
lish and maintain the long-term memory of innate immune cells122. COMPASS trial131 in patients with previous coronary artery disease
Epigenetic studies have identified specific histone modifications, includ- or peripheral artery disease showed that the use of both anticoagu-
ing histone 3 lysine 4 monomethylation (H3K4me1), H3K4 trimethylation lants and antiplatelet drugs provides superior protection against
(H3K4me3), H3K9 dimethylation (H3K9me2) and H3K27 acetylation adverse cardiovascular events than each drug type alone. The ASCEND
(H3K27ac), linked to trained immunity, indicating that epigenetic trial132 in individuals with diabetes without known atherosclerotic
mechanisms are at the core of this non-specific immunity123–126. CVD showed that the benefit of aspirin in preventing serious vascular
Additional studies have investigated the nature of epigenetic events was counterbalanced by an increased risk of major bleeding.
changes linked to hyperglycaemia-induced inflammation in peripheral In addition, a meta-analysis of 12 randomized controlled trials of
monocytes from patients with type 1 or type 2 diabetes. The changes aspirin compared with placebo or no treatment in individuals with
included decreased H3K4me2 levels on the IL8 gene and increased diabetes without known atherosclerotic CVD indicated potential
H3K9me2 levels on the IL1A gene, which are associated with chronic benefits of aspirin for the primary prevention of CVD in patients with
inflammation127. Furthermore, increased levels of histone acetylation at diabetes. These results emphasize the need for individualized aspirin
H3K9ac, H3K14ac, H4K5ac and H4K12ac have also been observed in the therapy strategies that are based on the individual’s baseline ath-
promoters of genes encoding inflammatory proteins, including tumour erosclerotic CVD status and bleeding risk133. Therefore, the current
necrosis factor and cyclooxygenase 2, in monocytes from patients with European Society of Cardiology (ESC) guidelines recommend using
type 1 or type 2 diabetes128. High glucose concentrations have been proton pump inhibitors to prevent ­gastrointestinal bleeding when
shown to induce pro-inflammatory and pro-atherogenic gene expres- antithrombotic drugs are used134.
sion in macrophages in a glycolysis-dependent manner in vitro118. Bone
marrow-derived macrophages from diabetic mice maintained a pro- Lipid-lowering therapies
inflammatory gene expression profile after being cultured in normal The cholesterol-lowering drugs statins, ezetimibe and PCSK9 inhibi-
glucose conditions118. This finding suggests that hyperglycaemia can tors are currently used to reduce plasma LDL cholesterol levels. Addi-
induce trained immunity. In non-diabetic atheroprone Ldlr−/− mice, tive benefits of these drugs have also been demonstrated, with the
transplantation of bone marrow from diabetic mice increased ath- IMPROVE-IT trial135 showing an incremental lowering of plasma LDL
erosclerosis burden compared with transplantation of bone marrow cholesterol levels and improved cardiovascular outcomes with com-
from mice without diabetes, which suggests a role of trained innate bined statin–ezetimibe therapy compared with statin therapy alone.
immunity in diabetes-associated atherosclerosis118. The transcription In addition to their lipid-lowering effects, statins exert pleiotropic
factor RUNX1 was identified as an important transcriptional regulator effects on inflammatory factors and lower plasma C reactive protein
in the context of hyperglycaemia-induced trained immunity in humans (CRP) levels136. Of note, an analysis of three randomized trials on statins
and in mice118. Gene targets of RUNX1 were enriched in both human showed that inflammation and high-sensitivity CRP (hsCRP) levels are

Nature Reviews Cardiology

Review article

better predictors of future cardiovascular events than plasma LDL hormone that is produced and released by L cells in the gut after food
cholesterol levels in individuals taking statin therapy137. ingestion149. GLP1 potentiates glucose-dependent insulin secretion from
In the FOURIER and ODYSSEY OUTCOMES studies therapy with pancreatic islets via the GLP1 receptor149,150. However, extra-pancreatic
the anti-PCSK9 monoclonal antibodies evolocumab and alirocumab, functions of GLP1 are being increasingly reported. Both experimental
respectively, improved CVD outcomes compared with placebo138,139. studies on GLP1 receptor agonists151,152 and clinical studies, such as the
A meta-analysis of randomized, controlled trials on alirocumab and LEADER153 and REWIND154 trials on the GLP1 receptor agonists liraglutide
evolocumab also showed that both drugs are well tolerated140. Suba- and dulaglutide, respectively, have demonstrated renoprotective and
nalyses of the population of patients with diabetes from the ODYSSEY cardioprotective actions of GLP1 receptor agonism. However, whether
OUTCOMES, FOURIER and PROFICIO phase III trials showed signifi- these effects are glucose-independent remains to be determined.
cant reductions in plasma LDL cholesterol levels in individuals with Multiple clinical trials have shown that SGLT2 inhibitors reduce the
diabetes in the treatment group versus the placebo group141. Clinical risk of cardiovascular events in patients with type 2 diabetes14,155,156.
evidence from these and other trials, including ODYSSEY DM-INSULIN The cardioprotective effects attributed to SGLT2 inhibitors extend to
and ODYSSEY DM-DYSLIPIDEMIA, showed that, in patients with hyper- lean individuals with normal blood pressure and without diabetes, sug-
cholesterolaemia and diabetes, the addition of PCSK9 inhibitors to gesting that the mechanism of action of this drug class goes beyond the
maximum-tolerated statin therapy significantly lowers plasma LDL effects on blood glucose levels, weight control and blood pressure157–159.
cholesterol levels and prevents CVD events without impairing glycaemic Furthermore, in a mouse model of heart failure, treatment with SGLT2
control compared with statin therapy alone138,139,142–145. inhibitors improved cardiac function in both wild-type mice and mice
New therapeutic approaches targeting dyslipidaemia involve with systemic SGLT2 deficiency (Sglt2−/− mice), suggesting that SGLT2
RNA-based gene-silencing technologies, such as siRNAs and anti-sense inhibitors have off-target effects, and these benefits were independent
oligonucleotides. siRNAs are double-stranded RNA molecules that of metabolic changes160. Despite the available evidence demonstrat-
mediate target mRNA degradation or inhibition of translation through ing multisystem benefits of SGLT2 inhibitors, detailed knowledge
the endogenous RNA-induced silencing complex. Anti-sense oli- on the mechanisms of action of SGLT2 inhibitors in addition to their
gonucleotides are longer single-stranded RNA molecules that are glucose-lowering effect is lacking. Of note, the clinical evidence with
complementary to target mRNA; anti-sense oligonucleotides bind respect to the reduction in atherosclerosis-related cardiovascular
to target mRNA and inhibit their translation into protein. The nucleic events, including myocardial infarction and stroke, is stronger for
acid therapeutics for dyslipidaemia that are currently in development GLP1 receptor agonists than for SGLT2 inhibitors 161,162. Evidence
target genes encoding proteins with important roles in lipoprotein from experimental models shows that SGLT2 inhibitors ameliorate
production or removal, including PCSK9, apolipoprotein (a), apoli- endothelial dysfunction and vascular injury163, which could translate
poprotein B, angiopoietin-related protein 3, angiopoietin-related into anti-atherosclerotic effects. However, clinical evidence is lacking.
protein 4 and apolipoprotein C-III, and are mostly conjugated to Clinical guidelines recommend the use of SGLT2 inhibitors in patients
N-acetylgalactosamine (GalNAc). GalNAc conjugation facilitates the with heart failure or kidney disease164 and GLP1 receptor agonists in
delivery of the RNA molecules specifically to liver cells, resulting in patients with type 2 diabetes and high risk of atherosclerotic CVD and
reduced toxicity. Inclisiran is a GalNAc-conjugated siRNA to PCSK9 stroke and in patients with kidney disease161,162,164.
that has been approved for the treatment of adults with heterozygous
familial hypercholesterolaemia or with clinical atherosclerotic CVD146. Anti-inflammatory therapies
The biannual dose regimen of inclisiran could make it a more suit- Large-scale, placebo-controlled clinical trials have assessed the efficacy
able therapy for PCSK9 inhibition by improving treatment adherence of anti-inflammatory therapies in patients with atherosclerotic CVD, such
compared with the fortnightly or monthly regimens of anti-PSCK9 as the CANTOS trial165 on canakinumab and the LoDoCo2 (ref. 164) and
antibodies. In the ORION-10 and ORION-11 trials146 in patients with COLCOT166,167 trials on colchicine. These trials confirmed that inflamma-
atherosclerotic CVD and high plasma LDL cholesterol levels, incli- tion has an important role in the pathogenesis of atherosclerosis. In the
siran therapy reduced plasma LDL cholesterol levels by 52.3% and phase III CANTOS trial165, treatment with the human monoclonal anti-IL-1β
49.9%, respectively, compared with placebo. Post hoc analyses of the antibody canakinumab was associated with improved CVD outcomes
ORION-9, ORION-10 and ORION-11 phase III trials showed that inclisiran compared with placebo in patients with previous myocardial infarction.
significantly reduces the levels of PCSK9 and other atherogenic lipids However, canakinumab was associated with an increased risk of fatal
and lipoproteins versus placebo in patients with diabetes mellitus or infection and thus has not been approved for cardiovascular indications.
obesity147. Another available lipid-lowering therapy is volanesorsen, Colchicine has broad anti-inflammatory properties, including inhibition
a GalNAc-conjugated antisense oligonucleotide targeting APOC3 mRNA of the NLRP3 inflammasome, and has been proposed for the preven-
that has been approved by the EMA for the treatment of patients with tion of atherosclerosis. The results of the COLCOT and the LoDoCo2
familial chylomicronaemia who are at increased risk of pancreatitis. clinical trials166,167 confirmed the cardiovascular benefits of colchicine
In patients with type 2 diabetes, volanesorsen significantly reduced fast- in patients after myocardial infarction and in patients with chronic sta-
ing plasma APOC3 concentrations and triglyceride levels and improved ble coronary artery disease, respectively. Colchicine was approved by
insulin sensitivity and glucose control compared with placebo148. the FDA in 2023 for the treatment of atherosclerotic CVD168. Of note,
a meta-analysis of the LoDoCo, LoDoCo2, COLCOT, COPS and CANTOS
Glucose-lowering therapies trials showed that anti-inflammatory therapy in patients with type 2
The approved novel therapeutic agents for GLP1 receptor agonists and diabetes mellitus was associated with reduced risk of major adverse
SGLT2 inhibitors have not only shown benefits for glucose control but cardiovascular events compared with placebo, indicating potential
also for reducing CVD events in patients with type 2 diabetes14. Interest- benefits of anti-inflammatory therapy in this patient population169.
ingly, these benefits were also evident in patients with heart failure with- Anakinra is a human recombinant IL-1 receptor antagonist that was
out both diabetes and kidney disease14. GLP1 is a gut-derived incretin evaluated in patients with ST segment elevation myocardial infarction,

Nature Reviews Cardiology

Review article

with a primary end point of change in hsCRP levels from baseline as a reduced LPS–Toll-like receptor 4–nuclear factor-κB signalling in aortic
marker of systemic inflammation170. The results showed a short-term tissues, thereby mitigating atherosclerosis development in these
decrease in inflammation compared with placebo. However, anak- mice183. Ideally, anti-inflammatory therapies should be directed spe-
inra has not been approved for the treatment of atherosclerotic CVD. cifically to the vascular endothelium to avoid the adverse effects of
Methotrexate is an inhibitor of purine metabolism that is used to sup- systemic inhibition of inflammation, such as observed with systemic
press the activity of the immune system in a range of conditions, includ- TGFβ inhibition, which has been associated with adverse effects on the
ing rheumatoid arthritis171. The CIRT trial172 evaluated the effects of a immune system and the promotion of atherogenic processes in mice184.
low-dose methotrexate on cardiovascular outcomes in patients with
multivessel coronary disease or patients with a history of myocardial Cell-specific therapies. Systemic drug administration has several limita-
disease with type 2 diabetes or metabolic syndrome. However, the tions, such as limited bioavailability, reduced stability, robust clearance
low methotrexate dose had no effect in reducing the levels of IL-1β, and off-target drug distribution. Drug delivery to specific cells at the site
IL-6 or hsCRP in plasma compared with placebo in patients with stable of injury is an alternative approach that could potentially overcome these
atherosclerosis with a history of myocardial infarction and diabetes173. limitations. However, cell-specific therapy still faces multiple challenges
Finally, ziltivekimab is a fully human monoclonal antibody to IL-6. In in atherosclerosis. The design of cell-targeted strategies needs to take
the RESCUE trial174 in patients with chronic kidney disease and inflamma- into account the role of different cell populations during the different
tion (as assessed by hsCPR ≥2 mg/dl), ziltivekimab reduced markers of stages of atherogenesis and should aim to target a unique antigen that is
inflammation and thrombosis related to atherosclerosis compared with only expressed on affected cells at the site of injury in a time-dependent
placebo, with better anti-inflammatory effects than those observed manner. Selecting an appropriate drug carrier is also important. Nano-
with IL-1β inhibition in the CANTOS trial in patients with myocardial particles are the most widely used drug carriers for targeted therapies.
infarction. Ziltivekimab is being further evaluated in the ZEUS trial175, Nanoparticles have multiple advantages, such as helping to improve drug
which includes 6,200 patients with chronic kidney disease stages 3–4 stability, bioavailability and delivery to the injured site in a cell-specific
and elevated hsCRP levels; the trial will assess atherosclerotic CVD manner and reduce off-target drug effects. Packaging of drugs in nano-
end points176. In addition, the effect of ziltivekimab on mortality and particles can also facilitate slow drug release where required, such as in
morbidity in patients with heart failure and inflammation is being the case of diabetes. For example, glucose-lowering drugs such as
evaluated in the HERMES trial177. GLP1 receptor agonists can be encapsulated in nanoparticles to con-
trol drug release and mimic natural GLP1 secretion in response to food
Emerging therapeutic strategies ingestion185,186. Available nanocarriers include polymeric, HDL-like and
Despite the mitigation of conventional CVD risk factors with currently LDL-like n­ anoparticles and liposomes.
available therapies, a substantial residual risk remains, particularly in Nanoparticles have been assessed in clinical trials in individuals
patients with diabetes, with a substantial number of patients having with diabetes but not in the context of atherosclerotic CVD187–189. Given
recurrent CVD events178. Furthermore, some of the current treatments that atherosclerosis development in diabetes involves the same cell
have important adverse effects in some patients179. Therefore, additional types as in individuals without diabetes, the development of nano-
mechanism-based therapies are urgently needed. therapies for patients with atherosclerosis without diabetes might
also be of benefit for those with diabetes. Each vascular cell population
Targeted therapies has different roles at distinct stages of atherosclerotic plaque develop-
Pathology-targeting, site-specific therapies. Atherosclerosis is ment. Although, advances in cell-specific technologies have helped to
associated with multiple local and cell-specific pathological processes identify cellular heterogeneity in vascular and immune cells, a com-
including oxidative stress, abnormal lipid metabolism and oxidation, mon consensus on the classification of subpopulations of specific cell
and an activated immune system. These pathological mechanisms types, such as endothelial cells or macrophages, is yet to be achieved.
have been targeted using site-specific therapeutics. For example, Cell surface proteins that are expressed on injured vascular cells can
the use of nanoparticles to deliver siRNA and antibodies to PCSK9 or be used for directing cell-specific drug delivery systems. Chronically
anti-PCSK9 siRNA conjugated with GalNAc have been shown to target activated endothelial cells express cell surface markers such as inter-
lipid metabolism specifically in the liver leading to significant reduc- cellular adhesion molecule 1, vascular cell adhesion protein 1, platelet
tions in plasma LDL levels146. Another example is 2-hydroxypropyl-β- endothelial cell adhesion molecule, E-selectin and P-selectin, which can
cyclodextrin, which has cholesterol-dissolving properties. Delivery of be targets for nanoparticle-based therapeutics190–192.
this compound via air-trapped polybutylcyanoacrylate nanoparticles Macrophages are key immune cells present in atherosclerotic
markedly improved its anti-atherosclerotic efficacy in the Apoe−/− plaques and are central regulators of the disease193. Macrophage-specific
mouse model of atherosclerosis180. therapeutic approaches have so far targeted several pathways, includ-
Oxidative stress caused by ROS overproduction has been impli- ing inflammation and impaired cholesterol efflux194. Preclinical studies
cated in vascular injury, including atherosclerosis in diabetes181. have implicated multiple macrophage-specific targets in atheroscle-
Although antioxidant therapies have largely had disappointing results rosis, including calcium–calmodulin-dependent protein kinase IIγ,
in clinical trials, targeted ROS-scavenging nanoparticles have been epsins and the CD47–signal regulatory protein-α signalling axis195–197.
shown to attenuate atherosclerosis progression in preclinical studies182. Despite exciting findings from these preclinical studies, the discovery
Systemic oxidative stress and inflammation have been targeted with of novel macrophage subpopulations in the past 5 years might mean
a specific class of nanocarriers made of fullerene, an allotrope of that drug delivery to specific macrophage populations is more com-
carbon183. Fullerene-based therapies were shown to improve intestinal plicated than previously thought. Information on macrophage hetero-
redox homeostasis and restore gut barrier integrity to prevent the trans- geneity provided by technologies such as single-cell RNA sequencing
location of bacterial lipopolysaccharide (LPS) into the bloodstream in can help in identifying and targeting the different macrophage popu-
Apoe−/− mice183. This reduction in plasma LPS levels contributed to lations. Advances in macrophage-directed nanotherapies that can

Nature Reviews Cardiology

Review article

target functions of macrophages within atherosclerotic lesions have those targeting inflammation and oxidative stress as well as epigenetic
shown great promise in preclinical studies by improving therapeutic pathways. Apabetalone is set to be tested in combination with an SGLT2
efficacy and minimizing off-target effects compared with systemic inhibitor in the BETonMACE2 study, which is expected to begin in 2024.
delivery194. Macrophage-specific therapies have also been tested in
clinical trials. Liposomal nanoparticles encapsulating prednisolone NOX inhibitors
were successfully internalized by macrophages in atherosclerotic The NOX1–NOX4 inhibitor setanaxib is the first NOX inhibitor to enter the
plaques in patients scheduled for endarterectomy owing to sympto- clinical trial arena. In preclinical studies, this dual NOX1–NOX4 inhibi-
matic iliofemoral atherosclerosis198. A small number of studies have tor reduced atherosclerosis and had renoprotective effects in Apoe−/−
also been performed to target VSMCs in atherosclerosis. The transient mice with STZ-induced diabetes21. In these mice, Nox1 deletion was
receptor potential cation channel TRPV1 is a thermosensitive channel atheroprotective, whereas Nox4 global deletion was associated with an
that has been shown to protect against foam cell formation by inducing increased atherosclerotic plaque area19,21. The dual NOX1–NOX4 inhibi-
autophagy in oxLDL-treated VSMCs199. Copper sulfide nanoparticles tor provided simultaneous renoprotection and atheroprotection210,
carrying a monoclonal antibody to TRPV1 were used as a photother- most probably mediated by balancing the effects on NOX1 and NOX4.
mal switch to activate TRPV1 signalling in mice. In Apoe−/− mice fed a This inhibitor is being investigated in several clinical settings of fibrosis,
high-fat diet, intravenous administration of the TRPV1-carrying nano- such as idiopathic pulmonary fibrosis211, primary biliary cholangitis212,213
particles and subsequent near-infrared-mediated activation of TRPV1 and squamous cell carcinoma214. The study investigating setanaxib in
triggered an influx of calcium ions into VSMCs, leading to the activa- patients with type 2 diabetes-associated kidney disease215 did not meet
tion of autophagy and a reduction in lipid accumulation in VSMCs the primary renal end point of albuminuria reduction. An ongoing
and a reduction in atherosclerosis compared with untreated mice200. phase II clinical trial216 will assess the effect of setanaxib on persistent
Several single-cell RNA sequencing studies have identified multiple albuminuria in patients with type 1 diabetes. The long-term effects of
VSMC-specific molecular targets, including transcription factor 21, setanaxib in clinical CVD are not known, but this drug has consistently
hyaluronan synthase 3, KLF4, growth differentiation factor 10 and shown anti-inflammatory effects and t­ herefore might potentially have
NOTCH, which might be attractive targets in atherosclerosis, including cardiovascular protective effects.
in the context of diabetes201. Upregulation of NOX5 seems to be deleterious in several disease
settings, including hypertension, diabetes, inflammation and fibrosis217.
Epigenetic drugs Therefore, NOX5 could represent an ideal target to improve CVD and
Inhibitors of enzymes associated with specific epigenetic mechanisms rep- renal disease outcomes in patients with diabetes. NOX5 is expressed
resent a new frontier in pharmacological research202. Advances in epige- in all cells of the vascular wall and at several stages of atherosclerosis
netic research have led to an increased understanding of chromatin biology development72. Furthermore, NOX5 activity seems to be regulated by
and have resulted in the development of novel chromatin-modifying epigenetic pathways. In human macrophages in vitro, histone acetyl-
drugs. These compounds mostly regulate the activity of enzymes that transferases, in particular p300 and HAT1, induced NOX5 overexpres-
write or read epigenetic marks, including DNA methylation or histone sion in inflammatory conditions through the recruitment to the site of
modifications, thereby affecting gene expression patterns linked to transcription within the NOX5 gene promoter218. Inhibition of HDACs
­diseases, including atherosclerosis203. Multiple epigenetic drugs have reduced the transcription of NOX enzymes and ROS production and
been approved for the treatment of malignancies. Apabetalone (RVX-208) ameliorated pulmonary arterial hypertension in a rat model of pul-
is the first inhibitor of bromodomain and extra-terminal (BET) proteins to monary arterial hypertension219. Another study in an experimental
receive breakthrough therapy designation from the FDA for the secondary model of diabetes had similar results220. High glucose levels induced
prevention of major adverse cardiac events in patients with type 2 diabetes H3K27ac enrichment at the promoters of the NOX1, NOX4 and NOX5
in combination with standard care204. BETs (BRD2, BRD3, BRD4 and the in human VSMCs in vitro220. This finding suggests that HDACs mediate
testis-restricted BRDT) are epigenetic readers that bind to specific acety- vascular NOX5 upregulation in the context of diabetes. In addition,
lated lysine residues on histone tails and facilitate the assembly of the given that HDAC inhibition reduced vascular ROS production via NOX5
transcriptional machinery205. Evidence in preclinical models indicates that inhibition, HDACs could represent a new target for the treatment of
BET inhibition can attenuate atherosclerosis and intimal hyperplasia205. diabetes-associated atherosclerosis.
Apabetalone has been shown to stimulate reverse cholesterol transport by
inducing apolipoprotein A-I expression and increasing plasma HDL levels Conclusions
in vivo in African green monkeys206,207. In addition, apabetalone has been The prevalence of diabetes is increasing worldwide at an alarming rate1,
shown to prevent hyperglycaemia-induced upregulation of IL-1β, IL-6 and and CVD remains the major cause of the increased morbidity and mortality
tumour necrosis factor in human endothelial cells in vitro and in aortic in patients with diabetes2. Currently available treatments target hyper-
atherosclerotic plaques in Apoe−/− mice208. Apabetalone has also been glycaemia, lipid profile and blood pressure, as well as the pro-thrombotic
shown to decrease systemic inflammation in patients with CVD compared state that is present in diabetes. Certain novel anti-glycaemic drugs,
with placebo208. In a pooled analysis of non-randomized studies, apabetal- such as SGLT2 inhibitors and GLP1 receptor agonists, have also shown
one treatment was associated with fewer CVD events than placebo in positive effects on cardiovascular outcomes even independently of
patients with coronary artery disease209. The phase III BETonMACE trial204 glucose-lowering effects221–224. Despite these advances, a large propor-
on the effect of apabetalone on cardiovascular outcomes in patients with tion of patients with diabetes still develop CVD. Several pathobiological
type 2 diabetes after a recent acute coronary syndrome showed a benefit mechanisms of diabetes-associated atherosclerosis have been identi-
for secondary end points, including hospitalization for heart failure. fied, including transcriptomic and phenotypic changes in vascular cells
The new guidelines from the ESC134 highlight the unmet need to reduce the that relate to turbulent blood flow, inflammation and oxidative stress.
CVD risk in patients with type 2 diabetes and emphasize the potential of Epigenetic changes have been implicated in vascular cell activation
combining multiple drugs with proven cardiovascular benefits, including and damage and in EndMT. These changes are reversible in principle and

Nature Reviews Cardiology

Review article

represent druggable targets related to epigenetic pathways, thereby 28. Kovacic, J. C. et al. Endothelial to mesenchymal transition in cardiovascular disease:
JACC state-of-the-art review. J. Am. Coll. Cardiol. 73, 190–209 (2019).
providing new avenues for additional treatment approaches in these 29. Krenning, G., Moonen, J.-R. A. J., van Luyn, M. J. A. & Harmsen, M. C. Vascular smooth
patients. The current trend is to develop cell-specific drug delivery muscle cells for use in vascular tissue engineering obtained by endothelial-to-
strategies by, for example, using nanoparticles to specifically target mesenchymal transdifferentiation (EnMT) on collagen matrices. Biomaterials 29,
3703–3711 (2008).
the injured or activated cells in the vascular wall that are undergoing 30. Chen, P. Y. et al. Endothelial-to-mesenchymal transition drives atherosclerosis
phenotypic changes and thereby avoid systemic adverse effects. progression. J. Clin. Invest. 125, 4514–4528 (2015).
31. Evrard, S. M. et al. Endothelial to mesenchymal transition is common in atherosclerotic
lesions and is associated with plaque instability. Nat. Commun. 7, 11853 (2016).
Published online: xx xx xxxx 32. Cooley, B. C. et al. TGF-β signaling mediates endothelial-to-mesenchymal transition
(EndMT) during vein graft remodeling. Sci. Transl. Med. 6, 227ra234 (2014).
References 33. Depuydt, M. A. et al. Microanatomy of the human atherosclerotic plaque by single-cell
1. Saeedi, P. et al. Global and regional diabetes prevalence estimates for 2019 and transcriptomics. Circ. Res. 127, 1437–1455 (2020).
projections for 2030 and 2045: results from the International Diabetes Federation 34. Alsaigh, T., Evans, D., Frankel, D. & Torkamani, A. Decoding the transcriptome of calcified
Diabetes Atlas, 9(th) edition. Diabetes Res. Clin. Pract. 157, 107843 (2019). atherosclerotic plaque at single-cell resolution. Commun. Biol. 5, 1084 (2022).
2. Wong, N. D. & Sattar, N. Cardiovascular risk in diabetes mellitus: epidemiology, 35. Lukowski, S. W. et al. Single-cell transcriptional profiling of aortic endothelium identifies
assessment and prevention. Nat. Rev. Cardiol. 20, 685–695 (2023). a hierarchy from endovascular progenitors to differentiated cells. Cell Rep. 27,
3. Low Wang, C. C., Hess, C. N., Hiatt, W. R. & Goldfine, A. B. Clinical update: cardiovascular 2748–2758.e3 (2019).
disease in diabetes mellitus: atherosclerotic cardiovascular disease and heart failure 36. Zhao, G. et al. Single-cell transcriptomics reveals endothelial plasticity during diabetic
in type 2 diabetes mellitus – mechanisms, management, and clinical considerations. atherogenesis. Front. Cell Dev. Biol. 9, 689469 (2021).
Circulation 133, 2459–2502 (2016). 37. Suzuki, T. et al. Isolation and characterization of endothelial-to-mesenchymal transition
4. Wang, X., Shen, Y., Shang, M., Liu, X. & Munn, L. L. Endothelial mechanobiology in cells in pulmonary arterial hypertension. Am. J. Physiol. Lung Cell Mol. Physiol. 314,
atherosclerosis. Cardiovasc. Res. 119, 1656–1675 (2023). L118–L126 (2018).
5. Davies, P. F., Civelek, M., Fang, Y. & Fleming, I. The atherosusceptible endothelium: 38. Manavski, Y. et al. Clonal expansion of endothelial cells contributes to ischemia-induced
endothelial phenotypes in complex haemodynamic shear stress regions in vivo. neovascularization. Circ. Res. 122, 670–677 (2018).
Cardiovasc. Res. 99, 315–327 (2013). 39. Welch-Reardon, K. M., Wu, N. & Hughes, C. C. A role for partial endothelial-mesenchymal
6. Chiu, J. J. & Chien, S. Effects of disturbed flow on vascular endothelium: transitions in angiogenesis? Arterioscler. Thromb. Vasc. Biol. 35, 303–308 (2015).
pathophysiological basis and clinical perspectives. Physiol. Rev. 91, 327–387 (2011). 40. Hall, I. F., Kishta, F., Xu, Y., Baker, A. H. & Kovacic, J. C. Endothelial to mesenchymal
7. Li, H., Zhou, W. Y., Xia, Y. Y. & Zhang, J. X. Endothelial mechanosensors for atheroprone transition: at the axis of cardiovascular health and disease. Cardiovasc. Res. 120,
and atheroprotective shear stress signals. J. Inflamm. Res. 15, 1771–1783 (2022). 223–236 (2024).
8. Lee, D. Y. & Chiu, J. J. Atherosclerosis and flow: roles of epigenetic modulation in vascular 41. Chen, P. Y. et al. Endothelial TGF-beta signalling drives vascular inflammation and
endothelium. J. Biomed. Sci. 26, 56 (2019). atherosclerosis. Nat. Metab. 1, 912–926 (2019).
9. Tamargo, I. A., Baek, K. I., Kim, Y., Park, C. & Jo, H. Flow-induced reprogramming 42. Souilhol, C., Harmsen, M. C., Evans, P. C. & Krenning, G. Endothelial-mesenchymal
of endothelial cells in atherosclerosis. Nat. Rev. Cardiol. 20, 738–753 (2023). transition in atherosclerosis. Cardiovasc. Res. 114, 565–577 (2018).
10. Zhou, M. et al. Wall shear stress and its role in atherosclerosis. Front. Cardiovasc. Med. 43. Chen, P. Y. et al. Smooth muscle cell reprogramming in aortic aneurysms. Cell Stem Cell
10, 1083547 (2023). 26, 542–557.e511 (2020).
11. Poznyak, A. et al. The diabetes mellitus–atherosclerosis connection: the role of lipid 44. Glaser, S. F. et al. The histone demethylase JMJD2B regulates endothelial-to-
and glucose metabolism and chronic inflammation. Int. J. Mol. Sci. 21, 1835 (2020). mesenchymal transition. Proc. Natl Acad. Sci. USA 117, 4180–4187 (2020).
12. Li, X. et al. Oxidative stress, endothelial dysfunction, and N-acetylcysteine in type 2 45. Lecce, L. et al. Histone deacetylase 9 promotes endothelial-mesenchymal transition and
diabetes mellitus. Antioxid. Redox Signal. 40, 968–989 (2024). an unfavorable atherosclerotic plaque phenotype. J. Clin. Invest. 131, e131178 (2021).
13. Wang, E., Wang, H. & Chakrabarti, S. Endothelial-to-mesenchymal transition: 46. Yang, D. R., Wang, M. Y., Zhang, C. L. & Wang, Y. Endothelial dysfunction in vascular
an underappreciated mediator of diabetic complications. Front. Endocrinol. 14, complications of diabetes: a comprehensive review of mechanisms and implications.
1050540 (2023). Front. Endocrinol. 15, 1359255 (2024).
14. Thomas, M. C., Coughlan, M. T. & Cooper, M. E. The postprandial actions of GLP-1 47. Hulshoff, M. S. et al. miR-132-3p and KLF7 as novel regulators of aortic stiffening-associated
receptor agonists: the missing link for cardiovascular and kidney protection in type 2 EndMT in type 2 diabetes mellitus. Diabetol. Metab. Syndr. 15, 11 (2023).
diabetes. Cell Metab. 35, 253–273 (2023). 48. Tian, J. et al. Dapagliflozin alleviates cardiac fibrosis through suppressing EndMT and
15. Irace, C. et al. Empagliflozin influences blood viscosity and wall shear stress in subjects fibroblast activation via AMPKα/TGF-β/Smad signalling in type 2 diabetic rats. J. Cell
with type 2 diabetes mellitus compared with incretin-based therapy. Cardiovasc. Diabetol. Mol. Med. 25, 7642–7659 (2021).
17, 52 (2018). 49. Yan, F. et al. Glucagon-like peptide 1 protects against hyperglycemic-induced
16. Li, S. et al. Recent advances of mechanosensitive genes in vascular endothelial cells endothelial-to-mesenchymal transition and improves myocardial dysfunction by
for the formation and treatment of atherosclerosis. Genes. Dis. 11, 101046 (2024). suppressing poly(ADP-ribose) polymerase 1 activity. Mol. Med. 21, 15–25 (2015).
17. Luo, J. Y. et al. Induction of KLF2 by exercise activates eNOS to improve vasodilatation 50. Yu, C.-H. et al. High glucose induced endothelial to mesenchymal transition in human
in diabetic mice. Diabetes 72, 1330–1342 (2023). umbilical vein endothelial cell. Exp. Mol. Pathol. 102, 377–383 (2017).
18. Haemmig, S. et al. Novel lesional transcriptional signature separates atherosclerosis with 51. Tang, R.-N. et al. Effects of angiotensin II receptor blocker on myocardial
and without diabetes in Yorkshire swine and humans. Arterioscler. Thromb. Vasc. Biol. 41, endothelial-to-mesenchymal transition in diabetic rats. Int. J. Cardiol. 162,
1487–1503 (2021). 92–99 (2013).
19. Gray, S. P. et al. NADPH oxidase 1 plays a key role in diabetes mellitus-accelerated 52. Liu, X. et al. Irisin inhibits high glucose‐induced endothelial‐to‐mesenchymal transition
atherosclerosis. Circulation 127, 1888–1902 (2013). and exerts a dose‐dependent bidirectional effect on diabetic cardiomyopathy. J. Cell.
20. Jude, E. B., Douglas, J. T., Anderson, S. G., Young, M. J. & Boulton, A. J. Circulating cellular Mol. Med. 22, 808–822 (2018).
adhesion molecules ICAM-1, VCAM-1, P- and E-selectin in the prediction of cardiovascular 53. Jha, J. C., Ho, F., Dan, C. & Jandeleit-Dahm, K. A causal link between oxidative stress
disease in diabetes mellitus. Eur. J. Intern. Med. 13, 185–189 (2002). and inflammation in cardiovascular and renal complications of diabetes. Clin. Sci. 132,
21. Gray, S. P. et al. Reactive oxygen species can provide atheroprotection via NOX4-dependent 1811–1836 (2018).
inhibition of inflammation and vascular remodeling. Arterioscler. Thromb. Vasc. Biol. 36, 54. Koulis, C., Watson, A. M. D., Gray, S. P. & Jandeleit-Dahm, K. A. Linking RAGE and Nox in
295–307 (2016). diabetic micro- and macrovascular complications. Diabetes Metab. 41, 272–281 (2015).
22. Khan, A. W. et al. The role of activator protein-1 (AP-1) complex in diabetes associated 55. Barlovic, D. P., Soro-Paavonen, A. & Jandeleit-Dahm, K. A. RAGE biology, atherosclerosis
atherosclerosis: insights from single cell RNA sequencing. Diabetes 73, 1495–1512 (2024). and diabetes. Clin. Sci. 121, 43–55 (2011).
23. Jiang, Y. Z., Manduchi, E., Jimenez, J. M. & Davies, P. F. Endothelial epigenetics in 56. Brownlee, M. The pathobiology of diabetic complications: a unifying mechanism.
biomechanical stress: disturbed flow-mediated epigenomic plasticity in vivo and in vitro. Diabetes 54, 1615–1625 (2005).
Arterioscler. Thromb. Vasc. Biol. 35, 1317–1326 (2015). 57. Brownlee, M. Biochemistry and molecular cell biology of diabetic complications. Nature
24. He, L., Zhang, C. L., Chen, Q., Wang, L. & Huang, Y. Endothelial shear stress signal 414, 813–820 (2001).
transduction and atherogenesis: from mechanisms to therapeutics. Pharmacol. Ther. 58. Giacco, F. & Brownlee, M. Oxidative stress and diabetic complications. Circ. Res. 107,
235, 108152 (2022). 1058–1070 (2010).
25. Kalluri, A. S. et al. Single-cell analysis of the normal mouse aorta reveals functionally 59. Du, X. L. et al. Hyperglycemia-induced mitochondrial superoxide overproduction
distinct endothelial cell populations. Circulation 140, 147–163 (2019). activates the hexosamine pathway and induces plasminogen activator inhibitor-1
26. Andueza, A. et al. Endothelial reprogramming by disturbed flow revealed by single-cell expression by increasing Sp1 glycosylation. Proc. Natl Acad. Sci. USA 97, 12222–12226
RNA and chromatin accessibility study. Cell Rep. 33, 108491 (2020). (2000).
27. Moonen, J.-R. A. J. et al. Endothelial progenitor cells give rise to pro-angiogenic smooth 60. Forbes, J. M. & Cooper, M. E. Mechanisms of diabetic complications. Physiol. Rev. 93,
muscle-like progeny. Cardiovasc. Res. 86, 506–515 (2010). 137–188 (2013).

Nature Reviews Cardiology

Review article

61. Haffner, S. M. Epidemiological studies on the effects of hyperglycemia and improvement pathological vascular remodeling. Arterioscler. Thromb. Vasc. Biol. 40, 1854–1869
of glycemic control on macrovascular events in type 2 diabetes. Diabetes Care 22, (2020).
C54–C56 (1999). 94. Cao, Q. et al. Inhibiting DNA methylation by 5-aza-2′-deoxycytidine ameliorates
62. Hsueh, W. et al. Recipes for creating animal models of diabetic cardiovascular disease. atherosclerosis through suppressing macrophage inflammation. Endocrinology 155,
Circ. Res. 100, 1415–1427 (2007). 4925–4938 (2014).
63. Touyz, R. M. & Montezano, A. C. Vascular Nox4: a multifarious NADPH oxidase. Circ. Res. 95. Svensson, E. C. et al. TET2-driven clonal hematopoiesis and response to canakinumab:
110, 1159–1161 (2012). an exploratory analysis of the CANTOS randomized clinical trial. JAMA Cardiol. 7,
64. Di Marco, E. et al. Pharmacological inhibition of NOX reduces atherosclerotic 521–528 (2022).
lesions, vascular ROS and immune-inflammatory responses in diabetic Apoe−/− mice. 96. Fuster, J. J. et al. Clonal hematopoiesis associated with TET2 deficiency accelerates
Diabetologia 57, 633–642 (2014). atherosclerosis development in mice. Science 355, 842–847 (2017).
65. Di Marco, E. et al. Differential effects of NOX4 and NOX1 on immune cell-mediated 97. Liu, Y. et al. TET2: a novel epigenetic regulator and potential intervention target for
inflammation in the aortic sinus of diabetic ApoE−/− mice. Clin. Sci. 130, 1363–1374 (2016). atherosclerosis. DNA Cell Biol. 37, 517–523 (2018).
66. Lewis, P. et al. Lack of the antioxidant enzyme glutathione peroxidase-1 accelerates 98. Cai, C. et al. TET2-mediated ECM1 hypomethylation promotes the neovascularization
atherosclerosis in diabetic apolipoprotein E-deficient mice. Circulation 115, 2178–2187 in active proliferative diabetic retinopathy. Clin. Epigenetics 16, 6 (2024).
(2007). 99. Liu, R. et al. Ten-eleven translocation-2 (TET2) is a master regulator of smooth muscle
67. Tan, S. M. et al. Derivative of bardoxolone methyl, dh404, in an inverse dose-dependent cell plasticity. Circulation 128, 2047–2057 (2013).
manner lessens diabetes-associated atherosclerosis and improves diabetic kidney 100. Yang, Q. et al. Low shear stress inhibited endothelial cell autophagy through TET2
disease. Diabetes 63, 3091–3103 (2014). downregulation. Ann. Biomed. Eng. 44, 2218–2227 (2016).
68. Jha, J. C. et al. Independent of renox, NOX5 promotes renal inflammation and fibrosis 101. Lee, M. K. S. et al. Interplay between clonal hematopoiesis of indeterminate potential
in diabetes by activating ROS-sensitive pathways. Diabetes 71, 1282–1298 (2022). and metabolism. Trends Endocrinol. Metab. 31, 525–535 (2020).
69. Jha, J. C. et al. Endothelial or vascular smooth muscle cell-specific expression of human 102. Libby, P. & Ebert, B. L. CHIP (clonal hematopoiesis of indeterminate potential): potent
NOX5 exacerbates renal inflammation, fibrosis and albuminuria in the Akita mouse. and newly recognized contributor to cardiovascular risk. Circulation 138, 666–668
Diabetologia 62, 1712–1726 (2019). (2018).
70. Ho, F. et al. Endothelial reactive oxygen-forming NADPH oxidase 5 is a possible player 103. Wu, D. et al. Glucose-regulated phosphorylation of TET2 by AMPK reveals a pathway
in diabetic aortic aneurysm but not atherosclerosis. Sci. Rep. 12, 11570 (2022). linking diabetes to cancer. Nature 559, 637–641 (2018).
71. Elbatreek, M. H. et al. NOX5-induced uncoupling of endothelial NO synthase is a causal 104. Nagareddy, P. R. et al. Hyperglycemia promotes myelopoiesis and impairs the resolution
mechanism and theragnostic target of an age-related hypertension endotype. PLoS Biol. of atherosclerosis. Cell Metab. 17, 695–708 (2013).
18, e3000885 (2020). 105. Agathocleous, M. et al. Ascorbate regulates haematopoietic stem cell function and
72. Guzik, T. J. et al. Calcium-dependent NOX5 nicotinamide adenine dinucleotide leukaemogenesis. Nature 549, 476–481 (2017).
phosphate oxidase contributes to vascular oxidative stress in human coronary artery 106. Cimmino, L. et al. Restoration of TET2 function blocks aberrant self-renewal and
disease. J. Am. Coll. Cardiol. 52, 1803–1809 (2008). leukemia progression. Cell 170, 1079–1095.e20 (2017).
73. Chen, Y. C., Jandeleit-Dahm, K. & Peter, K. Sodium-glucose co-transporter 2 (SGLT2) 107. Miranda, M. X. et al. The Sirt1 activator SRT3025 provides atheroprotection in Apoe−/−
inhibitor dapagliflozin stabilizes diabetes-induced atherosclerotic plaque instability. mice by reducing hepatic Pcsk9 secretion and enhancing Ldlr expression. Eur. Heart J.
J. Am. Heart Assoc. 11, e022761 (2022). 36, 51–59 (2015).
74. Libby, P. & Plutzky, J. Diabetic macrovascular disease: the glucose paradox? Circulation 108. Fick, R. J. & Trievel, R. C. An overview of chromatin modifications. Biopolymers 99, 95–97
106, 2760–2763 (2002). (2013).
75. Rahman, H. et al. Sodium-glucose cotransporter-2 inhibitors and primary prevention 109. Neele, A. E., Willemsen, L., Chen, H. J., Dzobo, K. E. & de Winther, M. P. J. Targeting
of atherosclerotic cardiovascular disease: a meta-analysis of randomized trials and epigenetics as atherosclerosis treatment: an updated view. Curr. Opin. Lipidol. 31,
systematic review. J. Am. Heart Assoc. 12, e030578 (2023). 324–330 (2020).
76. Kanter, J. E. et al. Increased apolipoprotein C3 drives cardiovascular risk in type 1 110. Hoeksema, M. A. et al. Targeting macrophage histone deacetylase 3 stabilizes
diabetes. J. Clin. Invest. 129, 4165–4179 (2019). atherosclerotic lesions. EMBO Mol. Med. 6, 1124–1132 (2014).
77. Hsu, C. C. et al. Hematopoietic NLRP3 and AIM2 inflammasomes promote diabetes- 111. Luque-Martin, R. et al. Targeting histone deacetylases in myeloid cells inhibits their
accelerated atherosclerosis, but increased necrosis is independent of pyroptosis. maturation and inflammatory function with limited effects on atherosclerosis.
Diabetes 72, 999–1011 (2023). Front. Pharmacol. 10, 1242 (2019).
78. Sharma, A. et al. Specific NLRP3 inhibition protects against diabetes-associated 112. Needham, L. A. et al. Drug targeting to monocytes and macrophages using
atherosclerosis. Diabetes 70, 772–787 (2021). esterase-sensitive chemical motifs. J. Pharmacol. Exp. Ther. 339, 132–142 (2011).
79. Khan, A. W., Paneni, F. & Jandeleit-Dahm, K. A. M. Cell-specific epigenetic changes 113. Bompada, P. et al. Epigenome-wide histone acetylation changes in peripheral blood
in atherosclerosis. Clin. Sci. 135, 1165–1187 (2021). mononuclear cells in patients with type 2 diabetes and atherosclerotic disease.
80. Bannister, A. J. & Kouzarides, T. Regulation of chromatin by histone modifications. Biomedicines 9, 1908 (2021).
Cell Res. 21, 381–395 (2011). 114. Bowes, A. J., Khan, M. I., Shi, Y., Robertson, L. & Werstuck, G. H. Valproate attenuates
81. Keating, S. T., Plutzky, J. & El-Osta, A. Epigenetic changes in diabetes and cardiovascular accelerated atherosclerosis in hyperglycemic apoE-deficient mice: evidence in support
risk. Circ. Res. 118, 1706–1722 (2016). of a role for endoplasmic reticulum stress and glycogen synthase kinase-3 in lesion
82. Mattei, A. L., Bailly, N. & Meissner, A. DNA methylation: a historical perspective. development and hepatic steatosis. Am. J. Pathol. 174, 330–342 (2009).
Trends Genet. 38, 676–707 (2022). 115. Huang, D. et al. Epigenetically altered macrophages promote development of
83. Iguchi-Ariga, S. M. & Schaffner, W. CpG methylation of the cAMP-responsive enhancer/ diabetes-associated atherosclerosis. Front. Immunol. 14, 1196704 (2023).
promoter sequence TGACGTCA abolishes specific factor binding as well as 116. van der Meer, J. W., Joosten, L. A., Riksen, N. & Netea, M. G. Trained immunity: a smart
transcriptional activation. Genes. Dev. 3, 612–619 (1989). way to enhance innate immune defence. Mol. Immunol. 68, 40–44 (2015).
84. Fuks, F. et al. The methyl-CpG-binding protein MeCP2 links DNA methylation to histone 117. Thiem, K. et al. Hyperglycemic memory of innate immune cells promotes in vitro
methylation. J. Biol. Chem. 278, 4035–4040 (2003). proinflammatory responses of human monocytes and murine macrophages. J. Immunol.
85. Meehan, R. R., Lewis, J. D. & Bird, A. P. Characterization of MeCP2, a vertebrate DNA 206, 807–813 (2021).
binding protein with affinity for methylated DNA. Nucleic Acids Res. 20, 5085–5092 (1992). 118. Edgar, L. et al. Hyperglycemia induces trained immunity in macrophages and their
86. Zhang, Y. et al. Analysis of the NuRD subunits reveals a histone deacetylase core complex precursors and promotes atherosclerosis. Circulation 144, 961–982 (2021).
and a connection with DNA methylation. Genes. Dev. 13, 1924–1935 (1999). 119. Keating, S. T. et al. Rewiring of glucose metabolism defines trained immunity induced
87. Zhang, Y. et al. DNA methylation in atherosclerosis: a new perspective. Evid. Based by oxidized low-density lipoprotein. J. Mol. Med. 98, 819–831 (2020).
Complement. Altern. Med. 2021, 6623657 (2021). 120. Bekkering, S. et al. Oxidized low-density lipoprotein induces long-term proinflammatory
88. Dai, Y., Chen, D. & Xu, T. DNA methylation aberrant in atherosclerosis. Front. Pharmacol. cytokine production and foam cell formation via epigenetic reprogramming of monocytes.
13, 815977 (2022). Arterioscler. Thromb. Vasc. Biol. 34, 1731–1738 (2014).
89. Zaina, S. et al. DNA methylation map of human atherosclerosis. Circ. Cardiovasc. Genet. 121. Thiem, K., Stienstra, R., Riksen, N. P. & Keating, S. T. Trained immunity and diabetic
7, 692–700 (2014). vascular disease. Clin. Sci. 133, 195–203 (2019).
90. Yin, Y. et al. Integrated investigation of DNA methylation, gene expression and immune cell 122. Robinson, T. M., Lee, J. W. & Roberts, V. C. Detection of early atherosclerosis by analysis of
population revealed immune cell infiltration associated with atherosclerotic plaque ultrasonic Doppler signals produced by mural flow disturbances. Med. Biol. Eng. Comput.
formation. BMC Med. Genomics 15, 108 (2022). 32, 702–703 (1994).
91. Hu, S. et al. DNA methylation profiling reveals novel pathway implicated in 123. Saeed, S. et al. Epigenetic programming of monocyte-to-macrophage differentiation
cardiovascular diseases of diabetes. Front. Endocrinol. 14, 1108126 (2023). and trained innate immunity. Science 345, 1578 (2014).
92. Nepali, K. & Liou, J. P. Recent developments in epigenetic cancer therapeutics: 124. Yoshida, K. et al. The transcription factor ATF7 mediates lipopolysaccharide-induced
clinical advancement and emerging trends. J. Biomed. Sci. 28, 27 (2021). epigenetic changes in macrophages involved in innate immunological memory.
93. Strand, K. A. et al. High throughput screen identifies the DNMT1 (DNA Nat. Immunol. 16, 1034 (2015).
methyltransferase-1) inhibitor, 5-azacytidine, as a potent inducer of PTEN (phosphatase 125. Arts, R. J. W. et al. Glutaminolysis and fumarate accumulation integrate immunometabolic
and tensin homolog): central role for PTEN in 5-azacytidine protection against and epigenetic programs in trained immunity. Cell Metab. 24, 807–819 (2016).

Nature Reviews Cardiology

Review article

126. Keating, S. T. et al. The set7 lysine methyltransferase regulates plasticity in oxidative 159. Filippatos, G. et al. Empagliflozin for heart failure with preserved left ventricular ejection
phosphorylation necessary for trained immunity induced by β-glucan. Cell Rep. 31, fraction with and without diabetes. Circulation 146, 676–686 (2022).
107548 (2020). 160. Berger, J. H. et al. SGLT2 inhibitors act independently of SGLT2 to confer benefit for HFrEF
127. Miao, F. et al. Genome-wide analysis of histone lysine methylation variations caused in mice. Circ. Res. 135, 632–634 (2024).
by diabetic conditions in human monocytes. J. Biol. Chem. 282, 13854–13863 (2007). 161. Kleindorfer, D. O. et al. 2021 guideline for the prevention of stroke in patients
128. Miao, F., Gonzalo, I. G., Lanting, L. & Natarajan, R. In vivo chromatin remodeling events with stroke and transient ischemic attack: a guideline from the American Heart
leading to inflammatory gene transcription under diabetic conditions. J. Biol. Chem. 279, Association/American Stroke Association. Stroke 52, e364–e467 (2021).
18091–18097 (2004). 162. Lee, M. M. Y., Kristensen, S. L., Gerstein, H. C., McMurray, J. J. V. & Sattar, N.
129. Michelson, A. D. Antiplatelet therapies for the treatment of cardiovascular disease. Cardiovascular and mortality outcomes with GLP-1 receptor agonists in patients with
Nat. Rev. Drug. Discov. 9, 154–169 (2010). type 2 diabetes: a meta-analysis with the FREEDOM cardiovascular outcomes trial.
130. Jourdi, G., Lordkipanidzé, M., Philippe, A., Bachelot-Loza, C. & Gaussem, P. Current and Diabetes Metab. Syndr. 16, 102382 (2022).
novel antiplatelet therapies for the treatment of cardiovascular diseases. Int. J. Mol. Sci. 163. Mone, P. et al. SGLT2 inhibition via empagliflozin improves endothelial function and
22, 13079 (2021). reduces mitochondrial oxidative stress: insights from frail hypertensive and diabetic
131. Coppens, M., Weitz, J. I. & Eikelboom, J. W. Synergy of dual pathway inhibition in chronic patients. Hypertension 79, 1633–1643 (2022).
cardiovascular disease: lessons from the COMPASS trial. Circ. Res. 124, 416–425 (2019). 164. Heidenreich, P. A. et al. 2022 AHA/ACC/HFSA guideline for the management of heart
132. ASCEND Study Collaborative Group Effects of aspirin for primary prevention in persons failure: a report of the American College of Cardiology/American Heart Association
with diabetes mellitus. N. Engl. J. Med. 379, 1529–1539 (2018). Joint Committee on Clinical Practice Guidelines. Circulation 145, e895–e1032 (2022).
133. Seidu, S. et al. Aspirin has potential benefits for primary prevention of cardiovascular 165. Ridker, P. M. et al. Antiinflammatory therapy with canakinumab for atherosclerotic
outcomes in diabetes: updated literature-based and individual participant data disease. N. Engl. J. Med. 377, 1119–1131 (2017).
meta-analyses of randomized controlled trials. Cardiovasc. Diabetol. 18, 70 (2019). 166. Nidorf, S. M. et al. Colchicine in patients with chronic coronary disease. N. Engl. J. Med.
134. Marx, N. et al. 2023 ESC Guidelines for the management of cardiovascular disease in 383, 1838–1847 (2020).
patients with diabetes. Eur. Heart J. 44, 4043–4140 (2023). 167. Tardif, J.-C. et al. Efficacy and safety of low-dose colchicine after myocardial infarction.
135. Cannon, C. P. et al. Ezetimibe added to statin therapy after acute coronary syndromes. N. Engl. J. Med. 381, 2497–2505 (2019).
N. Engl. J. Med. 372, 2387–2397 (2015). 168. Bonaventura, A. & Abbate, A. Colchicine for cardiovascular prevention: the dawn
136. Liao, J. K. & Laufs, U. Pleiotropic effects of statins. Annu. Rev. Pharmacol. Toxicol. 45, of a new era has finally come. Eur. Heart J. 44, 3303–3304 (2023).
89–118 (2005). 169. Masson, W., Lobo, M., Barbagelata, L., Lavalle-Cobo, A. & Molinero, G. Effect of
137. Ridker, P. M. et al. Inflammation and cholesterol as predictors of cardiovascular events anti-inflammatory therapy on major cardiovascular events in patients with diabetes:
among patients receiving statin therapy: a collaborative analysis of three randomised a meta-analysis. Diabetes Metab. Syndr. 15, 102164 (2021).
trials. Lancet 401, 1293–1301 (2023). 170. Abbate, A. et al. Interleukin‐1 blockade inhibits the acute inflammatory response
138. Sabatine, M. S. et al. Evolocumab and clinical outcomes in patients with cardiovascular in patients with ST‐segment-elevation myocardial infarction. J. Am. Heart Assoc. 9,
disease. N. Engl. J. Med. 376, 1713–1722 (2017). e014941 (2020).
139. Schwartz, G. G. et al. Alirocumab and cardiovascular outcomes after acute coronary 171. Lopez-Olivo, M. A. et al. Methotrexate for treating rheumatoid arthritis. Cochrane
syndrome. N. Engl. J. Med. 379, 2097–2107 (2018). Database Syst. Rev. 2014, CD000957 (2014).
140. Zhang, X.-L. et al. Safety and efficacy of anti-PCSK9 antibodies: a meta-analysis of 25 172. US National Library of Medicine. ClinicalTrials.gov www.clinicaltrials.gov/ct2/show/
randomized, controlled trials. BMC Med. 13, 123 (2015). NCT01594333 (2020).
141. Handelsman, Y. & Lepor, N. E. PCSK9 inhibitors in lipid management of patients with 173. Ridker, P. M. et al. Low-dose methotrexate for the prevention of atherosclerotic events.
diabetes mellitus and high cardiovascular risk: a review. J. Am. Heart Assoc. 7, e008953 N. Engl. J. Med. 380, 752–762 (2019).
(2018). 174. Ridker, P. M. et al. IL-6 inhibition with ziltivekimab in patients at high atherosclerotic risk
142. Leiter, L. A. et al. Efficacy and safety of alirocumab in insulin-treated individuals with type (RESCUE): a double-blind, randomised, placebo-controlled, phase 2 trial. Lancet 397,
1 or type 2 diabetes and high cardiovascular risk: the ODYSSEY DM-INSULIN randomized 2060–2069 (2021).
trial. Diabetes Obes. Metab. 19, 1781–1792 (2017). 175. US National Library of Medicine. ClinicalTrials.gov www.clinicaltrials.gov/ct2/show/
143. Muller-Wieland, D. et al. Design and rationale of the ODYSSEY DM-DYSLIPIDEMIA trial: NCT05021835 (2024).
lipid-lowering efficacy and safety of alirocumab in individuals with type 2 diabetes and 176. Ridker, P. M. From RESCUE to ZEUS: will interleukin-6 inhibition with ziltivekimab prove
mixed dyslipidaemia at high cardiovascular risk. Cardiovasc. Diabetol. 16, 70 (2017). effective for cardiovascular event reduction? Cardiovasc. Res. 117, e138–e140 (2021).
144. Ray, K. K. et al. Alirocumab vs usual lipid-lowering care as add-on to statin therapy 177. US National Library of Medicine. ClinicalTrials.gov www.clinicaltrials.gov/ct2/show/
in individuals with type 2 diabetes and mixed dyslipidaemia: the ODYSSEY NCT05636176 (2024).
DM-DYSLIPIDEMIA randomized trial. Diabetes Obes. Metab. 20, 1479–1489 (2018). 178. Warraich, H. J., Wong, N. D. & Rana, J. S. Role for combination therapy in diabetic
145. Cariou, B. et al. Efficacy and safety of alirocumab in insulin-treated patients with type 1 dyslipidemia. Curr. Cardiol. Rep. 17, 32 (2015).
or type 2 diabetes and high cardiovascular risk: rationale and design of the ODYSSEY 179. Fitchett, D. H., Hegele, R. A. & Verma, S. Cardiology patient page. Statin intolerance.
DM-INSULIN trial. Diabetes Metab. 43, 453–459 (2017). Circulation 131, e389–e391 (2015).
146. Ray, K. K. et al. Two phase 3 trials of inclisiran in patients with elevated LDL cholesterol. 180. Mehta, S. et al. An ultrasound-responsive theranostic cyclodextrin-loaded nanoparticle
N. Engl. J. Med. 382, 1507–1519 (2020). for multimodal imaging and therapy for atherosclerosis. Small 18, e2200967 (2022).
147. Leiter, L. A. et al. Inclisiran in individuals with diabetes or obesity: post hoc pooled 181. Paneni, F., Beckman, J. A., Creager, M. A. & Cosentino, F. Diabetes and vascular disease:
analyses of the ORION-9, ORION-10 and ORION-11 phase 3 randomized trials. Diabetes pathophysiology, clinical consequences, and medical therapy: part I. Eur. Heart J. 34,
Obes. Metab. 26, 3223–3237 (2024). 2436–2443 (2013).
148. Digenio, A. et al. Antisense-mediated lowering of plasma apolipoprotein C-III by 182. Dai, Y. et al. Targeted therapy of atherosclerosis vulnerable plaque by ROS-scavenging
volanesorsen improves dyslipidemia and insulin sensitivity in type 2 diabetes. Diabetes nanoparticles and MR/fluorescence dual-modality imaging tracing. Int. J. Nanomed. 17,
Care 39, 1408–1415 (2016). 5413–5429 (2022).
149. Baggio, L. L. & Drucker, D. J. Biology of incretins: GLP-1 and GIP. Gastroenterology 132, 183. Su, S. et al. Efficiently inhibiting systemic inflammatory cascades by fullerenes
2131–2157 (2007). for retarding HFD-fueled atherosclerosis. Adv. Health. Mater. 12, e2202161 (2023).
150. Dungan, K. A. B. Glucagon-like peptide 1-based therapies for type 2 diabetes: a focus 184. Mallat, Z. et al. Inhibition of transforming growth factor-β signaling accelerates
on exenatide. Clin. Diab 23, 56–62 (2005). atherosclerosis and induces an unstable plaque phenotype in mice. Circ. Res. 89,
151. Kodera, R. et al. Glucagon-like peptide-1 receptor agonist ameliorates renal injury 930–934 (2001).
through its anti-inflammatory action without lowering blood glucose level in a rat model 185. Li, C., Wan, L., Luo, J., Jiang, M. & Wang, K. Advances in subcutaneous delivery systems
of type 1 diabetes. Diabetologia 54, 965–978 (2011). of biomacromolecular agents for diabetes treatment. Int. J. Nanomed. 16, 1261–1280
152. Fujita, H. et al. The protective roles of GLP-1R signaling in diabetic nephropathy: possible (2021).
mechanism and therapeutic potential. Kidney Int. 85, 579–589 (2014). 186. Ismail, R. et al. Encapsulation in polymeric nanoparticles enhances the enzymatic
153. Marso, S. P. et al. Liraglutide and cardiovascular outcomes in type 2 diabetes. N. Engl. J. Med. stability and the permeability of the GLP-1 analog, liraglutide, across a culture model
375, 311–322 (2016). of intestinal permeability. Pharmaceutics 11, 599 (2019).
154. Gerstein, H. C. et al. Dulaglutide and renal outcomes in type 2 diabetes: an exploratory 187. Margolis, J. et al. Systemic nanoparticle paclitaxel (nab-paclitaxel) for in-stent restenosis
analysis of the REWIND randomised, placebo-controlled trial. Lancet 394, 131–138 (2019). I (SNAPIST-I): a first-in-human safety and dose-finding study. Clin. Cardiol. 30, 165–170
155. Wiviott, S. D. et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N. Engl. (2007).
J. Med. 380, 347–357 (2019). 188. Lecio, G. et al. Novel 20% doxycycline-loaded PLGA nanospheres as adjunctive therapy
156. Zinman, B. et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 in chronic periodontitis in type-2 diabetics: randomized clinical, immune
diabetes. N. Engl. J. Med. 373, 2117–2128 (2015). and microbiological trial. Clin. Oral. Investig. 24, 1269–1279 (2020).
157. Fernandez-Fernandez, B., Sarafidis, P., Soler, M. J. & Ortiz, A. EMPA-KIDNEY: expanding 189. Asadi, S. et al. Nano curcumin supplementation reduced the severity of diabetic
the range of kidney protection by SGLT2 inhibitors. Clin. Kidney J. 16, 1187–1198 (2023). sensorimotor polyneuropathy in patients with type 2 diabetes mellitus: a randomized
158. Anker, S. D. et al. Empagliflozin in heart failure with a preserved ejection fraction. N. Engl. double-blind placebo- controlled clinical trial. Complement. Ther. Med. 43, 253–260
J. Med. 385, 1451–1461 (2021). (2019).

Nature Reviews Cardiology

Review article

190. Li, X. et al. Synergistic effects of liposomes encapsulating atorvastatin calcium and 213. US National Library of Medicine. ClinicalTrials.gov www.clinicaltrials.gov/ct2/
curcumin and targeting dysfunctional endothelial cells in reducing atherosclerosis. show/NCT05014672 (2024).
Int. J. Nanomed. 14, 649–665 (2019). 214. US National Library of Medicine. ClinicalTrials.gov www.clinicaltrials.gov/ct2/
191. Mocanu, C. A. et al. P-selectin targeted RAGE-shRNA lipoplexes alleviate show/NCT05323656 (2024).
atherosclerosis-associated inflammation. J. Control. Rel. 338, 754–772 (2021). 215. US National Library of Medicine. ClinicalTrials.gov www.clinicaltrials.gov/ct2/
192. Wang, Q. et al. Inhibition of atherosclerosis progression by modular micelles. J. Control. Rel. show/NCT02010242 (2015).
354, 294–304 (2023). 216. Australian New Zealand Clinical Trials Registry. ANZCTR anzctr.org.au/Trial/
193. Moore, K. J. & Tabas, I. Macrophages in the pathogenesis of atherosclerosis. Cell 145, Registration/TrialReview.aspx?ACTRN=12617001187336 (2023).
341–355 (2011). 217. Touyz, R. M., Anagnostopoulou, A., Rios, F., Montezano, A. C. & Camargo, L. L. NOX5:
194. Chen, W. et al. Macrophage-targeted nanomedicine for the diagnosis and treatment molecular biology and pathophysiology. Exp. Physiol. 104, 605–616 (2019).
of atherosclerosis. Nat. Rev. Cardiol. 19, 228–249 (2022). 218. Vlad, M. L. et al. Histone acetyltransferase-dependent pathways mediate upregulation
195. Tao, W. et al. siRNA nanoparticles targeting CaMKIIγ in lesional macrophages improve of NADPH oxidase 5 in human macrophages under inflammatory conditions: a potential
atherosclerotic plaque stability in mice. Sci. Transl. Med 12, eaay1063 (2020). mechanism of reactive oxygen species overproduction in atherosclerosis. Oxid. Med.
196. Cui, K. et al. Epsin nanotherapy regulates cholesterol transport to fortify atheroma Cell Longev. 2019, 3201062 (2019).
regression. Circ. Res. 132, e22–e42 (2023). 219. Chen, F. et al. Inhibition of histone deacetylase reduces transcription of NADPH oxidases
197. Flores, A. M. et al. Pro-efferocytic nanoparticles are specifically taken up by lesional and ROS production and ameliorates pulmonary arterial hypertension. Free. Radic.
macrophages and prevent atherosclerosis. Nat. Nanotechnol. 15, 154–161 (2020). Biol. Med. 99, 167–178 (2016).
198. van der Valk, F. M. et al. Prednisolone-containing liposomes accumulate in human 220. Manea, S. A. et al. Epigenetic regulation of vascular NADPH oxidase expression and
atherosclerotic macrophages upon intravenous administration. Nanomedicine 11, reactive oxygen species production by histone deacetylase-dependent mechanisms
1039–1046 (2015). in experimental diabetes. Redox Biol. 16, 332–343 (2018).
199. Li, B. H. et al. TRPV1 activation impedes foam cell formation by inducing autophagy 221. Zhu, J. et al. Association of glucose-lowering medications with cardiovascular outcomes:
in oxLDL-treated vascular smooth muscle cells. Cell Death Dis. 5, e1182 (2014). an umbrella review and evidence map. Lancet Diabetes Endocrinol. 8, 192–205 (2020).
200. Gao, W. et al. Copper sulfide nanoparticles as a photothermal switch for TRPV1 signaling 222. Kosiborod, M. et al. Rates of myocardial infarction and stroke in patients initiating
to attenuate atherosclerosis. Nat. Commun. 9, 231 (2018). treatment with SGLT2-inhibitors versus other glucose-lowering agents in real-world
201. Basatemur, G. L., Jorgensen, H. F., Clarke, M. C. H., Bennett, M. R. & Mallat, Z. Vascular clinical practice: results from the CVD-REAL study. Diabetes Obes. Metab. 20, 1983–1987
smooth muscle cells in atherosclerosis. Nat. Rev. Cardiol. 16, 727–744 (2019). (2018).
202. Costantino, S. et al. Epigenetics and precision medicine in cardiovascular patients: 223. Kosiborod, M. et al. Cardiovascular events associated with SGLT-2 inhibitors versus
from basic concepts to the clinical arena. Eur. Heart J. 39, 4150–4158 (2018). other glucose-lowering drugs: the CVD-REAL 2 study. J. Am. Coll. Cardiol. 71, 2628–2639
203. Napoli, C. et al. Primary prevention of atherosclerosis: a clinical challenge for the (2018).
reversal of epigenetic mechanisms? Circulation 125, 2363–2373 (2012). 224. Peikert, A. & Vaduganathan, M. Sodium glucose co-transporter 2 inhibition following
204. Ray, K. K. et al. Effect of apabetalone added to standard therapy on major adverse acute myocardial infarction: the DAPA-MI and EMPACT-MI trials. JACC Heart Fail. 12,
cardiovascular events in patients with recent acute coronary syndrome and type 2 949–953 (2024).
diabetes: a randomized clinical trial. JAMA 323, 1565–1573 (2020).
205. Borck, P. C., Guo, L. W. & Plutzky, J. BET epigenetic reader proteins in cardiovascular Author contributions
transcriptional programs. Circ. Res. 126, 1190–1208 (2020). The authors contributed substantially to all aspects of the article.
206. Bailey, D. et al. RVX-208: a small molecule that increases apolipoprotein A-I and
high-density lipoprotein cholesterol in vitro and in vivo. J. Am. Coll. Cardiol. 55, Competing interests
2580–2589 (2010). The authors declare no competing interests.
207. Brandts, J. & Ray, K. K. Apabetalone – BET protein inhibition in cardiovascular disease
and type 2 diabetes. Future Cardiol. 16, 385–395 (2020). Additional information
208. Tsujikawa, L. M. et al. Apabetalone (RVX-208) reduces vascular inflammation in vitro and Peer review information Nature Reviews Cardiology thanks Peter Rossing, Geoff Werstuck
in CVD patients by a BET-dependent epigenetic mechanism. Clin. Epigenetics 11, 102 and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
(2019).
209. Nicholls, S. J. et al. Selective BET protein inhibition with apabetalone and cardiovascular Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
events: a pooled analysis of trials in patients with coronary artery disease. Am. J. published maps and institutional affiliations.
Cardiovasc. Drugs 18, 109–115 (2018).
210. Jha, J. C. et al. Genetic targeting or pharmacologic inhibition of NADPH oxidase nox4 Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this
provides renoprotection in long-term diabetic nephropathy. J. Am. Soc. Nephrol. 25, article under a publishing agreement with the author(s) or other rightsholder(s); author
1237–1254 (2014). self-archiving of the accepted manuscript version of this article is solely governed by the
211. US National Library of Medicine. ClinicalTrials.gov www.clinicaltrials.gov/ct2/ terms of such publishing agreement and applicable law.
show/NCT03865927 (2024).
212. US National Library of Medicine. ClinicalTrials.gov www.clinicaltrials.gov/ct2/ © Springer Nature Limited 2025
show/NCT03226067 (2022).

Nature Reviews Cardiology