Sphingolipids and Atherosclerosis

Summary:

Atherosclerosis begins when ApoB-containing lipoproteins accumulate and undergo modification in the subendothelial space and then progresses through endothelial dysfunction, leukocyte recruitment, foam-cell formation, and plaque formation. Sphingolipids participate in each of these stages both as structural membrane components and as signaling molecules. In this review, we highlight recent advances in sphingolipid biomarkers and discuss their clinical promise and limitations.

1. Introduction

Atherosclerosis is a chronic inflammatory disease of medium and large arteries initiated by the retention of ApoB-containing lipoproteins beneath the endothelium, which become trapped by arterial proteoglycans, triggering local inflammation and initiating plaque formation [1]. According to the World Health Organization, cardiovascular diseases, which mainly result from atherosclerotic coronary, cerebral, and peripheral artery disease, caused an estimated 19.8 million deaths in 2022 (≈32% of all global deaths). In the United States, the Centers for Disease Control and Prevention report that coronary heart disease alone accounted for 371,506 deaths in 2022. Therefore, slowing or preventing the progression of this disease has become a major global goal to reduce the large number of deaths it causes each year.

Atherogenesis is initiated in large and medium arteries when ApoB-containing low-density lipoproteins (LDL) enter the intima and are retained by glycosaminoglycan chains on arterial proteoglycans, which prolongs their residence time [2, 3]. The retained LDL undergoes biochemical modification and aggregation, promoting endothelial activation, up-regulation of adhesion molecules and chemokines, and leukocyte recruitment [3, 4]. Monocytes transmigrate, differentiate into macrophages, and take up modified LDL via scavenger receptors such as CD36 and SR-A, becoming foam cells that contribute to the lipid-rich core [5, 6]. In parallel, endothelial dysfunction, characterized by reduced nitric-oxide bioavailability and increased reactive oxygen species, impairs vasodilation and amplifies vascular inflammation and adhesiveness [7, 8].

This storyline naturally leads to sphingolipids. In the late 19th century, Johann Ludwig Thudichum described sphingosine and introduced the term “sphingolipid,” a lipid class now central to membrane architecture and cell signaling. Johann Ludwig Thudichum chose the name because the molecules’ complex and mysterious structure reminded him of the Sphinx’s riddles [9]. By modern human plaque lipidomics, sphingolipids—including sphingomyelin (SM), ceramides (Cer), dihydroceramides (dhCer), glucosylceramides (GluCer), lactosylceramides (LacCer), and sphingosine-1-phosphate (S1P)—occur at elevated levels in human atherosclerotic lesions and track with plaque inflammation and instability [10]. Decades of work have connected specific sphingolipids to key atherogenic processes. ApoE deficient mice, a well-known atherogenic mouse model is not only the hypercholesterolemia [11] but also hypersphingolipidemia [12]. We and others found that treatment with myriocin, an SPT-specific inhibitor, decreases atherosclerosis in ApoE deficient mouse models [13–16].

In this review, we integrate evidences from human and mouse model studies within an endothelium–lipoprotein–atherogenic plaque framework, with particular emphasis on species-resolved lipidomics to clarify disease mechanisms and guide biomarker development and targeted interventions that may improve patient outcomes.

1. Sphingolipid formation

1.1. De novo biosynthesis

Sphingolipid de novo biosynthesis in the ER starts with the condensation of serine and palmitoyl-CoA to form 3-ketosphinganine, catalyzed by serine palmitoyltransferase (SPT) (Figure 1, Step 1). In three subsequent steps, 3-ketosphingosine forms ceramide (Step 2). Ceramide molecules are subsequently transported from the ER to the Golgi apparatus by the ceramide transfer protein (CERT) or through vesicular routes, where they enter downstream metabolic branches. From the ceramide hub, multiple sphingolipids, including SM (Step 3), glucosylceramide (GluCer; Step 4), galactosylceramide (GalCer; Step 5), ceramide 1 phosphate (Step 6), and sphingosine-1-phosphate (S1P; Step 7), can be synthesized [17].

Fig.1. Sphingolipid biosynthesis: location and regulation.

SMS1&2, sphingomyelin synthase 1 &2; SMSr, sphingomyelin synthase related protein; PE-PLC, phosphtidylethanolamine-specific phospholipase C. SM, sphingomyelin; Sph, sphingosine; GluCer, glucosylceramide; S1P, sphingosine-1-phosphate; Cer-1P, ceramide-1-phosphage; P-ethanolamine, phosphoethanolamine; PE, phosphtidylethanolamine; DAG, diacylglycerol. SPT can be negatively regulated by ceramide, S1P, and SMSr activity-related PE. ER, endoplasmic reticulum; Golgi, golgi apparatus; PM, plasma membrane.

One major branch is the conversion of ceramide to SM via the enzyme family sphingomyelin Synthase (SMS; Figure 1, Step 3). There are at least three mammalian homologues: SMS1, localized mainly to the Golgi, SMS2 localized to both Golgi and plasma membrane, and SMS-related protein (SMSr), which has no SM synthase activity [18–20] (Figure 1, Step 8). SMS1 and SMS2 catalyze the transfer of the phosphocholine from phosphatidylcholine (PC) to ceramide, generating SM and diacylglycerol. The SM produced is then trafficked to various membrane compartments, including the plasma membrane, inner organelles and lipoprotein surfaces [21–23]. SMS family is also a group of phospholipase Cs. In particular, SMS1 and SMS2 are phosphatidylcholine-specific phospholipase Cs (PC-PLCs) [24] and SMSr is a phosphatidylethanolamine-specific phospholipase C (PE-PLC) [25].

In parallel, ceramide is converted by glucosylceramide synthase (GCS) to form GluCer, which is extended to lactosylceramide (LacCer) and further elaborated into complex gangliosides [26]. This glycosphingolipid branch supports cell–cell adhesion, immune signaling and lipid-raft architecture [27–29].

The catabolic and salvage pathways recycle sphingolipids, where sphingomyelinases hydrolyze SM to ceramide, ceramidases convert ceramide to sphingosine. Sphingosine-1-phosphate (S1P) is a bioactive sphingolipid produced by the phosphorylation of sphingosine, a reaction catalyzed by sphingosine kinase-1 (SphK1) and sphingosine kinase-2 (SphK2) [30]. Conversely, sphingosine phosphatases can dephosphorylate S1P to regenerate sphingosine, thereby maintaining a dynamic S1P–sphingosine–ceramide rheostat that regulates cellular sphingolipid homeostasis [31]. Together, these interconnected reactions enable the dynamic interconversion between bioactive and structural sphingolipids, forming the basis of the “sphingolipid rheostat” concept [32, 33].

1.2. Regulation of sphingolipid synthesis

The activity of mammalian SPT, the key enzyme for the de novo sphingolipid biosynthesis, is closely regulated by multiple factors. SPT is primarily a heterodimer of SPTLC1/SPTLC2 [34]; however, two low-molecular-weight proteins, ssSPTa and ssSPTb, enhance SPT activity and confer acyl-CoA substrate specificity on mammalian SPTs [35]. In yeast, orosomucoid (ORM)1 and ORM2, acute phase proteins which interact with SPTLC1, negatively regulate SPT [36, 37], as do ORMDLs, the human homologs of yeast ORMs [38, 39]. The cryo‐electron microscopy (EM) structure of the SPT-ORMDL3 complex [40] that SPTLC1 is not directly involved in catalysis but acts as an anchor to target SPTLC2 to the ER membrane [41]. Although SPT has a third subunit, SPTLC3, our knowledge about SPTLC3 is limited. SPTLC3 and SPTLC2 share 68% sequence identity and likely have similar functions, in terms of SPT activity [34], but act independently. SPTLC3 can mediate atypical sphingolipid formation [42, 43]. Human GWAS show that SPTLC3, but not SPTLC1 or 2, affects circulating lipid metabolism and cardiovascular disease [44–47]. Ceramide is an SPT regulator. Ceramide can directly bind to the SPT-ORMDL complex and stabilizes the ORMDL protein in an inhibitory conformation, thus, ceramide can serve as sensor for sphingolipid homeostasis.

S1P is another SPT activity regulator. Most sphingolipids can be recycled via a salvage pathway [17]. However, cleavage of S1P into E1P and hexadecenal by S1P lyase (S1PL), which also occurs in the ER, is a decisive step that enables sphingoid base substrates to irreversibly exit the sphingolipid metabolic pathway (Figure 1, Step 9).[48]. SPT activity can be regulated by S1PL activity. S1PL deficiency reduces SPT activity through S1P mediated downregulation of SPT [48].

Sphingomyelin synthase related protein (SMSr) activity is a novel regulator of SPT. We found that although SMSr cannot directly be involved in SM biosynthesis, however, its phosphatidylethanolamine-specific phospholipase C (PE-PLC) activity-mediated PE changes can serve as a sensor for SPT activity. SMSr activity positively correlated with SPT activity through its PE-PLC activity [49] (Figure 1, Step 8). The cryo-EM structure of human SMSr has been resolved, which assembles into a homohexamer by tandem trimerization of SMSr dimers [50]. SMSr catalyzes two separate reactions. First, it hydrolyzes PE to release the phosphoethanolamine (PE-PLC activity); Second, SMSr catalyzes re-formation of the phosphoester bond between generated phosphoethanolamine and ceramide, if the latter is available, to form a ceramide phosphoethanolamine. If ceramide is not available, only phosphoethanolamine is generated[50] (Figure 1; Step 8). So far, ceramide, S1PL-related S1P changes, and SMSr-related PE changes are three known factors to fine-turn SPT activity and contribute to sphingolipid homeostasis (Figure 1).

2. Sphingolipids as Biomarkers for the Detection and Progression of Atherosclerosis

Ceramides are increasingly viewed as clinically useful lipid biomarkers that capture both existing atherosclerotic cardiovascular disease (ASCVD) and the risk of future events. In large human cohorts, specific plasma ceramide species, mainly Cer(d18:1/16:0), Cer(d18:1/18:0), Cer(d18:1/24:1), and their ratios to Cer(d18:1/24:0) predict cardiovascular death and major adverse events beyond standard lipids, and these findings have been translated into validated ceramide-based risk scores now applied in clinical practice [51–53]. Plasma ceramides also relate to features of plaque vulnerability in vivo, such as thin cap fibroatheroma and plaque rupture measured by optical coherence tomography [54]. Within the arterial wall, sphingomyelinase hydrolyzes LDL-associated SM to ceramide, causing LDL aggregation and fusion and increases proteoglycan binding, thereby enhancing particle retention in the intima and supporting foam cell formation [55, 56]. A cross-sectional human study further linked ceramide (22:0 and 24:0) to arterial stiffness, adding a vascular-aging perspective and suggesting that these species may serve as biomarkers for this assessment [57].

Sphingomyelin (SM) has been connected to atherosclerotic lesion composition for many years. A study using hypercholesterolemic rabbits showed that SM accumulated in ASCVD lesions at higher levels than many other lipid classes [58]. The first clinical evidence came from a biethnic angiographic case–control study showing that patients with ASCVD had significantly higher plasma SM levels than controls, independent of LDL cholesterol, suggesting that SM may contribute to atherosclerosis risk beyond traditional lipids [59]. However, such case–control designs capture cross-sectional associations in patients who already have disease and may reflect disease-related metabolic changes and treatments rather than true prospective risk. Prospective cohort studies indicate that total plasma SM is not uniformly predictive after multivariable adjustment; for example, in the Multi-Ethnic Study of Atherosclerosis, total SM did not independently predict incident coronary heart disease [60]. In the Cardiovascular Health Study, targeted plasma sphingolipid profiling showed that ceramides and SM species differ in relation to heart failure [61]. Lemaitre et al. found out that higher plasma levels of Cer-16 and SM-16 were associated with an increased risk of incident heart failure, whereas higher levels of Cer-22, Cer-24, SM-20, SM-22, and SM-24 were linked to lower risk [61]. In a separate Cardiovascular Health Study analysis of sudden cardiac death, chain length again appeared important: in older adults, higher plasma levels of Cer-16 and SM-16 were associated with increased risk, whereas ceramide and SM species carrying very-long-chain saturated fatty acids (C20–C24) were not significantly associated with higher risk [62]. Similarly, in the Strong Heart Family Study, ceramides and SM species containing 16:0 were associated with higher all-cause mortality, whereas longer-chains species showed inverse associations with mortality [63]. Another multi-cohort analysis also emphasizes that both ceramides and SMs show outcome-specific associations that depend on acyl-chain length and saturation [64]. Taken together, these studies suggest that the biological and clinical effects of sphingomyelins and other sphingolipids are species-specific.

S1P, a bioactive sphingolipid carried largely by HDL in plasma, has attracted interest as a potential biomarker for ASCVD. In the circulation, most S1P is bound to HDL via apolipoprotein M (ApoM), which links S1P signaling to lipoprotein function and vascular health [65]. In a clinical study of patients with myocardial infarction, stable coronary artery disease, and healthy controls, HDL-bound S1P was reduced, whereas non-HDL-bound S1P was relatively increased, indicating a redistribution of S1P away from HDL and suggesting impaired HDL function in CAD [66]. Another study shows that serum S1P concentrations are consistently lower in patients with coronary, peripheral, or carotid atherosclerosis than in healthy individuals, and that these reductions correlate with disease severity and systemic inflammation [67]. These observations support the idea that lower circulating or HDL-bound S1P acts as a negative biomarker of vascular health.

For more complex sphingolipid, early human aortic tissue studies identified neutral glycosphingolipids in plaques, with GluCer and LacCer among the predominant species in diseased intima compared with uninvolved vessel wall [68, 69]. In a familial hypercholesterolemia swine model, spatial lipidomics revealed that phosphorylated and glycosylated ceramides such as CerP(36:1), CerPE(38:1) and HexCer(38:1) are enriched in advanced coronary plaques, supporting their role as tissue-level sphingolipid signatures of advanced atherosclerotic lesions [70].

In 200 human carotid plaques studies, mass-spectrometry showed six sphingolipid classes (dhCer, Cer, SM, S1P, GluCer, LacCer) are all elevated and associate with plaque inflammation, while histological markers of instability correlate with all except sphingosine-1-phosphate, supporting sphingolipids as plaque-level biomarkers [10].

3. Ceramide mechanisms in atherosclerosis

3.1. Endothelium: ceramide drives dysfunction

Ceramide activates protein phosphatase-2A (PP2A) in endothelial cells, which disrupts the Akt–Hsp90–eNOS complex and lowers eNOS phosphorylation and nitric-oxide (NO) output. Pharmacologic PP2A inhibition, by contrast, preserves the complex and vascular function in vivo [71, 72]. Ceramide also changes the mediator of flow-induced vasodilation from NO to mitochondrial H₂O₂ in human resistance vessels and inhibition of ceramide production restores NO-dependent dilation [73]. Oxidative stress then makes the problem worse. Reactive oxygen species activate acid and neutral sphingomyelinases, increasing intracellular ceramide, whereas ceramide itself facilitates the formation of membrane microdomains enriched in NADPH oxidase, thereby promoting more ROS generation and sustaining endothelial activation [74]. C16:0 ceramide can additionally function as a ligand for endothelial GPCRs by binding CYSLTR2 and P2RY6. This engagement activates Gq and inflammasome pathways, promoting endothelial activation in vivo. Blocking CYSLTR2 or P2RY6 reduces this response and limits lesion growth [75].

3.2. Lipoproteins and foam-cell formation: ceramide promotes LDL aggregation and retention

Neutral sphingomyelinases present in the intima can convert SM carried on LDL into ceramide. As ceramide accumulates on the particle surface, LDL tends to cluster and form larger aggregates, which increases their arterial retention and makes them more readily internalized by macrophages [55, 76]. Ceramide-rich or aggregated LDL further triggers macrophage activation responses, including enhanced matrix metalloproteinase-7 release and stimulation of T-cell, contributing to plaque remodeling and destabilization[77] [78].

3.3. Species and context matter: short and very-long-chain ceramides are different

Endothelial cells require a basal level of de novo–generated sphingolipids to regulate vascular tone; when SPTLC2 is deleted in the endothelium, eNOS signaling becomes impaired and systemic blood pressure rises, highlighting the physiological need for a minimal ceramide pool [79]. Studies further indicate that not all ceramide species behave the same. C16:0-ceramide appears more involved in receptor-mediated vasodilatory pathways, whereas very- long-chain species such as C24:0 or C24:1 are more important for flow-mediated vasodilation. This chain pattern is consistent with the substrate preferences of specific ceramide synthase: CerS4 and CerS6 mainly generate long-chain ceramides (C16–C20), have been linked to stress and apoptosis in multiple models, while CerS2, which responsible for C22–C24 Ceramide, tends to support more pro-survival or protective functions depending on context [80–83].

4. Sphingomyelin mechanisms in ASCVD

4.1. SMS in endothelial dysfunction

Raising SM at the endothelial surface through sphingomyelin synthase 2 (SMS2) enhances lipid rafts organization and makes the plasma membrane more permissive to inflammatory signaling. In human endothelial cells exposed to oxidative stress (H₂O₂), SMS2 overexpression increases the expression of ICAM-1, VCAM-1, and MCP-1 and promotes THP-1 adhesion while blocking SMS2 with the small molecule Dy105 reverses these changes and turns off Wnt/β-catenin signaling, indicating that SMS2 directly involved in stress-driven endothelial activation [84]. Genetic work supports the same direction. Loss of Sms2 reduces NF-κB activation and lowers induction of downstream genes after TNF or LPS stimulation, likely because fewer receptors such as TNFR1 can cluster within lipid rafts at the plasma membrane[85, 86]. Although these experiments were done mainly in macrophages or HEK293 cells, the underlying mechanism rafts-dependent receptor signaling is also applicable to endothelial membranes. Besides Wnt signaling, SMS2 can also impair endothelial health by triggering endoplasmic-reticulum stress [87]. SMS1 has been less directly studied in endothelium, but hematopoietic studies point in a similar pro-inflammatory role. In bone-marrow transplantation experiments, Sms1 deficiency lowers SM, dampens TLR4–NF-κB/MAPK signaling in macrophages, and reduces atherosclerotic lesion burden in susceptible mice [86]. Dietary trans-unsaturated fatty acids (TFAs) are incorporated into sphingolipids through SPT, producing SM and increasing circulating atherogenic lipoproteins. Blocking SPT through SPT, producing SM and increasing circulating atherogenic lipoproteins. Blocking SPT reverses these lipid changes and reduces atherogenesis in LDL receptor deficient mice [88].

4.2. Lipoproteins & foam-cell formation: SM makes ApoB particles more retention-prone

LDL becomes structurally altered when sphingomyelinases (SMases) hydrolyze its surface sphingomyelin to ceramide. This transformation modifies the particle surface and drives extensive aggregation and fusion among LDL, small VLDL, and IDL [89]. These structural changes markedly increase lipoprotein binding to human arterial proteoglycans, promoting subendothelial retention [56]. Experimental overexpression of SMS1 or SMS2 by adenovirus in mouse liver raises SM on ApoB-containing lipoproteins, making them highly sensitive to SMase-induced aggregation and may create a proatherogenic potential [90]. Conversely, lowering SM specifically in ApoB-containing lipoproteins of SMS2 and ApoE double knockout mouse reduces LDL aggregation, diminishes arterial retention, and significantly decreases atherosclerotic lesion area in [91]. Macrophage-specific SMS2 deficiency in ApoE deficient mice reduces aortic lesion burden, decreases macrophage accumulation and improves inflammatory markers [92]. Once LDL has been retained and aggregated, or modified by SMase, it is taken up readily by macrophages through macropinocytosis and other receptor-independent pathways, accelerating foam-cell formation [93, 94].

5. S1P mechanisms in ASCVD

5.1. Endothelium: S1P preserves barrier and NO, restrains adhesion

Sphingosine-1-phosphate (S1P) signals through five G-protein-coupled receptors (S1PR1–S1PR5), each with distinct expression patterns and downstream G-protein coupling. Through these receptors, S1P regulates cell migration, adhesion, proliferation, and inflammatory pathways [95, 96]. These receptors are functionally expressed in endothelial cells, vascular smooth-muscle cells, and macrophages, the major cell types involved in the initiation and progression of atherosclerosis [97]. S1P circulates in blood and lymph at high (0.1–1 mM) concentrations which are significantly higher than the nM concentrations required to activate S1P receptors [98]. S1P in blood is produced primarily by red blood cells (RBC) [99], platelets [100], and endothelial cells [98], and secreted by specific S1P transporter, major facilitator superfamily transporter 2b (Mfsd2b) [100] and S1P transporter spinster homolog 2 (Spns2) [101], respectively. Due to its hydrophobic nature, S1P is poorly water soluble and requires carrier proteins for efficient transport and circulation. According to the literature, most of plasma S1P (45–50%) is carried by ApoM and the remainder by albumin [102]. However, based on our most recent studies, we found that albumin is not SIP carrier, because albumin deficiency in mice does not influence plasma S1P levels [103]. Plasma ApoA4 [104] and phospholipid transfer protein activity [103] make contributions for S1P in the circulation.

ApoM-HDL-bound S1P plays an important protective role at the endothelium. It strengthens barrier function and limits inflammation by reducing VCAM-1 and E-selectin expression, thereby lowering monocyte adhesion [105]. Through S1PR1–Gi–PI3K/Akt/Rac signaling, HDL-bound S1P stimulates eNOS phosphorylation, promotes NO-dependent release, and supports endothelial survival [106, 107]. Conversely, loss of endothelial S1PR1 or depletion of circulating S1P increases vascular permeability, inflammation, and lesion formations [97, 108, 109]. Translational human population work also links lower plasma S1P to higher blood pressure and elevated inflammatory markers [110]. Hemodynamic forces also influence sphingolipid balance. Under disturbed flow, endothelial sphingolipid metabolism shifts away from ceramide and toward S1P production. Endothelial Nogo-B knockout maintains this S1P-favored state, reduces ASCVD, and stabilizes fibrous caps, consistent with an endothelial protective effect [111].

5.2. S1P receptor signaling in macrophages and smooth muscle cells (SMCs)

Beyond its endothelial role, S1P affects macrophage and smooth-muscle function within plaques. S1PR2, which primarily couples to Gα12/13, activates Rho–ROCK and PTEN pathways that promote inflammatory gene expression and suppress cell migration [112–115]. In bone-marrow transplantation studies, loss of S1PR2 in hematopoietic cells of ApoE-deficient mice markedly decreases lesion size and reduces macrophage accumulation l. S1PR2 deficiency also lowers expression of pro-inflammatory genes such as IL-18 and Il-1β, indicating less macrophage activation [116]. Although global S1PR3 deletion does not strongly change lesion area, double knockout of S1PR3 and ApoE reduces macrophage content and increases smooth-muscle accumulation, suggesting that S1PR3 also participates in monocyte recruitment and plaque remodeling [117].

5.3. Targeting S1P for Therapy

The S1P analogue FTY720 (Fingolimod; Figure 2) has been widely studied as a pharmacological modulator of S1P receptor signaling in atherosclerosis. Early studies demonstrated that systemic FTY720 administration significantly reduces atherosclerotic lesion size in ApoE-deficient mice, accompanied by decreased macrophage infiltration and lipid accumulation within plaques [118, 119]. Mechanistically, active metabolite FTY720-phosphate acts as a functional agonist at S1PR1,3,4,5, transiently activating followed by downregulating lymphocyte S1P receptors, which limits immune cell egress and systemic inflammation [120, 121]. In macrophages, FTY720 promotes cholesterol efflux by increasing ABCA1 expression through LXR–SphK2 signaling, thereby reducing foam-cell formation [122].

Fig.2.

Structure of S1P, Fingolimod (S1P receptor modulator), and Myriocin (a native inhibitor of SPT).

6. GluCer and LacCer: Glycosphingolipid Signaling in ASCVD

7.1. Endothelium: LacCer drives oxidative stress and adhesion

Glycosphingolipids are plasma-membrane lipid composed of a carbohydrate (glyco-) attached to a ceramide backbone that extend the sphingolipid pathway beyond structural membrane roles into cell signaling and inflammation [123, 124]. They accumulate substantially in human and experimental atherosclerotic plaques and contribute to vascular dysfunction, inflammation, and cellular proliferation [125]. Endothelial LacCer activates NADPH oxidase, increases superoxide (O₂⁻) and reduces eNOS-derived NO bioavailability, which together promote TNF-α–induced NF-κB signaling, upregulates ICAM-1 and drives monocyte adhesion to the endothelium [126–128]. LacCer further activates Ras–Raf–MEK–ERK signaling, resulting in endothelial proliferation and vascular wall thickening [129, 130]. Treatment with D-threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol (D-PDMP), an inhibitor of both glucosylceramide synthase and lactosylceramide synthase reverses these endothelial changes and restores vasorelaxation in ApoE knockout mice, indicating that elevated LacCer directly contributes to endothelial dysfunction [126, 131].

7.2. Glycosphingolipids in macrophage activation and foam-cell formation

During atherogenesis, modified lipoproteins such as oxidized LDL activate macrophage inflammatory signaling through TLR4–NF-κB pathways. In macrophages, endogenous glucosylceramide supports plasma-membrane microdomains that enhance LPS/TLR4 signaling, promoting inflammatory activation that may facilitate lipid uptake and foam-cell formation [132]. Reducing GluCer with a GCS inhibitor (D-PDMP) dampens IL-6, IL-1β, TNF-α and NF-κB activation [132]. In contrast to these endogenous effects, previous reports in other cell types have reported that exogenously added glycosphingolipids may inhibit LPS signaling, likely reflecting differences in lipid sources [133, 134].

7.3. Therapeutic targeting of GCS

In vivo, pharmacological inhibition of GCS with AMP-DNM markedly reduces atherosclerotic plaque size in ApoE*3-Leiden and LDLR knockout mice, accompanied by decreased hepatic cholesterol, increased biliary output and fecal sterol excretion [135]. In human carotid endarterectomy specimens, GluCer and LacCer (as well as Cer, dhCer, SM, and S1P) are elevated and correlate with plaque inflammation and instability. Earlier human aortic studies similarly found GluCer and LacCer enriched in atherosclerotic lesions and elevated neutral GSLs in cells from atherosclerotic intima [10, 136].

8. Sphingolipids and metabolic diseases

The sphingolipid biosynthesis pathway is associated with metabolic diseases, including insulin resistance, dyslipidemia, metabolic-associated fatty liver disease (MAFLD), which are strongly linked to atherosclerosis through shared underlying mechanisms. Ceramide is linked with insulin resistance, MAFLD, and atherosclerosis [15, 137]. Sphingosine-1-phosphate (S1P) modulates insulin signaling in various cell types [138–140]. Glycosphingolipids are also involved in insulin signaling [141]. A deficiency in ganglioside GM3, a key glycosphingolipid, enhances tyrosine phosphorylation of the insulin receptor [142, 143]. A reduction of sphingomyelin (SM), a major component of lipid rafts, in the plasma membrane improves insulin signaling [144–146] and decreases diet-induced obesity [144]. SMSr activity influences diet-induced fatty liver, liver fibrosis, and liver inflammation [24] through regulating SPT activity [49]. Deleting the enzyme dihydroceramide desaturase 1 (DES1) in the liver and/or adipose tissue resolved diet-induced liver steatosis and insulin resistance in mice through preventing lipid uptake and promoting glucose utilization [137].

9. Discussion

Perturbations of SPT activity have been linked to metabolic diseases and ASCVD. The mechanism could involve the reduction of plasma SM, ceramide, and glycosphingolipids. In mice, myriocin treatment protects against diet-induced insulin resistance [147], MAFLD [15], nonalcoholic steatohepatitis (NASH) [148], dyslipidemia [149] and ASCVD [13, 150]. Heterozygous SPTLC2 deficient mice were protected from high-fat-diet-induced obesity, insulin resistance [144] and atherosclerosis [151]. All these observations clearly indicated that inhibiting sphingolipid de novo synthesis is an ideal approach for the treatment of metabolic diseases and ASCVD. However, a big challenge was met in the field. Although myriocin can help prevent the diseases, it is toxic and has unfavorable physicochemical properties such as low solubility [152]. A modification of myriocin led to the development of FTY720, an agonist of the S1P receptor rather than an SPT inhibitor, to treat multiple sclerosis [153, 154]. The long-term goal is to develop a new treatment, through manipulation of sphingolipid biosynthesis, for human metabolic complications and ASCVD.

10. Conclusion

Sphingolipids influence atherosclerosis at three connected layers: endothelial homeostasis, lipoprotein retention and modification, and plaque inflammation/remodeling. Ceramide and sphingomyelin generally promote disease by reducing NO signaling, amplifying ROS and NF-κB pathways, and making ApoB-containing lipoproteins more retention- and aggregation-prone, which accelerates foam-cell formation. By contrast, HDL-bound S1P supports endothelial protection through S1PR1 signaling and, when favored over ceramide under disturbed flow, is associated with smaller, more stable coronary lesions. Glycosphingolipids further link vascular oxidative stress to inflammatory activation in both endothelium and macrophages, and their inhibition reduces lesion burden in vivo. Together, these findings suggest that targeting sphingolipid metabolism and signaling could provide complementary therapeutic strategies. Future work should emphasize species-resolved analysis, integrate human plaque and plasma readouts and develop translational studies that define when and where sphingolipid-directed interventions are most effective.