In cardiac tissue, the ER serves as the site of LD biogenesis, where neutral lipid synthesis precedes their assembly into nascent LDs (13). The ER harbors essential enzymes responsible for synthesizing the two major neutral lipid species: Acyl-CoA: cholesterol O-acyltransferases 1 (ACAT1, as ACAT2 is not expressed in the heart) primarily catalyzes CEs (14), and TAGs, synthesized by diacylglycerol acyltransferase (DGAT) 1 and DGAT2 (15). Whereby deficiency in specific lipid synthesis enzymes can be compensated by alternative pathways to maintain neutral lipid production and LD formation (16). DGAT1 localizes exclusively to the ER membrane, whereas DGAT2 displays dynamic subcellular distribution, localizing to ER membrane and LD interface. When intracellular FA concentrations increase, DGAT2 translocates to the LD surface, thereby facilitating TAG storage and LD expansion (15).

Beyond the canonical DGAT-mediated pathway, alternative mechanisms contribute to TAG synthesis, including transmembrane thioredoxin 1, which catalyzes storage lipid production independently of DGAT1/2, though the mechanism requires further investigation (17). LD nucleation occurs when TAG and CEs accumulate to concentrations of 5-10 mmol/l within the ER bilayer, forming lipid lenses that coalesce under the action of PL acid and diacylglycerol (DAG) (4,18-20). The subsequent LD budding is orchestrated by ER-intra TAG concentration and the scaffolding protein Seipin which localizes to nascent lipid lenses to drive LD growth and facilitate the detachment of the PL monolayer-enveloped LD through ER-to-cytoplasm (21). LD budding directionality and size are determined by the asymmetric distribution of the PL monolayer cap, which is governed by ER membrane asymmetry, lipid composition and the spatial positioning of budding site (22,23). Perturbations in neutral lipids trafficking to the ER membrane or aberrant retention on the cytoplasmic leaflet impair the clearance of misfolded proteins, thereby triggering ER stress and inflammatory responses that necessitate continuous PL supplementation to sustain budding (24). Notably, a fundamental distinction exists between mammalian and yeast LD biology; whereas mammalian LDs exist as independent cytoplasmic organelles, yeast LDs maintain persistent ER membrane continuity. This evolutionary divergence reflects species-specific adaptations in lipid storage compartmentalization and ER-LD membrane dynamics (25,26).

To meet cardiac energy demands, stored LDs undergo catabolism through lipolysis and lipohagy, processes that hydrolyze TAG into FA for subsequent oxidation. Lipolysis proceeds through a sequential three-step enzymatic cascade: (i) ATGL mediates rate-limiting TAG hydrolysis at the sn-2 position, generating DAG and FAs. (ii) Hormone-sensitive lipase (HSL) subsequently cleaves DAG into FAs and monoacylglycerol (MAG). (iii) MAG lipase (MGL) completes the process by liberating the final FAs and MAG (27,28). The specificity and efficiency of ATGL-mediated lipolysis are enhanced by its cofactor comparative gene identification 58/Alpha beta hydrolase domain-containing protein 5, which redirects ATGL activity toward the sn-1 position and cooperates with DGAT2 to generate SN1,3DAG. SN1,3DAG is a stereoisomer that bypasses protein kinase C activation but serves as the optimal HSL substrate for continued lipolytic flux (27,28). CGI 58 inhibited the lipolysis rate after interacting with Plin1, and the inhibitory effect disappeared after PKA phosphorylation of Plin1 (29,30). Similarly, Plin5 phosphorylation orchestrates a functional switch from a lipolysis barrier to a promoter of TAG breakdown, as phosphorylated PLIN5 recruits both CGI 58 and ATGL, facilitating the coordinated hydrolysis of TAG into glycerol and Fas (31,32). This phosphorylation-dependent regulation of perilipin-lipase interactions represents a critical control node integrating hormonal signals with cellular energy status to modulate cardiac lipid catabolism.

LDs are partially or completely decomposed into glycerol and FAs in a process known as lipophagy. Lipophagy mainly degrades neutral lipids such as TAGs and CEs through three autophagy modes, including molecular chaperon-mediated autophagy (CMA), macro-autophagy and micro-autophagy, each characterized by specific molecular mechanisms. In CMA, the 71 kDa heat shock cognate protein (HSC70) proteins directly recognizes Plin2, Plin3 and Plin5 via the five-peptide motif protein KFERQ like pentapeptide motifs, facilitating their selective degradation, and subsequently eliminating their protective shielding of ATGL at the LD surface (33). Following HSC70-mediated recognition, the KFERQ-bearing perilipins binds to lysosome-associated membrane protein 2A (LAMP2A), assembling into a translocation complex that delivers these cargo proteins into the lysosomal lumen for degradation. The CMA-mediated removal of Plin2 and Plin3 consequently enables ATGL accessibility to LDs, thereby promoting both ATGL-mediated lipolysis and subsequent lipophagy (34). During macro-autophagy, LDs recruit autophagy related proteins to initiate the biogenesis of autophagosomes, facilitating the selective sequestration and lysosomal degradation of lipid cargo (35). Micro-autophagy mediates direct LD engulfment by lysosomes through membrane contact site tethering mechanisms, wherein LD contents are transferred to the lysosomal lumen via direct injection or membrane transfer remains unclear, potentially regulated by membrane fusion associated small GTPases, and hydrolyzed by lysosomal acid lipase independent of cytoplasmic lipases and autophagy (36).

It should be noted that the degradation of LD through lipolysis or lipohagy is determined by the size of the LD. Lipohagy can only degrade smaller LDs in liver, while LDs exceeding the degradation capacity of lipohagy will trigger ATGL-mediated cytoplasmic neutral lipolysis under normal conditions, with starvation or cellular stress inducing acid lipolysis to release FAs for FAO. Thus, the liberation of lipids and TGs from LDs for energy production is facilitated by neutral and acidic lipolysis (37). Whether similar size-dependent and pathway-selective mechanisms operate across DCM warrants further investigation (Fig. 1).

The dynamic stability and regulatory function of LDs are significantly reliant on surface-binding proteins. The categorization of these LD-associated proteins remains insufficient: One perspective separates them according to the mechanism of metastasis. Class I proteins such as DGAT2, ACSL3 and GPAT4 are translocated from the ER to LDs through the ERTOLD and include hydrophobic hairpin motifs that directly engage with LDs (38).

The CYTOLD pathway mediates the direct transport of Class II proteins from the cytosol to LDs. These proteins include Plins, members of the Perilipin family that possess an amphipathic helix, as well as cell death-inducing DNA Fragmentation Factor Alpha (DFFA)-like effector (CIDE) family of proteins. Conversely, CGI58, which is associated with LDs via protein-protein interactions, engages with proteins, including histones and the small GTPase Rab18 (RAB18), which is anchored to LDs through lipid modification (39,40).

Based on their roles, these proteins can be categorized into LD structural proteins/resident proteins, lipid metabolism enzymes, and transport/signal transduction proteins. Such proteins are produced through fusion or localized lipid synthesis.

DGAT2: LD biogenesis involves functional coupling of FA transporter protein 1 on the ER membrane and DGAT2 at nascent LD surfaces, which coordinately promote TAG synthesis and lipid transfer (41). However, DGAT2 expression must be tightly controlled, as overexpression compromises ER membrane stability (42). Post-translational modification provides additional regulation, H₂S intervention enhances ubiquitin-mediated degradation of DGAT1 and DGAT2 and enhanced the expression of the E3 ligase Hrd1 to reduce LDs accumulation (43). The small GTPase RAB1B dynamically regulates DGAT2 subcellular distribution by facilitating its trafficking from ER to LDs via secretory pathways, promoting the expansion and growth of LDs (44). Collectively, these mechanisms encompassing protein interactions, expression control, proteolytic regulation and vesicular trafficking converge to fine-tune LD dynamics in response to metabolic demands.

The perilipin family represents the major LD-resident proteins in mammalian cells. In cardiac tissue, PLIN2 and PLIN5 predominate; whereas Plin1 is mainly located in adipocytes, Plin5 facilitates TAG storage. During the metabolic transition from glycolysis to β-oxidation, Plin2 upregulation activates the peroxisome proliferator-activated receptors (PPARs) signaling pathway (45,46). Plin2 regulates the shape, size and morphology of LDs, while both Plin2 and Plin5 inhibit ATGL mediated lipolysis, thereby modulating lipid accumulation (47,48).

RAB18: Plin2 harbors the RAB18 binding site within its C-terminal domain (49). Although this interaction exhibits cell type specificity. Genetic lack of RAB18 in C2C12 skeletal muscle cells does not significantly alter the expression level of Plin2, whereas RAB18 overexpression in HepG2 cells modulates Plin2 abundance (50). Functionally, RAB18 recruits the ER membrane-localized NAG-RINT1-ZW10 (NRZ) complex and associated SNARE proteins (Syntaxin18, USE1 and BNIP1) to form the RAB18-NRZ-SNARE multiprotein complex, which facilitates LDs' formation and mediates stable membrane contact sites (MCSs) between ER and LDs. The depletion of RAB18 does not affect early LD biogenesis but significantly impairs subsequent LD growth and maturation (50,51).

ABHD5/CGI 58: CGI 58, a highly conserved regulator, is characterized by its interaction with Plin and ATGL to function as a coactivator in ATGL-mediated lipolysis (52). PKA-mediated phosphorylation of Plin5 in the heart enhances the expression of CGI 58. By contrast, non-phosphorylated Plin5 binds CGI 58 and inhibits lipase activity. Beyond Plin regulation, cellular homeostasis also requires coordinated action of other key metabolic enzymes. Lipolysis is primarily mediated by three enzymes: ATGL, HSL and MGL. Upregulation of these key lipolysis enzymes ameliorate cardiac steatosis and myocardial fibrosis (53). Decreased expression of lipolysis enzymes aggravates mitochondrial dysfunction (54).

ATGL acts as the rate-limiting enzyme for TAG hydrolysis in LDs, ATGL deficiency reduces PPARα expression and FAO levels, resulting in human neutral lipid storage disorder and cardiomyopathy (55). Restricting lipid absorption alone fails to reverse ATGL deficiency induced cardiomyotoxicity (56). Yet ATGL overexpression ameliorates myocardial injury by enhancing lipolysis (55). This indicates that ATGL deficiency extends beyond mere lipid accumulation, as its mediated lipolysis liberates PPAR agonists that activate downstream PPARα/PGC-1α-regulated FAO and mitochondrial biogenesis (57). Although pharmacological interventions partially restore FAO, the signaling function of ATGL remains indispensable for cardiac protection.

HSL: Overexpression of Plin2 markedly reduces HSL expression, indicating that Plin2 as a functional inhibitor of HSL (53). HSL exerts a cardioprotective effect that precedes the overt development of cardiac lipotoxicity. In the context of DCM, overexpression of HSL does not influence VLDL absorption but effectively downregulates key profibrotic factors, including collagen, TGF-β and matrix metalloproteinase (MMP) 2. This suppression alleviates the lipotoxicity and LDs accumulation driven by TAG and DAG (58).

LDs sequester excess histones, preventing histone toxicity, and supplies histones for chromatin assembly during DNA replication (59,60). Multiple proteins govern stability of LDs while concurrently engaging in membrane transport and signal transduction, thereby influencing cellular metabolic remodeling (61). Proteins of the LD proteome related to LD biogenesis in DCM are included in Table I.

LDs establish contact sites with multiple organelles, most notably mitochondria; the adult heart derives ~95% of its ATP from mitochondrial oxidative phosphorylation, with FAO contributing 40-60% of reduction equivalents. Mitochondrial dysfunction generates lipotoxicity intermediates that promote intracellular lipid accumulation. LDs counteract lipotoxicity by sequestering excessive FAs and lipid intermediates. During energy stress such as fasting or β-adrenergic stimulation, increased cardiac energy demand and elevated adipocyte lipolysis lead to heightened expression of LDs, Plin5, and mitochondrial LDs contact sites in cardiac tissue. This metabolic demand promotes formation of 'peridroplet mitochondria (PDM)', a specialized mitochondrial subpopulation with unique functional, ridge-structured, and proteomically distinct mitochondrial compartment attached to LDs (5,69). Plin5 mediates these LD-mitochondria contacts (LDMCs) and preferentially targets PDMs (5). In cardiomyocytes, enhanced Plin5-dependent LDMC improve mitochondrial respiratory capacity and metabolic flexibility. Paradoxically, FAO remains independent of LDMC (70). This may relate to PDM primarily promoting FA activation and TAG synthesis with stronger oxidative phosphorylation and ATP supply capacity, which are not mainly responsible for FAO (71).

Previous studies have revealed a link between LDs and mitophagy. Mitophagy leads lysosomes to phagocytose PDM and release FAs, which aggravates intracellular lipid peroxidation and iron accumulation and eventually cause ferroptosis. Inhibiting mitophagy significantly reduces LD accumulation and FFA levels, yet DGAT1 expression remains elevated and continues to drive nascent LDs formation (72).

PLIN2 regulates LD homeostasis and mitochondrial remodeling through epigenetic mechanisms. By suppressing lipolysis, Plin2 elevates acetyl-CoA and histone acetylation levels, maintaining embryonic stem cell pluripotency (73). Loss of Plin2 shifts cellular metabolism from glycolysis to oxidative phosphorylation and accelerates pluripotency exit. Under these conditions, Plin2 is recognized by Hsc70 and degraded through CMA, promoting mobilization of LDs (74). This is similar to the metabolic transition of cardiomyocytes from embryonic to adult stages. Although short-term glucose deprivation alters Plin2 conformation, weakening its interaction with ATGL and enhancing lipolysis, though the impact on cardiac metabolism remains uncertain (75).

In fact, the link between LDs and mitochondria is largely established by specific molecules or pathways that serve multiple functions in both LD dynamics and mitochondrial operation. Under nutrient excess conditions, mitophagy becomes dysregulated and AMPK activity declines (76). Only part of AMPK-mediated phosphorylation of ACC1 and ACC2 fails to effectively inhibit FA synthesis, paradoxically promoting LD formation. AMPK activates PPARs γ coactivator 1 alpha (PGC1α) is blocked, disrupting the formation of complexes between PGC-1α, Plin5 and silent information regulator 1 (SIRT1), and impairing mitochondrial biosynthesis (77).

Concurrently, excessive mammalian target of rapamycin complex 1 (mTORC1) signaling stimulates lipid accumulation by activating its key downstream effector sterol regulatory element binding protein 1c (SREBP1c). Autophagy is inhibited by Akt-mediated suppression of FOXO transcription and further impaired by elevated branched-chain amino acids (BCAAs), which suppress the mTORC1/autophagy-related protein (ATG)/ULK pathway (78,79). SIRT1 mediated deacetylation activates FOXO1, whose deacetylation status regulates the expression of ATGL and PPARα (80).

Under energy-deficient conditions, FOXO1 also recruits Rab7 to sustain autophagy, a process involved in the micro-lipophagy degradation of LDs. In DCM, the inhibitory effect of FOXO on transcription factor EB (TFEB) is weakened (81). It is worth noting that PKA promotes lipolysis mediated by Plin5, CGI58 and ATGL (8). Therefore, LD metabolism is coordinately regulated by upstream kinases, for instance, PKA, mTOR, as well as AMPK and transcription factors including silent information regulator 1 (SIRT1), PGC-1α, FOXO, SREBP1, PPARs and TFEB (31).

LDs represent the principal cellular compartment specialized for neutral lipid storage, distinguished from other organelles by both structural architecture and functional capacity. The ER despite serving as the site of lipid biosynthesis, incorporates lipids within its PL bilayer rather than sequestering them in concentrated deposits, limiting its storage capacity and potentially compromising membrane function under lipid excess (4). Mitochondria maintain membrane lipids essential for bioenergetics but actively oxidize rather than store FAs, making them vulnerable to lipotoxic damage when substrate availability exceeds oxidative capacity. Peroxisomes metabolize very-long-chain and BCAAs through β-oxidation yet lack mechanisms for TAG or CEs accumulation (72). Lysosomes degrade lipids delivered via endocytosis or autophagy but function as catabolic rather than storage compartments.

By contrast, LDs possess a unique PL monolayer encompassing a hydrophobic neutral lipid core, an architecture that enables massive expansion without membrane stress or dysfunction of organelles (4). This specialized structure allows LDs to accumulate energy-dense TAGs and CEs while protecting other organelles from lipotoxic intermediates. The dynamic nature of LD expansion, lipophagy and lipolysis regulate lipid mobilization coordinated with metabolic demands, whereas lipid accumulation in other organelles typically signals pathological dysfunction rather than adaptive storage. Consequently, LDs serve as the primary defense against abnormal lipid deposition, protecting mitochondria, ER, and plasma membrane integrity from lipotoxicity-induced cellular damage.

Beyond their canonical storage function, LDs protect cells from lipotoxicity by limiting lipid peroxidation and buffering excess FAs during metabolic dysregulation. LDs regulate diverse metabolic substrates, including BCAAs and glucose, both central to DCM pathogenesis. LDs participate in intermembrane lipid trafficking, store precursor molecules for lipids, mitigate oxidative damage, and modulate inflammatory responses. By coordinating these diverse functions, LDs protect mitochondria and other organelles while regulating cellular processes such as autophagy, ferroptosis, and cell division (11,12).

Hyperglycemia induces cardiac lipid overload and metabolic dysregulation, with insulin resistance exacerbating lipotoxic injury to cardiomyocytes, disrupts metabolic homeostasis, manifesting as mitochondrial dysfunction, ER stress and calcium overload. LDs likely play multiple roles in modulating these pathological responses through their capacity to store and mobilize lipids. Under conditions of sustained hyperglycemia, excessive FA influx overwhelms oxidative capacity, triggering abnormal lipid accumulation that compromises cellular homeostasis. Insulin resistance further impairs metabolic flexibility, preventing adaptive substrate switching and intensifying reliance on FAO despite limited mitochondrial capacity (82). The resultant accumulation of lipotoxic intermediates, including DAG, ceramides and acyl-carnitines, directly impairs mitochondrial respiration, induce calcium dysregulation and ER stress.

LDs limit FAs' incorporation into membrane PLs and reducing susceptibility to peroxidative damage. This antioxidant function preserves organellar integrity and prevents ferroptosis driven by lipid peroxidation (72). Conversely, LDs mobilize stored lipids through two pathways: Neutral lipase-mediated lipolysis and autophagy-dependent acid lipolysis. Lipolysis liberates FAs for FAO during energy demand. Insulin resistance impairs autophagy, a mechanism essential for maintaining cellular homeostasis. Lipophagy, representing a cargo-selective autophagy pathway, specifically targets LDs for lysosomal degradation. Current evidence indicates that both CMA and micro-autophagy contribute to metabolic regulation in DCM, although the role of macro-autophagy remains controversial (81). The proteins and PLs on the surface of LDs interact with specific enzymes to exert important functions. The balance between LD storage and degradation thus determines lipid flux, organelle function, and influence cellular metabolic homeostasis in DCM.

Cardiomyocytes mainly rely on oxidative phosphorylation for energy supply and can also generate energy through glycolysis and other ways. Nowadays, the change of FAO in DCM remains controversial (83). It may be related to the stage of DCM. A multimeric complex consisting of extended synaptotagmin 1 (ESYT1), ESYT2, VAMP Associated Protein B and C localizes to ER-LD-mitochondria contact sites. Loss of this complex reduces LD-derived FAO by 67% and promotes LDs accumulation, indicating that the complex facilitates lipid mobilization and prevents lipotoxicity (84,85). This reduced oxidation capacity appears paradoxical given previous studies of elevated FAO in DCM, suggesting compensatory mechanisms may operate under different metabolic conditions (57). These findings demonstrate that LD accumulation and the resulting imbalance in FAO are important pathogenic mechanisms in DCM.

Ca²⁺ transfer at mitochondria-ER contact sites regulates FAO. The majority of Ca²⁺ stored in the ER is transferred to the mitochondria via Inositol-requiring enzyme 1α (IRE1α) regulated inositol trisphosphate receptor (IP3R) mediates Ca²⁺ influx from the ER to mitochondria activating the IP3R-75-kDa glucose-regulated protein (Grp75) and voltage-dependent anion channel (VDAC) complex (IP3R-Grp75-VDAC) complex, promoting mitochondrial respiratory chain and ATP synthesis. This complex recruits Seipin to accelerate LDs' biosynthesis and lipid transfer in the liver. Although regulating VDAC1 and GRP75 has a protective effect on the heart, their role in cardiac LDs biosynthesis remains unclear (86,87).

Normally, acyl-CoA synthetase long chain family member 1 activates FAs to fatty acyl-CoA for subsequent β-oxidation, which directly generates acetyl-CoA for the TCA cycle. Long-chain FAs require CPT1/2-mediated mitochondrial transport before oxidation (88). In DCM, long term overload of FAO promotes electron leakage, generating reactive oxygen species (ROS) and aggravates oxidative stress. Plin5 suppresses the ROS-activated Phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway and the mitogen-activated protein kinase pathway (p38/ERK/JNK), hence preventing oxidative stress and apoptosis (89,90).

LDs protect against ferroptosis through antioxidant mechanisms. FSP1, located in LDs, and maintains antioxidant capacity by reducing Coenzyme Q10, preventing peroxidation of neutral lipids. Impairment of FSP1 function induces the peroxidation of polyunsaturated FA-LDs (PUFA-LDs), triggering ferroptosis of cardiomyocytes (91). In type 1 diabetic patients, glucolipotoxicity reduces LAMP2A, Hsc70 and Hsc90 expression, inhibits the clearance of autophagosomes and inactivates TFEB, causing autophagy disorders and rendering the myocardium vulnerable to ER stress and metabolic injury (92).

Cardiomyocytes maintain lipid homeostasis by balancing the FAO rate with LD biosynthesis. Under nutrient excess, insulin activates acetyl-CoA carboxylase (ACC) to generate malonyl-CoA, which inhibits the acetylation level of CPT1 and MPC2, thereby restricting FAO (93,94). Studies in CPT1b-deficient skeletal muscle confirm that reducing FAO alleviates mitochondrial lipid burden during energy surplus (95). Conversely, ACC2 knockdown maintains FAO and reduced cardiac hypertrophy, demonstrating that preserving oxidative capacity protects cardiac function (96).

DGAT1 and GPAT4 are upregulated to enhance TAG synthesis, and the elevation of TAG levels is positively correlated with systolic dysfunction, suggesting that excessive accumulation of LDs adversely affects cardiac structure (97).

Downregulated TGR5 fails to inhibit DHHC4, which mediates the palmitoylation of CD36, consequently promoting CD36 overexpression, the abnormal FAO, and the elevated LDs' accumulation (98,99). Exceeding the LDs' buffering capacity results in surplus FAs, leading to cellular lipotoxicity and disruption of energy supply. Concurrently, the inhibitory effect of ATGL on ceramide will be weakened, resulting in increased ROS production and triggers mitochondrial dysfunction, ultimately inducing apoptosis and aggravating insulin resistance (100,101) (Figs. 2 and 3).

Macrophages primarily orchestrate inflammatory and immunological responses to sustain cellular metabolic homeostasis. While numerous studies have researched lipotoxicity in cardiomyocytes, studies on lipotoxicity in macrophages remain insufficient. Macrophages modulate their energy acquisition through dynamic remodeling of membrane receptors coupled with metabolic reprogramming of glucose and lipid pathways. These cells meet their lipid requirements by engulfing apoptotic cells, lipoprotein particles and LDs.

Low-density lipoprotein (LDL) is oxidized to OxLDL, which macrophages internalize via CD36-mediated endocytosis. Lysosomal hydrolysis releases FFAs and TC which are stored in LDs as CE. Reduced lipid autophagy and cholesterol efflux further lead to LDs accumulation, ultimately impairing phagocytic capacity and transforming into foam cells (102).

Macrophages form LDs under conditions of inflammation and oxidative stress. With different phenotypic switches, the expression levels of LD-related proteins are also different. Plin1 and Plin2 regulate both polarization of macrophages and LD size. Plin1 was found to promote larger LDs when overexpressed in M1 macrophages. With Plin1 overexpression promoting larger LDs in M1 macrophages, Plin1 associates with the inflammatory phenotype whereas Plin2 correlates with the anti-inflammatory phenotype and collectively governing lipid storage (103). Current investigations of LD biology in macrophage subtypes remain confined to classic M1 macrophages and M2 macrophages despite the identification of diverse macrophage phenotypes. Nevertheless, LDs serve critical functions in maintaining cellular homeostasis and macrophage function, processes fundamentally linked to atherosclerosis (104).

Advanced glycation end products in the diabetic cardiac microenvironment promote M1 macrophage polarization through the miR-471-3p/SIRT1 pathway. M1 macrophages enhance glycolysis while suppressing FAO, thereby activating the pentose phosphate pathway to generate pro-inflammatory and bactericidal mediators (105). M2 macrophages, by contrast, exhibit greater LD accumulation and elevated FAO levels, maintaining their anti-inflammatory phenotype through PPARγ (101). TGF-β promotes LDs synthesis via the ERK1/2 pathway and attenuates STAT1 to inhibit M1 polarization (106). Activation of the PPAR pathway upregulates CD36 and Plin2, thereby promoting LD formation, and influences immunophenotypic switching (107,108). CX3CR1 signaling appears to inhibit M1 polarization while promoting foam cell formation (109). The expanding lipid core subsequently elevates the levels of MMP, nitric oxide and endothelin, and eventually causes plaque rupture (110).

Macrophage-derived LD-associated hydrolase upregulates Abca1 and Abcg1 expression through an LXR dependent mechanism, thereby attenuating the inflammatory response induced by SE accumulation (111). Although conventional perspective considers that accumulation of LDs fosters insulin resistance or inflammation, protective regulatory mechanisms exist. Forkhead box protein C2 (foxc2) suppresses the angiopoietin-like 2 (angptl2)/NOD-like receptor thermal protein domain associated protein 3 (NLRP3) inflammasome pathway, reducing the expression of NLRP3, ASC and caspase-1. Foxc2 also modulates Bcl-2/Bax to diminish apoptosis, oxidative stress and LDs' accumulation in macrophages (112,113).

Foam cell formation is regulated by lipophagy factors including HSPA5, UBE2G2 and AUP1. Upon autophagy activation, LDs initiate lipophagy by recruiting LC3 and adaptor protein Sequestosome 1 (SQSTM1). Sirt6-dependent chromatin remodeling factor SNF2H inhibits Wnt family member 1/β-catenin pathway to accelerate lipophagy and autophagosome lysosome mediated lipolysis and accelerating autophagosome-lysosome-mediated lipolysis and FA transfer to mitochondria for FAO. A key unresolved question concerns the specific mechanism through which sirtuin6 regulates this transfer (114). In vascular smooth muscle cells, Plin2 similarly facilitates lipophagy through elevated LC3II levels and enhanced autophagosomes biogenesis, which impedes foam cell formation (115). Targeting lipophagy may represent a promising strategy for stabilizing vulnerable plaques (Fig. 4).

Cardiac fibroblasts exhibit region-specific distribution patterns and undergo distinct phenotypic transitions during pathological remodeling. Following acute myocardial infarction (MI) and HF, F9 (FAP/POSTN⁺) fibroblasts differentiate into the POSTN lineage through PI3K-Akt and AGE-RAGE signaling rather than adopting an ACTA2⁺ myofibroblasts phenotype (116). The transition of CD34⁺ cells to FABP4⁺ fibroblasts represents a critical fibrotic process, enhancing lipid storage via the PPARγ/Akt/Gsk3β pathway activation (117). Therapeutically targeting the NF-κB and Wnt/β-catenin/GSK3β pathways may mitigate pathological cardiac remodeling, including cardiac hypertrophy and fibrosis (118), underscoring how inflammatory signals govern fibroblast activation and differentiation. Lipotoxic intermediates, for instance, long-chain acyl-CoA, ceramide, acylcarnitine and DAG, can directly induce myocardial fibrosis (119). Hypoxia and cellular stress stimulate intracellular LD accumulation in fibroblasts. During early stage of myocardial fibrosis, cardiac fibroblasts proliferate excessively and excessive secretion of α-smooth muscle actin and alongside extracellular matrix (ECM) proteins generated in large quantities (116).

Fibroblasts-derived kinesin-like protein 23 (Kif23) activates the Ras homolog family member a pathway to enhance proliferation while suppressing LD hydrolase carboxylesterase 1d through ROCK1, thereby impairing FAO and promoting lipid accumulation in fibroblasts, and facilitating myofibroblast transformation (120). Thus, targeting Kif23 may represent a new direction for attenuating myocardial fibrosis. Nevertheless, additional investigation is required to define how LDs in fibroblasts contribute to myocardial fibrosis. Fully exploring the effects of cellular stress, inflammation, and LD accumulation on cardiac lipotoxicity represents a novel target for fibroblasts in DCM.

Abnormal cardiac LD accumulation constitutes a key feature of cardiac steatosis in diabetic patients. It is associated with a large accumulation of TAG and an enhanced lipolytic barrier in the heart after overexpression of Plin5 (64,183). This imbalance in LD dynamics accelerates ventricular remodeling and cardiac dysfunction. Studies in PPARα knockout mice have shown reduced TAG accumulation after fasting, while normal mice fasting for 16-48 h exhibit an upregulation of Plin2 expression and a concurrent increase in accumulation of LDs (184), suggesting that LDs can provide energy supply. Cardiac lipid accumulation is independent of systemic metabolism, and PPARα downregulation itself does not induce cardiac pathology (185).

During starvation, vacuolar protein sorting-associated protein 13D (VPS13D), endosomal sorting complex required for transport protein tumor susceptibility 101, along with adaptor-binding domain of VPS13, facilitate FAs transport at the LDMC site to achieve high-efficiency oxidation (186). This pathway may become dysregulated in diabetic hearts due to chronic energy excess.

LD-associated proteins form a complex regulatory network in the heart. Reduced expression levels of ATGL and CGI58 jointly impair lipolysis, which instead increases TG synthesis, promotes cardiac steatosis, and raises oxidative stress levels (65). Plin2 is involved in regulating LDL hydrolysis, and the increased TAG accumulation following Plin2 knockout may be associated with lipophagy. Overexpression of Plin2 reduces HSL levels, suggesting that energy stress during metabolic remodeling stimulates Plin2 to enhance lipase affinity for LDs, consequently exacerbating cardiac steatosis and dysfunction (53,187).

Plin5 inhibits lipolysis to preserve TAG storage within LDs, thereby limiting the rate of FAO and reducing excessive oxidative stress in heart. Upon Plin5 knockout, the upregulation of PPARα as well as PGC-1α induces mitochondrial proliferation, resulting in higher FAO and lipolysis, which can worsen cardiac hypertrophy potentially leading to HF (188). The deletion of Plin5 exerted no notable impact on heart function. During fasting, Plin5 associates with Rab8a in skeletal muscle and collaborates with ATGL, this process couples LCFAs transfer from LDs to mitochondria, indicating that Plin5 may have a cardioprotective effect through preserving mitochondrial integrity (64,189).

Myocardial steatosis and fibrosis are the basis of ventricular remodeling in DCM. Computed tomography scans of patients with MI have revealed substantial lipid deposition (190). Reduced expression of eNOS at the infarct site exacerbates abnormal myocardial lipid accumulation. This in turn promotes LD formation, increases oxidative stress, and activates the TGFβ/TβRII pathway, phosphorylating Smad2/3, inducing Smad4 to modulate transcription of ECM genes and hindering myocardial repair (190,191).

Macrophages also help regulate fibroblast activation through IL-1β and its downstream target mesenchyme homeobox 1 (116). In fact, the process of macrophage-to-myofibroblast transition (MMT) may involve LD regulation. In chronic pancreatitis, high expression of mitochondrial uncoupling protein 2 (UCP2) regulates ACSL3 to promote acinar cell release of LDs. Stimulation of SIRT1 and activation of downstream TGF-β/Smad3 signal transduction affect MMT and fibrosis process, while whether UCP2 has a similar regulatory effect still needs to be explored (192).

In this process, ATGL-dependent lipolysis promotes FA accumulation and inhibits ECM remodeling in fibroblasts (193). Low levels of CGI58 in the heart decrease the lipolysis rate and aggravate oxidative stress and myocardial steatosis (65). This indicates that the regulation of lipolysis is particular to cell type.

In addition to lipid metabolic disorders, alterations in cardiomyocyte energy metabolism are also considered to indirectly affect fibrosis progression (194). Wilms' tumor 1-associating protein promotes AR methylation in a YTHDF2-dependent way, subsequently promoting FAO proliferation, along with cardiac fibroblasts migration (195).

Direct investigations of cardiac fibroblasts and LDs are limited; however, alternative cellular models offer insights. In mouse embryonic fibroblasts, FAs generated by non-specific autophagy can be directed to LDs. This process is considered a protective regulatory technique, as LDs sequester excess FFAs to mitigate lipotoxicity (196). DGAT1 can channel FFAs generated during mTORC1-regulated autophagy to TAG-rich LDs, which then hold onto the FAs and reduce mitochondrial damage caused by acylcarnitine accumulation (160). TANGO2 is associated with LDs and the ER to modulate acyl-CoA metabolism. The cardiomyopathy and arrhythmia caused by the TANGO2 mutation may result in disturbance of LDs' morphology, function, and neutral lipid metabolism homeostasis. Consequently, lipid peroxide levels in TanGO2-deficient fibroblasts were raised metabolized, and LDs increased (197). Patients with neutral lipid storage disease with myopathy exhibit muscular, cardiac and hepatic impairment due to defective LD functions, similar to cells deficient in ATGL or CGI58, therefore providing a suitable model for investigating LD function (198). Collectively, these findings suggest that LDs may influence cardiac steatosis and myocardial fibrosis.