The smell phase is the first step to successful efferocytosis. Common forms of cell death include apoptosis, necroptosis, pyroptosis, and ferroptosis. Under steady-state conditions, cell apoptosis induced by gene-programmed suicide mechanisms is the most common (Newton et al., 2024). Unlike the pro-inflammatory response triggered by pathogen engulfment, the engulfment of ACs typically induces an anti-inflammatory response or immune tolerance (Zhang et al., 2019; Medina et al., 2020; Ampomah et al., 2022; Moon et al., 2023). Cells undergoing apoptosis and non-apoptotic cell death display and release soluble mediators, emitting various “find-me” signals to guide subsequent phagocytic processes. The forms of “find-me” signals vary, including nucleotides (ATP, UTP, AMP) (Elliott et al., 2009; Chekeni et al., 2010), membrane lipids (lysophosphatidylcholine (LPC), lysolecithin) (Lauber et al., 2003), chemokines (CX3CL1) (Truman et al., 2008). The release of nucleotides is closely related to the progression of apoptosis; the pan-connexin channel on the apoptotic cell membrane is cleaved and activated by caspase 3/7, leading to the release of ATP and UTP (Elliott et al., 2009). During the early stages of apoptosis, apoptotic cells (AC) release small amounts of nucleotides, which can serve as potent inducers for phagocyte migration towards the AC. Upon complete lysis of apoptotic cells, high concentrations of intracellular ATP are released (Mehrotra and Ravichandran, 2022). It is noteworthy that high levels of extracellular ATP can trigger a robust inflammatory response, during which other pathways are activated to promote a tolerogenic response when exposed to low doses of nucleotides for a long term (la Sala et al., 2003; Cauwels et al., 2014; Schilperoort et al., 2023). One such pathway involves the upregulation of the chloride transporter SLC12A2, which has been reported to inhibit the pro-inflammatory signaling induced by apoptotic cells (Perry et al., 2019). CX3CL1 protein belongs to indirect signaling molecules, rapidly released from ACs through caspase-3 and bcl-2-dependent regulation, and subsequently interacts with fractalkine receptors on macrophages to be attracted to apoptotic sites (Truman et al., 2008).

LPC and Sphingosine 1-phosphate (S1P) are apoptotic-specific discovery signals. Following apoptosis, activated caspase-3 mediates the activation of calcium-independent phospholipase A2, causing ACs to release “chemical attractants” such as lysolecithin (Lauber et al., 2003). S1P is produced by sphingosine via sphingosine kinase and regulates the chemotactic activity of phagocytic cells by participating in the G protein-coupled receptor signaling (Garris et al., 2013).

The actions of these soluble mediators are closely related to the types of phagocytes and ACs (Elliott et al., 2017). On one hand, they serve as attraction signals, recruiting phagocytes to migrate towards ACs. On the other hand, they stimulate the clearance potential of phagocytes, enhance the expression of cell surface phagocytic receptors, regulate cell cytoskeletons, and prepare for subsequent phagocytosis adequately (Medina et al., 2020). High efficiency in the release of find-me signals, sensitive signal perception, and the non-abnormal degradation of find-me signals are all crucial for the recognition and clearance of ACs.

The taste phase is the second step of efferocytosis. The accurate recognition and processing of ACs by phagocytes rely on two key factors: “eat me” signals on the surface of ACs and “do not eat me” signals on the surface of healthy viable cells (Kelley and Ravichandran, 2021), as well as specific phagocytic receptors on phagocytes (Arandjelovic and Ravichandran, 2015).

The most important “eat me” signal is phosphatidylserine (PS), which is evolutionarily conserved. Under normal physiological conditions, PS is usually restricted to the inner leaflet of the cell membrane by “flippases”. However, during stress or apoptosis, caspase three in ACs causes the breakdown and inactivation of flippases, leading to the activation of “scramblases”, which rapidly expose PS on the cell surface. PS is then recognized by ligands on adjacent phagocytes for engulfment (Suzuki et al., 2013; Segawa and Nagata, 2015). Phagocytes possess two types of ligand-receptor interactions for recognizing ligands and PS receptors: direct receptors and indirect receptors, which vary according to the ligand molecule involved. BAL1(Sokolowski et al., 2011), TIM(Kobayashi et al., 2007; Miyanishi et al., 2007; DeKruyff et al., 2010), and Stabilin2 (Lee et al., 2011) can directly bind to PS, while the receptor tyrosine kinase family TAM Receptors (Tyro3, Axl, and Mertk) (Lemke, 2017; Burstyn-Cohen and Fresia, 2023) bind to PS indirectly through soluble bridging molecules such as Gas6 and Protein S. PS, a lipid membrane component, enhances ligand binding affinity and facilitates TAM receptor-mediated signaling (Nakano et al., 1997; Graham et al., 2014; Zagórska et al., 2014; Kasikara et al., 2017). This effect is achieved through the calcium-dependent binding of PS to the Gla domain of either Gas6 or Protein S (Nelsestuen et al., 1978; Nakano et al., 1997). In this mechanism, Gas6 and Protein S function as bridging ligands that connect PS with the TAM receptor. The inclusion of PS-containing lipid membranes alongside TAM receptor ligands results in elevated receptor phosphorylation relative to the application of ligand alone (Zagórska et al., 2014; Kasikara et al., 2017). Such bridging-mediated signaling plays a critical role in the phagocytic clearance of apoptotic cells that externalize PS. In addition to PS, common “eat-me” signals on ACs include membrane-bound I proteins (Arur et al., 2003) and calcium-binding proteins (Gardai et al., 2005). Studies have shown that aside from serving as “find me” signal molecules, LPC on the membrane of dying cells can also interact with receptors on phagocytes (Kim et al., 2002). It is worth mentioning that MerTK and AXL play a vital role in anti-inflammation response and establishing tolerance to “self” antigens (Lemke and Rothlin, 2008). During the period of efferocytosis, MerTK functions to reduce inflammatory cytokine production and M1-like macrophage polarization through the inhibition of the NF-κB pathway (Tibrewal et al., 2008). Beyond suppressing the M1-like phenotype, MerTK also ​promotes​ macrophage ​differentiation​ into the alternative, reparative M2-like phenotype (Filardy et al., 2010; Zizzo et al., 2012). The expression and activation of MerTK promotes T cell tolerance by inhibiting the maturation of dendritic cells and the expansion of effector T cells (Wallet et al., 2008).

Healthy neighboring cells express “do not eat me” signals on their surface, which help avoid the unnecessary clearance of viable “healthy” cells (Kelley and Ravichandran, 2021). The cell surface proteins CD47 and CD24 are two classic signals that are recognized by the receptors SIRPα on macrophages and SIGLEC-10, respectively (Oldenborg et al., 2000; Barkal et al., 2019). CD47 is a crucial “do not eat me” molecule. However, upregulation of CD47 expression on diseased cells can lead to defects in phagocytosis and the occurrence of inflammation (Kojima et al., 2016).

The third step of efferocytosis involves the ingestion and degradation of dying cells by engulfing cells, which specifically includes the following steps: Ingestion of dying cells, degradation by lysosomes, maturation of phagosomes, and phagosome dissolution (Boada-Romero et al., 2020). The ingestion of dying cells occurs after engulfing cells recognize the dying cells, leading to the internalization of the dying cells through downstream signaling of AC engulfment mediated by engulfment receptors, forming “phagosomes (Levin et al., 2016). Actin polymerization is a critical component of phagosome formation and efficient capture of dead cells, while actin depolymerization is equally important for the detachment of phagosomes from the plasma membrane. The cargo ingested by phagocytes poses a metabolic burden, including amino acids, lipids, and nucleic acids. These metabolic byproducts expose phagocytes to a plethora of potentially cytotoxic macromolecules, necessitating their safe and efficient handling (Han and Ravichandran, 2011). As a mechanism for handling excess lipids within macrophages, the cholesterol transporter ATP-binding cassette transporter A1 (ABCA1) is upregulated following the detection of apoptotic cells during the feeding phase, facilitating the efflux of cholesterol to apolipoprotein A1 (Fond et al., 2015; Viaud et al., 2018). Phagosomes fuse with lysosomes to form “phagolysosomes”, which contain various proteases, lipases, and nucleases to digest the cargo of phagosomes to maintain homeostasis (Levin et al., 2016; Boada-Romero et al., 2020). In terms of energy metabolism, glycolysis within phagocytes facilitates actin polymerization and sustained uptake of dying cells. Sustained exposure to apoptotic cellular debris, phagocytes can efficiently undergo serial efferocytic processing to optimize efferocytosis, particularly crucial in tissues with high cell turnover (Trzeciak et al., 2021). Metabolites derived from AC metabolism in macrophages can achieve optimal sustained efferocytosis and resolution of injury (Yurdagul et al., 2020). Typical phagocytic pathways include the maturation from early to late phagosomes, ultimately fusing with lysosomal compartments. There is also an additional pathway involving microtubule-associated protein 1A/1B light chain 3 (LC3)-associated phagocytosis (Green et al., 2016). LC3-associated phagocytosis promotes the assembly and acidification of phagolysosomes (Martinez et al., 2011).

Upon completion of mature modifications, phagolysosomes begin to degrade AC remnants and release anti-inflammatory cytokines such as IL-10 (Xu et al., 2006), TGF-β (Fadok et al., 1998; Freire-de-Lima et al., 2006), and specialized pro-resolving lipid mediators (SPMs), including lipoxin A4, resolvin D1, D2, and E2 (Basil and Levy, 2016); simultaneously reducing the expression of pro-inflammatory prostaglandins and leukotriene B4, Lipoxin A4 and resolvin D1 further enhance phagocytosis, establishing a feedback loop that increases their own production (Dalli and Serhan, 2016; Kourtzelis et al., 2020). Phagocytosis promotes a pro-resolving phenotype by downregulating the expression of pro-inflammatory cytokines (Godson et al., 2000), increasing the levels of pro-resolving mediators, inhibiting inducible nitric oxide synthase (iNOS), and enhancing the production of vascular endothelial growth factor (VEGF) (Doran et al., 2020). Macrophages polarized towards a pro-resolving ARG1+ phenotype, which are more efficient phagocytes compared with pro-inflammatory macrophages (Yurdagul et al., 2020).

Recent studies utilizing murine models of T2DM have demonstrated defective efferocytosis in peritoneal macrophages, bone marrow-derived macrophages, and wound macrophages (Li et al., 2009; Khanna et al., 2010; Suresh Babu et al., 2016). In individuals with diabetes, macrophages undergo functional alterations characterized by heightened responsiveness to inflammatory stimuli and enhanced secretion of pro-inflammatory mediators. Concurrently, their phagocytic capacity for pathogens and efferocytic clearance of apoptotic cells are impaired. These deficits compromise pathogen elimination and, due to the role of efferocytosis as an inducer of phenotypic switching, result in a decreased population of M2-like reparative macrophages (Mao et al., 2025).

Diabetic macrovascular complications include atherosclerosis of large arteries and medium arteries, which is a common pathological feature of vascular disease in diabetic patients (Li Y. et al., 2023). Atherosclerotic cardiovascular disease (CVD) is the leading cause of death in adult diabetic patients (Kanter et al., 2020). High blood sugar is a significant risk factor for atherosclerosis, promoting endothelial dysfunction, and is considered an early event in the development of atherosclerotic lesions (Demir et al., 2021). Atherosclerosis is understood as a cholesterol storage disease and a lipid-driven inflammatory condition (Bäck et al., 2019). Hyperglycemia not only induces oxidative stress and modifies LDL-C into oxLDL form (Evans et al., 2002), but also promotes macrophage epigenetic and metabolic reprogramming, accelerating the inflammatory response (Russo et al., 2021). A study in mice demonstrated that the absence of low-density lipoprotein receptors leads to impaired phagocytosis of lesions and a larger necrotic core compared to healthy animals (Li et al., 2009). In the initial stage of atherosclerotic lesion development, circulating monocytes adhere to endothelial cells and enter the vessel wall, then differentiate into macrophages. These macrophages absorb excess oxLDL via scavenger receptors and become lipid-laden foam cells (Rawal et al., 2024), which must be cleared via efferocytosis. Inefficient clearance of dead cells is a major factor in the progression of atherosclerosis (Boada-Romero et al., 2020).

Defective efferocytosis is commonly considered the cause of improper clearance of ACs. Defective efferocytosis contributes to plaque expansion, plaque rupture, acute coronary syndrome, and stroke (Vucic et al., 2012; Poon et al., 2014). The NLRP3 inflammasome regulates vascular inflammation and atherosclerosis by modulating efferocytosis (Rochette et al., 2023). The NLRP3 inflammasome is a multimolecular complex that functions at the molecular level to mediate caspase-1 activation. Upon stimulation by various pathogen-associated molecular patterns (PAMPs) or DAMPs, it induces the release of interleukin-1β (IL-1β) and interleukin-18 (IL-18) within cells (Duewell et al., 2010). During the early pro-inflammatory stage of atherosclerosis, elevated TNF-α hinders the timely clearance of ACs, weakening efferocytic activity (Martinet et al., 2011). Additionally, mice lacking TIM-4, Mertk, MFGE8, or Pros exhibit failed AC clearance (Ait-Oufella et al., 2007; Cai et al., 2017), and increased inflammation has been observed (Kimani et al., 2014). EC is a unique barrier to inflammation (Shao et al., 2020). High levels of PCSK9 expression are observed in vascular cells, comprising endothelial cells, smooth muscle cells, and macrophages (Ding et al., 2015a; Ding et al., 2015b; Ding et al., 2020). Research has found that proprotein convertase subtilisin/kexin type 9 (PCSK9) levels increase in aortic endothelial cells during aging, which is associated with the loss of the phagocytic receptor MerTK. MerTK serves as a critical receptor for efferocytosis and is highly expressed in endothelial cells (ECs), indicating their potential capability to mediate efferocytosis (Li et al., 2019). Furthermore, the excessive production of reactive oxygen species (ROS) and chronic inflammation, characteristic of vascular aging, is alleviated in the aortas of PCSK9-deficient mice (Liu et al., 2023). Additionally, considering that PCSK9 promotes the clearance of oxidized LDL, subsequently increasing the synthesis of LDL receptors, the use of PCSK9 inhibitors can lower LDL concentrations in patients with diabetes (Chen et al., 2023). The factors exacerbate the progression of diabetic macrovascular complications by affecting different factors in the efferocytosis process.

Diabetic microvascular disease is caused by microvascular lesions in arterioles, capillaries, and venules. Microvascular disease can cause pathological and functional changes in many tissues, which are traditionally referred to as diabetic retinopathy (DR), diabetic kidney disease (DKD), diabetic cardiomyopathy (DCM), peripheral neuropathy, and autonomic neuropathy (Lu et al., 2023; Wang et al., 2023). It is noteworthy that DKD is the leading cause of end-stage renal disease, while DR is one of the major causes of blindness among various eye diseases. Unfortunately, due to the complex pathogenesis of these two diseases, effective treatment options are still lacking. Multiple studies have indicated that hyperglycemic states, inflammatory reactions, and macrophages play crucial roles in the development and progression of DKD, DR and DCM. Therefore, we have focused our review on the related mechanisms of inflammation and efferocytosis in diabetic microvascular disease, and look forward to future research and treatment prospects (Li Y. et al., 2023).

Several commonly used medications can enhance the efferocytosis of macrophages. Defective efferocytosis is considered a major risk factor for atherosclerosis, and enhancing efferocytosis can help reduce atherosclerotic lesions. Recent investigations have revealed that CD47, a critical ‘don’t-eat-me' signaling molecule, is paradoxically overexpressed across a spectrum of cancers (Majeti et al., 2009; Chao et al., 2011; Willingham et al., 2012). This upregulation enables malignant cells to evade established immune surveillance mechanisms, including elimination by tumoricidal macrophages, and is now considered a key mechanism underlying tumor progression (Perisic et al., 2013; Kojima et al., 2019). Studies suggest that statins increase macrophage “eat me” signals and induce a reduction in the “do not eat me” signal CD47 (N. Eberhardt, 2022; Tajbakhsh et al., 2022); On the other hand, statins can lower cholesterol and inflammation, inhibiting the overexpressed Ras homolog gene family member A, which is a negative regulator of phagocytosis in atherosclerotic lesions (Salman et al., 2008). The mineralocorticoid receptor (MR) is considered a potential target for the treatment of atherosclerosis. MR deficiency in apoE^−/− mice reduced the number of ACs in atherosclerotic lesions and increased effective efferocytosis, suggesting that MR antagonists like spironolactone could improve macrophage efferocytosis (Xie et al., 2022). The impairment of the p38 MAPK signaling axis is considered a fundamental mechanism underlying the reduced expression of TIM-4 in mononuclear phagocytes of the elderly, leading to compromised cytoplasmic function and inflammation resolution. To address this damage, the p38 inhibitor losmapimod is currently undergoing clinical trials. Losmapimod blocks p38 and restores TIM-4 expression and efferocytic activity in elderly mononuclear phagocytes under in vitro conditions. Elderly patients receiving oral losmapimod show improved resolution of skin inflammation, with a reduction in unresolved polymorphonuclear leukocytes (PMNs) at the inflammatory site and an increase in TIM-4 expression on mononuclear phagocytes. Currently, several small-molecule p38 MAPK inhibitors have completed Phase I clinical trials, demonstrating good safety profiles (De Maeyer et al., 2020). Losmapimod is also undergoing Phase III clinical trials for facioscapulohumeral muscular dystrophy (NCT05397470) (Anand et al., 2011; Patnaik et al., 2016).

Enhanced efferocytosis can reverse hypoxia in murine atherosclerosis, prevent necrotic core expansion (Marsch et al., 2014), and thus treat macrovascular complications of diabetes. Reducing the “do not eat me” signal is a new strategy for atherosclerosis treatment, with some drugs already researched or approved for clearing ACs, such as CD47 antibodies and PD1-/PD-L1 antibodies (Chen et al., 2021). A wealth of data has identified the TAM RTK family as a promising therapeutic target due to its multiple roles in cancer cell survival, metastasis, treatment resistance, and anti-tumor immune suppression. TAM receptor tyrosine kinases play an indispensable role in efferocytosis, the process in which apoptotic cells are engulfed and cleared by macrophages and other phagocytic cells (Scott et al., 2001; Seitz et al., 2007). This prospect has driven the development of various TAM-targeted agents for translational applications, many of which have progressed to clinical testing. Antibody-drug conjugates and decoy receptor fusion proteins have shown significant efficacy in enhancing efferocytosis in tumor cells (DeRyckere et al., 2023), and advancements in enhancing efferocytosis in cancer treatment have provided new directions for research into the treatment of diabetes and its complications.

Nanoparticles are nanoscale particulate structures that exhibit distinctive physicochemical and biological characteristics, enabling their application as versatile diagnostic agents and therapeutic modalities (Batista et al., 2015). Some researchers have used various regulatory mediators and agonists to enhance efferocytosis and promote the resolution of inflammation, which has been confirmed in many animal models. Nanoparticles can conveniently package various molecules that alter macrophage function, including siRNA, mRNA, and small molecule inhibitors, thus nanotechnology platforms have advantages in improving macrophage targeting at the lesion site over other technologies (Chen et al., 2022). For example, macrophage-targeted nanoparticles carrying siRNA against Ca2⁺/calmodulin-dependent protein kinase γ (CaMKIIγ), a plaque instability protein activated in late-stage human and mouse plaque macrophages, can mitigate necrotic core expansion by enhancing MerTK-mediated efferocytosis. Similarly, macrophage-targeted single-walled carbon nanotubes loaded with CD47-SIRPα signaling inhibitors can enhance efferocytosis in lesion macrophages and reduce atherosclerosis (Flores et al., 2020). ​Nanoparticles targeting processes associated with macrophage foam cells have been created, such as those that enhance reverse cholesterol transport or suppress LDL oxidation to limit foam cell development (Lewis et al., 2016).​The widespread acceptance of COVID mRNA vaccines has paved the way for exploring mRNA-based therapies. In diabetic mice, the levels of miR-126 are suppressed. A rational intervention for such mice could involve delivering miR-202 to areas burdened with apoptotic debris, such as diabetic wounds, using chitosan modified with nanoparticles or REDV peptides (Zhou et al., 2016). Although nanotechnology has demonstrated promising targeted therapeutic effects and partially mitigates systemic adverse reactions associated with conventional drug administration (Batista et al., 2015), current nanoparticle formulations still lack sufficient selectivity for target cells and intended organs (Smith and Gambhir, 2017; Flores et al., 2019). Furthermore, the clinical safety profile of these nanoscale drug delivery systems requires comprehensive evaluation. These inherent limitations explain why clinically approved nanotherapies remain relatively scarce to date. Nevertheless, this field continues to show substantial potential for advancing precision medicine through continued research and technological refinement.