Appropriate regulation of inflammation and its resolution is critical for host defense and maintaining tissue homeostasis. Communication between the innate arm of immunity, particularly myeloid cells, notably macrophages and dendritic cells, and the adaptive arm of immunity, notably T-lymphocytes, is essential. Dysregulation of myeloid-T-lymphocyte crosstalk can contribute to the pathogenesis and exacerbation of chronic inflammatory diseases such as atherosclerosis (1). Atherosclerosis is characterized by the formation of a fibro-fatty plaque in the arterial wall consisting of a necrotic core made up of deposited lipids and dead cells, immune cells including myeloid cells and T-lymphocytes, and a fibrous cap overlaying the necrotic core (2). Atherosclerosis progression is driven by elevated circulating LDL levels that cause increased lipid deposition and inflammatory cell infiltration into the plaque (2). Atherosclerosis regression on the other hand is a clinically relevant phenomenon that occurs with marked lowering of circulating LDL, which promotes inflammation resolution, reduced necrotic core size, and thickening/strengthening of the fibrous cap (3).

Macrophages and dendritic cells play key roles in atherosclerosis progression and regression. One critical role is related to the process by which apoptotic cells (ACs) are cleared by phagocytic cells such as macrophages and dendritic cells, called efferocytosis (4, 5). Efferocytosis prevents secondary necrosis of ACs and subsequent inflammation and contributes to the production of pro-resolving mediators, which can feedback to further enhance efferocytosis (efferocytosis-resolution cycle) (4, 5). As atherosclerosis progresses, the efferocytosis-resolution cycle becomes defective, characterized by increased inflammation, enhanced necrosis, and thinning of the protective fibrous cap of lesions (4, 5). In humans, these changes lead to plaques that are “unstable”, meaning they increase the risk for plaque rupture or erosion, acute luminal thrombosis, and major adverse cardiovascular events, notably myocardial infarction, stroke, and sudden cardiac death (4, 5). On the other hand, the efferocytosis-resolution cycle is reawakened during atherosclerosis regression, a process in which marked lipid lowering leads to atherosclerotic plaque stabilization through necrotic core reduction and fibrous cap thickening and, in humans, lower the risk of coronary artery disease (6–8).

There is also substantial evidence indicating that T-cells from the adaptive immune system play important roles in atherosclerosis pathogenesis (9–12). Under homeostatic conditions and during an immune response, myeloid cells and T-cells communicate with each other (13–15), with macrophages and dendritic cells serving as a critical bridge between the innate and adaptive immune systems through antigen presentation, secreted mediators, and cell-cell contact (13–15). T-cells can also feedback through secreted factors and cell-cell contact to regulate myeloid cell functions (16, 17).

While the roles of myeloid cells in atherosclerosis are varied and complex, we focus here on the key issue of how efferocytosis by lesional macrophages and dendritic cells can affect both plaque progression and regression through crosstalk with T-cells. Crosstalk between efferocytic myeloid cells and T-cells can promote a pro-resolving milieu that dampens chronic inflammation and promotes plaque stabilization, while failure of this process can result in the exacerbation of atherosclerosis and attenuation of regression. In this review, we will explore three critical aspects of this interaction and how it may affect atherosclerosis progression and regression: 1) the role of antigen cross-presentation by myeloid cells of AC-derived antigens during efferocytosis and its effect on T-cells; 2) the role of secreted cytokines, such as transforming growth factor-β1 (TGF-β1) and interleukin-10 (IL-10), and metabolites, such as lactate and polyamines in the crosstalk between efferocytic myeloid cells and T-cells; and 3) the role of cell-cell contacts through membrane proteins, including the immune checkpoint proteins programmed death ligand 1 (PD-L1), programmed cell death protein 1 (PD-1), and cytotoxic T-lymphocyte associated protein 4 (CTLA-4) in regulating efferocyte-T-cell crosstalk.

Myeloid cells are adept at a process called antigen presentation. Proteins taken up by myeloid cells through efferocytosis and other endocytic pathways are degraded, and peptides (antigens) are loaded onto major histocompatibility complex (MHC) molecules to present to adaptive immune cells such as T-cells. The recognition of presented antigens by T-cells through the engagement of the T-cell receptor (TCR) can either trigger immune activation towards the antigen or induce tolerance toward self-antigens (18, 19). Disruption of tolerance mechanisms can result in aberrant presentation of self-antigens to auto-reactive T-cells and a lack of restraint on T-cell activation, leading to autoimmunity. Efferocytosis helps facilitate tolerance by preventing self-antigens derived from ACs from triggering an autoimmune response and chronic inflammation (20). Accordingly, an impairment in efferocytosis, which is observed in advanced atherosclerosis, can result in an accumulation of uncleared self-antigen thus inducing autoimmunity (4, 21). In late atherosclerosis, defective efferocytosis can be attributed in part to the loss of cell surface efferocytic receptors, e.g., through the cleavage of the efferocytosis receptor, MER proto-oncogene tyrosine kinase (MerTK) (22). A direct example of autoimmunity triggered by impaired efferocytosis comes from a study showing that genetically deleting the efferocytic receptor CD300f in mice enhanced self-antigen presentation by dendritic cells, contributing to memory T-cell expansion and autoimmunity (23). Loss of tolerance due to impaired efferocytosis is likely relevant to atherosclerosis, which displays hallmarks of autoimmunity such as anti-apolipoprotein B autoantibodies and CD4⁺ T-cells (24, 25). This topic will be discussed in more detail in the following sections.

Antigen presentation is initiated when internalized proteins are degraded by lysosomes or by the proteasome and loaded onto MHC class I or II molecules and presented to a corresponding T-cell receptor (TCR). The proliferation and activation of T-cells can be initiated when TCRs on a CD4⁺ or CD8⁺ T-cell surpass the threshold binding affinity of a peptide presented on MHC class II or class I, respectively (26, 27). During efferocytosis, AC degradation generates self-antigens, but efferocytosing myeloid cells can restrain autoimmunity by suppressing the presentation of these antigens. The mechanism involves the secretion of pro-resolving mediators by efferocytes that (a) lower co-stimulatory signals needed for full activation of T-cells, e.g., those that mediate T-cell CD28 interaction with myeloid CD80/86; and (b) promote the formation of self-antigen-specific forkhead box P3 (FoxP3)⁺ regulatory T-cells (T_reg cells), which suppress auto-reactive effector T-cells, dampen inflammation, and promote tissue repair (28–30). In atherosclerosis, FoxP3⁺ T_reg cells are athero-protective (31) and are critical to atherosclerosis regression (32). In dendritic cells, AC-binding and efferocytosis mediated by MerTK triggers the suppression of nuclear factor kappa B (NF_κB) signaling to prevent the full maturation of these cells (33). This process results in low expression of MHC, co-stimulatory molecules like CD80, and the pro-inflammatory cytokine tumor necrosis factor-alpha (TNF-α), all of which inhibit the presentation of AC-derived self-antigens and prevent the activation of auto-reactive T-cells (33). Pro-resolving mediators from efferocytes, such as TGF-β1 and IL-10, can induce T_reg cells, which feedback to suppress CD80/86 and stop effector T-cell activity (34).

In atherosclerosis, antigen-presentation by myeloid cells to T-cells contributes to chronic inflammation by inducing effector CD4⁺ T-cells and the release of pro-inflammatory cytokines, e.g., IFN-γ and TNF-α, by activated T-cells (35). These cytokines in turn enhance the uptake of oxidized LDL (oxLDL) and minimally modified LDL by lesional macrophages, which promotes disease progression (36). Furthermore, pro-inflammatory cytokines such as IFN-γ and TNF-α are known to augment antigen presentation by myeloid cells, thus further activating pro-atherogenic effector T-cells (37, 38). Efferocytosis still occurs in inflammatory progressing atherosclerotic lesions, albeit reduced and with impaired induction of pro-resolution signaling. Thus, it is possible that in atherosclerosis progression, the suppression of antigen presentation and co-stimulatory molecules by efferocytes is impaired. For example, although antigenic peptides generated during efferocytosis can bypass being loaded onto MHC class II by the activity of Ras-related protein Rab-17 (Rab17) to avoid antigen presentation and activation of auto-reactive CD4⁺ T-cells (39), stimulation of toll-like receptors (TLRs) such as TLR4 can re-direct antigens to be loaded onto MHC molecules for antigen presentation (40–42). Thus, oxLDL, which accumulates in atherosclerotic plaques, can upregulate TLR4 expression and signal through TLR4 (43, 44), and other pro-inflammatory signals in atherosclerosis may enhance the presentation of self-peptide by efferocytosing myeloid cells to T-cells. Furthermore, chronic inflammation in atherosclerosis promotes the maturation of dendritic cells and expression of co-stimulatory molecules to enhance activation of self-reactive T-cells (45). Interestingly, MHC class II whole-body knockout in ApoE^-/- mice, designed to inhibit antigen presentation to CD4⁺ T-cells, worsened atherosclerosis despite reduced total CD4⁺ T-cells and circulating pro-inflammatory cytokines (46). This result may be related to the finding that antigen presentation by MHC class II of plaque antigens to T_reg cells is critical for immunosuppression (47), suggesting that a loss of T_reg cell activity is more consequential than reduced antigen presentation to effector CD4⁺ T-cells in atherosclerosis progression (46). Because these experiments were performed in whole-body MHC class II knockout mice and most cells can present antigen on MHC class II, it would be interesting to observe the effect on atherosclerosis when MHC class II is knocked out solely in myeloid antigen-presenting cells (46). In this context, mice with dendritic cell-specific knockout of the TLR signaling adaptor, myeloid differentiation primary response 88 (MyD88), attenuated dendritic cell maturation, antigen presentation to T_reg cells, T_reg cell maturation, and most importantly, worsened atherosclerosis (48). Thus, in this model of atherosclerosis, defective T_reg cell development trumped the benefits of reducing effector T-cells (48).

As mentioned above, both efferocytes and activated T-lymphocytes can produce a variety of cytokines and lipid-derived specialized pro-resolving mediators (SPMs). In addition, efferocytes release metabolites derived from the degradation of ACs and from shifts in cellular metabolism, and these molecules could have immunomodulatory effects on both the efferocyte and T-cells. The role of secreted molecules in efferocytosing myeloid cell-lymphocyte crosstalk is summarized in Figure 1 .

Macrophages can affect T cells through cell-cell contact involving interaction between the checkpoint proteins on the two cell types, notably PD-L1 and PD-1, and cytotoxic T-lymphocyte associated protein 4 (CTLA-4). The role of cell-cell contacts in efferocytosing myeloid cell-lymphocyte crosstalk is summarized in Figure 2 . PD-L1 is expressed more ubiquitously including on macrophages and dendritic cells, whereas PD-1 is primarily expressed on T-cells (116). Interaction between PD-L1 and PD-1 can suppress the activation of effector CD4⁺ and CD8⁺ T-cells while promoting the differentiation of naïve CD4⁺ T-cells to FoxP3⁺ T_reg cells and sustaining T_reg cell activity, in part by upregulating and sustaining FoxP3 expression (16, 117). Through MerTK-p38-STAT3 signaling, macrophages upregulate PD-L1 upon binding ACs (118). Moreover, PD-1 expression on FoxP3⁺ T_reg cells and cytotoxic CD8⁺ T-cells, and PD-L1 expression on macrophages can be upregulated by secreted factors from efferocytosing macrophages such as lactate (119), TGF-β1 (120), and tryptophan-derived metabolites (91).

There is also evidence that efferocytosis by myeloid cells may also induce the expression of another immune checkpoint protein, CTLA-4, on T_reg cells. CTLA-4 is a cell surface protein that is expressed principally on T-cells, but it can also be expressed by other immune and non-immune cell types such as myeloid cells and tumor cells. CTLA-4 expression is induced and sustained on T_reg cells and memory T-cells following the interaction of the T-cell receptor (TCR) and CD28 on activated T cells with the MHC-peptide complex and CD80/86 on antigen-presenting cells, such as macrophages and dendritic cells (121). CTLA-4 suppresses signaling downstream of the TCR to shut off T-cell proliferation and the T-cell response in effector T-cells (122, 123). On T_reg cells, CTLA-4 expression is critical to their tolerogenic and immunosuppressive function in several ways (124–126). Firstly, CTLA-4 can compete with CD28 for binding to CD80/86 on antigen-presenting cells to reduce T-cell activity (127). Secondly, CTLA-4 can mediate “transendocytosis” of CD80/CD86, which is the process by which T_reg cells engulf and remove CD80/86 from the antigen-presenting cell (125, 128, 129). This degradation can prevent CD28 co-stimulation as well as disrupt cis-CD80-PD-L1 dimer interactions on the antigen presenting cell (125, 128, 129). CD80 dimerization with PD-L1 prevents PD-1 on other cell types such as T-cells from interacting with PD-L1 to induce tolerogenic signaling, and thus disruption of the dimers allows for restoration of PD-L1-PD-1 signaling (125). Thirdly, administration of CTLA-4-Ig, which mimics CTLA-4 binding and its interaction with macrophage CD80/86, can suppress macrophage pro-inflammatory cytokine production and promote phenotypic switching to a pro-resolving phenotype (130, 131). Although not directly shown using T_reg cells, this may be a potential mechanism for T_reg cell-mediated inflammation resolution. Finally, T_reg cell CTLA-4 can induce dendritic cells to express IDO1 and increase tryptophan catabolism in a CD80/86-dependent manner (132). This increase in IDO1 (a) enhances the secretion of kynurenine and other tryptophan-derived metabolites by dendritic cells to modulate efferocytosing myeloid cells and T-cells as described previously; and (b) suppresses antigen presentation by dendritic cells (132). In the context of efferocytosis, CD169⁺ splenic marginal zone macrophages upregulated the secretion of CCL22, the chemotactic ligand for CCR4, following the uptake of apoptotic thymocytes in vitro (133). In vivo, injection of ACs into mice upregulated CCL22 secretion by CD169⁺ splenic macrophages, and T_reg cells were found to migrate into the spleen and increase CTLA-4 expression in a CCR4-dependent manner (133). These findings suggest that CCL22-CCR4 signaling or signaling by another secreted factor produced by efferocytosing macrophages, can increase CTLA-4 expression on T_reg cells. Moreover, CTLA-4 can be upregulated in T cells, including T_reg cells, by pro-resolving lactate and TGF-β (120, 134), but more work is needed to understand how CTLA-4 expression is affected by efferocytosing myeloid cells and how T-cell CTLA-4 signals back to efferocytosing myeloid cells.

In murine models of atherosclerosis progression, whole-body knockouts of either PD-L1 or PD-1 resulted in an exacerbation of atherosclerosis progression as measured by plaque area (135, 136). This corresponded with an increase in the number of macrophages, CD4⁺ T-cells, and CD8⁺ T-cells (135, 136). In vitro, myeloid cells from PD-L1 KO mice could more readily activate T-cells following antigen presentation, and T-cells from PD-1 KO mice were more inflammatory and cytotoxic in the case of CD8⁺ T-cells (135, 136). Interestingly, hypercholesterolemia increased PD-L1 expression on myeloid cells and increased PD-1 expression on T-cells, suggesting a possible compensatory response in which PD-L1-PD-1 interaction restrains the inflammatory response during atherosclerosis progression (135, 136). In this context, cancer patients who receive blocking antibodies targeting PD-L1 or PD-1 have an aggravation of atherosclerotic cardiovascular disease (137).

CTLA-4 was also found to have an anti-atherogenic effect in mice and humans (138–141). CTLA-4 blocking antibodies, similar to PD-L1/PD-1 blocking antibodies, are used to treat cancer, which is associated with an increased risk of coronary artery disease (138, 141). In Western diet-fed Ldlr^-/- mice, CTLA-4 blockade increased effector CD4⁺ and CD8⁺ T-cells in the blood, spleen, and plaque, increased plaque size, and decreased features of plaque stability based on increased necrotic core size and thinner fibrous caps (139). Conversely, transgenic T-cell overexpression of CTLA-4 resulted in an attenuation of atherosclerotic plaque size in the aortic root, with a reduced accumulation of macrophages and effector T-cells in the plaque (140). Furthermore, dendritic cell maturation was inhibited as determined by reduced CD80/86 expression, although this could at least in part be a result of enhanced transendocytosis of CD80/86 by T_reg cells (140).

Both myeloid cells and T-cells play important roles in affecting the plaque microenvironment in atherosclerosis progression and regression by modulating inflammation and other immune processes. Significant crosstalk occurs between myeloid cells and T-cells to influence the phenotype and behavior of both groups of immune cells. Efferocytosis by myeloid cells is a critical component of inflammation resolution and tissue repair, affecting the metabolism, secretome, and gene expression of macrophages and dendritic cells. These changes in efferocytes can lead to altered interactions between myeloid cells and T-cells and phenotypic alterations in T-cells. T-cells can also feedback to regulate efferocytes, thus contributing to the perpetuation or disruption of the efferocytosis-resolution cycle. In atherosclerosis, impaired efferocytosis exacerbates inflammation and increased plaque destabilization, whereas improved efferocytosis in atherosclerosis regression reversed these changes. Although the pro-tolerogenic function of myeloid cell efferocytosis and its associated downstream signals, e.g., secreted lactate or kynurenine, are beneficial in the context of chronic inflammation and atherosclerosis progression, this process can also contribute to immunosuppression in cancer, leading to cancer cell evasion of the immune system (142–145). Thus, conceptualizing methods to target the crosstalk between efferocytosing myeloid cells and T-cells for the treatment of atherosclerosis must also consider that excessive immunosuppression may result in impaired anti-tumor immunity. Thus, a deeper understanding of efferocyte-T-cell crosstalk is crucial to better understand how to achieve a balance between pro-inflammatory signaling and pro-resolution signaling. Future analysis of publicly available datasets or using tools such as CellChat to analyse cell-cell interactions could be used to identify candidate signaling pathways to target to better achieve the balance (146).

In this review, we introduce mechanisms by which efferocyte-T-cell crosstalk through antigen presentation, secreted factors, and cell-cell contact by ligand-receptor interactions could contribute to the benefits of efferocytosis on inflammation resolution and plaque stability. Much remains to be discovered about this interplay, and further work in this area has the potential to provide new insights into the pathogenesis of atherosclerosis and to suggest novel therapeutic targets.

DN: Conceptualization, Investigation, Writing – original draft, Writing – review & editing. SS: Investigation, Writing – review & editing. IT: Supervision, Writing – review & editing.