Functional Crosstalk Between T Cells and Monocytes in Cancer and Atherosclerosis

Abstract

Monocytes and monocyte-derived cells, including macrophages and dendritic cells, exhibit a diverse array of phenotypic states that are dictated by their surrounding microenvironment. These cells provide cues ranging from immunosuppressive to immunostimulatory cues to T cells, directing their activation and function. Solid tumors and atherosclerotic plaques represent two pathological niches with distinct immune microenvironments. While monocytes and their progeny possess a phenotypic spectrum found within both disease contexts, most within tumors are pro-tumoral and support evasion of host immune responses by tumor cells. In contrast, monocyte-derived cells within atherosclerotic plaques are usually pro-atherogenic, pro-inflammatory, and predominantly directed against self-antigens. Consequently, cancer immunotherapies strive to enhance the immune response against tumor antigens, whereas atherosclerosis treatments seek to dampen the immune response against lipid antigens. Insights into monocyte-T cell interactions within these niches could thus inform therapeutic strategies for two immunologically-distinct diseases. Here, we review monocyte diversity, interactions between monocytes and T cells within tumor and plaque microenvironments, how certain therapies have leveraged these interactions, and novel strategies to assay such associations.

Graphical Abstract

Summary Sentence:

This review summarizes our current understanding of monocyte heterogeneity and how monocytes differentially influence T cell responses within the tumor and plaque microenvironments.

Introduction

Monocytes and monocyte-derived cells, including macrophages and dendritic cells (DCs), display a wide range of phenotypic and functional diversity. This plasticity makes myeloid cells particularly responsive to their surrounding microenvironments and ideal for integrating and relaying complex signals [1]. Indeed, one of the primary functions of these cells is to instruct adaptive immune responses, with cues ranging from immunostimulatory to immunosuppressive. Here, we examine interactions between monocytes and T cells in two distinct pathologic niches that broadly represent opposite ends of the spectrum of immune activation: atherosclerotic plaques and tumors. Further, cancer and cardiovascular disease (CVD) are the two leading causes of mortality worldwide [2,3], highlighting the importance of understanding the immunobiology of these disease states.

In solid tumor cancers, mutations and epigenetic reprogramming dysregulate the expression of genes controlling normal cell growth, resulting in unrestrained cellular expansion. This is followed by malignant transformation and eventual metastasis of these cells to distal sites. Monocytes contribute to anti-tumoral immunity by recognizing and infiltrating established tumors, as well as preventing tumor metastasis [4–7]. However, cues from the tumor microenvironment can convert monocytes and their progeny into macrophages that aid tumor cells in evading cytotoxic T cells [8]. Increased numbers of CD8⁺ T cells in tumors are associated with positive outcomes in cancer [9]. Nevertheless, the events controlling monocyte recruitment and modulation of T cells within the tumor immune microenvironment (TIME) are still under investigation.

Conversely, the initiation of atherosclerotic plaque formation that frequently underlies CVD is characterized by deposition of lipoproteins within the vascular intima of arterial walls [10]. Subsequently, these lipoproteins undergo oxidation in the subendothelial space, which recruits inflammatory monocytes, macrophages, and DCs into the plaque via damage-associated molecular patterns [11,12]. Monocytes are one of the first immune cells recruited to the atherosclerotic plaque and they differentiate into either monocyte-derived macrophages or DCs once residing in this environment. In response to the inflammatory signaling initiated by monocyte-derived cells, CD4⁺ and CD8⁺ T cells infiltrate the plaque, where they further agonize plaque destabilization and increase the risk of myocardial infarction [12].

Monocytes are challenging to isolate from their macrophage/DC progeny because of their overlapping phenotypes, heterogeneity, and highly interdependent functions. These challenges are compounded in pathological settings, such as atherosclerosis and cancer, where new activation and differentiation states emerge [1]. For example, upon entry into lymph nodes and tissue, monocytes upregulate CCR7 and MHCII [13], which are also highly expressed on DCs and certain macrophages. While undifferentiated monocytes are abundant in lymph nodes and can present antigen, their relative contributions to T cell activation compared to macrophages and DCs remain unclear [13–18]. For example, compared to conventional DCs (cDCs), antigen-expressing monocytes elicit more robust T helper (Th) 1 and Th2 cell responses, during viral infection and airway inflammation, respectively [16,19]. Conversely, cDCs play dominant roles in activating both CD4⁺ and CD8⁺ T cells in murine tumor models [20,21]. Understanding the unique functions of monocytes in controlling T cell responses within diverse niches will help identify novel therapeutic targets for a range of diseases. In this review, we provide an overview of monocyte heterogeneity, the mechanisms coordinating interactions between particular subsets of monocytes and T cells, and how the distinctive plaque and tumor microenvironments impact these interactions. Understanding this crosstalk may be critical for improving treatment of cancer, CVD, and other diseases.

Monocyte Heterogeneity

Monocytes (Table 1) derive from a lineage-committed monocyte progenitor (cMoP; CD117⁺CD115⁺CD135⁻Ly6C⁺CD11b⁻ in mice [22] and CD34⁺CD135⁺CD64⁺CLEC12A⁺ in humans [23]) that resides in the bone marrow (BM). Human monocytes universally express CD11b, the major histocompatibility class II (MHC-II) receptor HLA-DR, and CD86, whereas murine monocytes universally express CD115, CD11b, and CD64. Monocytes have traditionally been classified into classical, nonclassical, and intermediate subsets, and this framework has provided the basis for functional studies to date. Classical monocytes (CD14⁺CD16⁻ in humans, Ly6C^hiCCR2^hiCX3CR1^lo in mice) rapidly extravasate into tissues during infection and inflammation [24] and are required for replenishment of macrophage populations in tissues such as the skin, intestines, heart, and liver [25]. Nonclassical monocytes (CD14^loCD16⁺ in humans, Ly6C^loCCR2^loCX3CR1^hi in mice) patrol the endothelium during homeostasis and extravasate into tissues less frequently than classical monocytes [26], although they can be detected in the lungs [27] and kidneys [28]. Intermediate monocytes (CD14⁺CD16⁺ in humans, Ly6C^intCCR2^hiCX3CR1^hi in mice) appear to represent a transition between classical and nonclassical subsets, contributing to the frequent contamination seen in monocytes during conventional flow cytometry [29,30]. Still, the contribution of intermediate monocytes to immune responses remains unclear, despite their increased frequency in patients with advanced CVD [31].

Table 1.

Monocyte Subsets and Potential T cell interactions in Tumor and Atherosclerotic Plaque Niches

Monocyte Subset	Mouse Markers	Human Markers
Classical monocytes	Ly6ChiCCR2hiCX3CR1lo; (CD115+CD11b+CD64+)	CD14+CD16−, (CD11b+HLA-DR+ CD86+)
Nonclassical monocytes	Ly6CloCCR2loCX3CR1hi (CD115+CD11b+CD64+)	CD14loCD16hi, (CD11b+HLA-DR+ CD86+)
	Tumors
Monocyte	Function	T cell interaction
Classical monocytes	Ly6ChiCD103+ monocyte-derived cells cross-present antigens with B16 [70].	Antigen Presentation
	F4/80hiCD24+ monocyte-derived cells cross-present antigens with B16 [69].
	Upregulate macrophage-associated genes at 5 days post injection into mammary tumor-bearing PyMT mice [47].
	Monocyte-derived TAMs from mammary tumors display increased antigen presentation transcripts [56].
	Monocytes from renal carcinoma, breast cancer, and endometrial cancer individuals display increased TNF, IL1B, IL6, and CCL3 [55].	Paracrine Signaling
	Recruit Tregs via CCR5 in SubQ RMA-S lymphoma tumors [59].	Treg Recruitment
	Employ CD40 [60], CD86 [61], and arginase-1 [62] to expand Treg numbers
Nonclassical monocytes	Display increased CCL3, CCL4, and CCL5 within lung tumor metastases [7].	Lymphocyte Recruitment
	Atherosclerotic plaques
Monocyte	Function	T cell interaction
Classical monocytes	Differentiate into CD11bhiCD11hi cells expressing CD80 and CD86; Some express F4/80 [86–90].	Antigen Presentation
	Macrophages secrete TNF-α, IL-12, iNOS, and IL-6. [98,99]	Paracrine Signaling
	Macrophages and DCs take up cholesterol and become foam cells [103,104,105].	DC-derived foam cells elicit Th1 CD4 T cells [106,107], potentially contributing to plaque rupture.
Nonclassical monocytes	Accumulate in vessel wall and under hypercholesterolemic conditions, display increased CD11c positivity [91–93].	Antigen Presentation (Possible acquisition of DC-like phenotype)

Work in both humans and mice suggests that classical monocytes develop in the BM and egress into peripheral blood in a CCR2-dependent manner [24]. Classical monocytes circulate in the periphery for approximately 1 day before entering tissues and then either differentiate or undergo apoptosis [32,33]. In contrast, nonclassical monocytes circulate for at least 2 days in mice (with a half-life of 2.2 days) and 7 days in humans [32,33]. Most nonclassical monocytes are thought to derive directly from classical monocytes [32,34]. However, the cues regulating the conversion of classical monocytes into nonclassical monocytes remain only partially characterized. Our laboratory has demonstrated that development of nonclassical monocytes requires the orphan nuclear receptor NR4A1, which is regulated by a superenhancer region specifically active in these cells [5,6]. NR4A1 expression is controlled by the transcription factors KLF2 and C/EBPβ [5], but additional studies are needed to identify the signals driving this genetic program.

High-dimensional technologies, such as single-cell RNA sequencing (scRNA-Seq) and mass cytometry by time-of-flight (CyTOF), have largely confirmed established monocyte subsets in humans and mice [34–36], while also identifying novel populations. For example, our laboratory recently employed CyTOF to identify a subset of nonclassical CD16⁺ monocytes that expresses the 6-Sulfo LacNAc (Slan) carbohydrate modification of P-selectin glycoprotein ligand-1 (PSGL-1) in humans [29]. CD16⁺Slan⁺ monocytes have an increased capacity for efferocytosis compared to CD16⁺Slan⁻ monocytes and are positively correlated with CVD severity [29]. In mice, scRNA-Seq has identified a subset of intermediate monocytes enriched for DC-related genes, including CD209a [34]. Whether this subset is related to the previously-identified CD209a-expressing Ly6C⁺ monocytes that differentiate into DCs following exposure to GM-CSF [37], or the CD74-expressing “monoDC” population detected in lung tumors [36], requires further investigation. High-dimensional technologies have revolutionized the profiling of immune cells, including simultaneous measurement of transcriptomes and epitopes [38], single cell assessment of chromatin accessibility [39], and in-depth metabolomic surveys [40]. Moreover, multiplexed spatial cytometry [41] and novel bioinformatics approaches to identify putative cellular interactions [42] have revealed new insight into the distribution of immune cells in tissues and revealed cellular networks. These and other high-dimensional approaches will likely continue to reveal new monocyte subsets and their interactions with cells in the surrounding microenvironment.

Monocyte-T Cell Interactions in Cancer

Monocytes and their progeny regulate the TIME by directly interacting with tumor cells, fibroblasts, endothelial cells, and other leukocytes, including T cells [43]. In a murine model of lung metastasis, both classical and nonclassical monocytes extravasate towards metastatic sites within hours of tumor cell seeding [7,44]. While monocytes initially contribute anti-tumoral functions [45], the TIME often promotes the differentiation of these cells into pro-tumoral tumor-associated macrophages (TAMs) [46]. Classical monocytes are recruited to mammary tumors and upregulate their expression of macrophage-associated markers such as F4/80, MHCII, and CD11c within 5 days after intravascular transfer [47]. Recruitment of Ly6C^hi monocytes from peripheral blood into the tumor, and the subsequent generation of TAMs, depends on the CCR2/CCL2 signaling axis [48], with M-CSF (CSF-1) also contributing to this process [49].

Tumor expression of CCL2 negatively correlates with intratumoral infiltration of CD8⁺ T cells in hepatocellular carcinoma patients [50] and, in combination with CD8, stratifies survival outcomes in pancreatic cancer patients [9]. Inhibition of CCR2 or CSF1R reduces TAM accumulation, increases CD8⁺ T cell abundance, and reduces tumor burden in mouse models of pancreatic and liver cancer [50,51]. Furthermore, the anti-tumoral activity of CCR2 inhibition in mice bearing Hepa1–6 liver tumors requires the presence of CD8⁺ T cells [50]. Interestingly, a higher frequency of classical CD14⁺ monocytes in peripheral blood is associated with increased T cell activation and improved survival in melanoma patients [52]. The precise relationship between monocytes and T cells likely depends on the immunostimulatory versus immunosuppressive capacity of monocytic cells within a specific TIME. Monocytic myeloid-derived suppressor cells (M-MDSCs) are defined by their ability to suppress T cell activation and proliferation [53]. While M-MDSCs share expression of markers such as CD14, CD33, and CD11b with monocytes, whether monocytes and M-MDSCs are distinct cell types remains unclear [54]. A better understanding of the relationship between monocytes, M-MDSCs, and other TIME myeloid cells

Paracrine Interactions

Monocytes isolated from the blood of renal cell carcinoma [55], breast cancer [56], and endometrial cancer patients [56] exhibit distinct transcriptional signatures, compared to monocytes from healthy individuals. In renal cell carcinoma specifically, transcripts associated with downstream activation of T cells, such as TNF, IL1B, IL6, and CCL3, are elevated [55]. TAMs derived from mammary tumors display increased expression of transcripts associated with antigen presentation, immune activation, and T cell costimulation [56]. Given that these transcripts are not similarly upregulated in cancer-derived monocytes, these gene expression programs may only be imprinted once monocytes begin maturation within the TIME. Blockade of monocyte recruitment via CCR2 inhibition reduces macrophage-based production of the cytokines IL-6, IL-13, IL-15, and TNFɑ that can differentially impact T cell responses [50,57,58]. Nonclassical monocytes within lung tumor metastases generate chemokines such as CCL3, CCL4, and CCL5, which are involved in lymphocyte recruitment to tumor sites [7] (Table 1). How T cells integrate and respond to diverse monocyte-derived signals will be important to understanding how to therapeutically target these interactions.

M-MDSCs display a monocyte-like phenotype and recruit T regulatory (Tregs) cells to lymphoma tumors in a CCL5/CCR5-dependent manner, which facilitates tumor growth [59] (Figure 1, Table 1). Consequently, monocyte-derived chemokines such as CCL5 may recruit both pro- and anti-tumoral T cell populations in a context-specific manner. Immunosuppressive M-MDSCs employ CD40 [60], CD80 [61], and arginase-1 [62] to expand Treg numbers and suppress anti-tumoral T cells at the tumor site, with accumulation of Tregs predicting poor prognosis in non-Hodgkin’s lymphoma and fibrosarcoma [63,64]. Targeting arginase has been explored in early phase clinical trials that demonstrated a small molecule inhibitor is well-tolerated and displays efficacy in inhibiting arginase in solid tumor patients [65]. Interestingly, Tregs also influence the fate of intratumoral monocytes by inhibiting their migration and/or differentiation and by eliciting an alternatively-activated phenotype via the anti-inflammatory cytokine IL-10 [66,67]. These observations warrant the continued exploration of Treg-monocyte interactions and potential for therapeutic targeting within the TIME.

Figure 1. Crosstalk between monocytes and T cells in the tumor niche.

Classical Ly6C^hi monocytes recruited through CCR2/CCL2 are a primary source of tumor-associated macrophages (TAMs) and dendritic cells (DCs). TAMs are primarily pro-tumoral in part through local immune suppression characterized by the production of cytokines that inhibit CD8⁺ T cell activation/proliferation and recruit T regulatory cells that further dampen innate and adaptive immune responses. Costimulatory and coinhibitory molecules are upregulated on TAMs compared to monocytes, which may play a role in modulating T cell responses. Monocytes and monocyte-derived DCs cross-present antigen to activate CD8⁺ T cells, particularly in the absence of conventional DCs. Nonclassical Ly6C^hi monocytes are important for anti-tumoral immunity in metastatic lesions, but their function in solid tumors remains largely unknown.

Antigen Presentation

Cross-presentation is particularly important for immune responses against tumor cells that express a high number of mutations but lack the machinery necessary for priming T cells directly. While cDCs are believed to be primarily responsible for both cross-presenting tumor antigens to CD8⁺ T cells and MHCII antigen presentation to CD4⁺ T cells [21,68], evidence suggests that other monocyte-derived cells also process and present tumor-derived antigens [69,70]. For example, F4/80^hiCD24⁺ cells within B16 melanoma and intestinal tumors cross-present tumor antigens and are CCR2-dependent, suggesting they derive from a monocytic origin [69]. In Pten^−/−Foxp3-Cre mice bearing B16 tumors, a population of Ly6C⁺CD103⁺ monocyte-derived cells cross-present antigens and re-activate anergic CD8⁺ T cells [70] (Figure 1, Table 1).

Maturation into macrophages/DCs is likely required for cross-presentation, as monocytes derived from human lung tumors are unable to present tumor antigens, while macrophages from the same tumors can cross-present and stimulate IFNɣ production by antigen-specific effector T cells [71]. Interestingly, tumor antigen in metastatic lung sites is redirected from macrophages to cDCs in CCR2-deficient mice, indicating that different APCs may compete for tumor antigen [44]. Additionally, monocytes may most effectively contribute to anti-tumoral immunity, especially in TIMEs with sufficient numbers of cDCs, by transporting antigen to lymphoid organs before transfer to APCs [72].

Costimulatory and Coinhibitory Molecules

Myeloid cells impact the strength of T cell receptor signaling and downstream T cell responses by surface expression of costimulatory and coinhibitory molecules [73]. In peripheral blood, the costimulatory molecule CD86 is universally expressed across monocyte subsets, while CD80 is lowly expressed at homeostasis [29]. In mice, expression of the coinhibitory molecule programmed death ligand 1 (PD-L1) is restricted to nonclassical Ly6C^lo monocytes at homeostasis [74], but appears to be broadly induced in both classical Ly6C^hi monocytes and myeloid progenitors in mice bearing B16 melanoma tumors [75]. Monocytes upregulate both the PD-L1/2 and CD80/CD86 pathways as they enter the TIME and differentiate into TAMs [76]. Tumor-derived RNA may serve as one of the signals regulating expression of coinhibitory molecules in monocytes, as RNA-loaded exosomes derived from leukemic cells increase PD-L1 expression in human monocytes [77]. Interestingly, the receptor for PD-L1/2, programmed cell death protein-1 (PD-1), is also absent from monocytes during homeostasis, but induced in tumor-bearing mice [75]

CD28 expressed on naive T cells binds to CD80 and CD86 expressed on APCs, and interactions between CD28 with CD80/CD86 are critical for facilitating memory and effector T cell formation [78]. Costimulation by CD86 generally promotes T cell activation, but CD86 can also inhibit this process through interaction with CTLA-4. In monocyte-derived TAM precursors recruited to lung metastases, CD86 suppresses CD8⁺ T cell-mediated tumor cell cytotoxicity through CTLA-4 [76]. Consequently, anti-CTLA-4 immunotherapy (currently approved for treatment of metastatic melanoma and renal cell carcinoma [79]) may act in part by interfering with interactions between immunosuppressive monocyte-derived cells and T cells, although this requires further investigation.

Recent work demonstrated that increased PD-1 expression on myeloid cells in tumor-bearing mice leads to enhanced production of myeloid progenitors and MDSCs that suppress T cell responses [75]. Immune checkpoint inhibitors targeting PD-1 and PD-L1/2 have been highly successful in subsets of non-small cell lung cancer, renal cell carcinoma, melanoma, and other solid tumor patients [79], but whether these therapies inhibit monocyte-T cell interactions remains unclear. Melanoma patients with higher baseline levels of classical monocytes display superior clinical responses and survival following anti-PD-1 treatment [52], providing evidence that monocyte-T cell interactions may contribute to therapies targeting PD-1:PD-L1/2 signaling. Additionally, monocytes can express OX40L, CD137L, and CD40 [80–82], which are currently under investigation as drug targets for cancer immunotherapy. Multiple Phase I and Phase II clinical trials are underway to examine the safety and efficacy of CD40 monoclonal antibodies in solid tumors [83]. Whether these molecules regulate crosstalk between monocytes and T cells in cancer, and the extent to which these interactions may be targeted clinically to increase anti-tumoral immunity will be of interest as further research is performed in this area.

Monocyte-T Cell Interactions in Atherosclerosis

Classical Ly6C^hi monocytes (Table 1, Figure 2) represent the first immune cell population to arrive at the atherosclerotic plaque via recruitment by CCR2-CCL2, CX3CR1-CX3CL1, and CCR5-CCL5 signaling [84,85]. Ly6C^hi monocyte frequencies double every month in atherosclerotic Apolipoprotein E-deficient (ApoE^−/−) mice fed a Western (high-cholesterol) diet [84]. Once recruited to the vessel wall, Ly6C^hi monocytes differentiate into CD11b^hiCD11c^hi cells and upregulate the costimulatory molecules CD80 and CD86 [86–89], suggesting acquisition of APC capacity [90]. Many of these cells also express F4/80, indicating that plaque monocytes may differentiate into both DCs and macrophages.

Figure 2. Crosstalk between monocytes and T cells in the atherosclerotic plaque niche.

Monocytes are the first immune cell type recruited to the atherosclerotic plaque. Ly6C^lo monocytes patrol the endothelium, yet whether these monocytes influence T cell responses in atherosclerosis is unknown. Ly6C^hi monocytes differentiate into macrophages and DCs that are CD11b⁺CD11c⁺, characterized by high expression of the costimulatory molecules. DCs, and likely monocytes, can egress from the plaque to the lymph nodes and present antigenic peptides to CD4⁺ T cells. However, as the atherosclerotic process progresses, DCs are maintained within the plaque. Macrophages secrete pro-inflammatory cytokines that increase Th1 differentiation. Macrophages and DCs also accumulate lipids and become foam cells that comprise the necrotic core and secrete IL-1β. DC-derived foam cells retain antigen presentation capabilities and drive pro-atherogenic CD4⁺ T cell responses.

In contrast to Ly6C^hi monocytes, nonclassical Ly6C^lo monocytes (Table 1, Figure 2) accumulate within the vessel wall via CCR5, but at a considerably slower rate [84,91]. Under hypercholesterolemic conditions, Ly6C^lo monocytes display increased CD11c surface expression, indicating potential acquisition of a DC-like phenotype [91–93]. Global knockout of Nr4a1 within ApoE^−/− mice results in loss of nonclassical Ly6C^lo monocytes and enhances atherosclerosis disease severity, which is accompanied by an increase in pro-inflammatory macrophages [94]. Western-diet feeding also increases the patrolling behavior of Ly6C^lo monocytes along the vascular endothelium in a CD36− and oxidized low-density lipoprotein (OxLDL)-dependent manner [95]. Unpublished data (Marcovecchio P.M.) from our laboratory suggests that disrupting patrolling behavior may impact endothelial cell homeostasis, which could have broad implications for leukocyte recruitment into the plaque [95]. Whether nonclassical monocytes interact with and influence atherogenic T cell responses remains unclear; however, recent evidence indicates that patrolling monocytes are able to present antigens within the vasculature to effector CD4⁺ T cells [96].

The persistence of undifferentiated monocytes within the plaque has yet to be determined, in part because the aforementioned studies have not included markers for distinguishing monocytes from macrophages and DCs. Thus, we now also focus on the interactions of T cells with monocyte-derived macrophages and DCs, as this constitutes the majority of atherosclerosis research.

Paracrine Interactions

Monocyte-derived macrophages (F4/80^hi CD11b⁺CD11c⁺; Table 1, Figure 2) represent the most abundant immune cell type within atherosclerotic plaques [97]. Following differentiation from monocytes, macrophages secrete pro-inflammatory cytokines, such as TNFɑ, IL-12, IL-6, and iNOS, which promote differentiation of naive CD4⁺ T cells into Th1 cells [98,99]. Th1 CD4⁺ T cells cell are pro-atherogenic, as deletion of the genes encoding Th1-associated factors IFNɣ or T-bet protects against atherosclerosis [100]. Plaque-associated macrophages instruct T cells to secrete pro-atherogenic cytokines and recruit additional leukocytes into the atherosclerotic lesion [101,102].

As atherosclerosis progresses, monocyte-derived macrophages ingest excess lipids and eventually become foam cells (reviewed in [103]). Foam cells contribute to the mass of the atherosclerotic plaque and are directly implicated in the development of the necrotic core, a hallmark of plaques vulnerable to rupture [97]. In addition to macrophages, monocyte-derived DCs can ingest lipids and differentiate into foam cells that retain their ability to prime T cell responses [104,105] (Figure 2). Cholesterol uptake by macrophages and DCs is characterized by activation of the inflammasome, concomitant with increased expression of IL-1β, and DC-derived foam cells have been implicated in the generation of pro-atherogenic Th1 CD4⁺ T cells [106,107]. Whether macrophage-derived foam cells retain their antigen presentation capabilities in a similar fashion to DCs and influence T cell responses is an unanswered question that will elucidate interactions between monocyte-derived cells and T cells within the plaque.

Monocyte-T cell interactions are not unidirectional, as T cells have a reciprocal impact on monocyte development in cardiovascular disease. A recent study demonstrated that CD8⁺ T cells promote medullary monocyte production and CD8⁺ T cell depletion reduces atherosclerosis and is accompanied by attenuated Ly6C^hi monocytes in blood, BM, and spleen, and lesional macrophages with atherosclerotic plaques [108]. These results indicate that CD8⁺ T cells modulate atherosclerosis severity by controlling myelopoiesis, similar to acute viral infection, and implicate GM-CSF as a key factor [108,109].

Paracrine signaling that impacts monocyte-T cell interactions has not been explicitly targeted in CVD. Instead, targets for CVD therapeutics have largely focused on lipids, as free and esterified cholesterol are robustly identified in plaques, and elevated blood cholesterol levels, particularly LDL-C, predict cardiovascular events [110,111]. Indeed, the inflammatory response was originally considered a byproduct of cholesterol accumulation in vessels and attributed to smooth muscle cell proliferation [110,112,113]. IL-1β, a cytokine synthesized by blood-derived monocytes, macrophages, and DCs, which provides essential cues for T cell activation, is the focus of a recent clinical trial to lessen atherosclerotic disease [114,115]. The double-blind, placebo Canakunimab Anti-inflammatory Thrombosis Study (CANTOS) treated individuals with high risk for myocardial infarction with the IL-1β monoclonal antibody canakinumab [116]. Canakinumab produces a 15% decrease in the primary composite endpoint (non-fatal myocardial infarction, stroke, CVD death) compared to placebo-treated individuals, which is accompanied by 60% reductions in C-Reactive Protein, a risk factor for future CVD events [116]. Interestingly, IL-1β inhibition is especially effective within individuals with lung cancer comorbidity, with a 77% decrease in fatality [117–119], highlighting the need to explore integrated therapeutic strategies.

Antigen Presentation

Interactions between T cells and APCs occur in the mouse aorta and elicit production of TNFɑ and IFNɣ [120–126]. Monocyte-derived DCs capture antigens in the vessel wall and traffic to draining lymph nodes for antigen presentation to T cells [127–129]. However, as atherosclerosis progresses, DCs are retained within the atherosclerotic plaque, where they prime effector T cells [130,131]. This is purportedly due to increased expression of chemokines such as CCL19, CCL21, P-Selectin, and V-CAM1 [132–134]. The cellular mechanisms of antigen presentation have primarily focused on DCs in the context of atherosclerosis, and whether these interactions occur between monocytes and T cells remains an open question.

The peptide antigens presented by APCs to T cells in atherosclerosis remain largely unknown, but numerous candidates have been proposed [135]. The best studied of these is LDL and its core protein ApoB [136]. CD4⁺ T cells displaying reactivity to oxLDL are located within human plaques [137,138]. In mice immunized with human oxLDL, MHC class II-restricted T cell clones reactive to native LDL possess a T cell receptor with the β chain TRBV31 and blockade of this receptor protects against atherosclerosis [139]. In addition to oxLDL, HSP60 [140] and ApoB100 [139] are also essential antigens for T cell activation during atheroprogression.

There are conflicting reports concerning the importance of cross-presentation by DCs in the development of atherosclerosis. Depletion of CD8ɑ⁺ and CD103⁺ cDCs via Batf3 deficiency has no effect on atheroprogression, despite inhibiting cross-presentation [141]. Moreover, atherosclerosis remains unchanged with deficiency of Antigen Peptide Transporter 1 (TAP-1), an essential component of the MHC class I presentation complex that participates in cross-presentation [142]. While these reports indicate that cross-presentation is dispensable for atherosclerosis, these studies do not rule out compensatory cross-presentation by monocytes, which occurs in cancer [70]. Therefore, further efforts should examine whether cross-presentation by monocytes occurs in atherosclerosis.

Costimulatory/Coinhibitory Molecules

Interactions between costimulatory molecules expressed on monocyte-derived cells and T cells are essential for atherosclerotic progression [143–146] and have been extensively reviewed elsewhere [147]. Mice deficient in CD80 and CD86 show marked reduction of atherosclerotic severity, but increased synthesis of the pro-atherogenic cytokine IFNɣ [148]. Expression of CD80 and CD86 is enhanced on monocyte derived-DCs from individuals with CVD [149]. Additionally, while the CD80/CD86 receptor CD28 was initially proposed as a promising therapeutic target, a CD28 superagonist antibody elicits cytokine release syndrome due to increased memory and effector T cell formation in healthy individuals [150].

PD-1 is upregulated in aorta-infiltrating T cells of Ldlr^−/− mice fed a cholesterol diet and expressed in human carotid plaque-infiltrating T cells. [151,152]. Interestingly, this transcriptional signature is similar to that of exhausted T cells in the TIME [152]. Myocardial infarction elicits an influx of PD-L1⁺ nonclassical monocytes into pericardial tertiary lymphoid organs within 5 days [74]. PD-L1 is localized at the interface of contacting nonclassical monocytes and T cells, and promotes T cell survival [74]. Whether PD-L1-expressing nonclassical monocytes are recruited to periaortic lymph nodes during atherosclerosis progression and impact atherosclerosis severity remains unknown. However, high fat diet-feeding of PD-L1- or PD-L2-deficient Ldlr^−/− mice increases atherosclerosis and aggravates effector CD8⁺ T cell responses, purportedly due to enhanced antigen presentation by DCs [153]. Thus, immune checkpoint blockade may have unintended consequences in treating individuals with CVD and cancer, and conversely, PD-1 agonists may prove beneficial in CVD [154].

Additional interactions between costimulatory molecules, such as CD40/CD40L and OX40/OX40L may mediate interactions between antigen-presenting monocytes and T cells. CD40 upregulation on APCs activates CD40L (CD154)-expressing CD4⁺ T cells and increases production of pro-atherogenic cytokines [155–157]. CD40/CD40L interaction also increases CD80 and CD86 expression on APCs and vice-versa [144]. Clinical trials targeting interactions between CD40 and CD40L in atherosclerosis were discontinued due to thrombotic complications; nevertheless, an antisense oligonucleotide approach targeting CD40 may represent a promising approach [149,158]. Signaling by OX40L expressed on myeloid cells is essential for memory T cell survival and implicated in atheroprogression [145]. Furthermore, global OX40L overexpression and deletion increases and reduces fatty streak formation, respectively, in high-fat diet-fed mice [145,146]. Plaque-resident CD4⁺ T cells and macrophages express OX40 and OX40L, respectively [145], and accumulation of TNFRSF4 and TNFSF4 transcripts, which encode OX40 and OX40L, correlate with CVD risk [159]. While there have been no clinical trials on these targets to date, an anti-OX40 antagonistic antibody that elicits increased OX40⁺ T cells is currently under trial for individuals with solid tumors [160].

Concluding Remarks

This review highlights the spectrum of monocyte and monocyte-derived macrophage phenotypes that are present within tumors and the atherosclerotic plaque. Monocytes are uniquely influenced by these niche microenvironments, with reparative and pro-inflammatory phenotypes prevailing in the tumor and plaque, respectively. Although T cells are influenced by diverse cell types other than monocytes, including endothelial cells, NK cells, smooth muscle cells, and neutrophils [135,161], monocytes and their progeny provide non-redundant cues that impact T cell-mediated immunity. Comparing the monocyte-T cell interactions within these divergent microenvironment niches in cancer and atherosclerosis provide essential insight regarding the dysregulated immune response in these diseases.

High-dimensional immunophenotyping via CyTOF and scRNA-Seq has uncovered striking heterogeneity in monocytes and T cells [34–36], paving the way for functional studies to examine interactions between new subsets. These methods have been championed in the cancer field, revealing novel insight into potential immunotherapeutic targets in the cancer setting [36,52,162]. Still, further study is needed in atherosclerosis to interrogate potential interactions between monocytes and T cells and the outputs of these interactions in atherosclerosis [152,163–166]. Recently, the algorithm Clustergrammer [162] was employed to predict the different potential macrophage-T cell interactions within atherosclerotic plaques that bring about either symptomatic or asymptomatic states in CVD patients [152]. Plaque-resident macrophages from asymptomatic individuals express IL1B, predicted to bind to IL1RAP, which encodes a component of the IL-1 receptor complex expressed by T cells [152]. In contrast, T cells within plaques from symptomatic patients express ligand-encoding transcripts predicted to elicit reparative macrophages [152]. Bioinformatic tools such as this, employed alongside multiplexed spatial immune profiling of tumors and atherosclerotic plaques, will be powerful tools to shed light on the monocyte-T cell interactions.

Improved fate-mapping studies to pinpoint the specific effects of monocyte-derived APCs on T cells versus DC subsets from independent lineages could greatly benefit atherosclerosis research [32,34,167]. However, fluorescence-based mouse transgenics, such as the Cx3cr1^gfp/+ strain, do not allow monocytes to be clearly separated from macrophages and DCs [167]. These technical limitations have contributed to the limited knowledge regarding whether monocyte-derived cells present lipid antigens within the atherosclerostic plaque.

Ultimately, insights on the influence of monocytes and their progeny on T cell responses in the setting of cancer and atherosclerosis are of great interest, as they will likely reveal novel therapeutic targets for a range of diseases. By comparing and contrasting the monocyte-T cell interactions within these distinct niches in cancer and atherosclerosis, we gain essential insights regarding the dysregulated immune response in these diseases. Importantly, these insights will identify new targets of innate and adaptive immunity that can improve current therapies for atherosclerosis and cancer.

Abbreviations

APC

Antigen-presenting cell

BM

Bone marrow

CANTOS

Canakunimab Anti-inflammatory Thrombosis Study

cDC

Conventional dendritic cell

CITE-Seq

Cellular indexing of transcriptome and epitopes by sequencing

CyTOF

Cytometry by Time-Of-Flight

CVD

Cardiovascular disease

DC

Dendritic cell

LDL

Low-density lipoprotein

MHC

Major histocompatibility complex

MDSC

Myeloid-derived suppressor cell

oxLDL

Oxidized low-density lipoprotein

PD-1

Programmed cell death protein-1

PD-L1/2

Programmed cell death ligand-1/2

PSGL1

P-selectin glycoprotein ligand-1

scRNA-Seq

Single-cell RNA-sequencing

TAM

Tumor-associated macrophage

TIME

Tumor immune microenvironment

Slan

6-Sulfo LacNac