Introduction

T-cells make up ~50–65% of all leukocytes in advanced human atherosclerotic lesions from carotid endarterectomies [1••, 2••]. After infiltration into atherosclerotic lesions, T-cells interact with macrophages and dendritic cells (DCs) [3••]. Upon recognition of their cognate antigen presented by DCs and dependent on the cytokine milieu, naïve T-cells differentiate into distinct subsets characterized by the expression of transcription factors (i.e., FoxP3 for T regulatory cells (Tregs); Tbet for T helper 1 (Th1) cells; GATA3 for Th2 cells; retinoic acid-related orphan receptor (ROR)γT for Th17 cells; and B-cell lymphoma (Bcl)6 for Tfollicular helper (Tfh) cells) [3••]. More than 80% of T-cells in atherosclerotic plaques express CD44, indicating that they are antigen experienced [2••, 3••]. Among the antigens that DCs present to T-cells are apolipoprotein B100 (apoB100), low-density lipoprotein (LDL), and oxidized LDL [4,5,6]. While initial studies have suggested that antigen presentation in atherosclerotic plaques induces production of the pro-atherogenic Th1 cytokines interferon γ (IFNγ) and tumor necrosis factor α (TNFα) [4, 7], later studies have shown an expansion of Tregs in response to antigens [6, 8, 9]. Tregs exert an anti-atherogenic role by secreting interleukin (IL)-10 and transforming growth factor β (TGF-β) [10, 11]. TGF-β induces smooth muscle cell (SMC) migration and collagen production by SMCs [12,13,14,15]. Recent single-cell RNA sequencing (sc-RNA-Seq) studies have revealed a high diversity of T-cells in human atherosclerotic plaques [1••, 2••]. The role of specific T-cell subsets in atherosclerosis has been reviewed previously [3••].

Even though individual T-cell subsets have pro- or anti-atherogenic effects, presumably via the cytokines they secrete, complete CD4+ or CD8+ T-cell ablation reduces atherosclerosis in mice [16,17,18,19]; however, in advanced atherosclerosis, CD8+ T-cell ablation increases plaque stability [20•], highlighting the complex role of T-cells in atherogenesis.

Recent studies have revealed that during atherosclerosis and cardiovascular disease (CVD) in humans, Tregs acquire markers of Th1, Th17, and Tfh cells, or switch to a more memory-like phenotype, which may render them pro-atherogenic [6, 9, 21, 22, 23•, 24]. Studies in mouse models have proposed that T-cell cholesterol accumulation critically contributes to this effect [23•]. In addition, T-cells from CVD patients lose their ability to proliferate and, therefore, to respond adequately to antigens [25•], a critical function of T-cells. The decreased proliferation may be the result of T-cell apoptosis downstream of excessive T-cell cholesterol accumulation [26••]. Aged T-cells also show increased cholesterol accumulation [27, 28]. Here, we will review how pathways that regulate T-cell cholesterol accumulation determine T-cell fate, atherosclerosis, and T-cell aging.

T-cell Receptor Stimulation and Cholesterol Accumulation

T-cells express high levels of the cholesterol transporters ATP Binding Cassette A1 (ABCA1) and ABCG1 that mediate cholesterol efflux to apolipoprotein A-I (apoA-I) and high-density lipoprotein (HDL), respectively [29]. T-cells mainly accumulate cholesterol in their plasma membrane, which is key to T-cell receptor (TCR) signaling and proliferation in response to interaction with their cognate antigen. TCR stimulation by anti-CD3, which mimics T-cell stimulation by antigen-presenting cells via major histocompatibility complex (MHC)I/II, decreases expression of the cholesterol transporters Abca1 and Abcg1 (Fig. 1) [30••, 31••]. The decreased expression of ABC cholesterol transporters is mediated by suppression of Liver X receptor (LXR) signaling due to upregulation of the enzyme sulfotransferase family cytosolic 2B member 1 (SULT2B1) that transfers sulfate groups to oxysterols, which inactivates oxysterols in terms of their ability to bind the transcription factor LXR and to activate it [30••, 32]. TCR stimulation also increases the expression of 3-hydroxy-3-methylglutaryl-CoA reductase (Hmgcr), the LDL receptor (Ldlr), and acetyl coA acyl transferase 1 (Acat1), which promote cholesterol synthesis, uptake, and esterification, respectively (Fig. 1) [30••, 31••].

Fig. 1
figure 1

Effects of anti-CD3 stimulation on expression of genes involved in cholesterol homeostasis. Gene transcription is shown in the nucleus. Created with BioRender.com

Full size image

T-cell Membrane Cholesterol Accumulation Induces T-cell Proliferation

Several lines of evidence indicate that cholesterol accumulation is key to T-cell proliferation and, as such, key to the T-cell response upon interaction with an antigen. Suppression of cholesterol synthesis due to deficiency of sterol regulatory element-binding protein (SREBP) cleavage-activating protein (SCAP) completely abolishes T-cell proliferation in response to anti-CD3 [33•]. Conversely, when cholesterol cannot be esterified due to deficiency of Acat1, plasma membrane cholesterol accumulation increases, as does T-cell proliferation [31••]. Similarly, deficiency of Abcg1-mediated cholesterol efflux promotes plasma membrane cholesterol accumulation and T-cell proliferation [30••, 34, 35]. T-cell cholesterol loading via methyl-β-cyclodextrin (MβCD)-cholesterol or LDL-cholesterol (LDL-c) also increases proliferation [34, 36].

Abcg1 deficient T-cells show high expression of Abca1 [34], presumably due to the accumulation of oxysterols that induce the activation of LXR and consequently Abca1 transcription [37,38]. Recent studies have revealed that T-cell Abca1 deficiency increases Abcg1 expression, reduces T-cell membrane cholesterol accumulation, and decreases T-cell proliferation in response to anti-CD3 [39]. These data suggest that, as initially proposed [30••], Abcg1 is the dominant cholesterol transporter in T-cells. We found that deficiency of both Abca1 and Abcg1 increases T-cell membrane cholesterol accumulation and proliferation in young mice [26••]. Conversely, incubation with reconstituted HDL (rHDL) that induces cholesterol efflux, shows the opposite [26••]. Recent studies revealed that histone deacetylase 3 (Hdac3) deficiency decreases T-cell proliferation, which was attributed to decreased membrane cholesterol accumulation and increased Abca1 and Abcg1 mRNA expression [40•]. These data substantiate the crucial role for cholesterol efflux pathways in regulating T-cell proliferation. An overview of pathways regulating cholesterol accumulation and T-cell proliferation is given in Table 1.

Table 1 Plasma membrane cholesterol and T-cell proliferation
Full size table

T-cell Proliferation During Aging and CVD

While combined T-cell Abca1/Abcg1 deficiency increased T-cell proliferation in young mice, T-cell Abca1/Abcg1 deficiency almost abolished T-cell proliferation in mice at 1 year of age, concomitant with an upregulation of the senescence marker p21 [26••]. These findings suggest that perhaps aged Abca1/Abcg1 deficient T-cells became senescent due to several rounds of homeostatic proliferation. In addition, Abca1/Abcg1 deficiency increased T-cell apoptosis, in both young mice and mice at 1 year of age [26••]. The increase in T-cell apoptosis may be more prominent during aging, as such contributing to the abolished T-cell proliferation in aged mice.

Interestingly, individuals over 70 years of age also show T-cell cholesterol accumulation compared to T-cells from individuals less than 25 years of age [27, 28], as do T-cells from wild-type mice at 2 years of age compared to T-cells from wild-type mice at 3 months of age [26••]. T-cells from aged mice (2 years) show increased apoptosis compared to T-cells from young mice (3 months) [26••]. Based on the findings in mice with T-cell Abca1/Abcg1 deficiency [26••], these data suggest that also during aging, T-cell cholesterol accumulation contributes to apoptosis and, consequently, the decline in total T-cells. T-cell proliferation was only minimally decreased in T-cells from aged mice compared to young mice [26••]. However, T-cells from Apolipoprotein e deficient (Apoe−/−) mice with advanced atherosclerosis due to 20 weeks of cholesterol-rich Western-type diet (WTD) feeding, show decreased T-cell proliferation and increased T-cell apoptosis compared to T-cells from Apoe−/− mice fed a chow diet [25•]. Even though this was attributed to impaired antigen presentation by DCs [25•], previous studies have shown that WTD feeding induces cholesterol accumulation in Apoe−/− T-cells [23•], and our studies in mice with T-cell Abca1/Abcg1 deficiency demonstrate that T-cell cholesterol accumulation may directly increase T-cell apoptosis [26••].

In line with the findings in Apoe−/− mice, patients with advanced coronary artery disease (CAD) show a decrease in proliferation and an increase in T-cell apoptosis compared to patients with early CAD, irrespective of age (n = 14 patients per group) [25•]. While this would need to be confirmed in a larger CAD cohort, the data suggest a direct link between advanced CAD and impaired T-cell functionality due to T-cell apoptosis. Our data show that T-cell cholesterol accumulation, which may be aggravated in advanced CAD, contributes to this impaired T-cell functionality.

Not all genes that affect T-cell membrane cholesterol accumulation and TCR signaling (Table 1) affect apoptosis. Acat1 deficiency decreased apoptosis in CD8+ T-cells [31••], perhaps due to Acat1 deficiency increasing T-cell proliferation and survival, which may offset potential effects on apoptosis. However, it should be noted that the effects of Abca1/Abcg1 or Apoe were most pronounced in CD4+ T-cells [25•, 26••], and are probably the consequence of an increase in intracellular T-cell membrane cholesterol accumulation that is more dramatic than reported for other genes listed in Table 1. Nonetheless, T-cell cholesterol accumulation induced by MβCD-cholesterol loading promotes endoplasmic reticulum (ER) stress and CD8+ T-cell exhaustion without affecting apoptosis [41•]. We found that also in aged T-cells from wild-type mice (2 years old), expression of SREBP2 was decreased compared to T-cells from young mice (3 months), suggestive of ER cholesterol accumulation [26••]. ER cholesterol accumulation may account for T-cell exhaustion during aging.

T-cell Membrane Cholesterol Accumulation and Differentiation into Cytotoxic T-cells

In addition to T-cell proliferation, TCR stimulation increases granzyme B, IFNγ, and TNFα positive CD8+ T-cells, which are required for killing of foreign cells or pathogens [42]. Similar to effects on T-cell proliferation, deletion of genes or treatments that favor cholesterol accumulation (Lxrβ deficiency [30••], Acat1 deficiency [31••], MβCD-cholesterol [31••], and LDL-c [36]) induce differentiation into these cytotoxic CD8+ T-cells, while a decrease in cholesterol synthesis by Scap deficiency [33•] or treatment with lovastatin [31••] or cholesterol depletion by MβCD [31••] does the opposite. Also, inhibition of Niemann-Pick C1 protein, which induces movement of cholesterol from lysosomes to the plasma membrane, by the U18666A compound, decreases differentiation into these cytotoxic T-cells [31••], presumably due to decreased plasma membrane cholesterol [43]. These findings are summarized in Table 2. In line, T-cell Abca1/Abcg1 deficiency induces differentiation into granzyme B and IFNγ expressing CD8+ T-cells [26••]. However, T-cell Abca1/Abcg1 deficiency decreased IFNγ secretion and T-cell mediated macrophage killing [26••]. We attributed these effects to increased T-cell apoptosis, and therefore these effects are simply the consequence of a lower number of T-cells [26••]. Similarly, T-cells from Apoe−/− mice fed WTD for 20 weeks show decreased IFNγ production compared to Apoe−/− mice fed a chow diet, concomitant with increased apoptosis [25•].

Table 2 Plasma membrane cholesterol and granzyme B+, IFNγ+, and TNFα+ CD8+ T-cells
Full size table

T-cell Membrane Cholesterol Accumulation, Atherosclerosis, and CVD

While effects of membrane cholesterol accumulation on T-cell proliferation and differentiation into cytotoxic T-cells seem to be relatively uniform, effects of T-cell cholesterol accumulation on downstream T-cell differentiation are not. T-cell Abcg1 deficiency increases membrane cholesterol accumulation and lipid droplet formation, indicative of increased cholesterol esterification [35]. T-cell Abcg1 deficiency increases formation of Tregs, with athero-protective effects [35].

Recent studies have revealed that during atherosclerosis and CVD, Tregs acquire markers of Th1, Th17, and Tfh cells, which may render them pro-atherogenic [6, 9, 21, 22, 23•, 24] (Fig. 2). Using a fluorescent tracing technique, current Tregs and exTregs (cells that were Tregs before) could be distinguished in Apoe−/− mice [23•]. This revealed that upon WTD feeding, Tregs underwent a phenotypic switch [23•]. Injections of apoA-I reversed this switch [23•], and therefore this switch was proposed to occur downstream of cholesterol efflux and thus to be cholesterol-dependent. In this model, Tregs lost their Foxp3 and CD25 expression and started to express IFNγ or Bcl6 and IL-21, suggesting differentiation into Th1 or Tfh cells, respectively [23•]. Previous sc-RNA-Seq studies have indeed shown that Tregs gain features of Th1 cells during atherosclerosis in Apoe−/− mice and that these cells are dysfunctional in terms of suppressing T-cell proliferation, a main characteristic of Tregs [22]. Deficiency of the specific Tfh transcription factor Bcl6 decreased atherosclerosis, indicating that Tfh cells are pro-atherogenic [23•], presumably because they induce B-cell activation and secretion of IL-21 [44,45]. One caveat to this atherosclerosis study was that Bcl6 is also expressed by germinal center B-cells that have a pro-atherogenic role [44,46]. Nonetheless, this study [23•] strongly suggests that cholesterol accumulation in Tregs compromises Treg function and enhances atherogenesis. This outcome is different from the mice with T-cell Abcg1 deficiency that showed cholesterol accumulation and increased Tregs. This may be due to a higher level of membrane cholesterol accumulation in T-cells from Apoe−/− mice fed a WTD than in WTD-fed Ldlr−/− mice with T-cell Abcg1 deficiency, simply because in the setting of T-cell Abcg1 deficiency cholesterol esters accumulate [35], which may not have been the case in Apoe−/− mice fed WTD.

Fig. 2
figure 2

Effects of cholesterol accumulation on regulatory T-cell (Treg) fate and on T-cell apoptosis and downstream effects on the production of interferon γ (IFNγ), inflammation, and atherosclerosis. Created with BioRender.com

Full size image

Interestingly, mice with T-cell Abca1 deficiency show a decrease in Tregs [47], attributed to increased Abcg1 expression [39], but T-cell Abca1 deficiency is athero-protective in Ldlr−/− mice fed WTD [39]. This athero-protective effect was attributed to a decrease in membrane cholesterol accumulation due to elevated Abcg1 expression, and a decrease in Tmemory effector cells that indeed may have a pro-atherogenic role [39]. In contrast, we recently found that combined T-cell Abca1/Abcg1 deficiency decreased Tmemory effector cells but did not affect atherosclerosis in young Ldlr−/− mice fed WTD, while decreasing atherosclerotic plaque size in Ldlr−/− mice fed a chow diet at 1 year of age [26••]. We attributed the latter to the higher number of T-cells in plaques of Ldlr−/− mice at 1 year of age than in young mice, and thus a more prominent role of T-cells in plaque formation in aged mice [26••]. Mechanistically, T-cell Abca1/Abcg1 deficiency increased T-cell apoptosis and, consequently, decreased IFNγ production, decreasing macrophage inflammation in lesions [26••] (Fig. 2). Even though Apoe−/− mice also show decreased T-cell IFNγ production after 20 weeks of WTD feeding, this does not compromise lesion growth [25•], presumably because pro-inflammatory effects of Apoe deficiency on other cell types, such as macrophages, are dominant. Apoe−/− mice fed WTD may resemble advanced CAD in humans [25•], and therefore these studies in Apoe−/− mice are most informative in providing mechanistic insights as to why T-cells in patients with advanced CAD lose their funtionality in terms of proliferation and IFNγ production, likely occurring downstream of increased T-cell apoptosis.

Conclusions and Future Directions

T-cell membrane cholesterol accumulation is key to T-cell proliferation and differentiation into cytotoxic T-cells both processes downstream of TCR signaling that are crucial to T-cell function [30••, 33•]. The exact mechanism for these findings is not yet clear. Membrane cholesterol accumulation may induce TCR clustering [31••], as such activating TCR signaling; however, studies employing artificial membranes have yielded conflicting data as to the role of membrane cholesterol in TCR signaling [48, 49], indicating that the exact mechanism remains to be elucidated.

During aging, T-cell numbers decline, and T-cells accumulate cholesterol [27, 28]. Cholesterol accumulation may induce T-cell apoptosis or T-cell exhaustion [26••, 41•], which both may contribute to the decrease in T-cell numbers.

The diminished T-cell functionality in terms of T-cell proliferation and IFNγ production in CAD patients may be the consequence of T-cell cholesterol accumulation [25•, 26••]. Similarly, cholesterol accumulation in Tregs of CVD patients may enhance differentiation into pro-atherogenic T-cell subsets [6, 9, 21, 22, 23•, 24] (Fig. 2). Although deficiency of T-cell cholesterol efflux pathways also increased T-cell apoptosis in atherosclerotic plaques [26••], it seems rather unlikely that high levels of cholesterol accumulation in T-cells from human atherosclerotic plaques have a similar effect. Even though the plaque environment is rich in cholesterol, plaques from human carotid endarterectomies show high numbers of T-cells [1••, 2••, 50••] that differentiate into specific T-cell subsets completely dependent on the local plaque environment [50••]. Triggers that regulate this differentiation remain to be determined. Recent single TCR sequencing studies suggest that atherosclerosis has an auto-immune component driven by autoreactive CD4+ T-cells [50••].

In conclusion, several findings, as summarized in Tables 1 and 2, indicate that T-cell membrane cholesterol accumulation is key to regulating the functionality of peripheral T-cells. This is particularly important in response to infections. Indeed, a lack of cholesterol synthesis in CD8+ T-cells resulted in an attenuated clonal T-cell expansion during viral infection [33•]. Excessive cholesterol accumulation compromises T-cell functionality by inducing T-cell apoptosis [26••]. This may contribute to the increase in T-cell apoptosis and impaired T-cell functionality in patients with advanced CAD [25•].