|Home | About | Journals | Submit | Contact Us | Français|
The presence of dendritic cells, antigen-presenting cells that link innate and adaptive immunity, is necessary to generate and maintain the production of antiphospholipid antibodies in response to exposed intracellular phospholipids on the outer surface of apoptotic cells. In turn, antiphospholipid antibodies enhance dendritic cell-induced inflammatory and proatherogenic responses in a number of conditions that are associated with accelerated atherosclerosis, including diabetes, chronic kidney disease, periodontal infections, and aging. While altering dendritic cells by modifying the ubiquitin-proteasome system enhances antiphospholipid antibody production and leads to development of accelerated atherosclerosis and autoimmune features, inducing tolerance by dendritic cell manipulation leads to decreased atherosclerosis and thrombosis. Therefore, further translational studies are needed to understand the interplay between dendritic cells and antiphospholipid antibodies, and to develop potential new therapies for antiphospholipid syndrome and atherosclerosis. Here we review current experimental and translational studies that have examined the role of dendritic cells in antiphospholipid antibody formation and in antiphospholipid-associated atherosclerosis and thrombosis.
Antiphospholipid antibodies (aPL Abs) are a heterogeneous group of Abs against phospholipids or phospholipid-binding proteins found not only in individuals with autoimmune diseases, but also in individuals without overt autoimmune diseases . Antiphospholipid syndrome (APS) is characterized by the presence of persistent aPL Abs associated with arterial and/or venous thrombosis and increased pregnancy morbidity [2; 3]. While not all aPL Abs are pathogenic, it has been well established that Abs against cardiolipin (aCL Abs) and particularly Abs against beta2 glycoprotein I (anti-β2-GPI Abs), a plasma protein that binds to negatively charged phospholipids, play an instrumental role in the pathogenesis of APS [4; 5; 6; 7]. While an association between aPL Abs and accelerated atherosclerosis with and without APS has also been proposed, the exact mechanisms are not well understood [8; 9; 10; 11; 12; 13; 14].
Atherosclerosis is a complex process that is initiated when low density lipoprotein (LDL) molecules are deposited in the arterial walls, where they become oxidized by reactive oxygen species or enzymes, such as myeloperoxidase or lipoxygenase. Oxidized LDL (oxLDL) particles are phagocytosed by macrophages and become foam cells that are subsequently deposited within arterial walls, where they release pro-inflammatory cytokines such as TNF-α and IL-1, matrix metalloproteinases and oxygen-activated radicals. In addition to macrophages, monocytes, neutrophils and dendritic cells (DCs) also contribute to endothelial activation and increased expression of adhesion molecules including E-selectin and VCAM-1, which leads to further immune system activation and plaque formation. Acute plaque rupture and thrombus formation can cause stroke or acute myocardial infarction [15; 16; 17; 18; 19; 20].
APS and atherosclerosis share several similar features: increased levels of Abs to oxLDL, phospholipids and heat shock proteins, endothelial dysfunction, platelet activation and thrombus formation, increased oxidative stress and increased immune cell activation (Figure 1). Several more potential contributions of aPL Abs in atherosclerosis have been proposed, including proatherogenic modification of LDL and high density lipoprotein (HDL), increased endothelial damage and immune system activation [8; 9; 10; 11; 13]. However, the epidemiologic evidence that the presence of aPL Abs may be a serologic marker or an independent risk factor for atherosclerosis is inconclusive [21; 22].
Here we review experimental and translational evidence suggesting that DCs may play an important role in the generation and maintenance of aPL Abs, and that illustrate a key connection between DCs, aPL Abs, innate and adaptive immunity and atherosclerosis.
DCs are antigen-presenting cells that link innate and adaptive immunity, and play an important role in the pathogenesis of atherosclerosis  and autoimmune diseases . Conventional DCs search peripheral tissues for pathogens, secrete cytokines, present antigen, and activate naive T-cells. In addition, conventional DCs play a role in inducing and maintaining central and peripheral self-tolerance and minimizing autoimmune reactions. Plasmacytoid DCs (pDCs) migrate from the blood into lymph nodes and predominantly modulate immune responses to viruses by expressing toll-like receptors (TLRs), especially TLR-7 and TLR-9, and producing type I interferons [24; 25; 26; 27].
DCs play a complex role in early and advanced atherosclerosis. DCs promote early atherosclerosis and inflammation primarily through macrophage activation and a relative shift toward Th1 responses. While Th1 responses are proatherogenic, Th2 responses have both proatherogenic and protective effects. Furthermore, in advanced atherosclerotic plaques, accumulation of pDCs contributes to plaque instability and endothelial damage . In addition, interactions between microbial pathogens and pDCs may link atherosclerosis to infections, explaining increased plaque vulnerability during acute infections .
The presence of apoptotic cells and DCs is needed to generate aPL Abs and to promote aPL-specific autoimmune responses. In intact cells, negatively charged phospholipids are not present on the outer cell surface and do not contribute to antibody formation. In inflammatory conditions such as autoimmune diseases and atherosclerosis, early endothelial damage in the arterial vessels can lead to the exposure of phospholipids to the outer cell surface and promote antibody formation [2; 29]. Indeed, several studies have demonstrated that apoptosis is an important initial step in formation of aPL Abs [29; 30; 31].
The initial evidence demonstrating the cooperative role of DCs and apoptotic cells in producing aPL Abs came from experiments showing that apoptotic cells were recognized by monoclonal Abs to β2-glycoprotein I (β2-GPI), and that DCs preferentially internalize apoptotic cells opsonized by anti-β2-GPI Abs. When apoptosis is induced in vitro in murine cells and analyzed with flow cytometry, apoptotic cells bound to anti-β2-GPI Abs are internalized more efficiently by DCs .
In contrast to necrotic cells, apoptotic cells usually produce only minimal immune responses and do not initiate inflammatory reactions. However, while exposure of anionic phospholipids serves as an anti-inflammatory and immunosuppressive signal, opsonization of apoptotic cells with aPL Abs skews phagocytosis toward inflammation. Presentation of cryptic epitopes from apoptotic cells in an inflammatory state such as atherosclerosis or an autoimmune disease may impair T cell tolerance, and increase survival of autoreactive T and B cells in the lymph nodes .
Similarly, the binding of β2-GPI to anionic phospholipids facilitates the processing and presentation of a cryptic epitope that activates pathogenic autoreactive T cells. When DCs and macrophages are pulsed with phospholipid-bound β2-GPI, an autoreactive T-cell response is induced, while β2-GPI alone or phospholipid alone do not. In addition, in vitro autoreactive T-cell responses were elicited when peripheral blood T cells from healthy individuals were stimulated with DCs pre-incubated with phospholipid-bound β2-GPI. Thus, activated β2-GPI-reactive T cells would subsequently stimulate B cells to produce pathogenic anti-β2-GPI Abs .
Additional evidence that DCs play an important role in developing Abs to β2-GPI came from a series of in vivo experiments. When NZB×NZWF mice were immunized with DCs that had macropinocytosed β2-GPI or phagocytosed apoptotic thymocytes that displayed β2-GPI, the mice developed anti-β2-GPI Abs and autoimmune features. However, apoptotic cells that were opsonized with β2-GPI did not induce anti-β2-GPI Abs or autoimmune features when DCs were not present .
Based on the data presented above, it is clear that the presence of DCs is important in aPL Ab formation. It would therefore follow that DCs are critical to the development of the clinical features of APS. Here we present evidence for an additional mechanism by which DCs may fulfill this role, demonstrating the synergistic association of aPL Abs with the Toll-like receptor (TLR) activation to promote pro-inflammatory functions of pDCs.
Stimulating human monocytes and pDCs in vitro with ligands of Toll-like receptor 7 (TLR7) and TLR8 leads to increased production of IL1-beta, an important mediator of inflammation associated with endothelial activation and damage. When human monoclonal IgG aPL Ab generated from a patient with APS is added, TLR7 mRNA expression is upregulated and production of IL1-beta is substantially increased, suggesting that the TLR7 signaling pathway synergistically increases IL-1beta production induced by aPL Abs in pDCs .
Further evidence of this synergistic effect is based on the observation that aPL Abs induce TLR7 mRNA expression in human pDCs in a dose-dependent manner. Prinz et al proposed a novel aPL Ab-dependent mechanism of TLR activation, whereby aPL Abs can accumulate in the endosomal compartment of pDCs and induce NADPH oxidation, most likely via interaction of these Abs with endosomal lysobisphosphatidic acid. NADPH oxidation stimulates activation of nuclear transcription factor NF-κB and a subsequent increase in TLR7 gene expression in mouse pDCs. Thus, aPL Abs sensitize pDCs to TLR7 ligands and promote increased production of IFN-alpha and TNF-alpha by pDCs in response to sub-threshold levels of aPL Abs . Notably, both of these cytokines play an important role in atherosclerosis and in autoimmunity: TNF-alpha is pathogenic in atherosclerosis  and in endothelial cell activation by anti-β2-GPI Abs in APS ; IFN-alpha is implicated in premature atherosclerosis in systemic lupus erythematosus (SLE) .
Although the evidence that β2-GPI generates strong proinflammatory and proatherogenic DC-mediated immune responses is convincing, the origin of these immunogenic properties remains elusive. One possible explanation of this phenomenon may involve oxidation of β2-GPI. In individuals with atherosclerosis, β2-GPI is found in carotid plaques , implying that oxidation of β2-GPI can take place in atherosclerotic vessels in a manner similar to LDL oxidation as described above. Incubation of oxidized β2-GPI with DCs from healthy human donors caused DCs to mature and initiate Th1 and Th2 responses, which are characterized by increased secretion of interleukin 12 (IL-12), IL-1beta, IL-6, IL-8, tumor necrosis factor alpha (TNF-alpha) and IL-10. However, DC maturation and NF-κB activation was prevented when DCs were incubated with β2-GPI that was first pretreated with the antioxidant alpha-tocopherol, suggesting that oxidation was an important step that potentially linked anti-β2-GPI Abs, DCs, and atherosclerosis. Oxidized β2-GPI has also been shown to generate neo/cryptic epitopes of β2-GPI, which caused maturation and activation of monocyte-derived DCs [41; 42].
The in vitro interaction of oxidized β2-GPI with DCs also results in interleukin-1 receptor associated kinase (IRAK) phosphorylation and NF-κB pathway activation . The NF-κB signaling pathway plays a central role in all stages of atherosclerosis, from early plaque formation to acute coronary syndromes (recently reviewed by Pamukcu et al ). Similarly, activation of the NF-κB signaling pathway leading to upregulation of tissue factor expression in monocytes and endothelial cells is a major component of pathogenesis in APS . Translational studies are needed to determine if physiologic oxidation of β2-GPI in individuals with atherosclerosis also leads to activation of the NF-κB signaling pathway in vivo.
Another proposed mechanism of DC activation involves nonenzymatic glycation of β2-GPI . Glycation is a nonenzymatic posttranslational modification in which a sugar molecule binds to a protein or a lipid molecule. Subsequent oxidation of these compounds generates advanced glycation end products (AGE). Deposition of AGE in the vessel wall may contribute to endothelial activation, inflammation, and plaque formation in various conditions including aging, diabetes, and chronic kidney disease [47; 48].
Glycated β2GPI has been shown to activate monocyte-derived immature DCs from healthy human donors in vitro. This, in turn, induces upregulation of the receptor for AGE (RAGE) on monocyte-derived DCs. When DCs are stimulated with AGE-β2GPI complexes, there is a shift in naive T lymphocyte responses toward both Th1 and Th2 responses. In addition, glycated β2GPI causes phenotypic and functional maturation of immature DCs that involve activation of p38 MAPK, ERK and NF-κB. Both p38 MAPK and ERK are involved in cardiovascular remodeling , while activation of p38 MAPK plays an important role in diabetes-associated cardiovascular disease (reviewed by Liu and Cao ). Evidence from these studies strongly suggests that in diseases associated with accelerated atherosclerosis such as diabetes and chronic kidney disease, as well as in the natural aging process, the vascular microenvironment can predispose local β2GPI to glycation and/or oxidation. These structural changes in β2GPI act as inflammatory stimuli that activate immunogenic DCs and alter tolerogenic DC function, resulting in autoreactive T-cell responses and the development of autoimmune features and atherosclerosis .
There is some evidence that DCs also serve as a link between aPL Abs and the development of atherosclerosis in individuals with chronic infections such as periodontal disease. Associations between various types of infections and increased production of aPL Abs with or without APS have been reported, suggesting that infectious agents may induce inflammatory responses and serve as the initial trigger of the production of Abs that cross-react with β2GPI . An association between periodontal disease and atherosclerosis has been demonstrated in several studies, although the evidence is controversial (as recently reviewed by Lockhart et al ). In particular, P. gingivalis infection has been shown to accelerate the development and progression of atherosclerosis in apoE-deficient mice . Patients with periodontal disease have been shown to have much higher levels of IgG and IgM aCL Abs in their serum samples compared with those from healthy controls. There was also a significant association between the presence of aPL Abs and increased levels of soluble adhesion molecules such as sICAM-1, sE-selectin and sVCAM-1 [54; 55], suggesting that these Abs are pathogenic and are associated with the endothelial activation and vascular inflammation seen in atherosclerosis. A similar process has been observed with aPL Abs from APS patients . Furthermore, significantly higher rates of spontaneous transformation of monocytes into DCs have been observed in monocyte cultures from people with periodontal disease compared with healthy controls .
Some bacterial pathogens contain antigens with peptide sequences similar to the TLRVYK sequence in β2GPI. These pathogens can induce cross-reactive aCL Abs that cause endothelial activation and pregnancy complications in mice . A search of the Swiss-Prot protein sequence database identified pathogens from several species of bacteria associated with periodontal disease with conserved TLRVYK sequences. Together, these studies suggest that cross-reactivity between bacterial pathogens and β2GPI leads to enhanced generation of anti-β2GPI Abs in individuals with periodontal infections . Another study demonstrated that aPL Abs induced by P. gingivalis and A. actinomycetemcomitans also cross react with oxidized LDL and form immune complexes. These immune complexes stimulate DCs and macrophages significantly more than bacteria or oxidized LDL alone . The association between various Abs, DCs, and periodontal disease was comprehensively examined in a recent review , which concluded that aPL Ab-mediated uptake of oxLDL or bacteria enhances DC maturation, NK-cell-IFN-γ responses and Th1 responses that sustain inflammation, thus providing an additional possible link between periodontal infections, aPL Abs and DCs.
While glycation, oxidation and bacterial infections represent environmental exposures that modify β2GPI and/or potentiate interactions between aPL and DCs, alterations in the ubiquitin pathway in DCs may explain why not all individuals develop aPL Abs, autoimmunity and atherosclerosis.
The ubiquitin-proteasome system is responsible for the nonlysosomal degradation of intracellular proteins via conjugation of these proteins with ubiquitin and formation of ubiquitin chains , and is thought to play an important role in autoimmunity and atherosclerosis. If ubiquitination is not terminated appropriately, either because of genetic alterations or because of an imbalance between ubiquitin conjugated products and proteasomal degradation, there is a shift towards vascular damage and autoimmunity [26; 60; 61].
The ubiquitin editing enzyme A20 (also known as TNF-α-induced protein [TNFAIP]-3) is a cytoplasmic protein that inhibits inflammatory responses by deubiquitinating intermediate proteins in the NF-κB signaling pathway [60; 61]. Associations between several single-nucleotide polymorphisms in the A20 genomic locus and various autoimmune diseases including SLE, rheumatoid arthritis and systemic sclerosis have been described previously . In a mouse model lacking A20 specifically in DCs (A20-deficient DCs), several abnormalities within A20-deficient DCs were observed, including activation of the classical NF-κB pathway and JNK phosphorylation and spontaneous maturation of these cells. Furthermore, A20-deficient DCs produce extremely high levels of pro-inflammatory cytokines that lead to increased B cell activation and differentiation into plasma cells. A20-deficient DCs are also associated with an increased T cell differentiation into Th1 and Th17 cells and a decrease in regulatory T cell (Treg) production compared with A20-sufficient DCs. Finally, mice lacking A20 in DCs developed high serum concentrations of IgG and IgA aCL Abs, along with auto Abs to double stranded DNA and RNP, and developed thrombocytopenia, recurrent abortions and other autoimmune features present in SLE and APS. These data strongly suggest that deregulation of the ubiquitin pathway leads to antibody formation and clinical features of APS .
While the above study did not evaluate the development of atherosclerosis in these mice, there have been several studies suggesting that the ubiquitin-proteasome system plays an important role in degradation of oxidized proteins, and that accelerated protein ubiquitination is associated with increased oxidative stress [63; 64]. However, high-levels of oxidative stress can also reduce proteasome activity. In unstable carotid plaques the presence of ubiquitin conjugates is significantly greater compared with stable plaques. The imbalance between an increase in ubiquitin conjugates and a decrease in proteasome activity results in cytotoxic protein aggregates in atherosclerotic plaques that become further oxidized and difficult to unfold or degrade .
In addition to promoting immune activation, DCs may also promote peripheral tolerance and decreased inflammation by inducing regulatory T-cell responses and T cell anergy following exposure to Il-10 [24; 25; 26; 27]. Therefore, DC manipulation to promote tolerance is a potential therapeutic target for treating APS and atherosclerosis. To test this theory in atherosclerosis-prone transgenic mice, DCs from these mice were pulsed with IL-10 and apolipoprotein B-100 (an LDL component) and then incubated with the murine spleen cells. As a result, the DCs induced decreased Th1 and Th2 responses and increased regulatory T cell responses. Furthermore, when hypercholesterolemic mice were injected with DCs that were pre-incubated with IL-10 and apolipoprotein B-100, there was a significant reduction in atherosclerotic plaques in the aorta, decreased CD4 T-cell deposition in the arterial wall and reduced signs of systemic inflammation [20; 66; 67].
Tolerogenic DCs may also induce tolerance to β2-GPI in effector/memory T cells from patients with APS. Human monocyte-derived DCs from 20 APS patients that were treated with IL-10 and transforming growth factor β-1 decreased proliferation of β2-GPI-specific effector/memory CD4 T-cells. Thus, tolerogenic DCs can induce β2-GPI -specific unresponsiveness in APS, perhaps by inducing apoptosis of autoreactive T cells or promoting generation of regulatory T cells .
We reviewed here the evidence that DCs play an important role in sustained production of aPL Abs triggered by endothelial damage in autoimmune diseases and in atherosclerosis. In turn, aPL Abs enhance DC-induced inflammatory responses by working synergistically with TLR and activating the NF-κB signaling pathway, which is important for the pathogenesis of both APS and atherosclerosis. Interactions between DCs and aPL Abs may be enhanced by various mechanisms including glycation, oxidation, and cross-reactivity with bacterial pathogens and oxLDL, providing a possible link between DCs, aPL Abs, APS and atherosclerosis. Genetic or pharmacologic alterations of DCs modulate interactions between DC and aPL Abs, and can either enhance or decrease the development of autoimmune features and accelerated atherosclerosis. However, additional translational studies are needed to understand the significance of these mechanisms in human disease. Further experimental studies should also evaluate the role of other targets of aPL Abs (i.e. phosphatidylserine, phosphatidylethanolamine, lysobisphosphatidic acid, and annexin A5) in atherosclerosis.