Apoptotic Cell-Derived Extracellular Vesicles: More Than Just Debris

1There are three main types of EVs formed by a cell, namely exosomes, microvesicles, and apoptotic bodies (ApoBDs). These three types of EVs vary in size, content, and mechanism of formation (Figure 1). To date, exosomes and microvesicles generated from healthy cells are more extensively characterized and the formation of these EVs is key in mediating intercellular communication and immune regulation. Exosomes and microvesicles have been shown to play an important role in processes including antigen presentation, immune suppression, antitumor immunity, and autoimmunity. This has been the subject of many reviews, highlighting how EVs modulate immune responses by a myriad of mechanisms. Of particular interest is how the contents of exosomes and microvesicles enable them to regulate immune cell functions. Notably, these EVs can exhibit immune activating or immune suppressing properties depending on the specific circumstances. For example, exosomes have been shown to either activate or dampen the overall cytokine response through regulation of gene expression in monocytes and release of soluble cytokine receptors, respectively. Exosomes derived from dendritic cells, B lymphocytes, and tumor cells have also been shown to regulate immunological memory through the surface expression of antigen-presenting MHC I and MHC II molecules, and subsequently eliciting T cell activation and maturation. Exosomes can also play a role in cross-presentation pathways and have been shown to promote dendritic cell activation and maturation. Furthermore, microvesicles can modulate immune responses by transporting cytokines such as IL-1β and proinflammatory microRNAs.

2Both exosomes and microvesicles are generally described as EVs released from healthy cells, however, dying cells can also release a variety of EVs, broadly known as apoptotic cell-derived EVs (ApoEVs) (Figure 1). Subtypes of ApoEVs include large membrane-bound vesicles like ApoBDs or smaller apoptotic microvesicles (ApoMVs), both of which are described in detail below. While it has been well established that EVs can exhibit immunomodulatory effects, most studies have focused on EVs released from healthy cells, with EVs released from dying cells largely understudied. Nevertheless, a number of studies have suggested that ApoEVs have similar functional importance as EVs released from healthy cells. ApoEV formation has two key proposed functions: (a) aiding apoptotic cell clearance and (b) means of intercellular communication, both of which have implications in immune regulation. Many cells in the body are constantly undergoing apoptosis, and while a large portion of these are healthy cells undergoing normal turnover, apoptosis also occurs in many immunological and disease settings including inflammation, infection, autoimmunity, and cancer. Here, we discuss how ApoEVs may act as an immunomodulatory mechanism for apoptotic cells.

3As first described by Kerr et al., during apoptosis a cell undergoes a series of morphological changes resulting in the dismantling of the dying cell. Recently, disassembly of the apoptotic cell is categorized into three distinct morphological steps, namely apoptotic membrane blebbing, thin membrane protrusion formation, and ultimately generation of ApoBDs that are generally defined as 1–5 µm in diameter (Figure 1). While less is known about the mechanisms driving the formation of ApoBDs compared to other types of EVs, recent studies suggest that it is a highly regulated process and has been reviewed in detail. Besides ApoBDs, cells can also release smaller EVs such as ApoMVs (<1 μm in diameter) during the progression of apoptosis, possibly through membrane budding. However, molecular regulators of ApoMVs formation are not well defined.

4It is important to note that in the literature there are striking discrepancies in the characterization and isolation of ApoEVs. Aside from size, currently there are no well-defined criteria to distinguish ApoBDs from other ApoEVs, in particular ApoMVs. Although proteomic studies comparing these ApoEV subtypes have been performed, clear standard for the characterization and purification of ApoEV subtypes is lacking (highlighted in Table 1). These discrepancies make it difficult to draw accurate conclusions regarding the functions of ApoEVs and caution should be taken when interpreting data involving ApoEVs. Taking these limitations into consideration, here we use the general term ApoEVs where it is unclear which subtype of ApoEVs is presented in a given study, and ApoBDs and ApoMVs to describe vesicles predominantly >1–5 µm and <1 μm in diameter, respectively.

5It has been well established that apoptotic cells coordinate a number of intercellular signals to aid in their detection and removal, and these signals are critical to ensure the immunologically silent characteristic of apoptosis. Defective apoptotic cell clearance has been identified as a key contributing factor to autoimmune disease, whereby cells that are unable to be cleared efficiently eventually undergo secondary necrosis and release potentially damaging proinflammatory contents and autoantigens. There has been mounting evidence suggesting that the release of ApoEVs during apoptosis can promote clearance of apoptotic material, with the mechanism underpinning this process discussed below.

6For efficient apoptotic cell clearance, the recruitment of phagocytic cells toward the site of cell death is essential. To this end, apoptotic cells can release molecular factors known as “find-me” signals to attract phagocytes. Traditional “find-me” signals include the release soluble factors such as ATP, UTP, CX3CL1/fractalkine, and lysophosphatidylcholine. However, there is also evidence of ApoEV-associated “find-me” signals being released from apoptotic cells.

7While few studies elucidated the specific molecules involved in ApoEV-mediated recruitment of phagocytes, they have demonstrated ApoEVs to exhibit chemoattractive properties. Nevertheless, the “find-me” signal CX3CL1/fractalkine was found to be released from apoptotic B lymphocytes in association with ApoMVs and the chemoattractive molecule ICAM-3 was associated with ApoEVs generated from apoptotic lymphoma cells. It is interesting to note that ApoEVs appear to have preferential recruitment of macrophages but not neutrophils. Such selective recruitment of different phagocytes by ApoEVs may be related to the subtype of ApoEVs being released by the apoptotic cell, in which one study comparing endothelial cell-derived ApoEVs of different size showed that only larger ApoEVs (1–3 µm in diameter, corresponding to ApoBDs) promoted neutrophil migration, whereas smaller EVs (<1 μm, corresponding to ApoMVs) could not. It is worth noting that this phenomenon has been observed in vitro as well as in vivo, where intraperitoneal administration of ApoBDs in a mouse model stimulated neutrophil infiltration. Thus, different subtype of ApoEVs may have distinct functions in apoptotic cell clearance.

8Besides attracting phagocytes, formation of ApoEVs, in particular cell fragmentation into ApoBDs has been suggested to enhance removal of apoptotic material, an effect probably attributed to the size of ApoBDs being smaller bite-size pieces that can easily be engulfed by phagocytes. Supporting this concept, it has been shown that dendritic cells can more readily engulf smaller ApoBDs than whole apoptotic cells. Furthermore, cells undergoing apoptotic cell disassembly and therefore producing ApoEVs are more efficiently engulfed by macrophages.

9It should be noted that exposure of “eat-me” signals, such as phosphatidylserine (PtdSer), on apoptotic cells label them for clearance by phagocytes. Likewise, ApoEVs can also expose “eat-me” signals like PtdSer on their surface and be recognized by macrophages for removal via phagocytic receptors such as CD36. Interestingly, ApoEVs can also expose ICAM-3, and specific sialylated and glycosylated ligands on its surface to trigger recognition and engulfment by macrophages.

10An important immunomodulatory property of EVs is their ability to aid antigen presentation, a fundamental process for adaptive immunity. As mentioned above, EVs like exosomes have been shown to mediate antigen presentation via direct and cross-presentation mechanisms. Similarly, ApoEVs can also regulate antigen presentation via these mechanisms in a number of disease settings including autoimmunity, antimicrobial immune responses, and organ/transplant rejection. Direct antigen presentation involves the vesicle carrying surface MHC molecules in complex with antigenic peptide to directly interact with naïve T cells. ApoEVs generated from dendritic cells and B16-F1 melanoma cells carried MHC II molecules suggesting the potential of ApoEVs to activate CD4+ T cells. Alternatively, cross-presentation relies on the vesicle transporting the antigen to professional antigen-presenting cells, in particular dendritic cells, for antigen processing and presentation to CD8+ T cells. In one study, ApoEVs generated from Mycobacterium tuberculosis-infected mouse macrophages were found to transfer bacterial-derived antigens to dendritic cells, and subsequently activate naïve CD8+ T cells. Furthermore, engulfment of ApoEVs by dendritic cells has also been shown to modulate their antigen-presenting capabilities. ApoEVs generated from lymphoblasts were found to suppress immune responses by downregulating MHC II molecules on dendritic cells.

11While the mechanisms underpinning the ability of ApoEVs to modulate antigen presentation are diverse, it is clear that ApoEVs can contribute to the development of autoimmunity, and establishment of antitumor and antimicrobial immunity by regulating the antigen presentation process, as discussed further in detail below.

12As discussed above, impaired clearance of dying cells is a major factor contributing to the development of autoantibodies in autoimmune conditions. Although the formation of ApoEVs has been shown to promote apoptotic cell clearance and thus limiting the release on intracellular antigenic and proinflammatory contents, ApoEVs formation has also been proposed as a mechanism of facilitating the transport of autoantigens to antigen-presenting cells and drive autoimmunity. In particular, ApoEVs have been implicated in the development of systemic lupus erythematous (SLE), whereby autoantigens such as histone H2B can be translocated into ApoEVs during the early stages of apoptosis in HeLa cells via a microtubule driven mechanism. Lymphoblast-derived ApoEVs containing histone were also more readily engulfed by monocyte-derived phagocytes. Furthermore, ApoEV-associated autoantigens like DNA can bind directly to antinuclear antibodies, a common feature of autoimmune conditions. In addition to autoantigens associated with SLE, Sjögren’s syndrome nuclear autoantigens, such as hy1-RNA, are also detectable in both ApoMVs and ApoBDs generated from epithelial cells and can be transferred to dendritic cells via these ApoEVs.

13With most cancer treatments focusing on inducing apoptosis in tumor cells, it becomes important to consider how the release of ApoEVs from dying tumor cells will impact the immune response toward the tumor. Recently, it has been shown that ApoMVs derived from tumorigenic apoptotic melanoma cells can promote antitumor immunity, in which mice immunized with ApoMVs generated from B16-F1 cells following doxorubicin treatment were protected against subsequent tumor challenges. Importantly, the tumor antigen PMEL was also found in ApoMVs, supporting the concept that ApoMVs can facilitate the transport of tumor antigens to antigen-presenting cells to promote antitumor immunity. It is interesting to note that despite ApoMVs having a relatively lower quantity of the tumor antigen PMEL as compared to other EVs like exosomes, the antitumor protective effect of ApoMV immunization was greater, suggesting that ApoMVs may aid antigen presentation via a different mechanism as other EVs and were able to promote a more robust antitumor immune response. As discussed earlier, “eat-me” signals such as PtdSer are present on ApoEVs. Interestingly, another “eat-me” signal, calreticulin, that are exposed on certain apoptotic tumor cells can play a key role in promoting antitumor immunity through dendritic cells. Therefore, it would be of interest to determine whether calreticulin is present on ApoEVs and whether exposure of calreticulin is important for ApoEV-mediated antitumor immunity.

14In addition to the presentation of self-antigens, it is important to note that under conditions where infected cells undergo apoptosis, the resultant ApoEVs may also harbor antigens from the infectious agent. The transfer of microbial-derived antigens via ApoEVs to antigen presentation cells like dendritic cells have been shown to provide a protective effect for the host. For example, ApoEVs released from apoptotic macrophages infected with the M. tuberculosis can be engulfed by peripheral monocyte-derived and splenic dendritic cells, which could subsequently activate the engulfing dendritic cells to prime naïve CD4+ or CD8+ T cells. Significantly, ApoEVs generated from M. tuberculosis-infected cells were able to be used to vaccinate naïve animals and provided protection against tuberculosis infection, highlighting the potential use of ApoEVs as vaccines. While these studies focus on tuberculosis infection, ApoEVs could play an important role in regulating antimicrobial immunity against other pathogens, however, these remain underexplored.

15In addition to antigens, ApoEVs can harbor a variety of biomolecules that could directly modulate immune cells, most commonly via vesicle-associated cytokines or damage-associated molecular patterns (DAMPs), which could drive inflammation and dictate the immune cell responses. For example, proinflammatory cytokine IL-1α was detected in ApoBDs but not ApoMVs generated from endothelial cells induced to undergo apoptosis by prothrombic and hypoxic conditions in vitro. In a mouse model, administration of these endothelial cell-derived ApoBDs into the peritoneal cavity was able to induce production of neutrophil chemokines and promote neutrophil infiltration to drive sterile inflammation. Furthermore, an increase in IFN-α production by plasmacytoid dendritic cells in response to DNA in lymphoblast-derived ApoMVs was comparatively more pronounced than DNA isolated from whole cells. In this case, vesicle-associated DAMPs were responsible in promoting dendritic cell maturation, with the potential to promote damaging inflammation and possibly autoimmune conditions. Besides DNA, other DAMPs such as HMGB1 can also be found in ApoEVs derived from peripheral blood mononuclear cells and T cells.

16The potential protective effects of ApoEVs in infection was discussed earlier, however, ApoEVs have also been implicated in facilitating the spread of infection via two different mechanisms. First, ApoEVs generated from infected cells could modulate the immune response and makes it favorable for the progression of infection. ApoMVs generated during HIV infection were able to modulate the dendritic cells response via binding to the CD44 receptor, resulting in decrease cytokine production from dendritic cells and inhibition of their ability to prime T cells or natural killer cells. Second, ApoEVs could directly aid viral propagation by mediating the transfer of infectious virions to neighboring cells. Chikungunya virus was shown to induce apoptosis and the formation of ApoBDs in infected HeLa cells and blocking ApoBD formation by targeting the apoptotic cell disassembly process pharmacologically limited infection spreading to neighboring cells. Thus, although the formation of ApoEVs by infected cells could be beneficial for the host by facilitating the antigen presentation process, certain viruses may hijack ApoEVs to aid viral propagation.

17Overall, there is compelling evidence to support the importance of ApoEVs in immune modulation, and ApoEVs can play a significant role across many aspects of immunity and disease settings. Therefore, ApoEVs are more than just debris or by-products of apoptosis and should be considered as a key mechanism for apoptotic cells to communicate with surrounding cells. The ability of ApoEVs to either activate or dampen immune responses demonstrates the fine balance between the beneficial effects of ApoEV generation and the potentially damaging implications. However, as highlighted in this review, there are marked discrepancies in the characterization and isolation of ApoEVs, making it difficult to accurately define their functions. To progress the field, it is critical to identify suitable criteria to distinguish different subtypes of ApoEVs and develop better experimental systems to modulate ApoEV formation under physiologically relevant conditions.

18All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

19The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.