Facts:

  • Apoptotic cells are potently immunosuppressive, eliciting a repertoire of immunomodulatory responses termed 'innate apoptotic immunity' (IAI).

  • IAI is triggered upon recognition of protein determinants externalized on the apoptotic cell surface, independent of phagocytosis.

  • IAI is exploited by pathogens to modulate host immunity, thereby enhancing their pathogenicity.

  • Pathogens use distinct strategies for the exploitation of IAI. Among these are apoptotic subversion, apoptotic mimicry, and death reprogramming.

Open Questions:

  • How widespread is the exploitation of IAI among the varied strategies of pathogenesis used by different pathogens?

  • Are other mechanisms (beyond apoptotic subversion, apoptotic mimicry, and death reprogramming) used by pathogens to exploit IAI?

  • What are the necessary and sufficient conditions (including expression of apoptotic-like determinants) that pathogens use for triggering IAI?

Infections by pathogenic microorganisms elicit host immune responses, which crucially limit those infections. Host innate immunity (especially including inflammation) is responsible typically for initial responses. At the same time, pathogens use various strategies to evade host immunity. Indeed, host immunity often is compromised by infection and associated stresses (see, for example, Lederer et al.1). Among specific strategies of immune evasion used by pathogens, some involve the exploitation of the potent immunosuppressive responses elicited normally by apoptotic cells.

Cells die physiologically throughout life in metazoans. This physiological process (apoptosis) serves to eliminate cells that have become functionally inappropriate, and assures their rapid clearance in a manner that does not elicit inflammatory and other immune responses.2, 3 During the process of apoptosis, new determinants for recognition appear on the dying cell surface.4, 5, 6 Recognition-dependent signaling leads both to the rapid engulfment and clearance of the apoptotic corpse,7 and to a profound repertoire of suppressive immune (especially anti-inflammatory) responses, which we have termed 'innate apoptotic immunity' (IAI8).

Apoptotic cell clearance occurs via a process of stimulated phagocytosis (macropinocytosis), which is characterized by the uptake of fluid-phase markers (such as Lucifer Yellow) and its sensitivity to inhibition by amiloride.7 The exposure of phosphatidylserine (PS), an anionic phospholipid cloistered normally in the inner leaflet of the plasma membrane of viable cells and externalized during the process of apoptosis, serves as a determinant for this engulfment process.4 Blockade of apoptotic cell engulfment can be affected by the masking of PS through the pre-binding of annexin V to exposed PS on the apoptotic cell surface.9 Immune pathologies attributed to the failure to promptly clear apoptotic cells, especially including chronic inflammation and autoimmunity characteristic of systemic lupus erythematosus and rheumatic diseases (see Savill et al.6 and Birge et al.8 and citations therein) may rather reflect deficiencies in apoptotic cell interactions and IAI.

IAI is triggered upon apoptotic recognition, independently of engulfment. Indeed, non-phagocytic cells recognize and respond to apoptotic corpses as well as phagocytic responders.5, 10 The apoptotic suppression of inflammation is exerted principally on the level of transcription,5 and involves both the repression of the expression of genes encoding pro-inflammatory cytokines (such as tumor necrosis factor-α (TNFα) and interleukin (IL)-6), and the induction of gene expression for anti-inflammatory cytokines (such as IL-10). Primary elements of this extensive repertoire are listed in Box 1. In contrast to its role in engulfment,4 PS is not a determinant for the recognition-dependent initiation of IAI.10, 11 The cell surface determinant(s) involved in initiating IAI are not masked by annexin V, rather, they are protease-sensitive.12 In addition, they are evolutionarily conserved, resident in all cells prior to cell death, and become surface-exposed (and susceptible to proteolytic digestion) during the process of apoptosis. We have referred to these characteristics as SUPER (surface-exposed (during apoptotic cell death), ubiquitously expressed (on all apoptotic cells), protease-sensitive, evolutionarily conserved, and resident normally in viable cells).12 (Subsequent responses, including the production of anti-inflammatory factors and cytokines (such as transforming growth factor-β; TGFβ), which may maintain and/or enhance IAI, are linked to PS-dependent apoptotic cell internalization13, 14, 15.)

Pathogens appear to exploit aspects of the host response elicited by apoptotic cells as a means of enhancing their pathogenic life cycle. Such apoptotic mimicry has been characterized rigorously with several viruses16, 17 and trypanosomatids.18, 19, 20 For example, viral internalization and the suppression of host immune responsiveness associated with vaccinia and hepatitis B viruses appear to follow apoptotic-like mechanisms of PS-dependent macropinocytosis and anti-inflammatory cytokine induction. Similarly, the internalization of the amastigotes from different species of Leishmania parasitic protists into mammalian macrophage hosts has been shown to occur via an amiloride-sensitive process, dependent on PS exposure and independent of immunoglobulin Fc receptors.21 Trypanosomatid parasites such as Leishmania alternate between a free-living, flagellated promastigote form in insect vectors and an intracellular, unflaggelated amastigote form in infected mammalian hosts. Just as the anti-inflammatory milieu elicited by apoptotic cells enhances leishmanial growth,22 Leishmania themselves also lead to the induction of anti-inflammatory cytokines in infected mammalian hosts.23 Among these, it is striking that TGFβ, in addition to suppressing host inflammatory responses, also serves as a stimulus for leishmanial growth.24 Especially under conditions of chronic Leishmania infection, systemic cytokine profiles in infected individuals resemble the apoptotic-like anti-inflammatory state.23 However, the specific triggering mechanism for anti-inflammatory cytokine expression has not been elucidated.

Several molecules that function intracellularly as enzymes involved in glycolysis, especially α-enolase and glutaraldehyde 3-phosphate dehydrogenase (GAPDH), become externalized on the surface of apoptotic cells and have been identified among the conserved, protease-sensitive apoptotic cell surface determinants responsible for apoptotic recognition and the triggering of IAI.12 The complete glycolytic pathway is delineated in Figure 1, and the glycolytic enzyme molecules that are externalized apoptotically are indicated. These glycolytic enzyme molecules exist among a surprisingly extensive array of cellular components, representing autoantigens, that are exposed on the surface of apoptotic cells25 it is likely that the absence of autoimmune reactivity normally is established and maintained through IAI.3 Biochemical analyses indicate that the externalized glycolytic enzyme molecules, including α-enolase and GAPDH, exist as peripheral membrane proteins in multimeric complexes (Palasiewicz, Pattabiraman, and Ucker, unpublished data).

Figure 1
figure 1

Apoptotic externalization of glycolytic enzyme molecules. Many of the protein species involved as enzymes in the intracellular glycolytic pathway of viable cells are enriched among membrane proteins of apoptotic cells and are exposed specifically on the apoptotic cell surface (shown in filled boxes). The externalization of these glycolytic enzyme molecules is a common early event during the process of apoptotic cell death. These externalized molecules fulfill SUPER criteria and are linked to the triggering of IAI (adapted from Ucker et al.12)

Full size image

The surface exposure of glycolytic enzymes has been noted previously with a variety of enteric bacteria and pathogens.26, 27, 28, 29, 30, 31, 32, 33 Independent studies also have implicated externalized α-enolase and GAPDH on mammalian and pathogen surfaces as sites for plasminogen binding and subsequent activation by host or pathogen activities,26, 27, 28, 30, 31, 34, 35, 36 and we have demonstrated that externalized glycolytic enzymes on the apoptotic cell surface, including α-enolase, also are sites for plasminogen binding.12 Although enhanced virulence has been attributed to the presumptive augmentation in mobility through matrix or fibrin clots associated with localized plasminogen activation following binding on the pathogen surface,37, 38 no compelling physiological rationale exists for plasminogen binding on non-invasive bacteria and apoptotic cells. We have argued that the exposure of glycolytic enzymes on microorganisms reflects an exploitation of IAI through apoptotic mimicry, and that consequent immune suppression facilitates pathogenesis, as well as commensalism.12

Some microbial pathogens, such as the food-borne bacterium Listeria monocytogenes,39 trigger host cell death directly. Although several insightful studies have implicated a functionally significant role for the involvement of apoptotic cells in instances of pathogenesis such as these,40, 41 the broader complexity and diversity of apoptotic and apoptotic-like events involved in enhancing microbial pathogenesis has not previously been appreciated fully.

Apoptotic Subversion

In the case of Listeria monocytogenes, a virulent and lethal facultative intracellular bacterial pathogen,39, 42 the induction of host cell apoptosis appears to be important for bacterial pathogenicity. Listeria monocytogenes long has been known to trigger the apoptotic death of mammalian cells. Although lymphocytes are not critical for the in vivo replication and pathogenesis of Listeria monocytogenes, they are extensively induced to undergo apoptotic death during Listeria monocytogenes infection.41 This has been of interest within the field of cell death, especially because of the involvement of a bacterial pore-forming molecule (listeriolysin O43), inflammatory cytokines, including Type I (especially β) Interferon (which potentiates listeriolysin O susceptibility44), and the relation to pyroptosis.45 It had not been clear, however, that host cell death was of relevance to the life cycle of the bacterium as a mammalian pathogen.

Previous work41 had suggested that Listeria monocytogenes-induced apoptotic lymphocyte cell death serves to facilitate pathogenicity. Mice lacking lymphocytes (and consequent apoptotic lymphocytes induced during infection) were found to be less susceptible to Listeria monocytogenes infection than were lymphocyte-replete, wild-type mice. The introduction of normal lymphocyte populations into genetically deficient animals restored pathogen susceptibility to normal (elevated) levels. These results, at first glance, seem paradoxical: lymphocytes are effectors of adaptive immunity and lymphocyte-deficient animals are immunocompromised generally. However, in light of the profound immunosuppressive responses elicited by apoptotic cells, it is reasonable to posit that the apoptotic lymphocytes induced by Listeria monocytogenes suppress host innate immune responses that would be directed against the bacterial infection (as modeled in Figure 2). Indeed, we have confirmed that exogenous apoptotic (and not necrotic) lymphocytes suppress host inflammatory responses and enhance the growth and pathogenicity of the bacterium in vivo (Pattabiraman, Palasiewicz, Visvabharathy, Freitag, and Ucker (2016), in preparation).

Figure 2
figure 2

Subversion of apoptosis and IAI by Listeria monocytogenes. A simplified model schematizes the competing, antagonistic responses elicited in the host upon infection by Listeria monocytogenes. The induction of host cell apoptosis by Listeria monocytogenes in the first phase of infection (Left panel) antagonizes pro-inflammatory responses subsequently, leading to an anti-inflammatory phase (right panel) during which Listeria monocytogenes replication and pathogenesis are enhanced. Left panel: Listeria monocytogenes secretes a pore-forming, apoptosis-inducing molecule, Listeriolysin O (LLO66). LLO is represented as small hook-shaped symbols, the long arm of which being the membrane-insinuating region, containing the cholesterol-binding motif. In addition to eliciting innate immune responses (such as the release of pro-inflammatory cytokines) via the triggering of extracellularly exposed innate immune receptors, such toll-like receptor 2 (TLR2, which senses bacterial lipoteichoic acid67), Listeria monocytogenes triggers intracellular innate immune receptors, such as Stimulator of Interferon Genes (STING; which senses released cyclic di-adenosine monophosphate68), stimulating the production and secretion of type I interferons (especially interferon-β). Extracellularly exposed and intracellular innate immune receptors are represented as simple structures with a concave binding pocket. The observed requirement for type I interferon (and the corresponding type I interferon receptor (IFN-αβR)) in the lymphocyte-dependent enhancement of Listeria monocytogenes pathogenesis43 reflects the ability of this family of cytokines to sensitize lymphocytes to LLO-dependent apoptosis.44 Right panel: Recognition by macrophages (among other cell types) of the apoptotic host lymphocytes induced to die upon Listeria monocytogenes infection (see left panel) via a putative innate apoptotic receptor leads to anti-inflammatory responses characteristic of IAI, including the repression of pro-inflammatory cytokine expression (Box 1). The induction of anti-inflammatory cytokine expression, which also ensues (Box 1), is not pictured in this figure. The innate apoptotic receptor is represented as a hypothetical heterodimeric structure

Full size image

Independent studies with a sepsis model of bacterial pathogenicity ('cecal ligation and puncture') also indicate that the induction of apoptotic cell death endogenously, or the introduction in vivo of exogenous apoptotic cells, furthers pathogenicity.40, 46 Apoptotic cells have been shown to exacerbate the proliferation of Trypanosome parasites, as well.47 These varied cases illustrate the subversion of apoptosis and apoptotic immunity in serving to enhance microbial pathogenesis.

Apoptotic Mimicry

The protist parasite Leishmania amazonensis exhibits what is perhaps the most revealing examples of apoptotic mimicry, where apoptotic-like PS exposure facilitates infective internalization21 and glycolytic enzyme exposure serves in host immune suppression. It is, as yet, unclear whether any one externalized molecule is sufficient for the triggering of IAI or whether multiple proteins necessarily function in concert for this purpose. Consequently, the detailed significance for IAI of the exposure of multiple glycolytic enzymes in mammalian cells and, correspondingly, how the process of protein externalization during apoptotic cell death functions selectively remain unresolved.

The biology of trypanosomatids presents a provocative context in which to explore this issue. The subcellular compartmentalization of glycolysis in trypanosomatids is fundamentally different than in all other eukaryotes, where glycolysis ensues cytoplasmically. In trypanosomatids, the majority of glycolytic enzymes are contained in unusual peroxisome-like organelles termed 'glycosomes'; only the terminal three glycolytic enzymes are cytosolic48, 49 (see Figure 3). Thus in trypanosomatids, GAPDH and enolase exist in distinct intracellular locales and are unassociated.

Figure 3
figure 3

Glycosomal localization of glycolytic enzymes. Enzymes and intermediates of glycolysis that are found within and without the glycosome48, 49 are diagrammed here. Glycosomes are intracellular peroxisome-like organelles of trypanosomatids, constrained by a single lipid bilayer membrane (represented by a double solid line), within which the proximal steps of glycolysis ensue. Glycosomal contents vary during the heteroxenous trypanosomatid life cycle, and other metabolic processes (including glycerol catabolism) also are localized within.49 Intraglycosomal glycolytic enzymes are shown as darker (purple) boxes; extraglycosomal (cytosolic) glycolytic enzymes are shown as lighter (blue) boxes. Filled boxes denote those glycolytic enzymes whose mammalian homologs are externalized on the apoptotic cell surface (see Figure 1)

Full size image

Glycolytic enzyme molecules have been reported to be externalized by Leishmania species and other trypanosomatids.50, 51 (The functional significance of plasminogen binding with respect to Leishmania pathogenesis has not been established definitively.52) Proteomic analysis indicates that enolase is a predominant surface-exposed Leishmania molecule.53 As in the case of glycolytic enzyme molecule exposure associated with apoptosis, externalized Leishmania enolase has no enzymatic activity.12, 50, 53 In this case, enolase appears to be regulated by posttranslational modification (dependent on oligopeptidase B53).

We have found that the infective form of the protist parasite Leishmania amazonensis externalizes extraglycosomal enolase on its cell surface in the absence of intraglycosomal glycolytic enzyme molecules (Wanderley, Gomes, Vieira, Barcinski, Meyer Fernandez, and Ucker (2016), in preparation). This is consistent with the lack of association of glycosomal and non-glycosomal enzymes reported in Trypanosoma cruzi, in which the externalization of extraglycosomal enolase50 appears to occur without the involvement of intraglycosomal triosephosphate isomerase.54 Beyond the evolutionary significance of enolase externalization, this may suggest that other glycolytic enzyme molecules are dispensable for the triggering of IAI.

This work, which documents the direct and extensive apoptotic mimicry accomplished by these protist pathogens, reinforces the role of glycolytic enzyme externalization in the triggering of IAI. Moreover, the evolutionarily distant trypanosomatid biology provides a fortuitous context with which to refine our view of essential IAI determinants.

Reprogrammed Cell Death

A growing list of microbes has been recognized that exploit apoptosis and IAI, including both Gram-positive and Gram-negative bacteria as well as non-bacterial pathogens.18, 21, 41, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64 Viruses exploit apoptotic processes as well, especially reducing the death of virally infected cells to facilitate intracellular viral replication. The viral reprogramming of cell death, in contrast, reflects a distinct approach.

As with many viral types, adenoviruses induce cell death in infected cells. These viruses encode a death-inhibitory gene product, which prevents rapid death, promotes viral replication, and results ultimately in a seemingly non-apoptotic form of cell death. The death-inhibitory protein, functionally equivalent to mammalian death-inhibitory Bcl-2, is the 19 kDa product of the adenovirus E1B gene (E1B 19 K).

Remarkably, the immunosuppressive activities of cells dying as a result of adenovirus infections were found to be exactly the opposite of 'typical' apoptotic and necrotic cells.65 That is, apoptotic-like cells, dying as a result of infection with an E1B 19 K-deficient adenovirus, were not immunosuppressive, and non-apoptotic cells, dying as a result of wild-type adenoviral infection, were immunosuppressive. This 'reversed' behavior was attributable genetically to E1B 19 K. That authentic Bcl-2 could substitute for E1B 19 K, but that it does not program such a reversed behavior in the context of uninfected mammalian cells suggests the additional contribution of other adenoviral gene(s).

It is clear that the inhibition of infection-induced death, as well as death reprogramming and associated immunomodulation, are beneficial to adenoviral pathogenicity.65 The complete genetic basis of adenoviral cell death reprogramming and its mechanism remain to be elucidated. Similarly, whether other viruses direct cell death reprogramming remains unexplored.

Conclusion

Whereas cell death once was considered to provide the answer to the issue of how the fate of cells that become inappropriate is resolved, we now understand that, more than cell death itself, it is the interaction of those dead cells with neighboring and other viable cells (leading to their clearance, the attenuation of inflammation, and tissue homeostasis generally (including wound healing)) that is determinative of resolution. Dysfunctions in those interactions are associated with immune (especially autoimmune and inflammatory) pathologies and neurodegenerative diseases. In this context, it may not be surprising that some microbes subvert these processes and enhance pathogenesis.

Although the above cases highlight distinct means used by microbial pathogens to exploit IAI and enhance pathogenicity, it is not clear that any single manipulative pathway ever is utilized in isolation. Microbial pathogenesis involves varied strategies including the utilization of a complexity of targeted apoptosis-specific and apoptotic-related manipulations. We speculate, as well, that apoptotic mimicry by indigenous microbes engages IAI locally and may facilitate commensalism. A detailed understanding of these modes of pathogenic exploitation of IAI is of profound importance for a more comprehensive appreciation of mechanisms of pathogenicity, and is likely to be particularly insightful for a fuller understanding of IAI as well. In particular, the molecular dissection of the mechanisms by which the pathogenic exploitation of IAI is affected holds promise for identifying novel targets for therapeutic intervention to enhance anti-microbial immune responsiveness.