Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Immunity. Author manuscript; available in PMC 2009 Jul 18.
Published in final edited form as:
Immunity. 2008 Jul 18; 29(1): 21–32.
doi:  10.1016/j.immuni.2008.05.013
PMCID: PMC2704496



The mammalian immune system discriminates between modes of cell death, with necrosis often resulting in inflammation and adaptive immunity, while apoptosis tends to be anti-inflammatory, promoting immune tolerance. In many systems immune tolerance can be established through cross presentation of antigens derived from apoptotic cells via the MHC class I pathway to CD8+ T cells. We have examined the features of apoptosis responsible for tolerance to cell-mediated immune responses in vivo, specifically the roles of caspases and the mitochondria. Our results show that caspase activation targets the mitochondria to produce ROS, which are critical to tolerance induction by apoptotic cells. ROS oxidizes the potential danger signal HMGB1 released from dying cells, thereby neutralizing its stimulatory activity and promoting tolerance. Apoptotic cells failed to induce tolerance and instead stimulated immune responses when caspase-dependent ROS activity was prohibited by scavenging or by mutation of a mitochondrial caspase target, p75 NDUSF1. Similarly blocking sites of oxidation in HMGB1 prevented tolerance induction by apoptotic cells. These results suggest that caspase orchestrated mitochondrial events determine the impact of apoptotic cells on the immune response.


The impact of dying cells on the immune system depends on the way in which cells die, and the current perception is that necrotic cells act as “danger” signals while apoptotic cells are “silent”. For adaptive immune responses, many studies have documented the immunogenic activity of necrotic cells and the tolerogenic influence of apoptotic cells (Chen et al., 2001; Ferguson et al., 2002; Griffith et al., 2007; Griffith et al., 1996; Kurts et al., 1998; Sauter et al., 2000). Other studies have shown, however, that in some cases apoptotic cells can be immunogenic, and this may depend, at least in part, on the inducing stimulus (Ronchetti et al., 1999) (Tesniere et al., 2008). Understanding this dynamic depends on discovering the factors elicited by dying cells that mediate these effects.

Clearly, the effect of dying cells on the adaptive immune response traces to their effects on dendritic cells (DC). This is of particular importance as it is now clear that apoptotic cells are not passive participants. Antigens associated with engulfed dying cells are channeled into the exogenous pathway of class I MHC presentation for stimulation of CD8+ T cells (Blachere et al., 2005; Dudziak et al., 2007; Ferguson et al., 2002; Steinman et al., 2003), but apoptotic cells, for the most part, fail to stimulate presentation to CD4+ T cells (Griffith et al., 2007). Production of IL-10 (Gao et al., 1998; Tomimori et al., 2000) and/or TGFβ (Chen et al., 2001) by apoptotic lymphocytes has been noted, and while it is unlikely that all apoptotic cells produce these cytokines, it is possible that macrophages and/or other phagocytic cells may produce them in response to apoptotic bodies (Fadok et al., 1998; Ronchetti et al., 1999). One distinction between apoptotic and necrotic cells is the production of danger signals from the latter. For example, the DNA binding protein HMGB1 (high mobility group protein B1) has been reported to be preferentially released from necrotic but not apoptotic cells (Scaffidi et al., 2002), and potently acts on DC to promote immunity (Dumitriu et al., 2005). Several recent studies, however, have question the preferential release of HMGB1 from necrotic cells (Bell et al., 2006; Choi et al., 2004; Obeid et al., 2007a; Tian et al., 2007). Other mediators such as uric acid (Shi et al., 2003), calreticulin (Obeid et al., 2007a; Obeid et al., 2007b), and HSP70 (Millar et al., 2003) can also stimulate the immune response.

A major distinction between apoptosis and other forms of cell death is the involvement of caspase proteases (Martin and Green, 1995). These enzymes are activated by signaling pathways in the apoptotic cells and orchestrate apoptosis through their cleavage of specific substrates. While often not required for cell death per se (Chipuk and Green, 2005), they are important for many of the cellular events associated with apoptosis, including DNA fragmentation, membrane blebbing, and fragmentation of the cell into membrane-bound apoptotic bodies (Taylor et al., 2007). Importantly for our discussion, inhibition of caspases during cell death using pharmacologic agents (Ferguson et al., 2002) or viral proteins (Chen et al., 2006) converted the tolerogenic signals provided by apoptotic cells into immunogenic ones. Therefore, caspase substrates or some downstream effect of their cleavage are important for tolerance induction by apoptotic cells.

The mitochondrial pathway of apoptosis is induced by a wide variety of stimuli which engage members of the Bcl-2 protein family which affects permeablization of the outer membranes of the mitochondria in the cell. This mitochondrial outer membrane permeablization (MOMP) results in the release of cytochrome c into the cytosol triggering the activation of the executioner caspases-3 and -7 (Bonfoco et al., 1998). Activated caspases then feed back on the permeable mitochondria cleaving a mitochondrial substrate, NADH dehydrogenase Fe-S protein-1 (p75 NDUSF1), the 75 kDa subunit of respiratory complex I and component of the electron transport chain. As a result, electron transport halts and reactive oxygen species (ROS) are produced in abundance (Ricci et al., 2004). We had previously observed that BclXL, which blocks MOMP and the mitochondrial pathway of apoptosis, prevents tolerance induction by apoptotic lymphocytes, even when apoptosis was induced by a pathway (Fas/Fas-ligand) (Griffith et al., 1996) which does not require MOMP for apoptosis (Huang et al., 1999). This led us to speculate that some aspect of MOMP, in addition to caspase activation, may be necessary for the tolerogenic effects of apoptotic cells on the immune system.

Reactive oxygen species (ROS) produced by the mitochondria can be involved in cell death. These toxic compounds are normally detoxified by the cell; however, during apoptosis ROS produced in the mitochondria participate in the apoptotic process (Fleury et al., 2002). Recently it was shown that caspase activity against mitochondrial complex I of the electron transport system lead to increased ROS production during mitochondrial mediated cell death (Ricci et al., 2004). ROS can modify cellular components and destroy their functions (Ott et al., 2007) so we reasoned that ROS might also affect components of apoptotic cells involved in tolerance and immunity. In this paper we utilize a well established system of tolerance induction to examine the consequences of mitochondrial changes mediated by caspases. Our results show that blocking caspase activity at the level of mitochondrial complex I converts tolerogenic apoptotic cells into immunogens. We further demonstrate that the caspase dependent production of ROS during apoptosis is responsible for destroying immunogenic signals released from dying cells. Thus caspase mediated events lead to the destruction of potential danger signals that might stimulate unwanted immune responses.


Executioner Caspases are Required for Induction of Immune Tolerance by Apoptotic Cells

To study what it is about apoptotic cells that influences the adaptive immune response we employed a system we have used extensively (Battisto and Bloom, 1966; Ferguson et al., 2003; Griffith et al., 2007). Cells (syngeneic, allogeneic, or xenogeneic) are coupled with hapten (trinitrophenol, TNP) and injected intravenously (i.v.) or uncoupled apoptotic cells (see Supplemental Figure 1) are fed to hapten modified DC which in turn are injected i.v. To assess priming (immunogenicity), animals are challenged 5 days later in the footpad with the antigen. Alternatively, to assess tolerance induction (tolerogenicity), mice injected i.v. with apoptotic cells or DC fed apoptotic cells are immunized by s.c injection of antigen on day 2 and then challenged with antigen in the footpad after an additional 4 days. In both cases DTH is measured with a micrometer 1 day (24 hrs) following footpad challenge. The basic design of these systems is illustrated in Figure 1A. (Henceforth Figures using these systems are labeled “immunogenicity” or “tolerogenicity” for clarity.

Figure 1
Figure 1A. Experimental systems. Immunogenicity and tolerogenicity of apoptotic cells were measured using a well defined in vivo model. Immunogenicity test: TNP-coupled apoptotic cells (107) or TNP-DC fed apoptotic cells (106) were injected i.v. into ...

Studies by us have shown that treatment of dying cells with pharmacologic inhibitors of caspase activation prevented tolerance induction by apoptotic cells (Ferguson et al., 2002). To confirm and extend these studies as the basis for the current work, we employed mouse embryonic fibroblasts (MEF) either heterozygous (caspase-3+/−, -7+/−) or null (caspase-3−/−, -7−/−) for the executioner caspases-3 and -7. Although caspase-3−/−, -7−/− cells die in response to pro-apoptotic stimuli, they do not show the phenotypes associated with apoptosis, including loss of mitochondrial function (unpublished observations and (Lakhani et al., 2006)). The cells were induced to die by treatment with ultraviolet (UV) radiation (see methods), then fed to TNP conjugated DC (TNP-DC), which were then injected i.v. In the experiment shown in Figure 1B, injection of the TNP-DC that had engulfed dead, caspase-heterozygous MEF (caspase-3+/−, -7+/−) failed to induce immunity to the hapten. In contrast, TNP-DC that had engulfed dying MEF deficient in executioner caspases (caspase-3−/−, -7−/−) induced immunity as efficiently as did conventional immunization (Immune Control). Similarly, mice injected with antigen-coupled apoptotic splenocytes (spleen cells, γ-irrad) did not stimulate immunity, but γ-irradiated spleen cells treated with a caspase inhibitor (zVAD-fmk) stimulated potent immunity.

Animals injected as above with DC fed apoptotic MEF were also examined for tolerance induction following s.c immunization with antigen (Figure 1C). Animals treated by injection of DC that had engulfed apoptotic, caspase-heterozygous cells (caspase-3+/−, -7+/−) were tolerant to subsequent immunization, while those mice given TNP-DC that had engulfed UV irradiated caspase-deficient cells (caspase-3−/−, -7−/−) showed no tolerance. Thus activation of caspases during apoptosis converts dying cells from potent immunogens (Figure 1B) to inducers of tolerance (Figure 1C).

ROS Contributes to Tolerance Induction by Apoptotic Cells

One consequence of caspase activation that depends on MOMP is the production of ROS by the dying cells due to the disruption of electron transport (Ricci et al., 2003; Ricci et al., 2004). We tested if such ROS production might contribute to tolerance induction by apoptotic cells. Antigen-coupled splenocytes were induced to undergo apoptosis in the presence or absence of the ROS scavenger butylated hydrozyanisole (BHA). The cells were then injected i.v. and animals immunized s.c with antigen 2 days later. While BHA had no effect on the extent of cell death (data not shown), it completely prevented the induction of tolerance by the apoptotic cells (Figure 2A).

Figure 2
ROS and tolerance by apoptotic cells. A. C57Bl/6 mice were injected i.v. with TNP-coupled apoptotic cells (107) (spleen cells, γ-irrad) that were untreated or treated with 100μM BHA. B. C57Bl/6 mice were injected i.v. with TNP-coupled ...

While there are several ways in which ROS could influence cellular events, an interesting possibility might be that the oxidation of some product of the dying cells affects its ability to stimulate immune responses, thereby producing tolerance. To test this idea, antigen-coupled spleen cells were induced to die by apoptosis (γ-irrad) or necrosis (freeze-thaw, F/T) and were then treated with reducing or oxidizing agents before injection into mice. The animals were subsequently immunized s.c and tolerance assessed by antigenic challenge (tolerogenicity, see Figure 1A). In the experiment shown in Figure 2B, antigen-coupled apoptotic cells induced tolerance, while antigen-coupled necrotic cells did not. Remarkably, treatment of the apoptotic cells with a reducing agent, dithiothreitol (DTT), destroyed their ability to induce tolerance. Conversely, treatment of the necrotic cells with an oxidant (H2O2) made them tolerogenic.

To extend these observations, we employed mice transgenic for a membrane-bound ovalbumin expressed from the β-actin promoter (Ehst et al., 2003) as a source of dead cells. Splenocytes from Act-mOva mice were γ-irradiated (γ-irrad) to induce apoptosis, or freeze/thawed (F/T) to cause necrosis, and inject i.v. into wild-type mice. In some cases the dead cells were treated with DTT or H2O2 prior to injection. To assess tolerance, mice were immunized by s.c. injection of OVA in Complete Freund’s adjuvant and 7 days later challenged by footpad injection of the soluble OVA. DTH was assessed 24 hrs later. Tolerance to OVA was induced by apoptotic, but not necrotic Act-mOVA spleen cells (Figure 2C). However, the effect of apoptotic cells was blocked by treatment with DTT, while necrotic cells treated with H2O2 became tolerogenic for the associated antigen.

In each case, the treatments with H2O2 or DTT were performed after the majority of cells had died (6 hrs after irradiation, e.g. Supplemental Figure 1). Therefore, it is likely that the effects were on a molecule in the dead cells that is capable of affecting immune responses, rather than on a metabolic process in the dying cell. This is explored further below.

Caspase Cleavage of Electron Transport Complex I in Apoptotic Cells is Essential for Immune Tolerance

During apoptosis, active executioner caspases gain access to the mitochondrial intermembrane space as a consequence of MOMP and there cleave specific substrates. Among them is p75 NDUSF1, a component of complex I of the electron transport chain. As a result of this cleavage event, electron transport, mitochondrial transmembrane potential and mitochondrial structure are all disrupted, and ROS are produced. A single amino acid substitution in p75 (D to A; called p75D255A or p75DA) prevents this cleavage and sustains mitochondrial function in cells undergoing apoptosis (Ricci et al., 2004). We therefore expressed p75wt or mutant p75DA proteins in HeLa cells and subjected them to apoptosis induction by UV. ROS production was dramatically reduced in the cells expressing the mutant (p75DA) versus the wild-type p75 (p75wt; Figure 3A). The reduction in ROS in p75wt was confirmed as caspase dependent since treatment of UV irradiated p75wt with zVAD-fmk eliminated ROS production (Figure 3A, p75wt/zVAD). Expression of the mutant p75DA did not influence phosphatidylserine externalization, but did delay loss of plasma membrane integrity in the dying cells (Figure 3B). Significantly, UV-treated p75wt- and p75DA-expressing cells were engulfed by DC to the same extent after 2 hours co-incubation (p75wt, 36.9% vs. p75DA 32.1%; Figure 3C).

Figure 3
Mitochondrial changes and immune tolerance

We then examined the consequences of p75 cleavage by caspases on the immunological effects of the dying cells. HeLa expressing p75wt or p75DA were treated with UV to induce apoptosis and cultured with TNP-DC. The DC were then injected i.v. into mice and immunity was assessed by challenge 5 days later. While DC that had engulfed apoptotic cells expressing p75wt did not induce immunity to the associated hapten, those that had engulfed apoptotic cells expressing p75DA effectively primed an anti-hapten response (Figure 3D). Tolerance induction by DC that had engulfed apoptotic HeLa cells was then assessed (Figure 3E). Tolerance was effectively induced by DC that had engulfed apoptotic HeLa cells expressing the control p75wt. This tolerance was blocked if the cells were pre-treated with a pancaspase inhibitor, zVAD-fmk. However, no tolerance was induced by DC that had engulfed apoptotic cells expressing the noncleavable p75DA mutant. Cleavage of p75 by caspases therefore appears to be important for the induction of immune tolerance by apoptotic cells in this system.

While apoptotic cells expressing p75wt induced tolerance, a mixture of apoptotic p75wt- and p75DA-expressing apoptotic cells, fed to DC, failed to do so (Figure 3F). Therefore, it is likely that cleavage of p75 by caspases leads to inactivation of an immunostimulatory activity in the dying cells, resulting in tolerance induction. Based on our observations using oxidizing and reducing agents (Figure 2), it is likely that this inactivation is via ROS resulting from p75 cleavage.

Immunostimulatory HMGB1 is Released from Apoptotic Cells with intact mitochondria and low ROS

One of the immunostimulatory molecules implicated in the differential effects of apoptotic and necrotic cells on the immune system is HMGB1. While necrotic cells release this molecule upon death, apoptotic cells reportedly do not (Scaffidi et al., 2002). We speculated that the mitochondrial effects we observed might influence the release of HMGB1. We found, however, that supernatants of UV-treated HeLa expressing control p75wt or mutant p75DA released significant amounts of HMGB1 into the supernatants by 24 hrs (Figure 4A). At an earlier time point (6 hrs), the cells expressing p75DA released less HMGB1, which was not surprising given the delay in loss of plasma membrane integrity in these cells (see Figure 3B, and (Ricci et al., 2004)). However by 24 hrs significant amounts of HMGB1 were present in the supernatants of either cell type.

Figure 4
HMGB1 release blocks tolerance

Remarkably supernatants of the dying cells expressing p75DA had a dramatic effect on the ability of apoptotic cells to induce tolerance. Apoptotic splenocytes were co-cultured with TNP-DC, with or without supernatants from the immunogenic HeLa expressing p75DA taken 24 hrs after induction of apoptosis in these cells. As shown in Figure 4B, DC that had engulfed apoptotic cells induced immune tolerance. However, if the DC were incubated with supernatants from the apoptotic cells expressing p75DA (p75DAs), tolerance induction was abrogated. This effect was neutralized by addition to the supernatants of an anti-HMGB1 (αHMGB1) antibody but not a control antibody (IgG). Supernatants from apoptotic HeLa expressing p75wt (p75wts) had no such effect (Figure 4C). In fact depleted HMGB1 from supernatants of the p75DA-expressing cells (which destroyed the tolerance-abrogating activity) could not be restored by addition of supernatants from the p75wt-expressing cells (p75wts; Figure 4C). Therefore, while HMGB1 appears to be important in preventing tolerance induction, and is released from cells regardless of p75 status, it appeared that only the HMGB1 from the cells expressing p75DA had this activity.

To further test the role of HMGB1 in the effects on dying cells on immunogenicity and tolerance, we employed MEF from animals lacking this protein (Figure 4D, E). As expected, UV-irradiated wt MEF (HMGB1+/+) fed to DC did not promote immunity unless caspases were inhibited with zVAD-fmk (Figure 4D). UV-irradiated HMGB1−/− MEF, however, did not induce immunity regardless of zVAD-fmk addition. Similarly, UV-irradiated wt MEF induced tolerance (Figure 4E) unless treated with the reducing agent DTT. In contrast, UV-irradiated HMGB1−/− MEF promoted tolerance induction irrespective of DTT treatment. Therefore, HMGB1 is central to the effects on immunity of caspases and ROS in dying cells. In addition, these results show that the anti-tolerogenic effect of DTT on apoptotic cells is due to its action on HMGB1, rather than on the antigen or some other aspect of antigen presentation.

Since a major difference between apoptotic cells expressing p75wt and p75DA is the production of ROS in the former (See Figure 3A), we tested if oxidation played a role in the activities of supernatants from these cells. In the experiment shown in Figure 5A, DC were again fed with apoptotic cells and cultured in supernatants from apoptotic p75DA HeLa cells. Supernatants from apoptotic cells expressing p75DA (p75DAs) again abrogated the induction of tolerance; while supernatants from apoptotic cells expressing p75wt (p75wts) did not. However, the ability of supernatants from the p75DA-expressing cells to abrogate tolerance was destroyed by H2O2. Conversely, reduction with DTT of supernatants from apoptotic p75wt-expressing cells conferred upon them the ability to block tolerance. This ability of supernatants from UV-irradiated p75wt cells treated with DTT to blocked tolerance induction by apoptotic cells was dependent on HMGB1, since addition of anti-HMGB1 abrogated this effect (Figure 5B).

Figure 5
HMGB1 function and ROS

Oxidative changes that alter covalent disulfide bonds in HMGB1 have recently been localized to cysteine (C) resides in the protein at positions C23, C45, and C106. Mutation of the C residues to serine (S) changed the ability of the protein to be affected by redox conditions (Hoppe et al., 2006). Because mutation of cysteines in HMGB1 affect its nuclear localization (Hoppe, et al., 2006) we added recombinant human HMGB1 (rHMGB1) to DC that had engulfed apoptotic cells (Figure 5C). As in our other experiments, DC that engulfed apoptotic cells induced tolerance when injected. Culture of these DC in DTT-reduced rHMGB1 completely prevented this tolerance induction, while culture with H2O2-oxidized rHMGB1 did not.

We then tested rHMGB1 in which all three cysteines had been mutated to serine (CS3). Culture of DC that had engulfed apoptotic cells with the mutant HMGB1-CS3 potently abrogated the ability of these DC to induce tolerance. However, in contrast to the effects on rHMGB1 activity, addition of either DTT or H2O2 to rHMGB1-CS3 had no effect on its activity. Thus, the oxidation/reduction status of cysteines in the protein has important consequences for the ability of HMGB1 to block tolerance induction.

Oxidation of HMGB1 can induce a mobility shift in SDS-PAGE (Hoppe, et al., 2006) and we have frequently observed the oxidized form of this protein in supernatants of UV-irradiated p75 wt cells, but not p75DA cells when analyzed in a non-reducing gel (Figure 6A). This presented a paradox, however, since a) this was not always observed (data not shown) and b) even when observed, the apparently unmodified form of HMGB1 was still present in the p75wts (e.g. Figure 6A) despite a lack of stimulatory activity in the supernatant. However, oxidation of the C106 does not cause such a mobility shift (Hoppe et al., 2006) and therefore we asked if this residue might be critical for the observed effects. We generated rHMGB1 with single serine substitutions at C23, C45, or C106. Addition of the reduced forms of each of these mutant proteins blocked tolerance induction by apoptotic cells fed to DC (Figure 6B). Oxidation of HMGB1C23S or HMGB1C45S abrogated this effect, while oxidation of HMGB1C106S did not. Thus, oxidation of C106 is necessary and sufficient to inactivate the immunostimulatory activity of HMGB1.

Figure 6
HMGB1 function and Cys modification

HMGB1 has been shown to induce maturation of DC (Messmer et al., 2004). However, no changes in class II MHC or CD86 levels were observed in DC fed apoptotic or necrotic splenocytes, supernatants from apoptotic p75wt or p75DA cells, or treated at the low concentrations of rHMGB1 we employed in our studies (Supplemental Figure 2). Therefore, despite profound effects on immunity and tolerance in vivo, the effects of HMGB1 on dendritic cell function cannot be readily explained by overt changes in DC maturation in our system.


Immune tolerance is essential for a functioning immune system and any approximation of self-nonself discrimination. One way this occurs is via negative selection during T cell development. However, tolerance to self antigens inefficiently presented in the thymus requires peripheral mechanisms, which also provide backup should cells escape the central tolerance process. A plausible mechanism for peripheral tolerance has been proposed wherein steady state homeostatic cell death (apoptosis) directs associated antigens into DC such that tolerance rather than immunity results. Should cell death due to damage or infection occur, antigen stimulation would be limited only to neoantigens present at the site since tolerance induced by ongoing apoptosis would eliminate responses to self (Matzinger, 2002) (Ferguson et al., 2003; Griffith et al., 2007).

During the course of an infection or other insult the cells of the host can undergo two general types of cell death, necrosis or apoptosis. It was originally thought that necrosis stimulates immunity while the immune system is “blind” to apoptotic cells such that they are cleared from the area without impact on the immune system (Ferguson and Griffith, 1997). It is now known that apoptotic cells can stimulate potent immunoregulatory mechanisms to avert the risk posed by exposure to self antigens (Ferguson et al., 2003) (Griffith et al., 2007) (Griffith et al., 1996). Indeed, apoptotic cells can engender potent immune tolerance to antigens perceived in the context of apoptosis (Griffith et al., 2007) (Griffith et al., 1996) (Wells et al., 1999). Mechanisms proposed to account for the tolerogenic nature of apoptotic cells include induction of peripheral deletion (Kurts et al., 1998; Steinman et al., 2003), anergy (clonal inactivation; (Pape et al., 1998), immune deviation (preferential induction of Th2 T cells over Th1; (Griffith et al., 1996), and active regulation (Treg or Ts cells; (Ferguson et al., 2002; Griffith et al., 2007).

The linkage between apoptosis and tolerance has focused on the dendritic cell (DC) as a number of groups have demonstrated that antigens from apoptotic cells are directed to the class I pathway for cross priming (or cross tolerizing) CD8+ T cells (Albert et al., 2001; Ferguson et al., 2002; Heath et al., 1998; Herndon et al., 2005). What is not clear is how the apoptotic cells influence the outcome of the response once they have encountered the DC. Several possibilities have been suggested. First, apoptotic cells might release something that influences the DC population. Indeed, dying lymphocytes can produce IL-10 (Gao et al., 1998; Tomimori et al., 2000) and/or TGFβ (Chen et al., 2001) during the process of apoptosis. While these studies are relevant to lymphocytes, it is not clear that all cells can make these cytokines in response to apoptotic stimuli. A second possibility is that apoptotic bodies may promote tolerance through engaging a receptor (or receptors) on the surface of the relevant antigen-presenting cells (e.g. CD36, phosphytidyl serine receptors; (Savill et al., 2002). Engagement of these receptors might change the program of the DC (i.e. immunity to tolerance) and potentially alter antigen handling and T cell activation. Recently it was shown that the loss of some receptors involved in the removal of apoptotic cells can lead to autoimmunity (Asano et al., 2004) (Miyanishi et al., 2007). However, it was recently shown that necrotic cells can induce tolerance if CD4+ T cell mediated help is abrogated (Griffith et al., 2007). This suggests that specific phagocytic requirements for apoptotic cells over necrotic cells may not be the only requirement for tolerance. A third mechanism is that engulfed apoptotic bodies induce the release of anti-inflammatory cytokines from DC (Fadok et al., 1998; Ronchetti et al., 1999). A fourth possibility is that subsets of DC handle antigen differentially. Indeed it was recently shown that tolerogenic CD8α+ DC present predominantly via the class I pathway while CD8α DC utilize the class II pathway (Dudziak et al., 2007). This fits with studies showing that dying cells deliver processed antigen to the class I pathway of the DC (Blachere et al., 2005). It also supports the idea that DC subsets can regulate tolerance or immunity by apoptotic cells (Ferguson et al., 2002). Here we describe another mechanism, wherein ROS produced in response to caspase cleavage at the mitochondria neutralize HMGB1, thereby preventing DC activation by this molecule.

An early model of immune tolerance involved conjugating antigen to cells and administering them i.v. (Battisto and Bloom, 1966). More recent studies have shown that the injected cells undergo apoptosis and are ingested by DC to cause tolerance (Ferguson et al., 2002). This ability of apoptotic cells to induce immune tolerance is dependent on caspase activation in the dying cells (Ferguson et al., 2002; Obeid et al., 2007b; Tesniere et al., 2008). It is likely that other mechanisms support tolerance when the caspase-dependent effects of apoptotic cells are absent. Our studies presented here show that the apoptotic cell, and specifically the pathway of apoptosis, can have a significant influence over the outcome of an encounter between dead cells and DC. The production of ROS resulting from caspase activation and mitochondrial destruction leads to the modification of an immunostimulatory signal, and this in turn leads to immune tolerance. Thus, an additional safeguard is provided by the process of apoptosis to avoid dangerous immune responses that could result from the release of molecules from dying cells promoting immunity.

Several mediators have been described to activate DC to promote adaptive immune responses, and/or activate macrophages to promote inflammation. These include uric acid (Shi et al., 2003), calreticulin (Obeid et al., 2007a; Obeid et al., 2007b), HSP70 (Millar et al., 2003), and HMGB1 (Andersson et al., 2000; Messmer et al., 2004). While these mediators can stimulate immunity, apoptotic cells can inhibit inflammatory and immune responses driven by macrophages and DC (Fadok et al., 1998). We reasoned that since apoptosis is tolerogenic, mechanisms may be in place to modify these danger signals during the apoptotic process. Our studies show that neutralization of ROS or prevention of oxidation of HMGB1 by mutation of susceptible cysteines is sufficient to block the induction of immune tolerance by apoptotic cells. While our studies demonstrate that HMGB1 is an important target in our system, it may not be the only target for ROS affecting immune function. For example, externalization of calreticulin, which promotes immunity in response to cell death (including apoptotic cell death induced by anthracyclins), is regulated by protein phosphatases (Obeid et al., 2007b) that can be targets of ROS (Kamata et al., 2005). Similarly, uric acid is known to be an anti-oxidant (Haberman et al., 2007) and therefore its pro-immune functions may be influenced by ROS.

Previously we showed that mutation of a conserved caspase cleavage site in p75 NDUSF1 delays loss of mitochondrial function during apoptosis, sustaining ATP levels in the cell and delaying loss of plasma membrane integrity (Ricci et al., 2004). Other caspase-dependent events, such as DNA fragmentation, were unaffected by the mutant p75. Here, we have shown that ROS production as a consequence of caspase cleavage of p75 contributes to the immunological effects of apoptotic cells by oxidation (and thereby neutralization) of HMGB1’s immunostimulatory activity. Release of the DNA binding protein HMGB1 has been shown to be important for several inflammatory conditions including LPS induced shock (Wang et al., 1999) and sepsis (Czura and Tracey, 2003). It has been widely published that HMGB1 is released during necrosis, but not apoptosis (Scaffidi et al., 2002). However, this idea has recently come into question for several reasons. First, it has been shown that nuclear DNA and associated proteins are released in a time dependent manner during apoptosis (Choi et al., 2004) and the binding of HMGB1 to DNA is increased during apoptosis, consistent with the idea that late-stage apoptotic cells can release both DNA and HMGB1 (Bell et al., 2006). Studies have also shown that apoptotic tumor cells can release HMGB1 (Bell et al., 2006; Tian et al., 2007). We have also found that cells undergoing apoptosis can release significant levels of HMGB1, however, only HMGB1 released from apoptotic cells that do not make significant ROS is functional in stimulating immune responses. This suggests that a mechanism is in place to control the potential danger that released HMGB1 could pose if it leaked from apoptotic cells. Consequently we would propose that it is not an inability to release a danger signal that determines whether a cell is tolerogenic but it is the ability of the cell to generate oxidative conditions that can destroy danger signals that is paramount. If so, then forms of cell death, including non-apoptotic cell death, that involve release of ROS may tend to be tolerogenic rather than immunogenic. At present, such forms of cell death have not been examined in this way.

It is of interest to note that only oxidation of the C106 of HMGB1 is required to block its immunogenic function. This residue lies in the B-box of the protein that has been shown to contain the biological activity of HMGB1 in other systems. HMGB1 is known to bind RAGE (receptor for advance glycation end products) to mediate its effects on macrophage and DC activation (Andersson et al., 2002; Li et al., 2003; Messmer et al., 2004). We have not addressed the issue of RAGE binding in our system however, since we do not observe DC maturation in the presence of rHMGB1 (see supplemental Figure 2) the role in binding to this receptor is not known. It is important to note that HMGB1 can bind immunostimulatory cytokines (Sha et al., 2008), bacterial products (Rouhiainen et al., 2007; Tian et al., 2007), and many proteins of diverse function (Dintilhac and Bernues, 2002). This may potentially explain many of the diverse effects on immunity for HMGB1 (Andersson et al., 2002; Li et al., 2003; Messmer et al., 2004), however, the role of Redox effects on HMGB1 and its ability to bind RAGE or other molecules will require further study.

Other factors may bypass the tolerogenic effects of apoptotic cells and/or convert their signals to immunogenic ones. Recently we have shown that one way in which apoptotic cells promote tolerance is by inducing DC to stimulate CD8+ T cells in the absence of CD4+ T cell help (Griffith et al., 2007). The resulting “helpless” CD8+ T cells produce the cytokine TRAIL in response to antigenic restimulation, thereby killing responding T cells and preventing immune responses. For this reason, TRAIL deficient animals are not tolerized by antigen-coupled apoptotic cells. However if alternative signals were to stimulate or mimic T cell help, e.g. by stimulating DC maturation, tolerance would be averted. Thus, while ROS are often associated with inflammatory conditions, other stimuli present in such settings are likely to override the effects on HMGB1 to promote immunity over tolerance during inflammation.



C57Bl/6 (B6) mice were purchased from the National Cancer Institute (Fredrick, MD). Act-mOVA mice were purchased from Jackson Labs (Bar Harbor, ME). All animal procedures were performed according to NIH guidelines and approved by the Washington University IACUC.

Antibodies and Reagents

BHA (2(3)-t-Butyl-4-hydroxyanisole), TNBS (2,4,6 trinitrobenzene sulfonic acid) were purchased from Sigma-Aldrich Co. (St. Louis, MO). ZVAD-fmk was purchased from R&D systems (Minneapolis, MN). A rabbit polyclonal antibody against HMGB1 was used for HMGB1 neutralization experiments. The antibody was produced by immunizing rabbits with synthesized peptide representing amino acids 166 to 181 of the human sequence. A monoclonal antibody against HMGB1 Sigma-Aldrich Co. (St. Louis, MO) or a polyclonal antibody (BD Biosciences, San Diego, CA) were used for western blotting. Fluorescent labeled Annexin V and antibodies to CD11c, MHC Class II, and CD86 were purchased from BD Biosciences (San Diego, CA). CFA (Complete Freund’s adjuvant) PKH26 and propidium iodide were purchased from Sigma-Aldrich Co. (St. Louis, MO).

Cell lines

HeLa cells expressing a modified NDUFS1 (NADH dehydrogenase Fe-S protein-1) were used in some experiments. NDUFS1 (called p75 henceforth) is the 75 kDa subunit of respiratory complex I and is a critical caspase substrate in the mitochondria. Caspase 3 cleavage of p75 is required mitochondrial changes associated with apoptosis. P75wt cells express the wild type protein and undergo typical changes following the induction of apoptosis, while p75DA cells express the non-cleavable mutant and maintain mitochondrial function during apoptosis. These cells have been previously described (Ricci et al., 2004). Caspase 3/7 deficient MEF (Caspase-3−/−, -7−/−) and heterozygous caspase 3/7 deficient MEF (caspase-3+/−, -7+/−) were obtained from Richard Flavell at Yale University (Lakhani et al., 2006). HMGB1 deficient MEF (HMGB1−/−) and control MEF (HMGB1+/+) were obtained from HMGBiotech, Inc. (Milan, Italy).

Production of recombinant HMGB1

Recombinant HMGB1 and rHMGB1-CS3 were prepared for in vivo studies. The CS3 mutant of HMGB1 was made by modifying the cysteine (C) residues at positions C23, C45 and C106 in human HMGB1 to serine (S) as described (Hoppe et al., 2006). These modifications prevent oxidation of these sites and the formation of intramolecular disulfide bonds. To make recombinant proteins CHO cells were transfected with the expression vector containing a histidine tagged HMGB1, HMGB1-CS3 (CS3), HMGB1-C23S, HMGB1-C45S, and HMGB1-C106S by Lipofectamine2000 (Invitrogen Carlsbad, CA). After 2 days incubation, cells were lysed in PBS 1% Triton-X 100 containing protease inhibitors. Recombinant HMGB1 was purified with EZview™ Red HIS-Select HC Nickel Affinity Gel (Sigma-Aldrich, St. Louis, MO). The beads were washed with PBS containing 10mM imidazole. Protein was eluted with PBS containing 500mM imidazole and the eluated material was desalted with a desalting column. Proteins were used at 500ng/ml.

Production and treatment of apoptotic cell and necrotic cells

Single cells suspensions of spleen cells were irradiated with 3000R γ-irradiation (apoptotic cells) or made necrotic by 5 alternating cycles of freeze/thawing (F/T) and used for i.v injection. Caspase-3−/−, -7−/− MEF cells, caspase-3+/−, -7+/− MEF cells, and HeLa cells (p75wt and p75DA) were irradiated to 800mJ/cm2 ultraviolet irradiation (UV). Plates were treated with trypsin and cells were harvested into a single cell suspension prior to use. In some experiments apoptotic cells or necrotic were coupled with TNP as described (Ferguson et al., 2002) or treated with antioxidant (100μM BHA), caspase inhibitors (100μM zVAD-fmk), DTT, or H2O2 prior to i.v. injection. HELA cells and MEF cells were first cultured with DC prior to injection of DC i.v. (see below).

Production and treatment of dendritic cells (DC)

Bone marrow derived DC were cultured for 10 days with 10ng/ml GM-CSF (Peprotech, Rocky Hill, NJ) in RPMI with 10% FBS. Non-adherent cells were collected and analyzed by flow cytometry for CD11c expression. Cells were typically 90–95% CD11c+. DC were then coupled with TNP as described [TNP-DC] (Ferguson et al., 2002). One million (106) TNP-DC were mixed with 5x106 apoptotic spleen cells, 106 apoptotic HeLa cells or 106 apoptotic MEF and incubated overnight. Prior to injection dead cells were removed by centrifugation over a Ficoll gradient (GE Healthcare, Piscataway, NJ).

ROS measurement

Apoptotic cells were stained with 10μM dihydroethidium (DHE; Invitrogen, Carlsbad, CA) for 30 min at 37ºC. Fluorescence intensity was measured by flow cytometry. Values are reported as % cells making ROS based on DHE staining.

Preparation and treatment of apoptotic cell supernatants

Supernatants from p75wt and p75DA cells were prepared by culturing UV-irradiated cells for 24 hrs at 2x106 cells/ml in RPMI-1640 + 10% FBS. Supernatants were used to treat TNP-DC that were fed apoptotic spleen cells for 6 hrs. In some experiments supernatants were treated with 50mM DTT or 50mM H2O2 for 30 minutes on ice. Irradiated apoptotic spleen cells were cultured for 6 hours at 107 cells/ml and then treated with 50mM DTT or 50mM H2O2 for 30 minutes on ice. Following treatment supernatants were loaded to NAP-25 gel column (GE healthcare, Piscataway, NJ) remove excess reagents. HMGB1 was neutralized by adding 10μg/ml of a polyclonal rabbit anti-HMGB1 just prior to the addition of the supernatant to DC cultures. Depletion of HMGB1 from supernatants was accomplished by adding 20μg/ml of polyclonal anti-HMGB1 along with protein G coupled beads for 30 min on ice. Supernatants were collected by centrifugation. Depletion was verified by western blotting.

Immune response

Immunogenicity and tolerogenicity of apoptotic cells were measured using a well defined in vivo model. Immunogenicity test: TNP-coupled apoptotic cells (107) or TNP-DC fed apoptotic cells (106) were injected i.v. into mice. Five days later mice were injected with TNBS in the right footpad and PBS in the left footpad. DTH was measured with a micrometer 24 hrs later. Tolerogenicity test: TNP-coupled apoptotic cells (107) or TNP-DC fed apoptotic cells (106) were injected i.v. into mice. Mice were immunized 2 days later by s.c injection of TNBS. Four (4) days following immunization mice were injected with TNBS in the right footpad and PBS in the left footpad. DTH was measured with a micrometer 24 hrs later.

Tolerogenicity of Act-mOVA apoptotic cells was assessed in a manner similar to TNP coupled apoptotic cells. Act-mOVA spleen cells were induced to undergo apoptosis by γ-irradiation and 1x107 cells were injected i.v. into recipient mice. Two days later mice were immunized s.c. with 100μg OVA emulsified 1:1 with CFA. Seven days later mice were injected with 100μg heat aggregated OVA (boiled 5 minutes) in the right footpad and PBS in the left footpad. DTH was measured with a micrometer 24 hrs later.

Western blots

Cell supernatants were centrifuged and centrifuged at 14,000g to remove cellular debris. The protein concentration of each sample was determined. Equal amounts of protein were separated on a reducing or non-reducing 10% Bis-Tris gel (Invitrogen, Carlsbad, CA), transferred to nitrocellulose membrane, and blocked with 5% skim milk in TBS/0.05% Tween 20 for 2 h. The membrane was incubated with anti-HMGB1 monoclonal antibody or polyclonal rabbit anti-HMGB1 at 4°C overnight. After washing, the membrane was incubated with HRP-conjugated anti-rabbit antibody in Tris-Tween 20 for 1 h. Following several washes, the blot was developed by chemiluminescence (ECL-plus; Amersham-Pharmacia, Piscataway, NJ). Control blots were processed without incubation of the primary Ab.

Supplementary Material


Supplemental Figure 1. Apoptosis of γ-irradiated spleen cells. A single cell suspension of spleen cells was subjected to 3000R of γ-irradiation and place in RPMI with 10% FBS at 37oC. At 2, 6, and 16 hr following irradiation samples were taken, stained with Annexin V/PI, and analyzed by flow cytometry.

Supplemental Figure 2. DC maturation. DC were harvested from 10 day bone marrow cultures and plated at 106 per ml in RPMI with 10% FBS. DC were treated overnight with supernatant from UV irradiated p75wt, supernatant from UV irradiated p75DA, 100ng/ml LPS, apoptotic cells (γ-irradiated spleen), necrotic cells (F/T), 250ng/ml HMGB1 or 3μg/ml HMGB1. Cells were harvested and stained for CD11c along with MHC class II or (A) CD86 (B-D). Data represent Class II+ or CD86+ CD11c+ cells.


This work was supported by a National Institutes of Health Grants EY06765 (TAF), EY015570 (TAF), AI44848 (DRG), AI40646 (DRG) and the Department of Ophthalmology and Visual Sciences core grant (EY02687). Support was also received from the Foundation for Fighting Blindness (Owings Mills, MD, USA) and from Research to Prevent Blindness (New York, N.Y., USA).


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Albert ML, Jegathesan M, Darnell RB. Dendritic cell maturation is required for the cross-tolerization of CD8+ T cells. Nat Immunol. 2001;2:1010–1017. [PubMed]
  • Andersson U, Erlandsson-Harris H, Yang H, Tracey KJ. HMGB1 as a DNA-binding cytokine. J Leukoc Biol. 2002;72:1084–1091. [PubMed]
  • Andersson U, Wang H, Palmblad K, Aveberger AC, Bloom O, Erlandsson-Harris H, Janson A, Kokkola R, Zhang M, Yang H, Tracey KJ. High mobility group 1 protein (HMG-1) stimulates proinflammatory cytokine synthesis in human monocytes. J Exp Med. 2000;192:565–570. [PMC free article] [PubMed]
  • Asano K, Miwa M, Miwa K, Hanayama R, Nagase H, Nagata S, Tanaka M. Masking of phosphatidylserine inhibits apoptotic cell engulfment and induces autoantibody production in mice. J Exp Med. 2004;200:459–467. [PMC free article] [PubMed]
  • Battisto JR, Bloom BR. Dual immunological unresponsiveness induced by cell membrane coupled hapten or antigen. Nature. 1966;212:156–157. [PubMed]
  • Bell CW, Jiang W, Reich CF, 3rd, Pisetsky DS. The extracellular release of HMGB1 during apoptotic cell death. Am J Physiol Cell Physiol. 2006;291:C1318–1325. [PubMed]
  • Blachere NE, Darnell RB, Albert ML. Apoptotic cells deliver processed antigen to dendritic cells for cross-presentation. PLoS Biol. 2005;3:e185. [PMC free article] [PubMed]
  • Bonfoco E, Stuart PM, Brunner T, Lin T, Griffith TS, Gao Y, Nakajima H, Henkart PA, Ferguson TA, Green DR. Inducible nonlymphoid expression of Fas ligand is responsible for superantigen-induced peripheral deletion of T cells. Immunity. 1998;9:711–720. [PubMed]
  • Chen M, Wang YH, Wang Y, Huang L, Sandoval H, Liu YJ, Wang J. Dendritic cell apoptosis in the maintenance of immune tolerance. Science. 2006;311:1160–1164. [PubMed]
  • Chen W, Frank ME, Jin W, Wahl SM. TGF-beta released by apoptotic T cells contributes to an immunosuppressive milieu. Immunity. 2001;14:715–725. [PubMed]
  • Chipuk JE, Green DR. Do inducers of apoptosis trigger caspase-independent cell death? Nat Rev Mol Cell Biol. 2005;6:268–275. [PubMed]
  • Choi JJ, Reich CF, 3rd, Pisetsky DS. Release of DNA from dead and dying lymphocyte and monocyte cell lines in vitro. Scand J Immunol. 2004;60:159–166. [PubMed]
  • Czura CJ, Tracey KJ. Targeting high mobility group box 1 as a late-acting mediator of inflammation. Crit Care Med. 2003;31:S46–50. [PubMed]
  • Dintilhac A, Bernues J. HMGB1 interacts with many apparently unrelated proteins by recognizing short amino acid sequences. J Biol Chem. 2002;277:7021–7028. [PubMed]
  • Dudziak D, Kamphorst AO, Heidkamp GF, Buchholz VR, Trumpfheller C, Yamazaki S, Cheong C, Liu K, Lee HW, Park CG, et al. Differential antigen processing by dendritic cell subsets in vivo. Science. 2007;315:107–111. [PubMed]
  • Dumitriu IE, Baruah P, Valentinis B, Voll RE, Herrmann M, Nawroth PP, Arnold B, Bianchi ME, Manfredi AA, Rovere-Querini P. Release of high mobility group box 1 by dendritic cells controls T cell activation via the receptor for advanced glycation end products. J Immunol. 2005;174:7506–7515. [PubMed]
  • Ehst BD, Ingulli E, Jenkins MK. Development of a novel transgenic mouse for the study of interactions between CD4 and CD8 T cells during graft rejection. Am J Transplant. 2003;3:1355–1362. [PubMed]
  • Fadok VA, Bratton DL, Konowal A, Freed PW, Westcott JY, Henson PM. Macrophages that have ingested apoptotic cells in vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms involving TGF-beta, PGE2, and PAF. J Clin Invest. 1998;101:890–898. [PMC free article] [PubMed]
  • Ferguson TA, Griffith TS. A vision of cell death: insights into immune privilege. Immunol Rev. 1997;156:167–184. [PubMed]
  • Ferguson TA, Herndon J, Elzey B, Griffith TS, Schoenberger S, Green DR. Uptake of apoptotic antigen-coupled cells by lymphoid dendritic cells and cross-priming of CD8(+) T cells produce active immune unresponsiveness. J Immunol. 2002;168:5589–5595. [PubMed]
  • Ferguson TA, Stuart PM, Herndon JM, Griffith TS. Apoptosis, tolerance, and regulatory T cells--old wine, new wineskins. Immunol Rev. 2003;193:111–123. [PubMed]
  • Fleury C, Mignotte B, Vayssiere JL. Mitochondrial reactive oxygen species in cell death signaling. Biochimie. 2002;84:131–141. [PubMed]
  • Gao Y, Herndon JM, Zhang H, Griffith TS, Ferguson TA. Antiinflammatory effects of CD95 ligand (FasL)-induced apoptosis. J Exp Med. 1998;188:887–896. [PMC free article] [PubMed]
  • Griffith TS, Kazama H, VanOosten RL, Earle JK, Jr, Herndon JM, Green DR, Ferguson TA. Apoptotic cells induce tolerance by generating helpless CD8+ T cells that produce TRAIL. J Immunol. 2007;178:2679–2687. [PubMed]
  • Griffith TS, Yu X, Herndon JM, Green DR, Ferguson TA. CD95-induced apoptosis of lymphocytes in an immune privileged site induces immunological tolerance. Immunity. 1996;5:7–16. [PubMed]
  • Haberman F, Tang SC, Arumugam TV, Hyun DH, Yu QS, Cutler RG, Guo Z, Holloway HW, Greig NH, Mattson MP. Soluble neuroprotective antioxidant uric Acid analogs ameliorate ischemic brain injury in mice. Neuromolecular Med. 2007;9:315–323. [PubMed]
  • Heath WR, Kurts C, Miller JF, Carbone FR. Cross-tolerance: a pathway for inducing tolerance to peripheral tissue antigens. J Exp Med. 1998;187:1549–1553. [PMC free article] [PubMed]
  • Herndon JM, Stuart PM, Ferguson TA. Peripheral deletion of antigen-specific T cells leads to long-term tolerance mediated by CD8+ cytotoxic cells. J Immunol. 2005;174:4098–4104. [PubMed]
  • Hoppe G, Talcott KE, Bhattacharya SK, Crabb JW, Sears JE. Molecular basis for the redox control of nuclear transport of the structural chromatin protein Hmgb1. Exp Cell Res. 2006;312:3526–3538. [PubMed]
  • Huang DC, Hahne M, Schroeter M, Frei K, Fontana A, Villunger A, Newton K, Tschopp J, Strasser A. Activation of Fas by FasL induces apoptosis by a mechanism that cannot be blocked by Bcl-2 or Bcl-x(L) Proc Natl Acad Sci U S A. 1999;96:14871–14876. [PMC free article] [PubMed]
  • Kamata H, Honda S, Maeda S, Chang L, Hirata H, Karin M. Reactive oxygen species promote TNFalpha-induced death and sustained JNK activation by inhibiting MAP kinase phosphatases. Cell. 2005;120:649–661. [PubMed]
  • Kurts C, Heath WR, Kosaka H, Miller JF, Carbone FR. The peripheral deletion of autoreactive CD8+ T cells induced by cross- presentation of self-antigens involves signaling through CD95 (Fas, Apo- 1) J Exp Med. 1998;188:415–420. [PMC free article] [PubMed]
  • Lakhani SA, Masud A, Kuida K, Porter GA, Jr, Booth CJ, Mehal WZ, Inayat I, Flavell RA. Caspases 3 and 7: key mediators of mitochondrial events of apoptosis. Science. 2006;311:847–851. [PMC free article] [PubMed]
  • Li J, Kokkola R, Tabibzadeh S, Yang R, Ochani M, Qiang X, Harris HE, Czura CJ, Wang H, Ulloa L, et al. Structural basis for the proinflammatory cytokine activity of high mobility group box 1. Mol Med. 2003;9:37–45. [PMC free article] [PubMed]
  • Martin SJ, Green DR. Protease activation during apoptosis: death by a thousand cuts? Cell. 1995;82:349–352. [PubMed]
  • Matzinger P. The danger model: a renewed sense of self. Science. 2002;296:301–305. [PubMed]
  • Messmer D, Yang H, Telusma G, Knoll F, Li J, Messmer B, Tracey KJ, Chiorazzi N. High mobility group box protein 1: an endogenous signal for dendritic cell maturation and Th1 polarization. J Immunol. 2004;173:307–313. [PubMed]
  • Millar DG, Garza KM, Odermatt B, Elford AR, Ono N, Li Z, Ohashi PS. Hsp70 promotes antigen-presenting cell function and converts T-cell tolerance to autoimmunity in vivo. Nat Med. 2003;9:1469–1476. [PubMed]
  • Miyanishi M, Tada K, Koike M, Uchiyama Y, Kitamura T, Nagata S. Identification of Tim4 as a phosphatidylserine receptor. Nature. 2007;450:435–439. [PubMed]
  • Obeid M, Panaretakis T, Joza N, Tufi R, Tesniere A, van Endert P, Zitvogel L, Kroemer G. Calreticulin exposure is required for the immunogenicity of gamma-irradiation and UVC light-induced apoptosis. Cell Death Differ. 2007a;14:1848–1850. [PubMed]
  • Obeid M, Tesniere A, Ghiringhelli F, Fimia GM, Apetoh L, Perfettini JL, Castedo M, Mignot G, Panaretakis T, Casares N, et al. Calreticulin exposure dictates the immunogenicity of cancer cell death. Nat Med. 2007b;13:54–61. [PubMed]
  • Ott M, Gogvadze V, Orrenius S, Zhivotovsky B. Mitochondria, oxidative stress and cell death. Apoptosis. 2007;12:913–922. [PubMed]
  • Pape KA, Merica R, Mondino A, Khoruts A, Jenkins MK. Direct evidence that functionally impaired CD4+ T cells persist in vivo following induction of peripheral tolerance. J Immunol. 1998;160:4719–4729. [PubMed]
  • Ricci JE, Gottlieb RA, Green DR. Caspase-mediated loss of mitochondrial function and generation of reactive oxygen species during apoptosis. J Cell Biol. 2003;160:65–75. [PMC free article] [PubMed]
  • Ricci JE, Munoz-Pinedo C, Fitzgerald P, Bailly-Maitre B, Perkins GA, Yadava N, Scheffler IE, Ellisman MH, Green DR. Disruption of mitochondrial function during apoptosis is mediated by caspase cleavage of the p75 subunit of complex I of the electron transport chain. Cell. 2004;117:773–786. [PubMed]
  • Ronchetti A, Rovere P, Iezzi G, Galati G, Heltai S, Protti MP, Garancini MP, Manfredi AA, Rugarli C, Bellone M. Immunogenicity of apoptotic cells in vivo: role of antigen load, antigen-presenting cells, and cytokines. J Immunol. 1999;163:130–136. [PubMed]
  • Rouhiainen A, Tumova S, Valmu L, Kalkkinen N, Rauvala H. Pivotal advance: analysis of proinflammatory activity of highly purified eukaryotic recombinant HMGB1 (amphoterin) J Leukoc Biol. 2007;81:49–58. [PubMed]
  • Sauter B, Albert ML, Francisco L, Larsson M, Somersan S, Bhardwaj N. Consequences of cell death: exposure to necrotic tumor cells, but not primary tissue cells or apoptotic cells, induces the maturation of immunostimulatory dendritic cells. J Exp Med. 2000;191:423–434. [PMC free article] [PubMed]
  • Savill J, Dransfield I, Gregory C, Haslett C. A blast from the past: clearance of apoptotic cells regulates immune responses. Nat Rev Immunol. 2002;2:965–975. [PubMed]
  • Scaffidi P, Misteli T, Bianchi ME. Release of chromatin protein HMGB1 by necrotic cells triggers inflammation. Nature. 2002;418:191–195. [PubMed]
  • Sha Y, Zmijewski J, Xu Z, Abraham E. HMGB1 develops enhanced proinflammatory activity by binding to cytokines. J Immunol. 2008;180:2531–2537. [PubMed]
  • Shi Y, Evans JE, Rock KL. Molecular identification of a danger signal that alerts the immune system to dying cells. Nature. 2003;425:516–521. [PubMed]
  • Steinman RM, Hawiger D, Nussenzweig MC. Tolerogenic dendritic cells. Annu Rev Immunol. 2003;21:685–711. [PubMed]
  • Taylor RC, Cullen SP, Martin SJ. Apoptosis: controlled demolition at the cellular level. Nat Rev Mol Cell Biol 2007 [PubMed]
  • Tesniere A, Panaretakis T, Kepp O, Apetoh L, Ghiringhelli F, Zitvogel L, Kroemer G. Molecular characteristics of immunogenic cancer cell death. Cell Death Differ. 2008;15:3–12. [PubMed]
  • Tian J, Avalos AM, Mao SY, Chen B, Senthil K, Wu H, Parroche P, Drabic S, Golenbock D, Sirois C, et al. Toll-like receptor 9-dependent activation by DNA-containing immune complexes is mediated by HMGB1 and RAGE. Nat Immunol. 2007;8:487–496. [PubMed]
  • Tomimori Y, Ikawa Y, Oyaizu N. Ultraviolet-irradiated apoptotic lymphocytes produce interleukin-10 by themselves. Immunol Lett. 2000;71:49–54. [PubMed]
  • Wang H, Bloom O, Zhang M, Vishnubhakat JM, Ombrellino M, Che J, Frazier A, Yang H, Ivanova S, Borovikova L, et al. HMG-1 as a late mediator of endotoxin lethality in mice. Science. 1999;285:248–251. [PubMed]
  • Wells AD, Li XC, Li Y, Walsh MC, Zheng XX, Wu Z, Nunez G, Tang A, Sayegh M, Hancock WW, et al. Requirement for T-cell apoptosis in the induction of peripheral transplantation tolerance. Nat Med. 1999;5:1303–1307. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...