Published online before print September 27, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,2
* Institute for Clinical Immunology, Department of Internal Medicine 3,
Department of Experimental Medicine I and
IZKF Research Group 2, Nikolaus Fiebiger Centre of Molecular Medicine, Friedrich-Alexander-University of Erlangen-Nuremberg, Erlangen, Germany;
Franz von Prümmer Klinik, Akutklinik für Rheumatologie und Allgemeinkrankenhaus, Bad Brückenau, Germany; and
¶ Chemotherapeutisches Forschungsinstitut Georg Speyer Haus, Frankfurt, Germany
2Correspondence: Institute for Clinical Immunology, Friedrich-Alexander-University of Erlangen-Nuremberg, Glueckstrasse 4a, 91054 Erlangen, Germany. E-mail: udo.gaipl{at}med3.imed.uni-erlangen.de
ABSTRACT
Apoptotic and necrotic cells expose phosphatidylserine (PS). This membrane modification ensures a swift recognition and uptake by phagocytes of the dying and dead cells. Annexin V (AxV) preferentially binds to anionic phospholipids and thereby, modulates the clearance process. First, we analyzed the influence of AxV on the immunogenicity of apoptotic cells. The addition to apoptotic cells of AxV prior to their injection into mice increased their immunogenicity significantly. Next, we studied the influence of endogenous AxV on the allogeneic reaction against apoptotic and necrotic cells. To preserve heat-labile, short-lived "danger signals," we induced necrosis by mechanical stress. Wild-type mice showed a strong, allogeneic delayed-type hypersensitivity (DTH) reaction. In contrast, AxV-deficient animals showed almost no allogeneic DTH reaction, indicating that endogenous AxV increases the immune response against dead cells. Furthermore, AxV-deficient macrophages had a higher immunosuppressive potential in vitro. Next, we analyzed the influence of AxV on chronic macrophage infection with HIV-1, known to expose PS on its surface. The infectivity in human macrophages of HIV-1 was reduced significantly in the presence of AxV. Finally, we show that AxV also blocked the in vitro uptake by macrophages of primary necrotic cells. Similar to apoptotic cells, necrotic cells generated by heat treatment displayed an anti-inflammatory activity. In contrast, mechanical stress-induced necrotic cells led to a decreased secretion of IL-10, indicating a more inflammatory potent-ial. From the experiments presented above, we conclude that AxV influences the clearance of several PS-exposing particles such as viruses, dying, and dead cells.
Key Words: immunogenicity • apoptotic cells • necrotic cells • clearance • macrophages • annexin
INTRODUCTION
Annexins have been in the spotlight of researchers for many years. One of their main biochemical characteristics is to bind anionic phospholipids in a Ca2+-dependent manner. Annexin V (AxV) represents a typical member of the annexin family and was formerly named anchorin CII, a collagen-binding protein isolated originally from chondrocyte membranes [1 ]. Analysis of the complete primary structure of this protein revealed that anchorin CII is another member of the calpactin/lipocortin/annexin family [2 ]. To date, annexins are defined as a multigene family of Ca2+-regulated proteins, which are characterized by a unique Ca2+- and membrane-binding module, namely the annexin core domain [3 ]. Via this domain, annexins can bind to membranes, which contain negatively charged phospholipids. Phosphatidylserine (PS), which is prominently located in membrane leaflets, is the preferential binding partner of AxV and is part of the membrane dynamics of apoptosis. Recently, it has been described that AxV can be internalized by apoptotic cells in vivo, and a PS-dependent novel way of cell entry was postulated [4 ].
One major "eat me" signal for phagocytes of dying and dead cells is the exposure of PS [5 ]. Several bridging molecules, such as the milk-fat globule protein MFG-E8 [6 ], the growth arrest-specific gene product, growth arrest-specific gene 6, which represents a ligand for the receptor tyrosine kinase MER [7 ], ß2-glycoprotein-1 [8 ], and annexin1 [9 ], have been described to bind to PS on the surfaces of dying cells. Thereby, they mediate the uptake by macrophages of the latter. Furthermore, several macrophage receptors such as class B scavenger receptors, SR-BI and CD36, might be able to interact directly with PS on the surfaces of apoptotic cells [10 ]. AxV preferentially binds PS with high-affinity and inhibits apoptotic and necrotic cell uptake by macrophages, most likely through interference with the availability of PS for recognition [11 , 12 ]. Additional eat me features present in membrane patches in the vicinity of PS were suggested just recently [13 ].
Apoptotic cells are normally cleared via an anti-inflammatory pathway [14 , 15 ]. In contrast, the in vivo uptake and removal of necrotic or lysed cells usually involve inflammation and an immune response. Cells dying by apoptosis may also become leaky if they are not cleared properly. Consecutively, they enter the late stages of apoptosis, often referred to as secondary necrosis. The high-mobility group B1 (HMGB1) protein, which is attached on the chromatin of apoptotic cells, remains immobilized even under conditions of secondary necrosis. However, in the case of primary necrotic cells, it is released and acts as an important inflammatory cytokine [16 ]. Primary and secondary necrotic cells therefore display different inflammatory signals [17 ].
Defects in the clearance process of dying cells have been shown to play a crucial role in the development of chronic autoimmune diseases. Not properly cleared, apoptotic cells lose their membrane integrity, may release "danger signals," and give access to intracellular autoantigens. Macrophages of some patients with systemic lupus erythematosus (SLE) show defects in the removal of dying cells [18 ]. We described for the first time the accumulation of cellular debris in germinal centers of the lymph nodes of humans with SLE [19 ]. Nuclear material derived from apoptotic cells might subsequently, positively select high-affinity, anti-DNA B cells and promote their survival [20 ].
In healthy situations, cells dying by apoptosis maintain their membrane integrity until they get cleared. Cell shrinkage leads to formation of apoptotic bodies containing organelles. Nevertheless, the dead corpse still has an important message: "No inflammation, please!"
We were asking the question of whether an interference with the anti-inflammatory clearance by macrophages of apoptotic cells, mainly mediated by PS recognition, consecutively restores the immunogenicity of apoptotic cells in vivo. For this purpose, mice were immunized with apoptotic cells in the presence or absence of the PS-binding protein AxV. Using AxV-deficient mice [21 ], we further analyzed the physiological function of AxV in the induction of an immune response against apoptotic and necrotic cells. PS is also expressed on monocytes as part of their differentiation program [22 ]. Many viruses including HIV cause extensive apoptosis, and infected monocytes/macrophages therefore express elevated levels of PS, which consecutively, can also be found in the outer membrane of the enveloped retrovirus [23 ]. Using an in vitro culture system, we examined whether, like the swift uptake of apoptotic cells, the AxV can also block the silent entry of HIV-1 into human macrophages. To better understand the mechanism how PS exposing AxV-binding particles modulates the immune response, we examined the cytokine secretion of activated human macrophages and peritoneal macrophages from AxV-deficient mice after contact with dying and dead cells.
MATERIALS AND METHODS
Animals
For the immunization experiments with viable, apoptotic, and AxV-treated, apoptotic human T cells, 6-week-old female Balb/c mice were used. Those mice were obtained from Charles River Wiga (Sulzfeld, Germany).
Ablation of the AxV gene by homologous recombination and generation of an AxV-lacZ fusion gene have been described recently [21 ]. Experiments were performed with 10- to 12-week-old AxV-deficient and wild-type (WT) mice displaying mixed genetic background of C57/BL6x129/SvJ.
Cells and induction of apoptosis and necrosis
Venous blood was drawn from normal, healthy volunteers according to institutional guidelines. The blood was anticoagulated by the addition of 20 U/ml heparin. PBMC were isolated by Ficoll density gradient centrifugation (LymphoprepTM, Gibco Invitrogen, Karlsruhe, Germany). Residual platelets were removed by density centrifugation through a cushion of FCS (Gibco, Eggenstein, Germany).
We established human T cell lines for the immunization experiments from PBMC using IL-2 and activation by PHA. We maintained and expanded the cells in RPMI-1640 medium (Gibco), supplemented with 30% CG medium (Vitromex, Vilshofen, Germany), 5% heat-inactivated (56°C for 30 min) FCS, gentamycin (100 µg/ml), and recombinant human (rh)IL-2 (100 U/ml; Eurocetus, Frankfurt, Germany) at 37°C and 5% CO2. The T cells were restimulated every 3 weeks using PHA (1 µg/ml) in the presence of irradiated (30 Gy), heterologous PBMC. Apoptotic T cells were obtained by irradiation with UV B light (UV-B; 180 mJ/cm2) and afterward, culture in medium without IL-2 for another 20 h.
Monocytes were isolated from platelet-depleted PBMC. The cells were suspended at 8 x 106 cells/ml in DMEM (Gibco), and 0.5 ml of the cell suspension was added per well of 48-well tissue-culture plates. After 2 h of incubation at 37°C in 5% CO2, the cell layer was washed twice with PBS to remove nonadherent PBL. The adherent monocytes were cultured in DMEM containing 10 % FCS and 5 ng/ml rhGM-CSF (Behringwerke, Marburg, Germany) plus 100 U/ml penicillin, 100 µg/ml streptomycin, and 200 mM glutamine (all from Gibco Invitrogen). Monocytes were cultured at 37°C in 5% CO2 for 8 days to generate human monocyte-derived macrophages (HMDM).
PBL were obtained from the same donor as the monocytes and stained with CFSE-diacetate (DA; Molecular Probes, Leiden, The Netherlands) as described previously [12 ]. For the induction of necrosis in CFSE-DA-labeled PBL, the cells were incubated at 56°C for 30 min and kept for another 30 min at 37°C. Necrosis was verified in each individual experiment before addition of the dead cells to the macrophages. More than 90% of the cells stained positive for trypan blue, propidium iodide (PI), and AxV-FITC.
The murine tumor cell line Sp2/0 was obtained from the American Type Culture Collection (Manassas, VA). For the induction of apoptosis, the cells were irradiated with UV-B (120 mJ/cm2) and cultured in medium for another 14 h. The cells (60–70%) were apoptotic, and 30–40% were secondarily necrotic. Necrosis was induced by forcing the cells for two to three times with high pressure through a narrow, hollow needle. We used a 20-G needle with a 11/2-inch length and further narrowed the aperture using a pincer. The final aperture was adjusted individually to the cell type, which has to be necrotized. It is important that the necrotizing procedure was performed rather quickly, and the necrotic status was checked by staining with AxV-FITC and PI. If more than 70% of the cells was primary-necrotic (positive for PI), they were injected immediately into the mice.
HMDM used for the experiments, dealing with the influence of AxV on the HIV-1 infection of cultured macrophages, were obtained as follows: PBMC were separated from buffy coats of healthy, HIV-1-seronegative donors by Ficoll density gradient centrifugation and cultured on hydrophobic Teflon foils (Heraeus, Hanau, Germany) for 7 days in RPMI-1640 medium supplemented with antibiotics (100 U/ml penicillin and 100 µg/ml streptomycin; Gibco Invitrogen), L-glutamine (2 mM; Gibco Invitrogen), and 4% pooled human AB serum. HMDM were then isolated by adhesion. Briefly, PBMC were plated on 24-well plates, and after 1 h, the nonadherent cells were removed by repeated washing with PBS.
Peritoneal macrophages were obtained 4 days after i.p. injection of 1.0 ml amylum (2% in PBS). Macrophages (1.5x106) were seeded in 24-well multi-wells in RPMI medium containing 10% FCS, 100 U/ml penicillin, 100 mg/ml streptomycin, and 200 mM glutamine (R10). The nonadherent cells were removed after 2 h, and the adherent macrophages were cultured for further 24 h in R10.
Immunization experiments
After induction of apoptosis in human T cells, the latter were harvested and washed twice with Ringer’s solution. The cells (106/ml) were incubated in Ringer’s solution in the presence or absence of AxV (1.2 µg/ml; responsif GmbH, Erlangen, Germany) on ice for 30 min. The bacterial endotoxin contamination of AxV was lower than 1.18 IU/ml. Each Balb/c mouse was then injected i.p. with 5 x 105 cells suspended in 500 µl Ringer’s solution without prior washing. Nonirradiated, control cells were cultured in the presence of IL-2 and washed twice and were also resuspended in Ringer’s solution 30 min before injection. Seventeen days after the primary immunization, all animals received a booster injection identical to their primary immunization. Murine sera were collected 10 days after each injection and stored at –20°C. Preimmune sera and sera from mice injected with Ringer’s solution only served as controls. Analysis of the antihuman lymphocyte antibodies was performed by indirect immunofluorescence analyses as described previously [24
].
AxV-deficient and WT animals were immunized with 5 x 106 apoptotic or necrotic Sp2/0 cells, respectively. For the i.p. injection, the cells were suspended in 500 µl Ringer’s solution. The injection of 500 µl Ringer only served as control. We immunized three times at Days 0, 21, and 63. The delayed-type hypersensitivity (DTH) test was carried out 6 days after the final immunization. Viable Sp2/0 cells (106) in 50 µl PBS were injected into the footpads. The diameter of the footpads was measured before and 18 h after the injections. The footpad swelling is indicated as index calculated by: [diameter of the footpad (DF) before the injection–DF 18 h after the injection]/(DF before the injection) x 100.
Assay for the phagocytosis by macrophages of necrotic cells
To quantify the uptake into 1,1'-dioctadecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate-labeled macrophages of the CFSE-labeled necrotic cells, we used two-color flow cytometry [12
]. To combine results from multiple experiments with macrophages from different donors displaying marked donor-to-donor variations in the basal uptake level, we normalized for the differences of the individual basal uptake level using the relative phagocytosis index.
Determination of cytokine secretion
HMDM (8 days old), differentiated from adherent monocytes, were cultured in DMEM containing 20% autologous serum. Heat-induced (56°C, 30 min) necrotic cells (1.2x106/well) were added to the HMDM in 50 µl medium 1 h prior to activation with LPS (100 ng/ml). Supernatants were harvested 16 h after activation. Cytokine concentrations were determined by ELISA using appropriate pairs of mAb specific for human TNF-
and IL-10, respectively (BD PharMingen, Germany). Mouse peritoneal macrophages were cultured for 24 h in R10. Mechanical stress-induced necrotic cells (1.0x106/well) were added to the macrophages in 500 µl R10 1–2 h prior to activation with LPS (100 ng/ml). Supernatants were harvested 16 h after activation. Cytokine concentrations were determined by ELISA using appropriate pairs of mAb specific for mouse TNF- and IL-10, respectively (BD PharMingen).
Chronic HIV infection of macrophages
On Day 7 after separation by adhesion, HMDM (4.0x105) were infected with 100 µl inoculum of a monocytotropic (R5) HIV-1 isolate from a perinatally infected child (HIV-1D117III; RT activity of the virus stock: 1.5x106 cpm/ml/90 min) [25
]. AxV (25 or 50 µg/ml) was added at Day 14. HIV production was measured with a p24 antigen ELISA (Innogenetics, Heiden, Germany) on Days 8, 15, and 22. As control, HMDM were cultured without AxV.
Statistical analysis
Statistical analyses were performed using two-tailed Student’s t-test. Results were considered statistically significant for P < 0.05 (*) and highly significant for P < 0.01 (**).
RESULTS
AxV heightens the immunogenicity of apoptotic cells
Balb/c mice were immunized with viable, apoptotic, and AxV-treated apoptotic cells. Those mice, which had received apoptotic cells, displayed reduced antihuman lymphocyte IgM and IgG titers compared with animals injected with viable cells, indicating a low immunogenicity of apoptotic cells. However, the incubation of apoptotic cells with AxV prior to the immunization significantly increased the immunogenicity of the cells undergoing programmed cell death. In Figure 1
, the amount of antihuman lymphocyte antibodies after booster injection induced by apoptotic or AxV-treated apoptotic cells in relation to those induced by viable cells is shown. To exclude that AxV antibodies are the reason for the increased antihuman lymphocyte reactivity in mice immunized with AxV-treated apoptotic cells, we performed a preabsorption of murine sera with AxV, which did not reduce the titers (data not shown).
 View larger version (19K): [in a new window] |
Figure 1. AxV influences the secondary immune response. After booster injection of viable, apoptotic (apo), or AxV-treated apoptotic human T cells into Balb/c mice, murine antihuman lymphocyte (a-huL) IgG and IgM antibodies (Ab) were measured with indirect immunofluorescence. The fluorescence of viable cells was set to 100. It should be noted that apoptotic cells show a decreased immunogenicity in comparison with viable ones. However, the immunogenicity of apoptotic cells could be increased by the addition of AxV. Results were considered statistically significant for *P < 0.05 and highly significant for **P < 0.01.
|
was similar in the case of AxV-deficient and WT macrophages. Taken together, macrophages of AxV-deficient animals display a more anti-inflammatory and immunosuppressive potential in response to mechanical stress-induced necrotic cells.
 View larger version (15K): [in a new window] |
Figure 2. Endogenous AxV influences the allogeneic immune response against necrotic cells. AxV-deficient and WT C57/BL6x129/SvJ were immunized three times i.p. with UV-B-irradiated apoptotic Sp2/0 cells, with mechanical stress-induced necrotic Sp2/0 cells, or with Ringer’s solution, only as placebo control. The DTH reaction against viable Sp2/0 cells was determined by measurement of the footpad swelling, which is indicated as footpad swelling index (A). It should be noted that WT mice, but not AxV-deficient mice, show a DTH reaction against fresh, primary necrotic cells and also a more pronounced reaction against late apoptotic cells. Values are the mean ± SD of three different mice. Murine TNF- and IL-10 secretion of AxV WT and AxV-deficient peritoneal macrophages, which were stimulated with LPS in the presence of mechanical stress-induced necrotic cells, was quantified by ELISA (B). Values are the mean ± SD of three assays. Results were considered statistically highly significant for **P < 0.01.
|
 View larger version (11K): [in a new window] |
Figure 3. AxV influences the HIV-1 replication in HMDM, which were infected with HIV-1 Strain D117III on Day 7 (arrow). AxV was added on Day 14 (arrow). A p24 antigen ELISA was used to measure the HIV production. HMDM cultured in the absence of AxV served as control. It should be noted that a single application of AxV was capable of reducing the HIV-1 replication significantly.
|
, IL-1ß, IL-10, and IL-12. We have shown previously that in the presence of apoptotic cells, the concentration of the anti-inflammatory cytokine IL-10 was raised, but the levels of the proinflammatory cytokines TNF-, IL-1 ß, and IL-12 were decreased [12
, 15
]. To examine the involvement of heat-stabile, long-living danger signals on the resolution of inflammation, we monitored the cytokine secretion of LPS-stimulated macrophages in the presence of primary necrotic cells, which were necrotized by heat treatment (56°C, 30 min). We found that under those conditions, necrotic cells led to a decreased secretion of TNF- and an increased secretion of IL-10 (Fig. 4A
). To achieve that putative, labile danger signals would still persist in the system, we induced primary necrosis via mechanical stress. In contrast to heat-necrotized cells, cells necrotized by mechanical shear stress showed a significant reduction of IL-10 secretion and therefore, have an increased inflammatory and immunostimulatory potential (Fig. 4B)
. Furthermore, we observed recently that several receptors involved in the uptake of apoptotic cells are also involved in the uptake of primary necrotic cells [12
]. Focusing on the immunomodulatory effects of AxV, we found that the phagocytosis by monocyte-derived macrophages of primary necrotic cells could also be blocked with AxV (Fig. 4C)
.
 View larger version (15K): [in a new window] |
Figure 4. Necrotic cells influence the cytokine secretion pattern of LPS-activated HMDM, and their phagocytosis is dependent on PS. TNF- and IL-10 secretion of HMDM, which were stimulated with LPS in the presence of medium or heat-induced primary necrotic cells (A) or mechanical stress-induced necrotic cells (B), was quantified by ELISA. Values are the mean ± SD of four assays. Furthermore, the phagocytosis of heat-induced necrotic PBL by HMDM in the presence or absence of AxV is shown (C). Values are the mean ± SD of four independent experiments. It should be noted that (A) heat-induced necrotic cells reduced the LPS-induced secretion of TNF- and increased the secretion of IL-10 in HMDM, and (B) mechanical stress-induced necrotic cells decreased the secretion of IL-10 in HMDM. (C) HMDM showed a decreased phagocytosis of primary necrotic cells in the presence of the PS-binding protein AxV. Co, Control. Results were considered statistically significant for *P < 0.05 and highly significant for **P < 0.01.
|
Apoptosis is defined as programmed cell death or cellular suicide, whereas necrosis arises as a result of a violent external stimulus. A hallmark of apoptotic cells, in contrast to necrotic ones, is that they maintain their membrane integrity over time. Thus, the release of intracellular components, which could damage the surrounding tissue, induce inflammation, or elicit immune responses, is avoided [26 ]. To ensure immediate recognition and uptake by phagocytes, apoptotic cells undergo early membrane modifications. One such event is the exposure of PS in the outer leaflet of the plasma membrane associated with a loss of phospholipid asymmetry. Phagocytes interact with PS on apoptotic cells mainly through secreted bridging adaptor proteins, also called opsonins. The latter are important for a high-capacity clearance of apoptotic cells [27 ].
The recognition of exposed PS triggers the release of immunosuppressive cytokines [14 , 15 , 28 ], which quench inflammation and prevent the maturation of antigen-presenting dendritic cells (DC). Apoptotic cells per se do influence the production of soluble pattern recognition receptors such as pentraxin 3 by maturing DC [29 ]. Furthermore, the opsonization of apoptotic cells by soluble factors influences their antigen presentation by DC and thereby, modulates their immunogenicity [30 ]. Recently, it was discussed that alterations of the plasma membrane phospholipid distribution have important influences, not only for cell clearance but also for the execution of apoptosis [31 ].
AxV is a natural-occurring, specific ligand for PS and may consequently interfere in vivo with the immunosuppressive effects of apoptotic cells. We showed that the immunogenicity of apoptotic cells could be restored by the addition of AxV. During apoptotic as well as necrotic cell death, autoantigens are cleaved or otherwise modified, and these modifications may render cryptic epitopes immune-dominant (reviewed in ref. [32 ]). When interfering with the clearance of dying cells, DC may acquire modified autoantigens such as apoptotic nuclei and chromatin, and consequently, autoreactive T cells can be activated. This may also lead to chronic autoimmunity, as is the case in SLE [24 ]. It was also shown that an impaired clearance of dying tumor cells can lead to tumor rejection. AxV decreased apoptotic cell uptake by peritoneal macrophages and concomitantly increased their uptake by CD8+/11c+ DC [33 ].
To analyze the physiological function of AxV in the induction of an immune response, we used an AxV-deficient mouse mutant [21 ]. We immunized the AxV-deficient and WT mice with apoptotic and necrotic Sp2/0 cells and monitored the DTH reaction. This reaction is characterized by an induction of a specific T cell response and can be used as a model for an induction of a specific immunization against, e.g., tumor antigens [34 ]. We used allogeneic cells for the immunization experiments to get a strong immune response in WT mice. Our results show that primary necrotic cells, which were prepared by mechanical stress and therefore, were still endowed with some of their internal danger signals, induced a DTH reaction only in WT but not in AxV-deficient mice (Fig. 2A) . Apoptotic cells led to a weaker allogeneic reaction in general, which, however, was again more pronounced in WT mice. In the case of apoptotic cells, the DTH reaction is most likely a result of secondary necrotic cells, which may contaminate the inoculate or emerge from apoptotic cells in situ after injection. We conclude that the proinflammatory properties of necrotic cells depend on the inductor of necrosis. The proinflammatory milieu finally contributes to the increased DTH reaction. It is interesting that endogenous AxV also skewed the cytokine secretion of primary necrotic cell clearance toward inflammation (Fig. 2B) , as peritoneal macrophages from AxV-deficient mice showed a higher secretion of the anti-inflammatory cytokine IL-10 in response to necrotic Sp2/0 cells. The AxV-deficient mouse strain was generated by homologous recombination containing a lacZ reporter gene cassette fused in-frame with Exon 3 (Pos. 178) of the AxV gene [21 ]. X-Gal staining of the macrophages clearly demonstrated the expression of AxV within the macrophage population (not shown). We propose that in the WT situation, AxV is expressed and released from macrophages and then binds to the dying and dead cells. Consecutively, the engulfment of apoptotic and/or necrotic cells is partially blocked.
When AxV binds to apoptotic cells, it "crystallizes" as an extended two-dimensional network. It has an autocrine function and activates a way of cell entry. This results in the internalization of the PS-expressing membrane patches from the surfaces of apoptotic macrophages [35
]. We have shown previously that AxV shows positive cooperativity for PS binding on membranes of apoptotic and necrotic but not of viable cells. We suggested that phagocytes can differentiate between dying and viable cells by means of PS clustering and consecutively, by the lateral mobility of the cellular membranes [36
]. PS-mediated phagocytosis of apoptotic cells suppresses inflammatory signals such as TNF-, IFN-
, and NO and also triggers the production of TGF-ß, an anti-inflammatory cytokine [15
]. We have previously shown that necrotic cells, like apoptotic ones, can also engage CD36 and a surface receptor recognized by mAb 217G8E9. They thereby mediate anti-inflammatory signals [12
]. Here, we show that necrotic cells induced by heat treatment promote the release of anti-inflammatory cytokines by LPS-activated macrophages, as is the case for apoptotic cells. Heat-induced necrosis destroys heat-labile and short-lived danger signals. The uptake of the necrotic cells then becomes anti-inflammatory. In contrast, mechanical stress-induced necrotic cells led to a decreased secretion of the anti-inflammatory cytokine IL-10 and consecutively, displayed a more immunostimulatory phenotype. We conclude that the proinflammatory properties of necrotic cells depend on the inductor of necrosis. Furthermore, we added AxV to this experimental setting and found a significant, further up-regulation of IL-10 secretion in the case of heat-necrotized cells but not in the case of stress-induced necrotic cells (not shown), further confirming the higher proinflammatory potency of mechanical stress-induced necrotic cells when compared with heat-induced necrosis.
The primary force of the immune system is the need to detect and protect against danger [37 ]. One mode of action of danger signals is to stimulate the maturation and activation of DC necessary for the initiation of primary and secondary immune responses. Endogenous danger signals, released by tissues undergoing stress, damage, or abnormal death, and also exogenous danger signals, elaborated by pathogens, can contribute to this stimulation. Some endogenous danger signals, which have been discovered recently are heat-shock proteins, nucleotides, reactive oxygen intermediates, extracellular matrix breakdown products, neuromediators, cytokines such as the IFNs [38 ], as well as uric acid [39 ], ATP [40 , 41 ], and HMGB1 [16 , 42 ], which as well as its best-characterized receptor for advanced glycation end products, is important for the maturation of human plasmacytoid DC [43 ] and also controls T cell activation [44 ]. Furthermore, the release of intracellular factors from necrotic tumor cells can promote reactive angiogenesis, stromal proliferation, and local immune suppression [45 ]. Future work has to be focused strongly on all such "alarmins" [46 ], which signal the body: "Attention, tissue damage has occurred!"
Our mechanical method for necrosis induction does not operate with chemicals or heat. It is faster than repeated cycles of freezing and thawing and mimics perfectly the situation in the body when tissue or cell damage occurs. Just recently, it was shown that the potency in DC activation produced by mechanical lysis is much stronger than thermal necrosis of the tumor cells [47 ]. The mechanical necrosis induction saves some of the labile, internal danger signals. However, short-lived danger signals, such as ATP, which are enzymatically degradable [48 ], and those that need a threshold concentration for their activity such as uric acid [39 ], may still be lost in our necrotizing process. The action of uric acid is dependent on the formation of crystals, which only appear above a certain threshold concentration of this compound [49 ]. We are currently working on a necrotizing procedure, which allows coincubation of responder cells and immunization of animals with necrotic cells after a lag time of only a few seconds.
Many viral infections (such as HIV) lead to the release of several proinflammatory cytokines and to extensive apoptosis. The latter may contribute to the impaired immune response accompanying such infections [50 , 51 ]. Recently, it has been shown that PS can be detected at the surface of HIV-1 and that AxV can be used to enrich these virus particles. Furthermore, the infection by HIV-1 strains of monocytes can be compromised upon the addition of AxV during the infection process [23 ]. Here, we demonstrated that clinically relevant, chronic infections of HMDM with HIV-1 isolates can be inhibited by a single application of AxV. In addition, Ma et al. [52 ] showed that Annexin II is necessary for HIV-1 uptake into human macrophages. This HIV-1 PS interaction with Annexin II could be disrupted by a secretory leukocyte protease inhibitor [52 ].
Taken together, AxV efficiently blocks the silent entry of HIV into macrophages as well as the swift clearance of apoptotic and necrotic cells. The disturbed, PS-dependent clearance by macrophages of apoptotic cells leads to the accumulation of the latter and to the occurrence of late apoptotic cells, which have lost their membrane integrities. The proinflammatory cytokine profile of the late clearance and endogenous danger signals released from cells, which have lost their membrane integrity, build a proinflammatory microenvironment. DC may then pick up antigens derived from the dying cells in a proinflammatory milieu and present the cell-derived antigens together with costimulation (Fig. 5 ). We showed that AxV increased the immunogenicity of apoptotic cells significantly. In addition, immunization with mechanical stress-induced necrotic cells was used to mimic the above-mentioned scenario. Only in the presence of endogenous AxV (WT animals) was a specific immune response against the dead cells to be observed. Furthermore, endogenous AxV led to a decreased secretion of IL-10 in peritoneal macrophages in response to mechanical stress-induced necrotic cells. Examinations of the in vitro mechanisms underlying the observed effects revealed that mechanical stress-induced necrotic cells led to an inflammatory modulation of macrophages in contrast to apoptotic cells and heat-induced necrotic ones. In conclusion, apoptotic and necrotic cells strongly influence the immune response. AxV is an important modulator of this interaction. It inhibits certain viral infections and increases the immune response against tumor cells and more general, that against dying and dead cells.
 View larger version (31K): [in a new window] |
Figure 5. AxV influences clearance and uptake of PS-exposing particles and cells.
|
ACKNOWLEDGEMENTS
This work was supported by "Deutsche Forschungsgemeinschaft" SFB 643 (Project B5), by the Interdisciplinary Center for Clinical Research (IZKF; Project Number A4 and N2) at the University Hospital of the University of Erlangen-Nuremberg, by the European Commissions [E.U. (QLK3-CT-2002-02017_APOCLEAR)], by the Lupus Erythemathodes Selbsthilfegemeinschaft e.V., by the responsif GmbH Erlangen, and by the Program Alban, the European Union Program of High Level Scholarships for Latin America, Scholarship No. "E04D047956VE" to L. E. M. S. F. was supported by the DFG Research Training Grant GK592, the University of Erlangen’s ELAN program, and by the Doktor Robert Pfleger Foundation, Bamberg.
FOOTNOTES
1 These authors contributed equally to this work. 
Received March 5, 2006; revised July 31, 2006; accepted August 6, 2006.
REFERENCES
vß3, CD14, CD36, and complement C1q in the phagocytosis of primary necrotic lymphocytes by macrophages Arthritis Rheum. 54,927-938[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  G. Liu, J. Wang, Y.-J. Park, Y. Tsuruta, E. F. Lorne, X. Zhao, and E. Abraham High Mobility Group Protein-1 Inhibits Phagocytosis of Apoptotic Neutrophils through Binding to Phosphatidylserine J. Immunol., September 15, 2008; 181(6): 4240 - 4246. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. G. Blankenberg In Vivo Detection of Apoptosis J. Nucl. Med., June 1, 2008; 49(Suppl_2): 81S - 95S. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
