Mouse Dendritic Cells Matured by Ingestion ofApoptotic Blebs Induce T Cells to
Produce Interleukin-17
Justin H. Fransen, Luuk B. Hilbrands, Jurjen Ruben, Monique Stoffels, Gosse J. Adema,Johan van der Vlag, and Jo H. Berden
Objective. Systemic lupus erythematosus (SLE) isan autoimmune disease characterized by the formationof antinuclear autoantibodies. Increased apoptosis andreduced clearance of apoptotic material have been as-signed a role in the pathogenesis of SLE, but theunderlying mechanisms remain elusive. During apopto-sis apoptotic blebs are formed in which autoantigens areclustered. The cellular remnants after blebbing arereferred to as apoptotic cell bodies. We undertook thisstudy to compare the effects of apoptotic blebs andapoptotic cell bodies on maturation of dendritic cells(DCs) and their T cell stimulatory capacity in a murinesetting.
Methods. The uptake by DCs of apoptotic blebsand apoptotic cell bodies was analyzed by flow cytom-etry and confocal microscopy. DC maturation and DC-induced T cell activation were determined by measuringexpression of costimulatory molecules using flow cytom-etry and by measuring production of cytokines usingenzyme-linked immunosorbent assay.
Results. DCs internalized apoptotic blebs moreefficiently than apoptotic cell bodies. Incubation of DCswith apoptotic blebs resulted in increased CD40 andCD86 expression and increased interleukin-6 (IL-6) andtumor necrosis factor � production, while apoptotic cell
bodies had no stimulatory effects. Using chloroquine,apoptotic bleb–induced DC maturation was shown to beindependent of Toll-like receptors 3, 7, and 9. Interest-ingly, in cocultures with allogeneic T cells, bleb-maturedDCs induced production of IL-2, interferon-�, and, inparticular, IL-17, suggesting a Th1/Th17 response.
Conclusion. Apoptotic blebs, in contrast to apo-ptotic cell bodies, induce DC maturation, thereby pro-viding DCs with increased Th17 cell stimulatory capac-ity. These data imply that apoptotic bleb–induced DCmaturation represents an important driving force in theautoimmune response in SLE.
Systemic lupus erythematosus (SLE) is an auto-immune disease characterized by the formation of auto-antibodies against nuclear self antigens. The formationof these autoantibodies was shown to be T cell depen-dent (1). SLE is a prototype of an immune complex–mediated disease, in which the deposition of chromatin–antichromatin immune complexes in the basementmembrane of, for example, skin and kidney can elicit alocal inflammatory reaction (2). Chromatin, in particu-lar mononucleosomes, can be detected in the circulationof patients with SLE and lupus mice (3,4). Most likely,this chromatin is released from apoptotic cells throughthe action of apoptosis-activated endonucleases (5).Both an increased rate of apoptosis and an insufficientclearance of apoptotic cells/material have been impli-cated as key processes leading to the presence of chro-matin in the circulation of SLE patients and to thesubsequent development of the antichromatin immuneresponse. Indeed, experimental interference with apo-ptosis or with the clearance of apoptotic material canlead to the development of antinuclear autoantibodiesand lupus disease manifestations like glomerulonephritisin mice (6). Accordingly, the clearance of apoptotic
Mr. Fransen’s work was supported by the Radboud UniversityNijmegen Medical Centre PhD program. Drs. van der Vlag andBerden’s work was supported by the Dutch Kidney Foundation (grantC05.2119).
Justin H. Fransen, MSc, Luuk B. Hilbrands, MD, PhD, JurjenRuben, MSc, Monique Stoffels, MSc, Gosse J. Adema, PhD, Johanvan der Vlag, PhD, Jo H. Berden, MD, PhD: Radboud UniversityNijmegen Medical Centre, Nijmegen, The Netherlands.
Address correspondence and reprint requests to Jo H. Ber-den, MD, PhD, Department of Nephrology (464), Radboud UniversityNijmegen Medical Centre, PO Box 9101, 6500 HB Nijmegen, TheNetherlands. E-mail: [email protected].
Submitted for publication August 22, 2008; accepted inrevised form May 8, 2009.
2304
material by phagocytes appears to be impaired both inlupus mice and in SLE patients (7,8).
Apoptosis leads to dramatic morphologic andbiochemical changes of cells. The formation of apoptoticblebs is one of the characteristic cellular features duringapoptosis. Autoantigens that are involved in SLE areclustered in the segregating apoptotic blebs (9), andthese blebs are specifically recognized by autoantibodies(10,11). We define the apoptotic cell body as the cellularremnant after the blebbing process has ended (seeFigure 1). Apoptotic cell bodies also have been termedlate apoptotic or secondary necrotic cells. During apo-ptosis, various modifications of autoantigens (e.g., chro-matin) may take place, including cleavage by protein-ases, caspases, and/or endonucleases and specificposttranslational modifications (6,11–15). These find-ings suggest that apoptotic blebs are a source of (mod-ified) autoantigens and that they may play a key role inthe development of the antichromatin response in SLE.Normally, early apoptotic cells are efficiently removedby macrophages, nonprofessional phagocytes, and den-dritic cells (DCs), which prevents the release of blebscontaining (modified) autoantigens. We hypothesizethat in SLE, modified chromatin derived from apoptoticcells, and in particular from blebs, has escaped a properclearance. Modified chromatin can then act as a dangersignal leading to maturation of DCs. These maturedDCs can present the altered self antigens to T cells(6,16–18).
Whether the exposure of the immune system toan (auto)antigen will result in tolerance or (auto)immu-nity depends on the maturation status of the DCs, inparticular that of myeloid DCs. In the immature stateDCs capture antigens, including self antigens like apo-ptotic cells. Immature DCs loaded with self antigens caninduce tolerance, which is associated with the produc-tion of antiinflammatory cytokines like transforminggrowth factor � (TGF�) and interleukin-10 (IL-10). In
contrast, fully matured and activated DCs are the mostefficient antigen-presenting cells that induce antigen-specific immunity, associated with the production ofproinflammatory cytokines like tumor necrosis factor �(TNF�) and IL-6 (18). Several reports describe that inparticular late apoptotic cells and/or secondary necroticcells lead to maturation of DCs, while early apoptoticcells do not exert such a maturation effect (19–21).
Recently, we have demonstrated that apoptosis-induced histone acetylation is a trigger for the immunesystem in SLE. Histone acetylation appeared to bepathogenic in prediseased lupus mice, whereas hyper-acetylated nucleosomes led to maturation of myeloidDCs and subsequent syngeneic T cell activation (11).Other studies have shown that administration of apopto-tic cells to lupus-prone mice, alone or in combinationwith DCs, leads to the development of autoimmunity(22–26). Notably, another type of DC, the plasmacytoidDC (PDC), responds specifically to RNA/DNA-containing immune complexes by producing the type Iinterferon interferon-� (IFN�), which seems cruciallyinvolved in the pathogenesis of SLE (27,28). However,ingestion of apoptotic cells by PDCs is unlikely to occur(29).
We hypothesize that the ingestion of apoptoticblebs and/or apoptotic cell bodies by myeloid DCs leadsto maturation of DCs and subsequent stimulation of Tcells. In this study in mice, we have separated apoptoticblebs from apoptotic cell bodies, and we have evaluatedtheir ingestion by myeloid DCs and their capacity toinduce maturation of myeloid DCs. Furthermore, wehave determined the T cell stimulatory capacity of theseDCs after exposure to apoptotic blebs or apoptotic cellbodies.
MATERIALS AND METHODS
Cell culture. BALB/c (H-2d) and CBA (H-2k) miceages 8–10 weeks were purchased from Charles River (Maas-tricht, The Netherlands) and maintained under specificpathogen–free conditions and handled according to the guide-lines of the local ethics committee of the Radboud UniversityNijmegen. Bone marrow–derived DCs were obtained by cul-turing bone marrow from BALB/c mice as previously described(30,31). Briefly, bone marrow was flushed from femur and tibiaand cultured for 8 days in 6-well plates (1 � 106 cells/well;Corning Costar, Badhoevedorp, The Netherlands) containingRPMI 1640 DM medium (Invitrogen Life Technologies,Breda, The Netherlands) with 10% fetal calf serum (FCS;Greiner Bio One, Alphen aan den Rijn, The Netherlands), 1%pyruvate, 1% Glutamax, 1% penicillin/streptomycin (all fromInvitrogen Life Technologies), and 20 ng/ml granulocyte–macrophage colony-stimulating factor (Pepro-Tech, London,
Figure 1. Scheme defining apoptotic blebs and apoptotic cell bodies.Left, Healthy cell. Middle, Actively blebbing apoptotic cell. Right,Remaining apoptotic cell body (ACB) that has finished blebbing withseparated apoptotic blebs.
UK). Murine 32D clone 3 (32Dcl3) cells (H-2k) (32,33) werecultured in complete RPMI 1640 DM medium supplementedwith 15% WEHI-3B–conditioned medium as a source ofmurine IL-3 (DSMZ, Braunschweig, Germany). Cells wereroutinely tested for mycoplasma (Gen-Probe, San Diego, CA),and the results were consistently negative.
Induction and measurement of apoptosis and isolationof apoptotic blebs and cell bodies. Apoptosis was induced in32Dcl3 cells by incubating the cells with 10 �M4-nitroquinoline 1-oxide (Sigma-Aldrich, Zwijndrecht, TheNetherlands) for 24 hours. Apoptosis was routinely deter-mined by staining cells with annexin V–fluorescein isothiocya-nate and propidium iodide (PI) (BioVision, Palo Alto, CA)and analyzed by flow cytometry using a FACSCalibur instru-ment (BD PharMingen, Alphen aan den Rijn, The Nether-lands) following the protocol of the manufacturer (11). Apo-ptotic cell bodies were isolated from apoptotic cell culture bycentrifugation for 10 minutes at 1,550g at room temperature.Subsequently, apoptotic blebs were isolated from the resultingsupernatant by centrifugation for 50 minutes at 15,700g atroom temperature. Pelleted blebs or apoptotic cell bodies weregently resuspended in RPMI 1640 DM medium. The concen-trations of blebs and apoptotic cell bodies were determinedwith the bicinchoninic acid protein determination assay(Sigma-Aldrich) and expressed as protein equivalents. FCSwas added to the bleb and apoptotic cell body preparations toachieve a final concentration of 10%. Equal amounts inprotein equivalents of blebs and apoptotic cell bodies con-tained approximately equal amounts of nucleic acids as deter-mined by measurements of absorbance at 260 nm (A260 nm)and A280 nm, respectively.
Phagocytosis of apoptotic blebs or apoptotic cell bod-ies. First, 32Dcl3 cells were labeled with 10 �M PKH26(Sigma-Aldrich) before the induction of apoptosis. Next,PKH26-labeled apoptotic cell bodies or blebs (10 or 100�g/ml) were added to 0.5 � 106/ml DCs labeled with 1 �M5,6-carboxyfluorescein succinimidyl ester (CFSE; InvitrogenLife Technologies) with a final volume of 200 �l, and subse-quently incubated at 0°C or 37°C for 2 hours. Internalization ofblebs or apoptotic cell bodies by DCs was visualized withconfocal laser scanning microscopy using a Leica TCS NTsystem (Leica Lasertechnik, Heidelberg, Germany). In addi-tion, the percentage of CFSE/PKH26 double-positive cells wasanalyzed by flow cytometry.
Determination of DC phenotype by flow cytometry andcytokine enzyme-linked immunosorbent assay (ELISA). Foranalysis of DC phenotype, 0.5 � 106 immature DCs per mlwere incubated in medium alone or medium supplementedwith blebs or apoptotic cell bodies. Addition of 1 �g/mllipopolysaccharide (LPS) (L4391; Sigma-Aldrich) was used asa positive control for maturation of DCs. After 14 hours ofincubation, cells were harvested and subjected to direct orindirect fluorescence staining of cell surface markers, essen-tially as described previously (31). Briefly, cells were stainedwith anti-CD40 (clone FGK45.5; Miltenyi Biotec, Utrecht, TheNetherlands) or isotype-matched control antibodies (R35-95;BD PharMingen), followed by phycoerythrin (PE)–conjugatedgoat anti-F(ab�)2 anti-rat (Beckman Coulter, Bedfordshire,UK) and Alexa 647–conjugated anti-CD11c (N418; Serotec,Oxford, UK), or with PE–conjugated anti-CD86 (PO3.1; eBio-science, Malden, The Netherlands) and Alexa 647–conjugated
anti-CD11c. Samples were analyzed using a FACSCaliburinstrument, and data were processed using CellQuest software(BD PharMingen). Supernatant was collected for determina-tion of levels of TNF�, IL-1�, IL-6, IL-23 (all from eBio-science), and IFN� (R&D Systems, Abingdon, UK) in sand-wich ELISA according to the protocols provided by themanufacturers.
Bleb-induced DC maturation was investigated for itsdependency on Toll-like receptors (TLRs) 3, 7, and 9 byincubating DCs with chloroquine (10 �M; Invivogen, SanDiego, CA) alone or in combination with apoptotic blebs (100�g/ml), LPS (1 �g/ml), or CpG-containing oligonucleotide1826 (1 �g/ml; Invivogen) as positive control. Maturation wasexamined by measuring levels of IL-6 in supernatants byELISA.
Mixed leukocyte reaction (MLR). Spleen cells fromCBA mice were isolated by passing spleen tissue through a70-�m Cell Strainer (BD PharMingen) followed by treatmentwith erythrocyte lysis buffer (0.15M NH4Cl, 10 mM KHCO3,0.1 mM Na2-EDTA [pH 7.4]) for 1 minute, and cells werefinally washed 3 times in medium (31). In MLRs, 2 � 104
BALB/c DCs were incubated with 1 � 105 CBA splenocytes, asa source for T cells, in 200 �l in 96-well plates at 37°C and 5%CO2. The effects of the addition of blebs or apoptotic cellbodies were analyzed. As controls, splenocytes with or withoutapoptotic blebs or cell bodies, DCs with or without apoptoticblebs or cell bodies, and LPS-matured DCs were included.Apoptotic cell bodies and blebs were matched for the spleno-cytes by major histocompatibility complex, which excludedallogeneic T cell stimulation by blebs or apoptotic cell bodies.On day 6 of the MLR, supernatant was collected for determi-nation of IL-2, IL-4, IL-5, IL-10, IL-17, and IFN� (eBio-science) production in sandwich ELISA according to thecorresponding protocols.
Statistical analysis. All data presented are obtainedfrom at least 3 different experiments using at least 3 differentmice. Values are expressed as the mean � SEM, and signifi-cance was determined by Student’s t-test using GraphPadPrism version 4 software (GraphPad Software, San Diego,CA). P values less than 0.05 were considered significant.
RESULTS
Mouse DCs ingest apoptotic blebs more effi-ciently than apoptotic cell bodies. Previously, it has beendemonstrated that in particular late apoptotic cells andnecrotic cells, in contrast to early apoptotic cells, inducematuration of DCs, which increases the T cell stimula-tory capacity of these DCs (19,34). However, in allstudies so far, apoptotic cell bodies or a mixture ofapoptotic blebs and cell bodies have been used, andconsequently the separate effects of blebs and bodies onDC phenotype and function remain elusive. Therefore,we first compared the uptake of purified blebs andapoptotic cell bodies by DCs.
Blebs and apoptotic cell bodies were isolatedfrom cultures of late apoptotic 32Dcl3 cells, as measured
2306 FRANSEN ET AL
by annexin V and PI staining (Figure 2). Apoptotic blebsand apoptotic cell bodies were separated by serial cen-trifugation. Using confocal laser scanning microscopy wecould visualize the uptake of PKH26-labeled apoptoticblebs (Figure 3A) and apoptotic cell bodies (Figure 3B)by CFSE-labeled DCs. The uptake and binding ofapoptotic blebs and apoptotic cell bodies was quantifiedby flow cytometry. Figure 3C depicts a representativeexample of a flow cytometric analysis of CFSE-labeledDCs that have ingested PKH26-labeled blebs or bodies.To distinguish uptake from binding, experiments wereperformed at 37°C and at 0°C, respectively, and theuptake was defined as the difference in percentages ofdouble-positive (both PKH26- and CFSE-labeled) cellsobserved at 37°C and at 0°C. Interestingly, DCs ingestedapoptotic blebs more efficiently than apoptotic cellbodies (Figure 3D). In particular, at the highest concen-tration tested (100 �g/ml), the percentage of DCs thathad ingested apoptotic blebs (46%; 54% uptake andbinding at 37°C minus 8% binding alone at 0°C) wassignificantly higher than the percentage that had in-gested apoptotic cell bodies (18%; 29% uptake andbinding at 37°C minus 11% binding alone at 0°C).
Apoptotic blebs enhance the expression of co-stimulatory molecules on DCs. Next, we analyzed theeffect of isolated apoptotic blebs and cell bodies onmaturation of mouse myeloid DCs by determining theexpression of the costimulatory molecules CD40 andCD86. Incubation of DCs with blebs led to an increasedexpression of CD40 (Figure 4A) and CD86 (Figure 4B)on CD11c-positive cells. In contrast, incubation of DCswith apoptotic cell bodies (10 or 100 �g/ml) did not leadto an increased expression of the costimulatory mole-cules CD40 and CD86 (Figures 4A and B). Interestingly,the expression of CD40 was even decreased after incu-bation of DCs with 100 �g/ml apoptotic cell bodies(Figure 4A). In summary, apoptotic blebs induce anincreased expression of costimulatory molecules onDCs, whereas apoptotic cell bodies do not result inmaturation of DCs.
Apoptotic blebs induce IL-6 and TNF� produc-tion by DCs, which is not mediated by TLRs 3, 7, and 9.Additionally, we examined the response of DCs toapoptotic blebs or apoptotic cell bodies at the level of
Figure 3. Dendritic cells (DCs) ingest apoptotic blebs more efficientlythan apoptotic cell bodies (ACB). DCs were incubated for 2 hours at0°C or 37°C with 10 or 100 �g/ml blebs or apoptotic cell bodies. A andB, Ingestion of PKH26-labeled blebs (A) and apoptotic cell bodies (B)by 5,6-carboxyfluorescein succinimidyl ester (CFSE)–labeled DCs wasvisualized using confocal microscopy (original magnification � 630).C, CFSE-labeled DCs that had ingested/bound PKH26-labeled blebsor apoptotic cell bodies were analyzed by flow cytometry. A represen-tative example is shown. D, Binding (0°C) and uptake and binding(37°C) were quantified by determination of the percentage of CFSE/PKH26 double-positive cells by flow cytometry. Percentage uptake isthe difference between uptake and binding (37°C) and binding alone(0°C). Values are the mean and SEM. � � P � 0.05.
Figure 2. Induction of apoptosis in culture of 32D clone 3 (32Dcl3)cells. Cells were analyzed for annexin V binding and propidium iodide(PI) staining by flow cytometry. Cells double-positive for annexin Vand PI can be considered late apoptotic. A, Analysis of 32Dcl3 cellsfrom a normal culture reveals hardly any binding of annexin V or PIstaining. B, Typical analysis of 32Dcl3 cells that were incubated with 10�M 4-nitroquinoline 1-oxide for 24 hours to induce apoptosis showsthat nearly 100% of the cells stain positively for annexin V and PI. Cand D, PKH26-labeled apoptotic blebs (C) and apoptotic cell bodies(D) were isolated from late apoptotic 32Dcl3 cell cultures (as pre-sented in B) by serial centrifugation (original magnification � 400).
cytokine production. Incubation of DCs with 10 �g/mland 100 �g/ml apoptotic blebs resulted in a dose-dependent increase in secretion of the proinflammatorycytokines IL-6 and TNF� compared with control DCs(Figures 5A and B). In contrast, apoptotic cell bodieswere not able to induce IL-6 or TNF� secretion by DCs(Figures 5A and B). Even incubation with a 100-foldexcess (1 mg/ml) of apoptotic cell bodies over blebs did
not lead to IL-6 or TNF� production (not shown). Theproinflammatory cytokines IL-1�, IL-23 and IFN� werenot detectable in supernatants of DCs incubated with
Figure 5. Apoptotic blebs induce enhanced production ofinterleukin-6 (IL-6) and tumor necrosis factor � (TNF�) by dendriticcells (DCs) without mediation by Toll-like receptors (TLRs) 3, 7, and9. DCs were incubated with 10 or 100 �g/ml blebs or apoptotic cellbodies (ACB), or with 1 �g/ml lipopolysaccharide (LPS) or withmedium alone. A and B, The production of the proinflammatorycytokines IL-6 (A) and TNF� (B) was determined by enzyme-linkedimmunosorbent assay. C, The possible involvement of TLRs 3, 7, and9 in apoptotic bleb–induced DC maturation was examined using theinhibitor chloroquine, with CpG-containing oligonucleotide 1826 as apositive control. Values are the mean and SEM. � � P � 0.05.
Figure 4. Apoptotic blebs induce enhanced expression of the costimu-latory molecules CD40 and CD86 by dendritic cells (DCs). DCs wereincubated with 10 or 100 �g/ml blebs or apoptotic cell bodies (ACB),or with 1 �g/ml lipopolysaccharide (LPS) or with medium alone.Expression of CD40 (A) and CD86 (B) on CD11c-positive cells wasanalyzed by flow cytometry. Data are expressed as the mean and SEMmean fluorescence intensity (MFI). � � P � 0.05.
2308 FRANSEN ET AL
either apoptotic blebs or apoptotic cell bodies (data notshown).
Subsequently, we evaluated the possible involve-ment of TLRs in apoptotic bleb–induced maturation ofDCs. We focused on TLRs 3, 7, and 9, since normallythese TLRs can be triggered by various (endogenous)nucleic acid–containing compounds, which may bepresent in these apoptotic blebs. Therefore, we appliedthe endosome acidification inhibitor chloroquine toblock signal transduction through TLRs 3, 7, and 9.However, addition of chloroquine in combination with100 �g/ml apoptotic blebs did not inhibit IL-6 produc-tion by DCs, while the IL-6 production induced byCpG-containing oligonucleotide 1826 (a TLR-9 ligand)was completely inhibited (Figure 5C).
In summary, apoptotic blebs, but not apoptoticcell bodies, induce the production of the proinflamma-tory cytokines IL-6 and TNF� by DCs. Bleb-inducedcytokine production by DCs appears to be independentof TLRs 3, 7, and 9.
Apoptotic bleb–matured DCs stimulate the pro-duction of IL-2, IFN�, and IL-17 by activated T cells. Asa functional test, we compared T cell stimulatory capac-ities between DCs incubated with apoptotic blebs, DCsincubated with apoptotic cell bodies, or DCs alone in anallogeneic MLR with splenocytes as responders. Weused IL-2 production as a measure of T cell prolifera-tion, and we measured the production of cytokinesspecific for Th1 (IFN�), Th2 (IL-4, IL-5, IL-10), or Th17(IL-17) responses. IL-10 production may also indicatethe involvement of Treg cells.
Bleb-matured DCs showed an increased ability toactivate T cells as measured by IL-2 production (Figure6A). In contrast to bleb-matured DCs, DCs exposed toapoptotic cell bodies induced IL-2 production in MLRsimilar to that induced by control DCs. Moreover, IFN�and IL-17 production were significantly increased inMLR with bleb-matured DCs compared with their pro-duction in MLR either with DCs exposed to apoptoticcell bodies or with control DCs (Figures 6B and C).Notably, IL-17 production in MLR with bleb-maturedDCs was several times higher than that in MLR withLPS-matured DCs, while IFN� production was aboutthe same in both conditions. As a result, the IL-17:IFN�ratio was �5 after stimulation with bleb-matured DCscompared with a ratio of only �1.5 after stimulationwith LPS-matured DCs. These data suggest a mixedTh1/Th17 response of responder cells toward bleb-matured DCs. No secretion could be detected of the Th2cytokines IL-4, IL-5, and IL-10 in any of the conditionstested in MLR (data not shown), which suggests the
absence of a Th2 response. The absence of IL-10 alsosuggests that Treg cells are not involved. In summary,
Figure 6. Apoptotic bleb-matured dendritic cells (DCs) are able toinduce Th1/Th17 T cell cytokine production in a mixed leukocytereaction (MLR). BALB/c (H-2d) DCs (20,000) were incubated with 10or 100 �g/ml blebs or apoptotic cell bodies (ACB), or with 1 �g/mllipopolysaccharide (LPS) or with medium alone, and were used in anMLR with CBA (H-2k) splenocytes (100,000) as a source for T cells.After 6 days, supernatants were isolated and examined by enzyme-linked immunosorbent assay for production of interleukin-2 (IL-2) (A),interferon-� (IFN�) (B), and IL-17 (C). No IL-2, IFN�, or IL-17production was found when splenocytes or DCs alone were incubatedwith apoptotic blebs (data not shown). Values are the mean and SEM.� � P � 0.05.
bleb-matured DCs induce in vitro a mixed Th1/Th17response, whereas DCs exposed to apoptotic cell bodieshave no effect on cytokine production in MLR.
DISCUSSION
We observed in a murine system that the percent-age of DCs that ingested apoptotic blebs was about2–3-fold higher than the percentage of DCs that in-gested apoptotic cell bodies. Our data extend some ofthe observations made by Frisoni et al (35), who de-scribed an increased uptake by DCs of a preparationenriched in rather large apoptotic blebs compared with anonfractionated apoptotic cell preparation, especially inthe presence of opsonizing antibodies. Seemingly incontrast with Frisoni et al’s data and ours, Ip and Laushowed that late apoptotic cells induced maturation ofDCs (19). However, in their study apoptotic cell bodiesand apoptotic blebs were not separated or defined,suggesting that the observed effects were mediated byapoptotic blebs and not by apoptotic cell bodies.
We found that the addition of apoptotic blebs,but not of apoptotic cell bodies, to mouse DC culturesresulted in maturation of DCs, as measured by anup-regulated expression of the costimulatory moleculesCD40 and CD86 and an increased production of theproinflammatory cytokines IL-6 and TNF�. Interest-ingly, elevated levels of endogenous IL-6 have beendescribed in SLE (36). There are several candidatemolecules residing in the blebs that may specificallytrigger the maturation of DCs. It has been shown thatbesides nucleosomes, which are comprised of histonesand DNA, many other SLE-associated autoantigens areclustered in blebs, such as Sm, RNP, and Ro 60,containing U1 and Y1–Y5 RNAs, respectively (9,37).Indeed, maturation of DCs is a known effect of nucleo-somes (38), mammalian DNA (39), and RNA (37).
The failure of chloroquine to inhibit bleb-induced maturation of DCs, as shown here, indicatesthat TLRs 3, 7, and 9 are not involved, suggesting thatDNA and RNA do not mediate the bleb-induced mat-uration of DCs. However, nucleosomes could very wellbe responsible for the observed DC maturation, sincenormal nucleosomes can lead to maturation of DCs in amyeloid differentiation factor 88/TLR–independent way(38). It has also been suggested by others that theinitiation phase of SLE, mediated by uptake of apoptoticmaterial by DCs, is TLR independent, while the ampli-fication phase, with uptake of TLR ligands derived fromself antigens (principally nucleic acids) complexed withautoantibodies, seems to be TLR dependent (40). Once
maturation of DCs has been induced, it can amplifyfurther uptake of apoptotic blebs, since an increasedantigen uptake has been described in the early phase ofDC maturation (41).
An additional factor triggering the maturation ofDCs may be the presence of apoptosis-induced modifi-cations on the nucleosomes residing in apoptotic blebs.We have recently identified apoptosis-induced acetyla-tion of nucleosomes as a pathogenic factor in SLE andfound that hyperacetylated nucleosomes were superiorin maturation of DCs when compared with normalnucleosomes (11). These apoptosis-induced chromatinmodifications were predominantly located in the apo-ptotic blebs and to a lesser extent in apoptotic cellbodies, which could explain the higher maturation po-tential of the blebs compared with the apoptotic cellbodies. Even at extremely high concentrations of apo-ptotic cell bodies, we did not observe any maturation ofDCs.
There are several ways to explain how DC mat-uration by blebs can be a pathogenic factor in thedevelopment of SLE. As shown here, uptake of apopto-tic blebs will lead to an immunogenic presentation ofnative and modified autoantigens by DCs to T cells,possibly resulting in a Th17 response (see below). Tworoutes may then lead to the development of full-blownautoimmunity: 1) direct activation of autoreactive T cellsthat recognize the autoantigens presented by bleb-matured DCs, and 2) activation of T cells that recognizecryptic epitopes, or modified self antigens, with subse-quent activation of autoreactive B and T cells via epitopespreading.
However, a remaining key question is whetherapoptotic blebs can also be found in vivo. Indeed,particles representing apoptotic blebs, sometimes re-ferred to as microparticles, can be found in the circula-tion of SLE patients and also in healthy controls (42,43).Whether apoptotic blebs found in SLE patients are ableto induce maturation of DCs is currently unknown.Nevertheless, the concentrations of nucleosomes, alsoresiding in the apoptotic blebs, that are found in thecirculation of patients with SLE correspond to the blebconcentrations that induce maturation of DCs in ourstudy (38). This suggests that the in vivo concentrationsof apoptotic blebs/nucleosomes are sufficient for induc-tion of maturation of DCs.
We also observed that DCs matured by apoptoticblebs, in contrast to DCs exposed to apoptotic cellbodies, obtained an increased ability to stimulate allo-geneic T cells. Upon stimulation with bleb-maturedDCs, T cells produced IFN� and especially high levels of
2310 FRANSEN ET AL
IL-17, which suggests a mixed Th1/Th17 response. Pre-viously, most autoimmune diseases were thought to beTh1 mediated. However, the recently discovered Th17cell population, characterized by the production of IL-17, has been shown to be involved in several auto-immune diseases.
Th17 cells are present in both mice and humans;however, along with many similarities of these specieswith regard to induction and characteristics, there aresome differences. The development of Th17 cells in micedepends on the presence of IL-6 and TGF�, after whichIL-21 produced by the Th17 cells acts as an autocrinefactor inducing expansion. However, in humans, IL-1�seems to be an important factor in the development ofTh17 cells from naive T cells. IL-23 is suggested to beimportant in sustaining the Th17 response (44). A recentreport also indicates a mixed Th1/Th17 response inpatients with SLE, as measured by the concentrations ofthe Th1-promoting cytokine IL-12, the Th17-maintaining cytokine IL-23, and the Th1 chemokineCXCL10 (45). Th17 cells have been linked to thedevelopment of autoimmunity (44,46), and there is alsoevidence for a role of these cells in patients with SLE(45). The tolerance that can be induced in lupus-pronemice with low-dose peptide is associated with a reduc-tion of Th17 cells (47). Furthermore, a genetic associa-tion has been found between SLE and polymorphisms ofIL-21, an important cytokine for the development ofTh17 cells (48). Blocking IL-21 in a lupus-prone mousemodel reduces disease progression (49).
Altogether, these data indicate that Th17 cellsmay have a central role in the development of SLE, andpossibly they are activated by bleb-matured DCs. It istempting to speculate which constituent of the blebsinduces DCs to produce high amounts of IL-6 and toactivate and differentiate T cells into IL-17–producingcells in this system. Recently, it was shown that DCsmatured by peptidoglycans are potent inducers of Th17differentiation, compared with DCs matured by LPS orCpG (44). It is a challenging task to determine whethermodified chromatin is part of this causative constituent.Irrespective of the causative constituent, extrapolationof our in vitro data toward an in vivo setting suggeststhat a defective clearance of apoptotic cells by phago-cytes results in the release of apoptotic blebs that can betaken up by DCs. This can lead to a Th1/Th17-drivenimmune response against (modified) self antigens in-cluding chromatin, finally resulting in SLE. However,the physiologic relevance of bleb-induced DC matura-tion in the development of SLE should be examined invivo and in a human setting and is the subject of our
ongoing research. In addition to the activation of Th17cells and the inhibition of Treg cells, IL-6 produced bybleb-matured DCs may facilitate autoantibody produc-tion by autoreactive B cells.
It was recently described that apoptotic blebs, asdefined by us (Figure 1), stimulate human PDCs toproduce IFN� (50). This may indicate a maturing effectof blebs on PDCs in addition to that on myeloid DCs.Nevertheless, uptake of apoptotic bodies/blebs by PDCshas never been demonstrated (29). Notably, we did notobserve any IFN� production (data not shown), exclud-ing the involvement of PDCs in our experiments.
In conclusion, our in vitro data show that apo-ptotic blebs generated by apoptosis induce the matura-tion of DCs that subsequently acquire a Th1 and strongTh17 stimulatory capacity. This bleb-induced DC matu-ration, and the following Th1/Th17 response, may rep-resent an important driving force in the autoimmuneresponse in SLE and a novel target for developingtherapeutic strategies.
ACKNOWLEDGMENTS
We thank Dr. J. Greenberger (University of PittsburghCancer Institute, Pittsburgh, PA) and Dr. S. Baker (TempleUniversity, Philadelphia, PA) for providing the 32D clone 3cell line. Mrs. Claudia Koeter (Department of Nephrology,Radboud University Nijmegen Medical Centre, Nijmegen, TheNetherlands) is acknowledged for her technical assistance.
AUTHOR CONTRIBUTIONSAll authors were involved in drafting the article or revising it
critically for important intellectual content, and all authors approvedthe final version to be published. Dr. Berden had full access to all ofthe data in the study and takes responsibility for the integrity of thedata and the accuracy of the data analysis.Study conception and design. Fransen, Hilbrands, van der Vlag,Berden.Acquisition of data. Fransen, Ruben, Stoffels.Analysis and interpretation of data. Fransen, Hilbrands, Adema, vander Vlag, Berden.
REFERENCES
1. Mortensen ES, Fenton KA, Rekvig OP. Lupus nephritis: thecentral role of nucleosomes revealed. Am J Pathol 2008;172:275–83.
2. Van Bavel CC, Fenton KA, Rekvig OP, van der Vlag J, BerdenJH. Glomerular targets of nephritogenic autoantibodies in sys-temic lupus erythematosus [review]. Arthritis Rheum 2008;58:1892–9.
3. Licht R, van Bruggen MC, Oppers-Walgreen B, Rijke TP, BerdenJH. Plasma levels of nucleosomes and nucleosome–autoantibodycomplexes in murine lupus: effects of disease progression andlipopolysaccharide administration. Arthritis Rheum 2001;44:1320–30.
4. Amoura Z, Piette JC, Bach JF, Koutouzov S. The key role ofnucleosomes in lupus [review]. Arthritis Rheum 1999;42:833–43.
5. Hengartner MO. The biochemistry of apoptosis. Nature 2000;407:770–6.
6. Dieker JW, van der Vlag J, Berden JH. Triggers for anti-chromatin autoantibody production in SLE. Lupus 2002;11:856–64.
7. Zykova SN, Seredkina N, Benjaminsen J, Rekvig OP. Reducedfragmentation of apoptotic chromatin is associated with nephritisin lupus-prone (NZB � NZW)F1 mice. Arthritis Rheum 2008;58:813–25.
8. Herrmann M, Voll RE, Zoller OM, Hagenhofer M, Ponner BB,Kalden JR. Impaired phagocytosis of apoptotic cell material bymonocyte-derived macrophages from patients with systemic lupuserythematosus. Arthritis Rheum 1998;41:1241–50.
9. Casciola-Rosen LA, Anhalt G, Rosen A. Autoantigens targeted insystemic lupus erythematosus are clustered in two populations ofsurface structures on apoptotic keratinocytes. J Exp Med 1994;179:1317–30.
10. Radic MZ, Shah K, Zhang W, Lu Q, Lemke G, Hilliard GM.Heterogeneous nuclear ribonucleoprotein P2 is an autoantibodytarget in mice deficient for Mer, Axl, and Tyro3 receptor tyrosinekinases. J Immunol 2006;176:68–74.
11. Dieker JW, Fransen JH, van Bavel CC, Briand JP, Jacobs CW,Muller S, et al. Apoptosis-induced acetylation of histones ispathogenic in systemic lupus erythematosus. Arthritis Rheum2007;56:1921–33.
12. Casciola-Rosen L, Andrade F, Ulanet D, Wong WB, Rosen A.Cleavage by granzyme B is strongly predictive of autoantigenstatus: implications for initiation of autoimmunity. J Exp Med1999;190:815–26.
13. Rosen A, Casciola-Rosen L. Autoantigens as substrates for apo-ptotic proteases: implications for the pathogenesis of systemicautoimmune disease. Cell Death Differ 1999;6:6–12.
14. Utz PJ, Gensler TJ, Anderson P. Death, autoantigen modifica-tions, and tolerance. Arthritis Res 2000;2:101–14.
15. Dieker J, Cisterna B, Monneaux F, Decossas M, van der Vlag J,Biggiogera M, et al. Apoptosis-linked changes in the phosphory-lation status and subcellular localization of the spliceosomalautoantigen U1-70K. Cell Death Differ 2008;15:793–804.
16. Matzinger P. The danger model: a renewed sense of self. Science2002;296:301–5.
17. Savill J, Dransfield I, Gregory C, Haslett C. A blast from the past:clearance of apoptotic cells regulates immune responses. Nat RevImmunol 2002;2:965–75.
18. Banchereau J, Steinman RM. Dendritic cells and the control ofimmunity. Nature 1998;392:245–52.
19. Ip WK, Lau YL. Distinct maturation of, but not migrationbetween, human monocyte-derived dendritic cells upon ingestionof apoptotic cells of early or late phases. J Immunol 2004;173:189–96.
20. Steinman RM, Turley S, Mellman I, Inaba K. The induction oftolerance by dendritic cells that have captured apoptotic cells. JExp Med 2000;191:411–6.
21. Basu S, Binder RJ, Suto R, Anderson KM, Srivastava PK. Necroticbut not apoptotic cell death releases heat shock proteins, whichdeliver a partial maturation signal to dendritic cells and activatethe NF-�B pathway. Int Immunol 2000;12:1539–46.
22. Bondanza A, Zimmermann VS, Dell’Antonio G, Dal Cin E,Capobianco A, Sabbadini MG, et al. Cutting edge: dissociationbetween autoimmune response and clinical disease after vaccina-tion with dendritic cells. J Immunol 2003;170:24–7.
23. Shoshan Y, Mevorach D. Accelerated autoimmune disease inMRL/MpJ-Fas(lpr) but not in MRL/MpJ following immunizationwith high load of syngeneic late apoptotic cells. Autoimmunity2004;37:103–9.
24. Bondanza A, Zimmermann VS, Dell’Antonio G, Dal Cin E,Balestrieri G, Tincani A, et al. Requirement of dying cells and
environmental adjuvants for the induction of autoimmunity. Ar-thritis Rheum 2004;50:1549–60.
25. Mevorach D, Zhou JL, Song X, Elkon KB. Systemic exposure toirradiated apoptotic cells induces autoantibody production. J ExpMed 1998;188:387–92.
26. Tzeng TC, Suen JL, Chiang BL. Dendritic cells pulsed withapoptotic cells activate self-reactive T-cells of lupus mice both invitro and in vivo. Rheumatology (Oxford) 2006;45:1230–7.
27. Lovgren T, Eloranta ML, Kastner B, Wahren-Herlenius M, AlmGV, Ronnblom L. Induction of interferon-� by immune com-plexes or liposomes containing systemic lupus erythematosusautoantigen– and Sjogren’s syndrome autoantigen–associatedRNA. Arthritis Rheum 2006;54:1917–27.
28. Banchereau J, Pascual V. Type I interferon in systemic lupuserythematosus and other autoimmune diseases. Immunity 2006;25:383–92.
29. Dalgaard J, Beckstrom KJ, Jahnsen FL, Brinchmann JE. Differ-ential capability for phagocytosis of apoptotic and necrotic leuke-mia cells by human peripheral blood dendritic cell subsets. J Leu-koc Biol 2005;77:689–98.
30. Lutz MB, Suri RM, Niimi M, Ogilvie AL, Kukutsch NA, RossnerS, et al. Immature dendritic cells generated with low doses ofGM-CSF in the absence of IL-4 are maturation resistant andprolong allograft survival in vivo. Eur J Immunol 2000;30:1813–22.
31. Emmer PM, van der Vlag J, Adema GJ, Hilbrands LB. Dendriticcells activated by lipopolysaccharide after dexamethasone treat-ment induce donor-specific allograft hyporesponsiveness. Trans-plantation 2006;81:1451–9.
32. Guchhait P, Tosi MF, Smith CW, Chakaraborty A. The murinemyeloid cell line 32Dcl3 as a model system for studying neutrophilfunctions. J Immunol Methods 2003;283:195–204.
33. Greenberger JS, Sakakeeny MA, Humphries RK, Eaves CJ,Eckner RJ. Demonstration of permanent factor-dependent multi-potential (erythroid/neutrophil/basophil) hematopoietic progeni-tor cell lines. Proc Natl Acad Sci U S A 1983;80:2931–5.
34. Gallucci S, Lolkema M, Matzinger P. Natural adjuvants: endoge-nous activators of dendritic cells. Nat Med 1999;5:1249–55.
35. Frisoni L, McPhie L, Colonna L, Sriram U, Monestier M, GallucciS, et al. Nuclear autoantigen translocation and autoantibodyopsonization lead to increased dendritic cell phagocytosis andpresentation of nuclear antigens: a novel pathogenic pathway forautoimmunity? J Immunol 2005;175:2692–701.
36. Linker-Israeli M, Deans RJ, Wallace DJ, Prehn J, Ozeri-Chen T,Klinenberg JR. Elevated levels of endogenous IL-6 in systemiclupus erythematosus: a putative role in pathogenesis. J Immunol1991;147:117–23.
37. Kelly KM, Zhuang H, Nacionales DC, Scumpia PO, Lyons R,Akaogi J, et al. “Endogenous adjuvant” activity of the RNAcomponents of lupus autoantigens Sm/RNP and Ro 60. ArthritisRheum 2006;54:1557–67.
38. Decker P, Singh-Jasuja H, Haager S, Kotter I, Rammensee HG.Nucleosome, the main autoantigen in systemic lupus erythemato-sus, induces direct dendritic cell activation via a MyD88-indepen-dent pathway: consequences on inflammation. J Immunol 2005;174:3326–34.
39. Martin DA, Elkon KB. Intracellular mammalian DNA stimulatesmyeloid dendritic cells to produce type I interferons predomi-nantly through a Toll-like receptor 9–independent pathway. Ar-thritis Rheum 2006;54:951–62.
40. Baccala R, Hoebe K, Kono DH, Beutler B, Theofilopoulos AN.TLR-dependent and TLR-independent pathways of type I inter-feron induction in systemic autoimmunity. Nat Med 2007;13:543–51.
41. Reis e Sousa C. Dendritic cells in a mature age. Nat Rev Immunol2006;6:476–83.
2312 FRANSEN ET AL
42. Abid Hussein MN, Nieuwland R, Hau CM, Evers LM, MeestersEW, Sturk A. Cell-derived microparticles contain caspase 3 invitro and in vivo. J Thromb Haemost 2005;3:888–96.
43. Nieuwland R, Berckmans RJ, McGregor S, Boing AN, Romijn FP,Westendorp RG, et al. Cellular origin and procoagulant propertiesof microparticles in meningococcal sepsis. Blood 2000;95:930–5.
44. Bettelli E, Korn T, Oukka M, Kuchroo VK. Induction and effectorfunctions of TH17 cells. Nature 2008;453:1051–7.
45. Wong CK, Lit LC, Tam LS, Li EK, Wong PT, Lam CW.Hyperproduction of IL-23 and IL-17 in patients with systemiclupus erythematosus: implications for Th17-mediated inflamma-tion in auto-immunity. Clin Immunol 2008;127:385–93.
47. Kang HK, Liu M, Datta SK. Low-dose peptide tolerance therapy
of lupus generates plasmacytoid dendritic cells that cause expan-sion of autoantigen-specific regulatory T cells and contraction ofinflammatory Th17 cells. J Immunol 2007;178:7849–58.
48. Sawalha AH, Kaufman KM, Kelly JA, Adler AJ, Aberle T,Kilpatrick J, et al. Genetic association of interleukin-21 polymor-phisms with systemic lupus erythematosus. Ann Rheum Dis 2008;67:458–61.
49. Herber D, Brown TP, Liang S, Young DA, Collins M, Dunussi-Joannopoulos K. IL-21 has a pathogenic role in a lupus-pronemouse model and its blockade with IL-21R.Fc reduces diseaseprogression. J Immunol 2007;178:3822–30.
50. Heyder P, Bekeredjian-Ding I, Parcina M, Blank N, Ho AD,Herrmann M, et al. Purified apoptotic bodies stimulate plasmacy-toid dendritic cells to produce IFN-�. Autoimmunity 2007;40:331–2.
