HLA-DR-Mediated Susceptibility Discriminates Differentiation Stages of Dendritic/Monocytic APC1

Nicolas Bertho,* Bernard Dre´nou,* Be´atrice Laupeze,* Claudine Le Berre,† Laurence Amiot,* Jean-Marc Grosset,* Olivier Fardel,‡ Dominique Charron,§ Nuala Mooney,2§ and Rene´e Fauchet*

Professional APC are characterized by their ability to present peptide via HLA class II in the presence of costimulatory molecules (CD40, CD80, and CD86). The efficiency of Ag presentation can be classed as follows: mature dendritic cells (DC) are most efficient, immature DC and are intermediate, and are considered poor APC. There is a large body of evidence demonstrating that HLA-DR transmits signals in the APC. In this study, we have addressed the question of the outcome of HLA-DR signals on APC of the /DC lineages throughout their differentiation from immature to mature APC. DC were .generated from both monocytes and CD34؉ cells of the same individual, macrophages were differentiated from monocytes Immunophenotypical analysis clearly distinguished these populations. HLA-DR-mediated signals led to marked apoptosis in mature DC of either CD34 or monocytic origin. Significantly less apoptosis was observed in immature DC of either origin. Nonetheless, even immature DC were more susceptible to HLA-DR-mediated apoptosis than macrophages, whereas monocytes were resistant to HLA-DR-mediated apoptosis. The mechanism of HLA-DR-mediated apoptosis was independent of caspase activation. Taken together, these data lead to the notion that signals generated via HLA-DR lead to the demise of mature professional APC, thereby providing a means of limiting the immune response. The Journal of Immunology, 2000, 164: 2379– 2385.

rofessional APC represent a subset responsible for thymic membrane ligands (CD40-L), they migrate from tissues to periph- selection and peripheral activation (1). Ag-specific eral lymph nodes. In the course of their migration, they undergo P immune responses of CD4 T cells depends on the success- maturation and are considered mature when localized in lymph ful presentation of processed peptide Ags by HLA class II mole- nodes. Mature DC express CD83 (6) and have enhanced HLA cules (HLA-DR, -DP, and -DQ) (2). These molecules are consti- class I, class II, CD80, and CD86 expression (7). Immature DC can tutively expressed on professional APC including dendritic cells be generated in vitro from either peripheral blood monocytes (8) 3 (DC) (3), macrophages, and B lymphocytes. using GM-CSF and IL-4 or from CD34ϩ hemopoietic stem cells DC are the most potent APC (reviewed in Ref. 4), with the (9) using GM-CSF and TNF-␣. Maturation of cultured immature unique ability to prime naive T cells (5). Tissular DC are consid- DC may be obtained by a 48-h incubation with TNF-␣, bacterial ered as immature and can be identified by their expression of HLA LPS (10), or CD40-L (11). DC differentiation and maturation can class I and II molecules, CD1a, CD40, CD80 (.1), and CD86 therefore be reproduced in vitro. The success of Ag presentation by (B7.2) but not CD83. They capture and process Ag with high DC has led to them being considered as promising tools for im- efficiency. When DC receive inflammation signals mediated by munotherapy (12, 13). chemokines (-inflammatory 3␣, macrophage- Macrophages express HLA class I and HLA class II Ags as well inflammatory protein 1␣, and RANTES) and (TNF-␣)or as CD80 and CD86 accessory molecules (14). They are character- ized by their high ability to phagocytose, which allows them

*Laboratoire Universitaire d’He´matologie et de Biologie des Cellules Sanguines, In- to process and to present particulate Ags. However, macrophages stitut National de la Sante´ et de la Recherche Me´dicale CRI 9606-UPRES EA 22-33, are less effective than DC in Ag presentation to T lymphocytes † Rennes, France; Etablissement de Transfusion Sanguine de Bretagne, Rennes, (15, 16). France; ‡Institut National de la Sante´ et de la Recherche Me´dicale U456 Universite´de Rennes I, Rennes, France; and §Institut National de la Sante´ et de la Recherche B lymphocytes are considered to be professional APC (17) and Me´dicale U396 Institut des Cordeliers, Paris, France express the requisite accessory molecules for T lymphocyte acti- Received for publication July 12, 1999. Accepted for publication December 20, 1999. vation (such as CD86, CD80, and CD40 molecules (18)) as well as The costs of publication of this article were defrayed in part by the payment of page HLA class II molecules. There is ample evidence for HLA-DR- charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. mediated in B lymphocytes (reviewed in Ref. ␤ 1 This work was supported by grants from the Agence Franc¸aise du Sang. N.B. is a 19). The intracytoplasmic region of the HLA-DR -chain is crit- recipient of a fellowship from Rennes I University. B.L. is a recipient of a fellowship ical for the translocation of protein kinase C ␣ and ␤II (20). Con- from la Ligue contre le cancer (Comite´ des Coˆtes d’Armor). We are grateful for the sequences of these signals are either proliferation and differentia- financial support of Association pour la Recherche sur le Cancer, La Ligue Contre le Cancer (LLCLC), and LLCLC (Comite´ de Paris), EC grants B70-4-98-0458 and tion (21), cell-cell adhesion (22), or cell death (23). Activated ESF-IGA. B lymphocytes are more susceptible to HLA-DR-mediated 2 Address correspondence and reprint requests to Dr. Nuala Mooney, Laboratoire apoptosis than resting cells (24, 25). The mechanism of HLA-DR- d’Immunoge´ne´tique Humaine, Institut Biome´dical des Cordeliers, 15 rue de l’e´cole de me´decine/Bat A, 75006 Paris, France. mediated cell death has not been elucidated, although enhancement 3 Abbreviations used in this paper: DC, ; SCF, stem cell factor; L, ligand; of sensitivity to CD95/Fas (26) and expression of Fas-L (27) has PI, propidium iodide. been described after HLA-DR cross-linking. Fas, in common with

Copyright © 2000 by The American Association of Immunologists 0022-1767/00/$02.00 2380 PROFESSIONAL APC ARE TARGETS FOR HLA-DR-MEDIATED APOPTOSIS other death receptors, initiates activation of cysteinyl aspartate- were composed of 84.5 Ϯ 5.8% CD14ϩ cells when analyzed by flow cy- ϩ specific proteinases (caspases) which are considered as the prin- tometry. Nonadherent cells were removed for isolation of CD34 cells. Nonadherent cells were allowed to adhere for another 30 min in 45-cm2 cipal executioners of Fas-induced apoptosis (28, 29). ϩ flasks to remove remaining monocytes before isolation of CD34 cells. Ag presentation needs TCR and CD4 interaction with HLA Hemopoietic progenitors were positively selected using CD34-conjugated class II molecules but also activation of cells by costimulation immunomagnetic beads (Dynabeads M-450 CD34; Dynal, Oslo, Norway) molecules (30). Numerous mechanisms to control the T cell re- according to the manufacturer’s instructions. The final purity of CD34ϩ sponse have been described (anergy, apoptosis). Mechanisms con- cells was 88.9% Ϯ 5.7% as demonstrated by flow cytometry analysis. trolling the APC response have not yet been elucidated. Since Generation of DC and macrophages HLA-DR mediated more cell death in activated than in resting B cells, it has been postulated to play a role in the termination of the Macrophages. Monocytes (adherent cells) were cultured for 7 days in immune response (23). If this were the case, HLA-DR-mediated IMDM (Life Technologies) and 10% FCS supplemented with 800 U/ml GM-CSF. Cultures were fed by replacing half of the medium at day 3. cell death could be also expected to occur in other APC. Tyrosine Monocyte-derived DC. Adherent cells were cultured in IMDM, 10% protein kinase activation via HLA-DR on DC has been reported FCS supplemented with 800 U/ml GM-CSF, and 1000 U/ml IL-4. Half of (31); however, the question of apoptosis has not been addressed. the medium was replaced at days 3, 5, and 7 by fresh medium with 800 We have examined the role of HLA-DR-mediated signals U/ml GM-CSF and 500 U/ml IL-4. At day 7, 100 U/ml TNF-␣ was added to culture to provided mature monocyte-derived DC. throughout differentiation of primary professional APC. Hemopoi- ؉ etic progenitors, monocytes, macrophages, as well as immature Hemopoietic progenitor (CD34 cells)-derived DC. Isolated CD34ϩ cells were cultured in IMDM, 10% FCS supplemented with 500 and mature DC were examined. We have compared DC derived ␣ ϩ U/ml GM-CSF, 2.5 U/ml SCF, 10 U/ml Flt3-L, and 200 U/ml TNF- . from both monocytes and CD34 cells of the same individual to Cultures were fed by replacing half of the culture medium by fresh medium ensure that the results obtained were representative of DC regard- at days 5, 9, and 14. At day 14, 100 ng/ml LPS was added to provide less of their origin. We report HLA-DR-mediated apoptosis of mature CD34-derived DC. mature differentiated APC which did not appear to depend on Phenotypic analysis of cells by flow cytometry caspase activation. An in vitro model of APC differentiation and maturation allowed us to demonstrate that the apoptotic function of Before labeling, cells were incubated for1hinhuman AB serum, at 4°C, MHC class II is differentially regulated in professional APC sub- to avoid nonspecific mAb binding. Several mAbs were used for immuno- labeling: FITC- or PE-conjugated mouse mAb against CD1a, CD14, CD34, sets. Finally, these results strongly suggest an “autoregulation” of and HLA-DR, and FITC-labeled isotype controls were purchased from MHC class II-mediated Ag presentation. Immunotech whereas mAbs against CD40, CD83, CD86, and PE-labeled isotype controls were purchased from PharMingen/San Diego, CA). After Materials and Methods immunolabeling, cells were analyzed by flow cytometry using a Cytoron Reagents and cytokines for cell cultures (Ortho Diagnostics, Raritan, NJ) equipped with an argon laser operating at 488 nm. Data were acquired with the Immunocount II software (Ortho The following mAbs were used in cell cultures: Agonistic anti-CD95 mAb Diagnostics). (CH11, IgM␬) was purchased from Immunotech (Marseille, France). Monomorphic anti-HLA-DR mAb (L243, D1.12), anti-HLA-DR and Detection of HLA-DR-induced cell death -HLA-DP mAbs (L227) were purified from ascitic fluid. The W6.32 mAb was used to study HLA class I-mediated apoptosis. IgG1 and IgG2a iso- For studies of HLA-induced apoptosis, cells were plated in 1 ml of culture type-matched controls were purchased from PharMingen (San Diego, CA). medium consisting of IMDM supplemented with 10% FCS. Cells were ϫ 5 All mAbs used for cell cultures were conditioned without sodium azide. cultured at a density of 1 10 cells/well, in 24-well plates, in the presence GM-CSF, IL-4, stem cell factor (SCF), Flt3-L and TNF-␣ were pur- or absence of mAb L243, D1.12, L227, W6.32, or the relevant isotype- chased from Life Technologies (Rockville, MD). LPS from Escherichia matched control at different concentrations as specified in Results. Cultures coli was purchased from Sigma (St. Louis, MO). were maintained in a humidified atmosphere with 5% CO2 at 37°C. In some experiments, to determine the role of caspases, cells were seeded at Caspase inhibitors Z-VAD-fmk (32) and Z-DEVD-fmk (33) were pur- ␮ chased from Calbiochem (La Jolla, CA). They were resuspended in DMSO 37°C for 30 min with 250 M Z-VAD-fmk, Z-DEVD-fmk, or DMSO and used at a final concentration of 250 ␮M. before adding mAbs. Samples were harvested after 20 h of culture. Flasks containing adher- Cell lines ent cells (monocytes and macrophages) were maintained for1hinice before scraping. Cells were stained with PE-labeled CD14 (for monocytes Cell lines from different lineages were obtained from the American Tissue and macrophages) or CD1a (for DC), for 30 min at 4°C, and then incubated Culture Collection (Manassas, VA). Raji is a line isolated from a with FITC-labeled annexin V to determine apoptosis. Flow cytometry Burkitt lymphoma (34). Jurkat is a T cell line isolated from an acute T cell gating allowed us to select cells that exhibit CD14 or CD1a Ag and to leukemia (35). THP-1 is a monocytic cell line isolated from an acute mono- analyze their binding of annexin V, which is a membrane marker of apo- cytic leukemia (36). KG1 is a myeloid cell line isolated from acute myeloid ptotic cells (39). leukemia (37). TF-1 (38) is a myeloid cell line isolated from erythroleu- We also analyzed apoptosis of mature monocyte-derived DC using pro- kemia. KG1 and TF-1 express CD34 and HLA-DR Ags, whereas Raji and pidium iodide (PI). PI binds to DNA and is actively excluded by live cells. THP-1 express HLA-DR and Jurkat does not express either HLA-DR or Cell viability was determined by trypan blue exclusion. CD34. All cell lines were cultured in RPMI supplemented with 10% FCS, 2mML-glutamine, and 1 mM pyruvate. TF-1 culture medium was sup- Statistics plemented with 300 U/ml GM-CSF. The comparison between variables was analyzed using the Student t test for Cells comparisons of percentages of cell death. DC and macrophages were derived from monocytes isolated from buffy coats of normal healthy donors. Cytapheresis products were collected at the Results end of the chemotherapy-induced aplasia from patients with nonmy- Immunophenotyping characterization elomonocytic malignancies (non-Hodgkin lymphoma or multiple my- eloma) to have access to both CD34-derived and monocyte-derived DC The indicated markers were used to define the following subsets from the same donor. of APC. Cell purification Monocytes. Monocytes expressed HLA-DR molecules and a low level of CD40 and CD86 molecules (Fig. 1a). They did not After separation of PBMC on Ficoll, cells were washed twice in HBSS (Life Technologies) and 10% FCS. They were resuspended at 3 ϫ106 express CD80. cells/ml and incubated in 6-well microplates (3 ml/well) for DC and mac- Macrophages. After 7 days of culture with GM-CSF, to induce Ϯ rophage cultures. After a 2-h incubation at 37°C in 5% CO2, adherent cells macrophage differentiation, 79.8% 14.0 of cells expressed The Journal of Immunology 2381

FIGURE 1. Phenotypic analysis of monocytes, macrophagic cells, immature and mature monocyte-derived DC and CD34-derived DC. CD34ϩ hemo- poietic progenitors and monocytes were analyzed. Macrophages and monocyte-derived DC were differentiated for 7 days with suitable cytokines (see Materials and Methods). CD34-derived DC were differentiated for 14 days with suitable cytokines (see Materials and Methods). Immature DC of either origin were cultured for 2 additional days with TNF-␣ or LPS to provide mature DC. Dual-color flow cytometry was performed using cells stained with FITC-labeled mAbs and PE-labeled lineage-specific mAb (CD14, CD1a, and CD34). Electronic gating of lineage-positive cells was performed before displaying the immunofluorescence profiles, as single parameter histograms, of selected cells. a, Freshly isolated monocytes gated by CD14 (cells were selected by CD14 staining). b, Macrophages gated by CD14 (cells were selected by CD14 staining). c, Immature monocyte-derived DC (cells were selected by CD1a staining). d, Mature monocyte-derived DC gated by CD1a (cells were selected by CD1a staining). e, Freshly isolated CD34ϩ cells gated by CD34 (cells were selected by CD34 staining). f, Immature CD34-derived DC gated by CD1a (cells were selected by CD1a staining). g, Mature CD34-derived DC gated by CD1a (cells were selected by CD1a staining). Filled histograms indicate the control with irrelevant mAb; open histograms represent specific staining. Data are representative of four different experiments.

CD14 Ag; they also expressed CD40, CD80, CD86, and HLA-DR pressed the CD83 Ag (Fig. 1f); such a phenotype is characteristic (Fig. 1b). of immature DC. Immature monocyte-derived DC. DC differentiation was in- Mature CD34-derived DC. In response to a 48-h incubation with duced by 7 days of culture with GM-CSF and IL-4, 83.1% Ϯ 11.1 LPS, CD34-derived DC displayed induction of CD83: 51.7% Ϯ of cells exhibited CD1a Ag. They also expressed HLA-DR, CD40, 15.2 of CD1aϩ cells expressed this Ag (Fig. 1g). CD80, and CD86 but not CD83 Ag (Fig. 1c), in agreement with Macrophages, immature DC, and mature DC were also morpho- previous reports of the immature DC phenotype. logically identified after May-Gru¨nwald- Giemsa staining (data not Mature monocyte-derived DC. After a 48-h incubation with shown). TNF-␣, monocyte-derived DC had up-regulated HLA-DR, CD80, and CD86 Ag expression and also expressed CD83 (Fig. 1d), such Mature DC are susceptible to HLA-DR-mediated cell death in a an expression of these Ags is characteristic of mature DC. dose-dependent manner and regardless of their origin CD34؉ hemopoietic progenitors. Peripheral blood CD34ϩ cells We examined apoptosis induction of monocyte-differentiated DC do not express any of the following Ags: CD40, CD80, CD86, and from peripheral blood of healthy individuals using increasing CD83 (Fig. 1e). doses (from 0.5 to 10 ␮g/ml) of isotypic control IgG1, D1-12, or Immature CD34-derived DC. After 14 days of culture with GM- anti-HLA class I mAb W6.32. Cell death was detected using an- CSF, TNF-␣, SCF, and Flt3-L, to induce DC differentiation, 44.7% nexin V staining. This methodology allow us to assess specific Ϯ 3.7 of cells expressed CD1a as well as HLA-DR and CD40 and HLA-DR-mediated cell death and to exclude the implication of a low level of CD80 and CD86 but Ͻ15% of CD1aϩ cells ex- Fc␥ receptors. As shown in Fig. 2, anti-HLA-DR mAb specifically 2382 PROFESSIONAL APC ARE TARGETS FOR HLA-DR-MEDIATED APOPTOSIS

lation via HLA-DR and the number of trypan blue cells increased. PI or annexin V-positive cells of all samples exhibit morphologic changes typical of apoptotic cells, including a low forward scatter. Similar results (data not shown) were obtained with DC differen- tiated from monocytes of patients who had undergone chemother- apy or with DC differentiated from CD34ϩ cells of the same in- dividuals. Identical results were obtained using the L227 mAb. Taken together, these results demonstrate that mature DC are sus- ceptible to apoptosis via HLA-DR regardless of their origin.

HLA-DR-induced cell death of mature DC was not inhibited in the presence of caspase inhibitors FIGURE 2. HLA-DR signals mediate cell death of mature DC in a dose-dependent manner. Mature DC differentiated from peripheral blood Having observed HLA-DR-mediated apoptosis of DC, we exam- monocytes of normal donor (see Materials and Methods) were seeded at ined the role of caspases in this pathway. Because of the limited 1 ϫ 105 cells/ml in IMDM plus 10% FCS and IgG1 irrelevant mAb, anti- number of cells available, the approach that we have taken was to HLA-DR (D1-12), or anti-HLA class I (W6.32) mAbs. Percentages of use cell-permeable irreversible inhibitors of caspases. Z-VAD-fmk apoptotic cells after a 20-h exposure to mAbs was measured by annexin V inhibits a broad spectrum of IL-1-converting enzyme-like pro- binding using flow cytometry. Specific apoptosis was calculated as follows: teases (32), whereas Z-DEVD-fmk is specific for caspase 3 (33). specific apoptosis ϭ 100 ϫ (% of apoptotic cells with mAb (D1-12, Because we did not observe any inhibition of HLA-DR-mediated W6.32, or IgG1) Ϫ % apoptotic cells in culture without mAb)/(100 Ϫ % apoptotic cells in culture without mAb). The experiment shown is repre- apoptosis in the presence of low concentrations of Z-VAD-fmk or sentative of three experiments using DC from different donors. Z-DEVD-fmk (data not shown), in the experiments shown, high concentrations were used (250 ␮M) to ensure that the absence of inhibition was not due to the effect of dose. Whereas Fas-induced induces cell death of mature DC in a dose-dependent manner. Spe- apoptosis of Jurkat cells was dramatically inhibited by either pep- cific apoptosis was not significantly increased by signaling via tide, neither Z-VAD-fmk nor Z-DEVD-fmk altered HLA-DR HLA class I Ags (even when saturating levels, 10 ␮g/ml, of W6.32 apoptosis of mature DC (Fig. 4). were used). To confirm apoptosis of HLA-DR-stimulated mature DC, we performed dual fluorescence staining using annexin VFITC and Immature DC, mature DC, and macrophages were differentially CD1a-PE (Fig. 3a). The cell death was further corroborated by PI susceptible to HLA-DR-mediated cell death binding (Fig. 3b). Moreover, apoptosis of cells was further estab- We then compared HLA-DR-mediated apoptosis of APC of the lished by counting the absolute cell number detected after trypan monocytic lineage. Positive and negative controls were provided blue staining. The absolute number of DC decreased after stimu- by Raji B cells which underwent apoptosis in the presence of HLA-DR mAb, whereas viability of the HLA class II-negative Jurkat T cell line was unchanged.

Monocytes, macrophages, and monocyte-derived DC We failed to detect significant apoptosis of monocytes in the pres- ence of HLA-DR mAbs. Since HLA-DR-mediated apoptosis of monocytes has been described (40), we tested the monocytic cell line THP-1, which expresses the same level of HLA-DR molecules as normal monocytes. THP-1 like normal monocytes did not apoptose after HLA-DR cross-linking (data not shown). However, after primary monocyte differentiation to macrophages, sensitivity to HLA-DR-mediated cell death was clear (Fig. 5): 17.0% Ϯ 6.1 of annexin V after culturing in the presence of L243 mAb com- pared with 8.6% Ϯ 2.0 culturing with an isotype control (n ϭ 4, p Ͻ 0.05). Even immature monocyte-derived DC were sensitive: 34.0% Ϯ 14.9 of annexin V staining cells after culturing with L243 mAb compared with 7.4% Ϯ 1.2 after culture with an isotype control (n ϭ 4, p Ͻ 0.05; Fig. 5). Finally, HLA-DR-mediated signaling led to death in mature monocyte-derived DC (Fig. 5): 50.8% Ϯ 17.0 of annexin V stain- ing cells using L243 mAb vs 11.52% Ϯ 3.5 in the presence of an isotype control (n ϭ 4, p Ͻ 0.01). FIGURE 3. HLA-DR stimulation triggered cell death of mature DC. Differentiated mature DC were seeded at 1 ϫ 105 cells/ml in IMDM plus ␮ 10% FCS and 5 g/ml IgG2a irrelevant mAb or anti-HLA-DR mAb CD34ϩ cells and CD34-derived DC (L243). The cytogram and the histogram depict the percentages of apopto- ϩ tic cells after exposure to mAbs for 20 h. a, CD1a and annexin V staining. Primary CD34 cells were resistant to HLA-DR-mediated death Percentage of cells is depicted for each quadrant. b, PI staining. (Fig. 5). Because of the high level of spontaneous apoptosis in The Journal of Immunology 2383

FIGURE 4. HLA-DR-mediated cell death of monocyte-derived DC is not inhibited by caspase inhibitors. Mature monocyte-derived DC were seeded for 30 min at 37°C in IMDM, 10% FCS with 250 ␮M caspase inhibitors Z-VAD.fmk or Z-DEVD.fmk, or the same quantity of DMSO before incubation with mAbs. Similar results were obtained in two experiments. The percentages of apoptotic cells are indicated on the histograms. primary CD34ϩ cells, we confirmed these data using two CD34ϩ Discussion cell lines, KG1 and TF1, which do not undergo spontaneous ap- Stimulation of APC using HLA-DR mAb is a model which repro- optosis. Both of these cell lines express levels of HLA-DR com- parable to those of CD34ϩ primary cells. In agreement with the duces at least some of the signals generated via HLA-DR during results from the primary CD34ϩ cells, neither cell line was sus- Ag presentation. Since it has been postulated that HLA-DR-me- ceptible to HLA-DR-mediated apoptosis (data not shown). diated apoptosis down-regulates the immune response, we have Similar to monocyte-derived DC, differentiation of CD34ϩ cells focused on apoptosis of professional APC during their differenti- to immature DC led to sensitivity to HLA-DR-mediated cell death: ation and maturation. This model has allowed us to observe the 32.7% Ϯ 8.0 of annexin V staining using L243 mAb vs 13.2% Ϯ emergence of differences in HLA-DR signaling during APC cell 1.2 using an isotype control (n ϭ 4, p Ͻ 0.01). The level of HLA- differentiation. HLA-DR triggers significant apoptosis of mature DR-mediated apoptosis was therefore comparable in either CD34 DC as has been previously described in B lymphocytes. ϩ or monocyte-derived DC (Fig. 5). In this study, we differentiated DC from CD34 hemopoietic Furthermore, maturation of CD34-derived DC with LPS signif- progenitors and from monocytes of the same individual, therefore icantly increased their sensitivity to HLA-DR-mediated cell death: permitting valid comparison of DC of either origin. HLA-DR Ag 50.7% Ϯ 5.9 of apoptotic cells were detected vs 10.9% Ϯ 2.5 expression was higher on monocyte-derived DC than on CD34- using an isotype control (n ϭ 4, p Ͻ 0.001). The difference be- derived DC. Despite this phenotypic difference, DC differentiated ϩ tween the apoptotic response induced via HLA-DR in either im- from monocytes or CD34 cells were remarkably similar with mature or mature CD34-derived DC was significantly different: regard to their sensitivity to HLA-DR-mediated apoptosis and this p Ͻ 0.05 (Fig. 5). remained the case even after maturation. 2384 PROFESSIONAL APC ARE TARGETS FOR HLA-DR-MEDIATED APOPTOSIS

FIGURE 5. HLA-DR stimulation triggered cell death of mature and immature DC. Differen- tiated DC were seeded at 1 ϫ 105 cells/ml in IMDM plus 10% FCS and 5 ␮g/ml anti-HLA-DR mAb (L243) or 5 ␮g/ml IgG2a irrelevant matched mAb. The histograms depict the percentages of apoptotic cells, determined by annexin V, after a 20-h exposure to mAbs. imm. mono. DC, imma- ture monocyte-derived DC; mat. mono. DC, ma- ture monocyte-derived DC; imm. CD34 DC, immature CD34-derived DC; mat. CD34 DC, mature CD34-derived DC. Probability values demonstrating significant differences between HLA-DR and an isotype control-induced apopto- p Ͻ ,ءء ;p Ͻ 0.05 ,ء :sis are indicated as follows p Ͻ 0.001. A significant difference in ,ءءء ;0.01 the level of apoptosis via HLA-DR was found be- tween immature and mature DC: #, p Ͻ 0.05. Data indicate means Ϯ SD of at least four sam- ples.

CD34ϩ cells and monocytes are resistant to HLA-DR-mediated the mitochondrial intermembrane space and could be released into apoptosis, whereas macrophages were sensitive and DC were even the cytosol and nucleus upon pore opening by increasing intracel- more sensitive and particularly after maturation. A correlation be- lular Ca2ϩ concentration. The apoptotic activity of apoptosis-in- tween the differentiation state of B cells and their ability to trans- ducing factor is maintained in the presence of Z-VAD-fmk caspase mit HLA-DR-mediated signal has been described (24, 41). We inhibitor. We are currently examining the mechanisms of HLA- show that, although macrophages and immature DC are fully dif- DR-mediated apoptosis in APC of the DC/monocytic cell lineage. ferentiated cells, macrophages scarcely undergo apoptosis whereas The importance of this study is not only in the revelation of immature DC do so clearly. We therefore propose that HLA-DR HLA-DR signals leading to apoptosis of the APC but also the signaling for apoptosis correlates with the ability to present Ag determination of the hierarchy of susceptibility which reflects the rather than with simply the differentiation state of the cells. We hierarchy of Ag presentation capacity. These data provide strong have previously reported (3) that HLA-DR expression is 4.6-fold support for the notion that HLA-DR-induced apoptosis reflects a higher in immature DC than in monocytes and 1.5-fold higher in mechanism of down-regulation of the immune response. Further immature DC than in macrophages. HLA-DR-mediated death was support from this notion came from the following observations of most significant on populations that had acquired expression of the disappearance, in lymph node, of DC after interaction with Ag- costimulatory molecules CD80 and CD86. Indeed, CD34ϩ cells do specific T cells (48), signal transduction in DC after T cell contact not express these two Ags; monocytes lack CD80 and express a (49), and the induction of apoptosis of DC after an Ag-specific T low level of CD86. Macrophages express CD80 and a low level of cell interaction (50). CD86 and immature monocyte-derived DC like CD34-derived DC express a higher level of both these Ags, which are again up- References regulated during DC maturation. Indeed, our data show that HLA- DR-mediated cell death correlates with the acquisition of requisite 1. Mellman, I., S. J. Turley, and R. M. Steinman. 1998. Antigen processing for amateurs and professionals. Trends Cell Biol. 8:231. molecules for optimal Ag presentation. Moreover, it was demon- 2. Watts, C. 1997. Capture and processing of exogenous antigens for presentation strated that macrophages are less effective than DC in inducing T on MHC molecules. Annu. Rev. Immunol. 15:821. cell responses (15, 16). Taken together, these data support the hy- 3. Laupeze, B., O. Fardel, M. Onno, N. Bertho, B. Drenou, R. Fauchet, and L. Amiot. 1999. Differential expression of major histocompatibility complex pothesis that HLA-DR-mediated apoptosis is linked to Ag presen- class Ia, Ib, and II molecules on monocytes-derived dendritic and macrophagic tation capacity. cells. Hum. Immunol. 60:591. In activated B lymphocytes, it has been shown that HLA-DR- 4. Hart, D. N. 1997. Dendritic cells: unique leukocyte populations which control the primary immune response. Blood 90:3245. mediated cell death involves Fas/Fas-L interaction. However, a 5. Young, J. W., and R. M. Steinman. 1990. Dendritic cells stimulate primary hu- caspase-independent mechanism of HLA-DR-mediated apoptosis man cytolytic lymphocyte responses in the absence of CD4ϩ helper T cells. has been recently reported in mature normal and tumoral B cells J. Exp. Med. 171:1315. 6. Zhou, L. J., and T. F. Tedder. 1995. Human blood dendritic cells selectively (42). We have used cell-permeable and irreversible caspase inhib- express CD83, a member of the immunoglobulin superfamily. J. Immunol. 154: itors to establish that HLA-DR apoptosis of DC occurs in a 3821. caspase-independent manner. Nonetheless, it is clear that these ex- 7. Rescigno, M., M. Martino, C. L. Sutherland, M. R. Gold, and P. Ricciardi-Castagnoli. 1998. Dendritic cell survival and maturation are regu- periments only allow us to exclude caspases specifically inhibited lated by different signaling pathways. J. Exp. Med. 188:2175. by either Z-VAD-fmk or Z-DEVD-fmk. There is increasing evi- 8. Sallusto, F., and A. Lanzavecchia. 1994. Efficient presentation of soluble antigen dence that apoptosis can occur in a caspase-independent manner. by cultured human dendritic cells is maintained by granulocyte/macrophage col- ony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis Overexpression of the PML (43) product triggers apoptosis in factor ␣. J. Exp. Med. 179:1109. the absence of caspase 3 activation and caspase inhibitors are in- 9. Caux, C., C. Dezutter-Dambuyant, D. Schmitt, and J. Banchereau. 1992. GM- efficient in inhibiting this apoptotic pathway (44). In the same way, CSF and TNF-␣ cooperate in the generation of dendritic Langerhans cells. Nature 360:258. NO-induced cell death is not inhibited by YVAD-mca and DEVD- 10. Sallusto, F., M. Cella, C. Danieli, and A. Lanzavecchia. 1995. Dendritic cells use mca caspase inhibitors (45). However, overexpression of bax in- macropinocytosis and the to concentrate macromolecules in the duced certain, but not all, features of apoptosis even in the pres- major histocompatibility complex class II compartment: downregulation by cy- tokines and bacterial products. J. Exp. Med. 182:389. ence of caspases inhibitors (46). A recently described 11. Cella, M., D. Scheidegger, K. Palmer-Lehmann, P. Lane, A. Lanzavecchia, and mitochondrial apoptosis-inducing factor (47) is contained inside G. Alber. 1996. Ligation of CD40 on dendritic cells triggers production of high The Journal of Immunology 2385

levels of interleukin-12 and enhances T cell stimulatory capacity: T-T help via 32. Zhu, H., H. O. Fearnhead, and G. M. Cohen. 1995. An ICE-like protease is a APC activation. J. Exp. Med. 184:747. common mediator of apoptosis induced by diverse stimuli in human monocytic 12. Hsu, F. J., C. Benike, F. Fagnoni, T. M. Liles, D. Czerwinski, B. Taidi, THP.1 cells. FEBS Lett. 374:303. E. G. Engleman, and R. Levy. 1996. Vaccination of patients with B-cell lym- 33. Nicholson, D. W., A. Ali, N. A. Thornberry, J. P. Vaillancourt, C. K. Ding, phoma using autologous antigen-pulsed dendritic cells. Nat. Med. 2:52. M. Gallant, Y. Gareau, P. R. Griffin, M. Labelle, Y. A. Lazebnik, et al. 1995. 13. Reid, C. D. 1998. The biology and clinical applications of dendritic cells. Trans- Identification and inhibition of the ICE/CED-3 protease necessary for mammalian fus. Med. 8:77. apoptosis. Nature 376:37. 14. Vecchiarelli, A., M. Dottorini, D. Pietrella, C. Monari, C. Retini, T. Todisco, and 34. Pope, J. H., M. K. Horne, and E. J. Wetters. 1969. Significance of a complement- F. Bistoni. 1994. Role of human alveolar macrophages as antigen-presenting cells fixing antigen associated with herpes-like virus and detected in the Raji cell line. in Cryptococcus neoformans . Am. J. Respir. Cell Mol. Biol. 11:130. Nature 222:186. 15. Miyazaki, H., and T. Osawa. 1983. Accessory functions and mutual cooperation 35. Weiss, A., R. L. Wiskocil, and J. D. Stobo. 1984. The role of T3 surface mole- of murine macrophages and dendritic cells. Eur. J. Immunol. 13:984. cules in the activation of human T cells: a two-stimulus requirement for IL 2 16. Toujas, L., J. G. Delcros, E. Diez, N. Gervois, G. Semana, G. Corradin, and production reflects events occurring at a pre-translational level. J. Immunol. 133: F. Jotereau. 1997. Human monocyte-derived macrophages and dendritic cells are 123. comparably effective in vitro in presenting HLA class I-restricted exogenous peptides. Immunology 91:635. 36. Tsuchiya, S., M. Yamabe, Y. Yamaguchi, Y. Kobayashi, T. Konno, and K. Tada. 17. Van Der Pouw Kraan, T. T., F. E. Stiekema, M. B. Teunissen, J. D. Bos, and 1980. Establishment and characterization of a human acute monocytic leukemia L. Kapsenberg. 1989. The selective antigen-presenting cell capacity of activated cell line (THP-1). Int. J. Cancer 26:171. B lymphocytes in HLA-II-restricted responses of CD4ϩ T lymphocytes. Immu- 37. Koeffler, H. P., L. Lowe, and D. W. Golde. 1980. Amygdalin (Laetrile): effect on nology 68:147. clonogenic cells from human myeloid leukemia cell lines and normal human 18. Liu, Y. J., C. Barthelemy, O. de Bouteiller, C. Arpin, I. Durand, and marrow. Cancer Treat. Rep. 64:105. J. Banchereau. 1995. Memory B cells from human tonsils colonize mucosal ep- 38. Kitamura, T., T. Tange, T. Terasawa, S. Chiba, T. Kuwaki, K. Miyagawa, ithelium and directly present antigen to T cells by rapid up-regulation of B7-1 and Y. F. Piao, K. Miyazono, A. Urabe, and F. Takaku. 1989. Establishment and B7-2. Immunity 2:239. characterization of a unique human cell line that proliferates dependently on 19. Scholl, P. R., and R. S. Geha. 1994. MHC class II signaling in B-cell activation. GM-CSF, IL-3, or erythropoietin. J. Cell. Physiol. 140:323. Immunol. Today 15:418. 39. Koopman, G., C. P. Reutelingsperger, G. A. Kuijten, R. M. Keehnen, S. T. Pals, 20. Rich, T., S. E. Lawler, J. M. Lord, V. M. Blancheteau, D. J. Charron, and and M. H. van Oers. 1994. Annexin V for flow cytometric detection of phos- N. A. Mooney. 1997. HLA class II-induced translocation of PKC␣ and PKC␤ II phatidylserine expression on B cells undergoing apoptosis. Blood 84:1415. ␤ isoforms is abrogated following truncation of DR cytoplasmic domains. J. Im- 40. Thibeault, A., H. Zekki, W. Mourad, D. Charron, and R. Al-Daccak. 1999. Trig- munol. 159:3792. gering HLA-DR molecules on human peripheral monocytes induces their death. 21. Lane, P. J., F. M. McConnell, G. L. Schieven, E. A. Clark, and J. A. Ledbetter. Cell. Immunol. 192:79. 1990. The role of class II molecules in human B cell activation: association with 41. Ahsmann, E. J., S. E. Boom, H. M. Lokhorst, G. Rijksen, and A. C. Bloem. 1997. phosphatidyl inositol turnover, protein tyrosine phosphorylation, and prolifera- Anti-adhesive signals are mediated via major histocompatibility complex class II tion. J. Immunol. 144:3684. molecules in normal and neoplastic human B cells: correlation with B cell dif- 22. Ramirez, R., J. Carracedo, N. Mooney, and D. Charron. 1992. HLA class-II- ferentiation stage. Eur. J. Immunol. 27:2688. mediated homotypic aggregation: involvement of a protein tyrosine kinase and protein kinase C. Hum. Immunol. 34:115. 42. Dre´nou, B., V. Blancheteau, D. H. Burgess, R. Fauchet, D. Charron, and 23. Truman, J. P., M. L. Ericson, C. J. Choqueux-Seebold, D. J. Charron, and N. A. Mooney. 1999. A caspase-independent pathway of MHC class II antigen- N. A. Mooney. 1994. Lymphocyte programmed cell death is mediated via HLA mediated apoptosis of human B lymphocytes. J. Immunol. 163:4115. class II DR. Int. Immunol. 6:887. 43. Mu, Z. M., K. V. Chin, J. H. Liu, G. Lozano, and K. S. Chang. 1994. PML, a 24. Naitoh, K., Y. Ichigi, K. Miyake, A. Muraguchi, and M. Kimoto. 1994. Signal growth suppressor disrupted in acute promyelocytic leukemia. Mol. Cell. Biol. transmission through MHC class II molecules in a human B lymphoid progenitor 14:6858. cell line: different signaling pathways depending on the maturational stages of B 44. Quignon, F., F. De Bels, M. Koken, J. Feunteun, J. C. Ameisen, and H. de The. cells. Microbiol. Immunol. 38:967. 1998. PML induces a novel caspase-independent death process. Nat. Genet. 20: 25. Garban, F., J. P. Truman, J. Lord, B. Drenou, J. Plumas, M. C. Jacob, J. J. Sotto, 259. D. Charron, and N. Mooney. 1998. Signal transduction via human leucocyte 45. Okuno, S., S. Shimizu, T. Ito, M. Nomura, E. Hamada, Y. Tsujimoto, and antigen class II molecules distinguishes between cord blood, normal, and malig- H. Matsuda. 1998. Bcl-2 prevents caspase-independent cell death. J. Biol. Chem. nant adult B lymphocytes. Exp. Hematol. 26:874. 273:34272. 26. Yoshino, T., L. Cao, R. Nishiuchi, Y. Matsuo, I. Yamadori, E. Kondo, 46. Miller, T. M., K. L. Moulder, C. M. Knudson, D. J. Creedon, M. Deshmukh, N. Teramoto, K. Hayashi, K. Takahashi, N. Kamikawaji, et al. 1995. Ligation of S. J. Korsmeyer, and E. M. Johnson, Jr. 1997. Bax deletion further orders the cell HLA class II molecules promotes sensitivity to CD95 (Fas antigen, APO-1)- death pathway in cerebellar granule cells and suggests a caspase-independent mediated apoptosis. Eur. J. Immunol. 25:2190. pathway to cell death. J. Cell Biol. 139:205. 27. Truman, J. P., C. Choqueux, J. Tschopp, J. Vedrenne, F. Le Deist, D. Charron, 47. Susin, S. A., H. K. Lorenzo, N. Zamzami, I. Marzo, B. E. Snow, G. M. Brothers, and N. Mooney. 1997. HLA class II-mediated death is induced via Fas/ J. Mangion, E. Jacotot, P. Costantini, M. Loeffler, et al. 1999. Molecular char- interactions in human splenic B lymphocytes. Blood 89:1996. acterization of mitochondrial apoptosis-inducing factor. Nature 397:441. 28. Salvesen, G. S., and V. M. Dixit. 1997. Caspases: intracellular signaling by pro- 48. Ingulli, E., A. Mondino, A. Khoruts, and M. K. Jenkins. 1997. In vivo detection teolysis. Cell 91:443. ϩ 29. Raff, M. 1998. Cell suicide for beginners. Nature 396:119. of dendritic cell antigen presentation to CD4 T cells. J. Exp. Med. 185:2133. 30. Greenfield, E. A., K. A. Nguyen, and V. K. Kuchroo. 1998. CD28/B7 costimu- 49. Montes, M., D. McIlroy, A. Hosmalin, and A. Trautmann. 1999. Calcium re- lation: a review. Crit. Rev. Immunol. 18:389. sponses elicited in human T cells and dendritic cells by cell-cell interaction and 31. Kuhn, U., P. Brand, J. Willemsen, H. Jonuleit, A. H. Enk, soluble ligands. Int. Immunol. 11:561. R. van Brandwijk-Petershans, J. Saloga, J. Knop, and D. Becker. 1998. Induction 50. Matsue, H., D. Edelbaum, A. C. Hartmann, A. Morita, P. R. Bergstresser, of tyrosine phosphorylation in human MHC class II-positive antigen-presenting H. Yagita, K. Okumura, and A. Takashima. 1999. Dendritic cells undergo rapid cells by stimulation with contact sensitizers [Published erratum appears in 1999 apoptosis in vitro during antigen- specific interaction with CD4ϩ T cells. J. Im- J. Immunol. 162:3106]. J. Immunol. 160:667. munol. 162:5287.