A Novel Pathway of Alloantigen Presentation by Dendritic Cells

A Novel Pathway of Alloantigen Presentation by Dendritic Cells
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A Novel Pathway of Alloantigen Presentation byDendritic Cells 1 Osquel Barroso Herrera, Dela Golshayan, Rebecca Tibbott, Francisco Salcido Ochoa,Martha J. James, Federica M. Marelli-Berg, and Robert I. Lechler 2 In the context of transplantation, dendritic cells (DCs) can sensitize alloreactive T cells via two pathways. The direct pathway isinitiated by donor DCs presenting intact donor MHC molecules. The indirect pathway results from recipient DCs processing andpresenting donor MHC as peptide. This simple dichotomy suggests that T cells with direct and indirect allospecificity cannotcross-regulate each other because distinct APCs are involved. In this study we describe a third, semidirect pathway of MHCalloantigen presentation by DCs that challenges this conclusion. Mouse DCs, when cocultured with allogeneic DCs or endothelialcells, acquired substantial levels of class I and class II MHC:peptide complexes in a temperature- and energy-dependent manner.Most importantly, DCs acquired allogeneic MHC in vivo upon migration to regional lymph nodes. The acquired MHC moleculeswere detected by Ab staining and induced proliferation of Ag-specific T cells in vitro. These data suggest that recipient DCs, dueto acquisition of donor MHC molecules, may link T cells with direct and indirect allospecificity.  The Journal of Immunology, 2004, 173: 4828–4837. A lloreactive T cells recognize alloantigens via two dis-tinct, but not mutually exclusive, pathways: direct andindirect. In the direct pathway of allorecognition, re-sponder T cells recognize intact foreign MHC:peptide complexeson the surface of donor cells (1–5). This pathway is characterizedby the high precursor frequency of responder T cells (  100-foldgreater than for a response to conventional protein Ags) and iscritical in the initiation of the alloresponse and acute graft rejection(3, 6–8). By the indirect pathway, in contrast, recipient T cellsrecognize peptides derived from foreign MHC molecules after pro-cessing and presentation by self-MHC molecules on recipientAPCs (8–10). The indirect pathway, in other words, is the normalmechanism of self MHC-restricted T cell stimulation. During recentyears,increasingexperimentalevidencehassupportedtheroleforthispathway in both acute and chronic graft rejection (11, 12).Several lines of evidence suggest that T cells with indirect al-lospecificity can regulate, positively or negatively, direct pathwayT cells. Indeed, Lee et al. (13) have shown, using MHC classII-deficient skin grafts, that CD4  T cells with indirect anti-donorspecificity can amplify direct pathway CD8  T cell responses.Similarly, tolerant indirect pathway T cells can suppress the re-sponse of direct pathway T cells in some models (14, 15). It hasbeen shown that the induction of tolerance to minor mismatchedskin grafts in mice using nondepleting anti-CD4 and anti-CD8mAbs involves the reprocessing of minors on host APCs and theinduction of regulatory CD4  T cells with indirect pathway spec-ificity (14). In addition, elimination of CD4  T cells from tolerantmice resulted in the rejection of long-standing grafts, suggestingthat direct pathway CD8  T cells had been under continuous reg-ulation by tolerant, indirect pathway CD4  T cells (15). Theseobservations point to a four-cell, unlinked, model for interactionsbetween direct and indirect pathway T cells during the course of graft rejection: helper, or suppressor, CD4  T cells with indirectspecificity are activated by recipient dendritic cells (DCs) 3 thatreside in secondary lymphoid organs, whereas direct pathway ef-fector CD8  T cells must recognize determinants expressed on thecells of the donor graft.We propose in this study that the recently described phenome-non of MHC transfer between cells may provide a mechanism toresolve this conundrum. Several studies have shown that DCs arecapable of acquiring intact MHC molecules from other cells (otherDCs, macrophages, activated T cells, B cells, and tumor cells) invitro (16–18). Harshyne et al. (18) have reported that individualDCs, especially immature DCs, physically extract plasma mem-brane and, to a lesser extent, intracellular proteins from other DCsin a cell contact-dependent fashion. In addition, DCs are capable of shedding soluble MHC molecules and membrane vesicles, calledexosomes, that express MHC class I and class II molecules andcostimulatory molecules that can be captured by other DCs (16,19). As a consequence, there could be a third, semidirect pathwayof allorecognition by which direct pathway T cells would recog-nize allogeneic MHC molecules after being transferred, intact,from the surface of donor cells to the surface of recipient DCs. Inthis study the acquisition and presentation of MHC:peptide com-plexes by DCs has been investigated in vitro and in vivo. Materials and Methods  Mice C57BL/6, BALB/c, CBA/Ca, AKR/J, B10.A(2R), and B10.A(4R) mice,6–10 wk of age, were purchased from Harlan Olac (Bicester, U.K.). MHCclass II and   2 -microglobulin double-knockout mice (MHC 0 ) (20) were Department of Immunology, Imperial College, Hammersmith Hospital, London,United KingdomReceived for publication December 4, 2003. Accepted for publication August 5, 2004.The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked  advertisement   in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by a grant (DMIMM PN0559) from GlaxoSmithKlineResearch and Development. 2 Address correspondence and reprint requests to Dr. Robert I. Lechler, Departmentof Immunology, Imperial College, Hammersmith Hospital, Du Cane Road, London,U.K. W12 ONN. E-mail address:  3 Abbreviations used in this paper: DC, dendritic cell; BMDC, bone marrow-derivedDC; DNP, dinitrophenol; EC, endothelial cell; FSC, forward scatter; MFI, mean flu-orescence intensity; MLN, mesenteric lymph node; TRITC, tetramethylrhodamineisothiocyanate. The Journal of Immunology Copyright © 2004 by The American Association of Immunologists, Inc. 0022-1767/04/$02.00  bred at the Biological Services Unit, Hammersmith Hospital, and weredonated by Dr. M. Merkenschlager. RAG   /   TCR-transgenic mouse strainF5 (CD8  T cells specific for influenza strain A/HK/8/68 nucleoprotein-derived peptide NT68/H-2D b , sequence ASNENMDAM), were obtainedfrom Dr. D. Kioussis (National Institute for Medical Research, London,U.K.). Spleens from RAG   /   , TCR-transgenic OT-II mice (CD4  T cellsspecific for OVA 323–339  /H-2A b ) were obtained from mice kept in a patho-gen-free environment at GlaxoSmithKline Research and Development(Stevenage, U.K.). Mice of the same sex were used within experiments.  DC cultures Mouse bone marrow-derived DCs (BMDCs) were generated as previouslydescribed (21), with slight modifications. Briefly, bone marrow was flushedfrom femurs, passed through a 200-  m pore size mesh to remove fibroustissue, and RBCs were lysed using ACK buffer. Cells were cultured at 10 6 cells/ml in RPMI 1640 medium (Invitrogen Life Technologies, Paisley,U.K.) supplemented with 10% FCS, 2 mM glutamine, 50   M 2-ME, 100IU/ml penicillin, 100   g/ml streptomycin, and 6 ng/ml mouse rGM-CSF(produced by Dr. M. Sims, GlaxoSmithKline Research and Development).On day 3 of culture, floating cells were gently removed, and fresh rGM-CSF-containing medium was added. On day 5 of culture, BMDCs wereeither left untreated or induced to mature by adding 1   g/ml LPS (Sigma-Aldrich, Gillingham, U.K.) to the cultures. After an overnight incubation,nonadherent cells and loosely adherent proliferating BMDC aggregateswere collected, washed, and replated for 1 h at 37°C to remove contami-nating macrophages. Subsequently, resulting cell populations were en-riched for CD11c-positive BMDCs by positive selection after incubationwith anti-CD11c-coated magnetic microbeads (Miltenyi Biotec, Bisley,U.K.) and passing the bead-bound BMDCs through a separation column(MS  columns; Miltenyi Biotec) placed on the separation unit, accordingto manufacturer’s instructions.  Endothelial cell (EC) cultures ECs derived from hearts of C57BL/6 and CBA/Ca mice were cultured aspreviously described (22) in DMEM (Invitrogen Life Technologies, Gaith-ersburg, MD) supplemented with 20% heat-inactivated FCS, 2 mM glu-tamine, 100 IU/ml penicillin, 100  g/ml streptomycin, 1 mM sodium pyru-vate, 20 mM HEPES, 1% nonessential amino acids, 50   M 2-ME, 150  g/ml EC growth supplement (Sigma-Aldrich), and 12 U/ml heparin in 2%gelatin-coated tissue culture flasks. At confluence, the ECs were detachedfrom the culture flasks using a solution of 0.125% trypsin in 0.2% EDTAand passaged. For phenotypic analysis, the ECs were used between pas-sages 4 and 10. CFSE staining Cells were resuspended at 5–10  10 6 cells/ml in a 2-  M solution of CFSE(Molecular Probes, Leiden, The Netherlands) in PBS and incubated at37°C for 10 min in the dark. At the end of the incubation period, the cellswere immediately washed once in cold PBS/8% FCS and twice more incold PBS/2% FCS. In the case of DCs to be injected, cells were labeledwith 5   M CFSE and finally resuspended in PBS before injection. Transfer cultures CFSE-labeled C57BL/6 or CBA/Ca DCs were pulsed for 4 h with NT68(10   M) and OVA 323–339  (20   M) peptides, washed twice, and mixed atequal numbers (5  10 5 ) with BALB/c DCs. Cells were cocultured at 4 or37°C in 24-well plates in 1 ml of complete RPMI 1640 medium for 20 hand harvested for Ab staining. In some experiments DCs were coculturedin the presence or the absence of titrated concentrations of DNP (1–100  g/ml) or sodium azide (1–10   M).Confluent ECs were detached from the culture flasks by trypsin/EDTAtreatment and added at 10 6 cells/flask (25 cm 2 ; Nunc, Roskilde, Denmark)in 3 ml of complete EC culture medium supplemented, or not, with 1000U of mouse rIFN-    (PeproTech, London, U.K.) for 96 h. Cells were re-covered by trypsin/EDTA treatment, washed twice, stained or not, withCFSE, and replated at 2    10 5 cells/well in 24-well plates overnight toallow the formation of a monolayer. In some experiments the cells werepulsed for 4 h with NT68 and OVA 323–339  peptides and washed in the wellsthree times with PBS before the addition of 3    10 5 BALB/c DCs/well.Cells were cocultured in 1 ml of complete RPMI 1640 medium for 20 h andcollected by pipetting off nonadherent cells, followed by trypsinization of adherent ECs. For trans-well studies, peptide-pulsed labeled DCs or ECswere added to 0.4-  m pore size trans-well chambers (Costar, Cambridge,U.K.) inserted into wells containing unlabeled DCs. After 20 h of culture,cells in the lower wells were collected and analyzed by flow cytometricstaining or were used as stimulators of T cells. Flow cytometry All the mAbs used, unless otherwise stated, were purchased from BDPharmingen (Cowley, U.K.). For analysis of DC purity and phenotype,cells were washed in cold PBS supplemented with 2.5% FCS and 0.05%sodium azide. BMDCs were first incubated with an anti-CD16/CD32 (anti-FcR    III/FcR    II, clone 2.4G2) mAb for 10 min and subsequently double-stained for 30 min with a PE-conjugated anti-CD11c mAb (clone HL3) inconjunction with FITC-conjugated mAbs to MHC class II (anti-H-2A b ,clone AF6-120.1; anti-H-2A d , clone AMS-32.1; anti-H-2A k  , clone 11-5.2),MHC class I (anti-H-2K b , clone AF6-88.5; anti-H-2K d , clone SF1-1.1),anti-H-2K k  (Caltag Medsystems, Silverstone, U.K.), CD80 (clone 16-10A1), CD86 (clone GL1), or CD40 (clone 3/23). In each case, an FITC-conjugated mAb of the same isotype as the marker-specific mAb was usedin conjunction with the anti-CD11c mAb as a negative control. The purityof DCs was consistently between 80 and 90% CD11c-positive cells.The purity of responder T cells was assessed by staining with FITC- orPE-conjugated mAbs to CD3 (clone 145-2C11), CD4 (clone RM4-5), CD8(clone 53-6.7), and H-2A d or H-2A b . Adherent ECs were detached from theculture flasks with trypsin/EDTA and resuspended in the same FCS-con-taining buffer before staining with PE- or FITC-conjugated mAbs specificfor H-2A b , H-2K b , H-2D b (clone KH95), H-2A k  , H-2K k  , CD80, CD86,CD40, or CD106 (clone 429). For analysis of MHC transfer in cultures of mixed cells, cells were incubated with the anti-FcR   mAb and stained withthe specific PE-conjugated mAbs as indicated. Analyses were performedon a FACSCalibur flow cytometer (BD Biosciences, Mountain View, CA)using CellQuest acquisition and analysis software on cells gated for ho-mogeneous forward scatter (FSC) and side scatter characteristics. FACS  DCs in the mixed cultures were purified to no less than 99% purity fromCFSE-labeled DCs or ECs on a FACStar cell sorter (BD Biosciences). Preparation of responder T cells Responder T cells were purified from splenocytes of normal BALB/c orTCR-transgenic F5 or OT-II mice. Briefly, cell suspensions were preparedby mashing spleens through a cell strainer, and RBC were lysed using RBClysis buffer. Cell preparations were passed through a nylon wool column toremove most contaminating macrophages and B cells. For further purifi-cation of CD4  or CD8  T cells, remaining B cells, macrophages, NKcells, and CD8  or CD4  T cells were removed after incubation with amixture of rat mAbs to B220 (clone RA3-6B2; BD Pharmingen), MHCclass II (culture supernatant from MS/114.15.2 hybridoma), CD16/CD32(clone 2.4G2; BD Pharmingen), and anti-CD8 (culture supernatant from53-6.72 hybridoma) or anti-CD4 mAbs (clone YTS191; purified at Glaxo-SmithKline Research and Development), respectively, followed by an in-cubation period with sheep anti-rat IgG-coated Dynabeads (Dynal Biotech,Oslo, Norway). The bead/mAb-bound cells were selected using a magnet,and the purified T cell populations were recovered from the fluid phase.The purity of the T cell preparations obtained was consistently  85%, asdetermined by flow cytometric analysis of cell phenotype as describedabove. T cell proliferation assays  MLR.  Purified irradiated (30 Gy) DCs (10 4 ) were used as stimulators of 10 5 purified CD4  or CD8  BALB/c T cells in triplicate wells of 96-well,flat-bottom plates. T cell proliferation was measured by [ 3 H]thymidineincorporation after 5 days in culture (1  Ci/well for the last 18 h of culture;Amersham Biosciences, Little Chalfont, U.K.). Results are presented as themean cpm of triplicate determinations  SD.  Peptide-specific T cell responses.  Purified TCR-transgenic OT-II CD4  or F5 CD8  T cells (2.5    10 4 ) were stimulated with titrated numbersof irradiated DCs in triplicate wells of 96-well plates. Alternatively, 10 5 responder R2.2 T cells (23) were stimulated with 2  10 4 irradiated DCsin the presence of increasing amounts (0.01–1   M) of H-Y peptide. T cellproliferation was measured by [ 3 H]thymidine incorporation after 3 daysin culture. Results are presented as the mean cpm of triplicate determina-tions  SD.  In vivo mouse model To investigate the acquisition of MHC molecules in vivo, different exper-imental approaches were used (Fig. 1). First, 6- to 8-wk-old C57BL/6 micewere injected i.p. with 200  l of PBS or 1000 U of rIFN-   in PBS to inducelocal inflammation. Forty-eight hours later, mice within each group re-ceived an i.p. injection of 2.5    10 6 CFSE-labeled immature or matureMHC 0 DCs. Twenty hours after DC inoculation, mesenteric lymph nodes(MLNs) were collected for immunofluorescence staining (Fig. 1  A ). 4829The Journal of Immunology  In the second method, naive B10.A(2R) or AKR/J mice were injectedi.p. with PBS or CFSE-labeled mature DCs derived from B10.A(2R) orB10.A(4R) mice. Twenty hours later, cell suspensions were prepared fromMLNs and depleted of T and B cells by negative selection using rat mAbsspecific for B220 and CD3 and sheep anti-rat IgG-coated Dynabeads asdescribed above. Cells were double-stained with a PE-conjugated anti-H-2E  k  (clone 14-4-4S; BD Pharmingen) and a Cy5-conjugated anti-CD8  (clone 53-6.7; BD Pharmingen) mAb and were analyzed by flow cytometry(Fig. 1  B ).Finally, mature DCs were prepared from B10.A congenic strainsB10.A(2R) and B10.A(4R), labeled with CFSE, and transferred i.p. intomatched (negative controls, B10.A(4R) DCs into 4R hosts) and mis-matched recipients (positive controls, B10.A(2R) DCs into 4R hosts; ex-perimental group, B10.A(4R) DCs into 2R hosts). Thirty hours after thetransfer, the recipients were killed, and cell suspensions were preparedfrom the spleen and pooled abdominal lymph nodes. CFSE-positive cellswere selected by FACS and used in an in vitro proliferation assay as stim-ulators of the HY-specific and H-2E k  -restricted R2.2 T cell clone (Fig. 1 C  ).  Immunofluorescent staining of tissue sections MLNs were removed, placed in compound-embedding medium (OCT;BDH, Dorset, U.K.), snap-frozen in liquid nitrogen, and stored at  80°Cfor no more than 96 h. Ten-micron cryostat sections were cut, collected onpoly- L -lysine-coated slides (VWR International, Lutterworth, U.K.), andallowed to air-dry. Slides were stored at 4°C for 24 h before staining.Sections were fixed with cold acetone for 5 min, air-dried, and incubatedwith the appropriate dilutions in 5% FCS/PBS of rabbit anti-FITC (1/1000;DakoCytomation, Ely, U.K.) and biotinylated anti-H-2A b (5  g/ml; cloneAF6-120.1; BD Pharmingen) or anti-H-2K b (5  g/ml; clone AF6-88.5; BDPharmingen) mAbs for 1 h. Isotype-matched mIgG2a (BD Pharmingen)and rabbit Ig (DakoCytomation) served as controls. The slides werewashed three times (once for 1 min, once for 2 min, and once for 3 min)in PBS under stirring and then incubated with an FITC-conjugated swineanti-rabbit mAb (1/40; DakoCytomation) in conjunction with ExtrAvidin-tetramethylrhodamine isothiocyanate (ExtrAvidin-TRITC; 5   g/ml; Sig-ma-Aldrich) in 5% FCS/PBS for 1 h. Slides were washed again four times(once for 1 min, once for 2 min, once for 3 min, and once for 4 min) andmounted in mounting medium (Citifluor, Kent, U.K.). Slides were visual-ized with a Coolview 12-cooled CCD camera (Photonic Science, Newbury,U.K.) mounted over an Axiovert S100 microscope equipped with Meta-morph software (Zeiss, Welwyn Garden City, U.K.). Photomicrographswere taken under equal exposure conditions to obtain a permanent record.Final image processing was performed using Photoshop 5.0 (Adobe Sys-tems, San Jose, CA). Results Transfer in vitro of surface MHC molecules between liveallogeneic DCs Based on the assumption that DCs are the most efficient donors of MHC molecules due to exosome release, we first examined thetransfer of allogeneic MHC molecules in cocultures of mouse BM-DCs. Both immature and LPS-treated, mature DCs were generatedfrom three different mouse strains: BALB/c, C57BL/6, andCBA/Ca (see Table I). The populations obtained were then en-riched (up to 90%) for CD11c-expressing DCs by positive selec-tion using anti-CD11c-coated magnetic beads before being used inthe cocultures. The phenotypic characterization of DCs by flowcytometric staining showed a marked up-regulation of MHC classI and class II molecules and the costimulatory ligands CD80, FIGURE 1.  Experimental designs to investigate the acquisition of allo-geneic MHC molecules in vivo.  A , C57BL/6 mice were injected i.p. with200   l of PBS or 1000 U of rIFN-    in PBS to induce local inflammation.Forty-eight hours later, mice were injected with 2.5    10 6 CFSE-labeledimmature or mature MHC 0 DCs. Twenty hours after DC inoculation,MLNs were collected for immunofluorescent staining.  B , B10.A(2R) orAKR/J mice were injected i.p. with PBS or CFSE-labeled mature DCs derivedfrom B10.A(2R) or B10.A(4R) mice. Twenty hours later, cell suspensionswere prepared from MLNs, depleted of T and B cells by negative selection,and double-stained with a PE-conjugated anti-H-2E   and a Cy5-conjugatedanti-CD8  mAbs and analyzed by flow cytometry.  C  , Recipient B10.A(2R) orB10.A(4R) mice were injected i.p. with 1000 U of rIFN-   . Forty-eight hourslater, 10 7 CFSE-labeled mature DCs were transferred i.p. into i) matched (neg-ative controls, B10.A(4R) DCs into 4R hosts), ii) H-2E k  -positive into H-2E k  -negative mismatched recipients (positive controls, B10.A(2R) DCs into 4Rhosts), and iii) H-2E k  -negative into H-2E k  -positive mismatched recipients (ex-perimental group, B10.A(4R) DCs into 2R hosts). Thirty hours after the trans-fer, the recipients were killed, and cell suspensions were prepared from thespleen and pooled abdominal lymph nodes. CFSE-positive cells were selectedby FACS and used in an in vitro proliferation assay as stimulators of theHY-specific and H-2E k  -restricted R2.2 T cell clone.Table I.  Mouse strains used  Strain HaplotypeH-2 moleculesK A   A   E   E   D BALB/c  d d d d d d d  C57BL/6  b b b b b b CBA/Ca  k k k k k k k  AKR/J  k k k k k k k  B10.A(2R)  h2 k k k k k b B10.A(4R)  h4 k k k k b 4830 SEMIDIRECT PATHWAY OF ALLOANTIGEN PRESENTATION BY DCs  CD86, and CD40 in LPS-treated, as compared with nontreated,immature DCs (not shown).We devised a system to single out DC populations by labelingdonor cells with an intracellular dye, CFSE, before coculture withallogeneic DCs. Labeling of the donor cells and subsequent anal-ysis of CFSE-negative recipient cells excludes the possibility of mistakenly analyzing doublets comprising cells from the two inputpopulations. DCs obtained from BALB/c mice (H-2 d ) were cocul-tured with CFSE-labeled C57BL/6 (H-2 b ) or third-party, controlCBA/Ca DCs (H-2 k  ). After coculture for 0 and 20 h, CFSE-neg-ative BALB/c DCs were analyzed by flow cytometry using PE-conjugated mAbs specific for H-2A b (Fig. 2,  B–E  ,  left panels ) andH-2K b (Fig. 2,  B–E  ,  right panels ). For comparison, the expressionof H-2A b and H-2K b on C57BL/6 cells, acting as donors of MHCmolecules, is shown in Fig. 2  A . The results showed acquisition of allogeneic H-2 b MHC molecules by a significant number of BALB/c DCs. The efficiency of transfer was higher for MHC classII molecules than for MHC class I molecules. Mature DCs ac-quired higher levels of allogeneic MHC class II and class I mol-ecules (Fig. 2  E  ) compared with immature cells (Fig. 2 C  ). MatureDCs were also much better donors of surface MHC molecules thanimmature DCs. For example,   42% of immature acceptorBALB/c DCs became H-2A b -positive when cocultured with ma-ture C57BL/6 DCs compared with 10% when the donor cells wereimmature (Fig. 2 C  ,  left panel ). This may reflect the higher levelsof MHC expression on mature DCs. Similar results were obtainedwhen the acceptor and donor DC strain combinations were in-verted (data not shown). Labeled DCs were viable during the entirecoculture period as measured by lack of staining using a PE-con- jugated annexin V-specific mAb to measure early apoptosis in con- junction with phosphatidylinositol incorporation to assess induc-tion of necrosis (not shown), suggesting that transfer of MHCmolecules was not the result of uptake of dead or dying cells.Parallel experiments were performed in which the donor andacceptor DCs were physically separated by a 0.4-  m pore sizesemipermeable membrane that allows the traffic of soluble mole-cules and exosomes, but not of intact cells. After 20 h of culture,a considerable percentage of CFSE-negative DCs in the lowerwells expressed the MHC of DCs in the upper wells. The transfer,however, was   3-fold less efficient than when the cells werecocultured (data not shown), suggesting that direct cell-cell contactis not essential for MHC transfer between DCs. This correlateswith the findings reported by Emerson and Cone (24) describingthe shedding of mouse MHC molecules within membrane-derivedlipid vesicles (exosomes). Transfer efficiency is affected by temperature and levels of available ATP To test whether the observed transfer of MHC molecules is a tem-perature- and/or an energy-dependent phenomenon, allogeneicDCs were cocultured at either 4 or 37°C in the presence or theabsence of titrated concentrations of sodium azide (1 or 10   M) orDNP (1, 10, and 100   g/ml). Azide ions inhibit the mitochondrialrespiratory chain, whereas DNP is an uncoupler of oxidative phos-phorylation. In both cases, the synthesis of ATP is inhibited, thusdecreasing the energy supply of the cell.The results in Fig. 3  A  showed that the coculture of DCs at 4°Cresulted in a notable decrease in molecule transfer. At 20 h, a mere7.6% of DCs had acquired the H-2A b molecule compared with27.6% of DCs cocultured at 37°C (not shown). A reduction in thenumber of molecules acquired per cell was also observed, as dem-onstrated by the drop in mean fluorescence intensities (MFIs) at4°C to over half that seen at 37°C (Fig. 3  A ). A similar pattern wasobserved after 40 h of coculture. These results confirm the srci-nally reported observations showing that MHC shedding and ab-sorption between splenic mouse cells were temperature-dependent(25, 26).The presence of either azide or DNP in the cocultures alsocaused a significant decrease in the efficiency of allogeneic MHCcapture by DCs (Fig. 3,  B  and  C  ). Interestingly, the azide appearedto exert a greater inhibitory effect on transfer, with different con-centrations seemingly irrelevant. Only 3.6–3.8% of recipient DCsacquired the molecules in the presence of azide compared with27.6% under normal conditions (not shown). MFIs were also 4- to5-fold lower, indicating that far fewer molecules per cell had been FIGURE 2.  Acquisition of intact allogeneic MHC molecules by DCs.  A ,The expression of self MHC class II (H-2A b ;  left panel ) and class I (H-2K b ; right panel ) molecules on immature (thin lines) and mature (heavy lines)CFSE-labeled donor C57BL/6 DCs was assessed after staining with PE-conjugated specific mAbs. As a negative control, cells were stained with aPE-conjugated, isotype-matched, irrelevant mAb (filled profiles).  B–E  ,Equal numbers of immature (  B  and  C  ) or mature (  D  and  E  ) BALB/c DCs(H-2 d ) were mixed with immature (thin lines) or mature (heavy lines)CFSE-labeled C57BL/6 DCs. After 0 h (  B  and  D ) or 20 h ( C   and  E  ) of coculture, cells were stained with either a PE-conjugated anti-H-2A b ( left  panels ) or anti-H-2K b mAb ( right panels ). As controls for the staining,BALB/c DCs were cocultured with third-party CFSE-labeled CBA/Ca DCs(H-2 k  ; filled profiles) and stained with the same PE-conjugated mAb. Allanalyses shown on  B–E   were on populations with a homogeneous, singlepeak on FSC characteristics, which were CFSE-negative ( F  ; regionR1  R4; i.e., BALB/c DCs acting as acceptors of allogeneic H-2 b MHCmolecules). Numbers within figures correspond to percentages of BALB/cDCs expressing the H-2 b MHC molecules after coculture with CFSE-la-beled mature CBA/Ca or immature or mature C57BL/6 DCs, respectively. 4831The Journal of Immunology  transferred (Fig. 3 C  ). Although not as marked as with azide, ad-dition of DNP also had some effect (Fig. 3  B ). Overall, transfer wasbetween 7.4 and 13.6% of DCs, with a 2- to 3-fold drop in MFIs.  Acquired foreign MHC-peptide complexes allow recognition byT cells To confirm that the acquired allogeneic MHC molecules were fullyfunctional, acceptor mature BALB/c DCs, displaying both endog-enous H-2 d and acquired H-2 b molecules, were used as allo-stimulators of polyclonal BALB/c CD4  and CD8  T cells andH-2 b -restricted peptide-specific TCR transgenic T cells from F5and OT-II mice. The donor C57BL/6 DCs had previously beenpulsed with the influenza nucleoprotein peptide NT68 and theOVA 323–339  peptide presented by H2-D b and H-2A b , respectively.CFSE-negative DCs were purified from the cocultures by cellsorting to no less than 99% of purity (Fig. 4  A ). Cells that hadacquired foreign MHC molecules in the trans-well system werealso recovered and used as stimulators of T cells. BALB/c DCs,cultured in medium alone, were used as negative controls. TheCFSE-labeled C57BL/6 DCs, selected by cell sorting from the FIGURE 3.  Transfer of MHC molecules in vitro is temperature- and energy-dependent. Equal numbers of recipient BALB/c DCs (H-2 d ) were mixedwith CFSE-labeled C57BL/6 DCs (H-2 b ) and cocultured at 4 or 37°C for different periods of time (0, 20, or 40 h;  A ). Alternatively, cells were coculturedat 37°C for 20 h in the presence or the absence of titrated concentrations of DNP (1, 10, and 100   g/ml;  B ) or sodium azide (1 and 10   M;  C  ). Aftercoculture, cells were harvested and stained using a PE-conjugated anti-H-2A b mAb. Cells cocultured under normal conditions (37°C in medium alone) andstained with a PE-conjugated, isotype-matched, irrelevant mAb served as negative controls. FIGURE 4.  T cell recognition of foreignMHC-peptide complexes acquired by DCs.Equal numbers of LPS-treated, matureBALB/c DCs (H-2 d ) were mixed with matureCFSE-labeled C57BL/6 DCs (H-2 b ) that hadpreviously been pulsed for 4 h with NT68 pep-tide and OVA 323–339  peptide. Alternatively,these allogeneic DCs were physically sepa-rated using a 0.4-  m trans-well chamber. Ascontrols, BALB/c DCs were also pulsed withthe NT68 and OVA 323–339  peptides and cul-tured in medium alone. After 20 h of culture,CFSE-negative BALB/c DCs in the mixed cul-tures were separated from CFSE-expressingC57BL/6 DCs by FACS (  A ), and BALB/c DCswere recovered from the lower wells of thetrans-well cultures or the control single cul-tures. DCs (10 4 ) were used as stimulators of 10 5 purified BALB/c CD8  (  B ,  left panel ) andCD4  T cells (  B ,  right panel ). Alternatively,various numbers of DCs were used as stimu-lators of 2.5    10 4 TCR-transgenic, RAG   /   NT68-specific F5 CD8  T cells ( C  ,  left panel )and OVA 323–339 -specific OT-II CD4  T cells( C  ,  right panel ). T cell proliferation was mea-sured by [ 3 H]thymidine incorporation after 5days (  B ) or 3 days ( C  ) in culture. Results arepresented as the mean cpm of triplicate deter-minations  SD. 4832 SEMIDIRECT PATHWAY OF ALLOANTIGEN PRESENTATION BY DCs
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks