Skip to main content
NCBI home page
UKPMC Funders Author Manuscripts logoLink to UKPMC Funders Author Manuscripts
. Author manuscript; available in PMC: 2013 Dec 1.
Published in final edited form as: J Immunol. 2013 Apr 29;190(11):5829–5838. doi: 10.4049/jimmunol.1300458

Co-presentation of intact and processed MHC alloantigen by recipient dendritic cells enables delivery of linked help to alloreactive CD8 T cells by indirect-pathway CD4 T cells

Siva Sivaganesh 1,1, Simon J Harper 1,1, Thomas M Conlon 1, Kourosh Saeb-Parsy 1, Margaret C Negus 1, Chris J Callaghan 1, Reza Motallebzadeh 1, Eleanor M Bolton 1, J Andrew Bradley 1, Gavin J Pettigrew 1
PMCID: PMC3736307  EMSID: EMS52731  PMID: 23630361

Abstract

In transplantation, direct-pathway CD8 T cells that recognize alloantigen on donor cells require CD4 help for activation and cytolytic function. The ability of indirect-pathway CD4 T cells to provide this help remains unexplained, because a fundamental requirement for epitope linkage is seemingly broken. The simultaneous presentation, by host dendritic cells (DCs), of both intact MHC class I alloantigen and processed alloantigen would deliver linked help, but has not been demonstrated definitively.

Here, we report that following in vitro co-culture with BALB/c DCs, small numbers (~1.5%) of C57BL/6 DCs presented acquired H-2d alloantigen both as processed allopeptide and as unprocessed antigen. This re-presented class I alloantigen provides a conformational epitope for direct-pathway allorecognition, because C57BL/6 DCs isolated from co-cultures and transferred to naïve C57BL/6 mice provoked cytotoxic CD8 T cell alloimmunity. Crucially, this response was dependent upon simultaneous presentation of class II-restricted allopeptide, because despite acquiring similar amounts of H-2d alloantigen upon co-culture, MHC class II-deficient C57BL/6 DCs failed to elicit cytotoxic alloimmunity. The relevance of this pathway to solid organ transplantation was then confirmed by the demonstration that CD8 T cell cytotoxicity was provoked in secondary recipients by transfer of DCs purified from wild-type, but not from MHC class II-deficient, C57BL/6 recipients of BALB/c heart transplants.

These experiments demonstrate that re-presentation of conformationally-intact MHC alloantigen by recipient APC can induce cytotoxic alloimmunity, but simultaneous co-presentation of processed allopeptide is essential, presumably because this facilitates linked recognition by indirect-pathway helper CD4 T cells.

Introduction

Transplant alloantigens are distinct because they can be recognized by T cells through two pathways: the direct-pathway, whereby MHC alloantigen is recognized intact on the surface of donor APC; and the indirect-pathway, whereby alloantigen is recognized as self-restricted peptide fragments, after internalization, processing and presentation by recipient APCs (1-3). Both pathways are applicable to CD4 and to CD8 T cell alloimmune responses, but for cytotoxic CD8 T cell responses against vascularised allografts, only the direct-pathway is relevant, because despite some replacement of the endothelium by recipient-derived haemopoietic progenitor cells (4), graft parenchymal cells remain overwhelmingly of donor origin and thus cytolytic injury of allogeneic target cells is limited to effector CD8 T cells that bind through direct recognition of the allogeneic MHC (alloMHC) class I (5).

CD8 T cell alloimmune responses are generally dependent upon CD4 T cell help for their development, although there may be exceptions (6). This raises an important, but yet unanswered, question as to how such help is provided, because, due to the complexities of allorecognition, a number of different mechanisms may be envisaged. In contrast to humoral responses where help is delivered directly to the B cell (Figure 1A), models of conventional responses against non-transplant antigen suggest that help for CD8 T cell immunity is delivered to an intermediary APC, which is then licensed to prime CD8 T cells (6-8). The key consideration for such delivery is that both the CD4 and CD8 T cell epitopes are expressed on the same APC, and an analogous three-cell cluster model (Figure 1B) is only possible in transplantation if the APC is of donor origin and CD4 T cell help is provided via direct recognition of MHC class II alloantigen on its surface. Studies incorporating MHC class II deficient recipients that are unable to recognize processed alloantigen have confirmed that robust CD8 T cell cytotoxic responses are generated when CD4 T cell help is restricted to the direct-pathway (9). Nevertheless, Auchincloss has demonstrated unequivocally that CD4 T cell allorecognition exclusively via the indirect-pathway can also provide sufficient help for generating cytotoxic CD8 T cell responses that effect graft rejection (10). How this occurs is not clear, because it implies the formation of a four-cell cluster, comprising CD4 and CD8 T lymphocytes and recipient and donor APC (Figure 1C). Moreover, there is no mechanism in this cluster for physical contact between the donor APC / recipient CD8 T cell couplet and the recipient APC / CD4 T cell couplet, which raises concerns regarding inappropriate and uncontrolled CD8 T cell activation, because similar ‘unlinked’ help could theoretically be provided by concurrent encounter with any unrelated antigen.

Figure 1. Pathways of delivery of CD4 T cell help for cytotoxic and humoral alloimmune responses.

Figure 1

Open in a new tab

(A) Conventional indirect-pathway CD4 T cell help for B cell allorecognition. Alloreactive B cells that recognize MHC alloantigen on donor cells present processed allopeptide in the context of self MHC Class II for cognate recognition by indirect-pathway CD4 T cells. (B) Three-cell cluster model of direct-pathway CD4 T cell help. Direct-pathway CD4 T cells that recognize MHC Class II alloantigen ‘license’ the donor APC to activate direct-pathway CD8 T cells that recognize MHC Class I alloantigen on the same donor APC. (C) Four-cell model of unlinked help. Indirect-pathway CD4 T cells that recognize self-MHC II-restricted allopeptide presented by recipient APCs provide help to direct-pathway CD8 T cells recognizing MHC Class I alloantigen on donor APCs. (D) Proposed three-cell cluster model of provision of linked help by indirect-pathway CD4 T cells. Recipient APC present both intact MHC Class I alloantigen and processed allopeptide for simultaneous recognition by, respectively, direct-pathway CD8 T cells and indirect-pathway CD4 helper T cells.

The growing appreciation that many cell types, including dendritic cells (DCs), are able to capture membrane proteins from neighboring cells (11-14) has led to proposal of an alternative pathway for allorecognition, in which recipient APCs acquire intact MHC class I alloantigen on their surface for subsequent presentation to CD8 and B lymphocytes (15-18). For example, Herrera described capture of host MHC by allogeneic DCs upon i.p. injection (15) and we have reported that DCs pulsed in vitro can re-present soluble MHC for alloantibody generation (16). Similarly, monocytes, when cultured with allogeneic endothelial cells, acquire MHC alloantigen for presentation to T cells (17). However, this alternative pathway of allorecognition has not yet been validated, because demonstration is required that recipient DCs not only present intact alloMHC following challenge with a solid organ allograft (18), but also that this re-presentation primes for humoral and cellular alloimmunity. Moreover, this pathway is teleologically advantageous over conventional allorecognition pathways only if the same DC presents both intact MHC and processed allopeptide, as this provides a mechanism for formation of a three-cell cluster and delivery of ‘linked’ help from indirect-pathway CD4 T cells to direct-pathway cytotoxic CD8 T cells through interaction on a single cell (Figure 1D). Co-presentation of intact and processed alloantigen following murine kidney transplantation has been described recently (19), but the functional significance of such dual presentation in priming cytotoxic alloimmune responses has yet to be addressed.

Here we show that DCs purified following either in vitro culture with allogeneic cells or in vivo challenge with a heart allograft express intact alloMHC and upon adoptive transfer into naive recipients prime for humoral and cellular alloimmunity. Crucially, these responses were CD4 T cell dependent and were abrogated if the transferred DCs were genetically-deficient in MHC class II expression, presumably because this prevented recognition and provision of help by indirect-pathway CD4 T cells. These experiments thus provide the first evidence that acquisition and re-presentation of intact allogeneic MHC by recipient APC can function to provoke cytotoxic alloimmunity, but highlight that simultaneous co-presentation of processed allopeptide by the same APC is critical.

Materials And Methods

Animals

Wild-type (WT) C57BL/6 (H-2b, B6), BALB/c (H-2d), and (BALB/c × B6) F1 mice were purchased from Charles River Laboratories, UK. MHC class II−/− mice B6.129S2-H2dlAb1-Ea/J (H-2b) (20) and BALB/c CD11c-DTR (21) mice were purchased from the Jackson Laboratory (Bar Harbor, ME). TCR-transgenic RAG2−/− TEa mice (H-2b), specific for I-Ab-restricted I-Eα52-68 peptide (22), were gifted by Prof A. Rudensky (University of Washington, Seattle, WA). All animals were maintained in specific-pathogen-free facilities and all experiments approved by the United Kingdom Home Office under the Animal (Scientific Procedures) Act 1986.

Dendritic cell purification and culture

Bone marrow-derived DC (BMDC) were prepared as described previously (16). Briefly, bone marrow (BM) was flushed from femurs and tibias with HBSS (Invitrogen Ltd, UK). Cells were disaggregated by passing through a 40μm mesh, and BM cells cultured in 6-well plates at 3 ×106 per ml in 6 mls complete medium (RPMI 1640, 10% FCS, 100 IU/ml penicillin, 100 μg/ml streptomycin, 2mM L-glutamine (all Invitrogen), supplemented with murine granulocyte-macrophage colony stimulating factor (prepared in house from a B-cell hybridoma) at 20 ng/ml and recombinant murine IL-4 (Peprotech Ltd, UK) at 10 ng/ml. Cells were maintained by replacing half the culture medium with fresh medium on alternate days. Non-adherent cells were discarded on day 4 and DC used on days 7 – 8 for flow cytometric analysis and co-culture.

Splenic DCs were purified by first injecting the recently-excised spleen with digestion buffer (HEPES buffered RPMI + 2% FCS + Collagenase D 1 mg/ml + DNase 0.02 μg/ml, both purchased from Roche, Welwyn Garden City, UK), and incubating for 15 minutes at 37°C, before disruption with sterile forceps and incubation for a further 15 minutes at 37°C. A single-cell suspension was prepared by filtration through a 40 μm nylon cell strainer (Becton Dickinson, Franklin Lakes, NJ) and DCs labeled by incubation at 4°C for 15 minutes with anti-CD11c (N418) coated magnetic microbeads (Miltenyi Biotec, Bergisch, Germany) at 10 μl per 107 cells in MACS buffer (PBS + 0.5% BSA + 2mM EDTA + 0.09% azide). The DCs were extracted by positive selection on an Automacs™ (Miltenyi Biotec) machine according to manufacturer’s instructions.

Co-culture experiments

Single cell suspensions of BALB/c thymocytes were labeled with PKH26 red fluorescent dye (Sigma-Aldrich Co., Germany) according to manufacturer’s instructions, mixed at a ratio of 1:1 with B6 BMDC that had been labeled with PKH67 green fluorescent dye (Sigma-Aldrich) and cultured for 4 hours in complete medium at 2 × 106 cells ml−1. Apoptosis was induced in some experiments by pre-culture of thymocytes with 1mM dexamethasone (VWR, Lutterworth, UK) for 6 hours prior to mixing.

Dendritic cell co-culture was achieved by mixing BALB/c and B6 preparations of either splenic or BM DCs at a ratio of 1:1 and incubating in 75 cm2 ultra-low attachment vented cell culture flasks (Corning, NY) for 20 hours at 37°C in 5% CO2 in complete medium at 2 × 106 cells ml−1. Following co-culture, cells were re-suspended, stained for alloantigen acquisition and assessed by flow cytometric analysis or sorted and adoptively transferred into naïve B6 mice as detailed below. Control DCs and thymocytes were cultured separately for the same period before mixing and immediate labeling and analysis. In certain experiments, splenocytes from BALB/c-DTR mice were co-cultured with WT B6 splenocytes (2 × 106 cells ml−1), but with diphtheria toxin (DT, 12.5 nmol ml−1, List Biological Laboratories, Inc. Epsom, UK) added to culture. B6 DCs were then purified by passage through a dead cell removal kit (Miltenyi Biotec), followed by positive selection with CD11c beads (Miltenyi Biotec) and i.v. transfer to naïve B6 mice which also received a single dose of DT (64 ng g−1 body-weight, i.p.). As positive and negative controls, BALB/c and BALB/c-DTR splenocytes, respectively, were cultured alone with DT for 20 hours, and DCs purified as above then injected into naïve B6 mice.

Flow cytometry and cell sorting

FITC-conjugated anti-mouse CD11c (clone HL3), PE-conjugated anti-mouse TCR Vβ6 (clone RR4-7), biotinylated anti-mouse CD4 (clone GK1.5), FITC-conjugated anti-H-2Kb (clone AF6-88.5), PE-conjugated anti-H-2Dd (clone 34-2-12), FITC-conjugated anti-I-Ab (clone AF6-120.1), and PE-conjugated anti-I-E (clone 14-4-4S) were purchased from BD Pharmingen (San Diego, CA). YAe antibody (anti-I-Ab+I-Eα52-68) (23) was prepared in house from hybridoma supernatants and biotinylated using Biotin Tag Micro Biotinylation Kit (Sigma-Aldrich). Co-cultured DCs, splenic single cell suspensions and peripheral blood (depleted of erythrocytes by incubating with 0.17M NH4Cl red cell lysis buffer) were Fc receptor-blocked with anti-mouse CD16/CD32 (clone 2.4G2, BD Pharmingen, San Diego, CA), before staining with the relevant antibodies and dead cell exclusion dye 7-AAD (BD Pharmingen, San Diego, CA). Biotinylated antibodies were detected by allophycocyanin - conjugated streptavidin (Invitrogen, Paisley, UK) or allophycocyanin -Cy7-conjugated streptavidin (BD Pharmingen, San Diego, CA) and all cells analyzed on a FACSCanto II flow cytometer with FACSDiva software (BD Biosciences, San Jose, CA).

B6 DCs were purified from the co-culture mix following staining for self H-2Kb and donor H-2Dd MHC class I antigen by fluorescence activated cell sorting using either a Mo-Flo high speed cell sorter (DAKO / Beckman Coulter, High Wycombe, UK) or a FACSVantage Diva (Becton Dickinson, San Jose, CA) cell sorter and gating on the predominantly H-2Kb-expressing population. Purity of samples after sorting was typically 89-94%.

Heterotopic cardiac transplantation and DC adoptive transfer

Vascularized cardiac allografts were transplanted intra-abdominally (24). Splenic DCs were purified after transplantation as detailed above and 3 × 106 cells (the approximate total recovered from one recipient) transferred i.v. into naïve B6 mice; in co-culture experiments, 5 × 106 purified B6 DCs were transferred. CD8 T cell ELISPOT was performed 10 days after injection as detailed below. In certain experiments CD4 T cell depletion was achieved (typically > 99%, as confirmed by flow cytometry analysis of peripheral blood lymphocytes) as described previously (25), by i.p injection with two doses of 1.0 mg depleting anti-CD4 mAb (YTS 191.1, hybridoma from the European Collection of Animal Cell Cultures), administered 4 days before and 5 days after adoptive transfer of DCs.

CD8 T cell IFN-γ ELISPOT and alloantibody analysis

CD8 T cell ELISPOT was performed as described (26). Briefly, purified CD8 T cells were mixed with irradiated BALB/c stimulator splenocytes, and added to Multiscreen™ HTS filtration system plates (Millipore Corporation, Billerica, MA) that had been coated with anti-mouse IFN-γ (BD Pharmingen, Franklin Lakes, NJ) in carbonate-bicarbonate buffer. Plates were incubated at 37°C and 5% CO2 for 20 hours and after washing, spots were detected with biotinylated rat anti-mouse IFN-γ (BD Pharmingen). Plates were read (Autoimmun Diagnostika GmbH, Straβberg, Germany), and data expressed as spot counts per 106 responder CD8 T cells for each well.

Alloantibody production against H-2d alloantigen was assessed by flow cytometric quantification of binding of test sera to target BALB/c thymocytes, as recently described (26). Briefly, after non-specific antibody binding was blocked with anti-CD16/CD32 FcγR block (BD Pharmingen), serial tripling dilutions of heat-inactivated test sera were added and bound alloantibody was detected with FITC-conjugated antibodies directed against pan IgG (STAR70, Serotec, Oxford, UK) using FACSCalibur™ or FACSCanto II™ flow cytometers with CellQuest or FACSDiva software (all BD Biosciences, San Jose, CA, USA). Alloantibody was quantified by calculating the area under the titration curve and expressed as percentage of positive control (pooled immune serum from B6 recipients of BALB/c heart transplants).

CD4 T Cell Proliferation

Single cell suspensions of splenocytes obtained from TEa mice were stained with 5mM CFSE (Invitrogen, Molecular Probes, Paisley, UK) in the dark for 5mins and then quenched with PBS + 5% FCS. 2-5×106 CFSE stained splenocytes were injected intravenously, spleens harvested 10 days later and flow cytometry performed using allophycocyanin-conjugated anti-CD4 plus PE-conjugated anti-TCR Vβ6 to identify TEa cells.

Statistical Analysis

Data were presented as mean ± SD where appropriate. Statistical analyses were performed using GraphPad prism (GraphPad Software Inc., San Diego, CA). The normality of data sets was assessed using the D’Agostino-Pearson omnibus test and by observing their frequency-distribution pattern. Non-parametric data were analyzed using either the Mann Whitney U test or the Kruskal-Wallis test with Dunn’s multiple comparison test where appropriate. Paired non-parametric data were analyzed using the Wilcoxon matched pairs signed rank test.

Results

In vitro acquisition of membrane proteins

The potential for recipient DCs to prime direct-pathway allo-cytotoxic CD8 T cell responses (Figure 1) was first assessed by testing the ability of DCs to capture membrane proteins from allogeneic cells, using, initially, in vitro transfer of fluorescent membrane dyes as a marker of acquisition. Compared to control cultures in which green-labeled B6 (H-2b) DCs were mixed with red-labeled, BALB/c (H-2d) thymocytes and then analyzed immediately by flow cytometry, four hours of co-culture resulted in capture of appreciable quantities of fluorescent dye by the DCs (Figure 2A). Acquisition was greater if thymocyte apoptosis was induced by 6 hour culture with dexamethasone before co-culture with the DCs (Figure 2A).

Figure 2. Acquisition of membrane proteins from allogeneic cells.

Figure 2

Open in a new tab

(A) Acquisition of thymocyte membrane by BMDCs in co-culture. B6 BMDCs and BALB/c thymocytes were labeled with green and red membrane dyes respectively and co-cultured for 4 hours. Flow cytometric plots of gated B6 BMDCs are shown (i) immediately before co-culture, (ii) 4 hours after co-culture with red labeled, untreated BALB/c thymocytes and (iii) 4 hours after co-culture with red labeled, dexamethasone-treated BALB/c thymocytes. Percentage of double staining DCs is indicated in right upper quadrant. (iv) Representative fluorescence microscopy image (× 400 magnification) of membrane transfer after 20 hours co-culture of B6 and BALB/c BMDCs, labeled with green and red membrane dyes, respectively. (B). Representative flow cytometric plots of B6 (H-2b) and BALB/c (H-2d) BMDCs, double labeled with (i) anti-MHC I (Kb-FITC & Dd-PE) or (ii) anti-MHC II (I-Ab-FITC & I-Ed-PE) antibodies either immediately after mixing (control, upper plot) or 20 hours after mixing and co-culture (lower plot). Number shown is the percentage of B6 BMDCs staining positive for BALB/c MHC (mean ± SD from eight experiments). Representative overlay histograms showexpression of acquired MHC I Dd or MHC II I-Ed by B6 DCs gated from BMDC co-culture (green) compared with control (red). (iii) Corresponding histogram for DCs labeled with isotype control antibody (IC) in place of Dd-PE. (C) (i) and (ii) As for (B), but with co-culture of splenic B6 and BALB/c DCs. The additional blue line on the histogram demonstrates absence of acquisition of alloantigen by B6 DCs following culture with the two DC populations separated by a semi-permeable membrane (transwell). Representative of at least three separate experiments.

Flow cytometric demonstration of acquisition of fluorescent membrane does not distinguish expression of the captured membrane on the cell surface from simple phagocytotic entry into the intracellular degradative pathways. To analyze cell surface expression of MHC alloantigen, H-2d BMDC and LPS-matured H-2b BMDC were co-cultured for 20 hours and stained with monoclonal antibodies against H-2Kb and H-2Dd MHC class I proteins. Compared to control DCs that were mixed and stained immediately, co-cultured H-2b DCs expressed significant amounts of H-2Dd on their surface (mean geometric fluorescence of 8 experiments; 134.5 vs. 251.6, p<0.05, Figure 2B). There was no difference in staining with isotype control antibody between freshly mixed DCs and DCs cultured for 20 hours (Figure 2B), excluding non-specific ‘stickiness’ due to DC activation while in culture as explanation for the increased staining with anti-MHC class I antibody. B6 DCs also acquired intact MHC class II I-E alloantigen from the H-2d DCs (Figure 2B). Because, uncertainties persist as to how representative BMDC are of naturally-occurring DC subsets, the experiments were repeated using freshly-isolated splenic DCs, with similar results (Figure 2C). MHC transfer did not occur when the two DC populations were separated in transwell culture (Figure 2C), demonstrating that MHC acquisition requires cell contact and is not due, for example, to capture of exosomes released from allogeneic DCs (27).

Simultaneous presentation of processed alloantigen

Given their main role as scavenging cells for antigen presentation, DCs would generally be expected to direct acquired exogenous antigen (irrespective of how it is captured) to the MHC class II processing pathway for presentation as processed peptide. To analyze indirect presentation of alloMHC as allopeptide, co-cultured DCs were stained with YAe mAb, an antibody that recognizes dominant I-Eα peptide presented in the context of MHC class II I-Ab antigen (23). As shown in Figure 3A, YAe mAb bound consistently more strongly to co-cultured DCs than to control DCs (mean channel fluorescence of 6 experiments, 149.0 vs. 328.2 p<0.05). The percentage that stained positive was difficult to determine, because, as reported previously (28), there was an element of background staining of naïve B6 DCs. Nevertheless, a consistent and significantly greater number of co-cultured DCs (9.2 ± 6.74%) stained above background observed in the control culture. As discussed above, presentation of intact alloMHC by recipient APC is advantageous over the four-cell-cluster model only if processed alloantigen is presented simultaneously in the context of self-MHC class II, as this permits formation of a three-cell cluster in which indirect-pathway CD4 and direct-pathway CD8 T cells interact on the same cell. Notably, 12.6 ± 3.3% of the B6 DCs that expressed processed I-Eα allopeptide after co-culture with allogeneic DCs also expressed intact MHC alloantigen on their surface (Figure 3B). The relatively small proportion of B6 DCs that expressed both unprocessed and processed alloantigen (~1.5%) suggests that re-presentation of intact antigen is either transient or confined to a select subset of DCs; concerning this, expression of intact MHC class I alloantigen was greater in those DCs that also expressed processed allopeptide, although this difference was not statistically significant (Figure 3B).

Figure 3. Simultaneous presentation of intact and processed alloantigen by recipient DCs.

Figure 3

Open in a new tab

B6 and BALB/c BMDCs were mixed and, after 20 hours co-culture, labeled with anti-MHC class I antibodies (Kb-FITC &Dd-PE) and YAe mAb (an antibody that recognizes dominant I-Eα peptide presented in the context of I-Ab MHC class II antigen). (A) Flow cytometric plots represent YAe expression on gated B6 DC population after 20 hours co-culture (bottom panel). As control, DCs were labeled immediately after mixing (top panel). Number shown is the percentage increase in YAe positive B6 BMDCs above background control levels (mean ± SD, from six experiments). Also depicted is an overlay flow cytometric histogram of YAe staining, gating on the B6 DC population after co-culture (green line) or stained immediately after mixing (red line). Isotype control stained B6 DCs, post co-culture or immediately after mixing, are represented by the dotted brown and blue lines respectively.(B) Representative flow cytometric plots of H-2Dd expression by B6 DCs according to high (pink) or low (blue) staining for YAe antibody. Values indicate the percentage H-2Dd population (mean ± SD) from six different experiments.

Presentation of acquired intact alloantigen provokes cellular and humoral alloimmune responses

The ability of DCs that have acquired intact alloantigen to prime humoral and cellular alloimmune responses in vivo was tested by sorting B6 BMDCs from the co-culture and assessing alloantibody and cytotoxic CD8 T cell responses 10 days after adoptive transfer into naïve B6 recipients. Representative sort profiles are shown in Figure 4A. Adoptive transfer of sorted B6 DCs provoked modest CD8 T cell responses (assayed by IFN-γ ELISPOT assay), weaker than in mice that received BALB/c DCs, but nevertheless significantly higher than naive controls (Figure 4B) or in mice that received naive B6 DC (not shown). It should be noted that the ELISPOT assay only detects direct-pathway CD8 T cell alloresponses against intact alloantigen, and not self-class I-restricted responses against processed alloantigen, because B6 APC were not present in the ELISPOT culture. Mice were sacrificed at day 10 for assessment of CD8 T allo-cytotoxicity and longer term assay of humoral alloimmunity was not possible, but nevertheless, small but detectable IgG alloantibody responses had also developed (Figure 4C).

Figure 4. Presentation of intact alloantigen by syngeneic DCs primes cellular and humoral alloimmunity.

Figure 4

Open in a new tab

B6 BMDCs were purified following co-culture with BALB/c BMDCs by fluorescence activated cell sorting, using the gating strategy as illustrated in flow cytometry plot (A) and 5 × 106 sorted DCs injected into wild-type (WT) B6, or MHC class II deficient (MHC II−/−), or CD4 T cell-depleted B6 mice. Cytotoxic CD8 T cell (B) and humoral (C) alloimmune responses were quantified ten days later by IFN-γ ELISPOT assay against target BALB/c splenocytes and flow cytometric assay of binding of test sera to BALB/c thymocytes, respectively. Data presented represent the values for each individual mouse with mean values indicated. Also shown for comparison are values obtained from naïve mice and following injection of 5 × 106 BALB/c BMDCs. * P< 0.05, Kruskal-Wallis, Dunn’s multiple comparison test.

One of the main concerns with this experiment is that the cytotoxic CD8 T cell response observed may not be due to recognition of MHC class I alloantigen on the surface of recipient APC, but instead to recognition of small numbers of contaminating BALB/c DCs; in this regard, flow cytometric analysis of the DCs after sorting confirmed that approximately 5% were now outside the initial selection gate, albeit these were overwhelmingly B6 DCs adjacent to the periphery (not shown). As a control to exclude contamination with BALB/c DC, purified B6 DCs that had been mixed with BALB/c DCs immediately prior to sorting were thought inappropriate, because of the potential for significant cross-over of membrane protein during the sorting process (which typically last several hours). Two alternative approaches were adopted. Firstly, following co-culture with BALB/c DCs, B6 DCs were sorted into two populations according to high or low expression of acquired H-2Dd antigen (Figure 5A), and transferred into naive B6 mice. Despite the expectation that the same potential for contamination with BALB/c DCs would apply to both groups, cytotoxic CD8 T cell responses were lower in the H-2Dd-low group (Figure 5B), although this difference did not reach statistical significance. The confounding effect of contaminant BALB/c DCs was next investigated using BALB/c-CD11c-DTR mice, in which treatment with DT leads to specific depletion of BALB/c DCs. After co-culture of WT B6 and BALB/c-CD11c-DTR splenocytes in the presence of DT, transfer of the surviving DCs (after dead cell exclusion and CD11c-bead selection) provoked appreciable cytotoxic CD8 T cell responses in naïve B6 mice (Figure 5C). Notably, transfer of DCs selected from BALB/c-CD11c-DTR splenocytes that were cultured alone with DT did not elicit cytotoxicity (Figure 5C). These two experiments thus provide further evidence that CD8 T cell recognition upon DC transfer following in vitro co-culture is dependent on expression of acquired H-2Dd alloantigen by recipient DCs, rather than contamination of the sorted cells with BALB/c DCs.

Figure 5. Cytotoxic CD8 T cell alloimmunity is not due to contaminating BALB/c DCs.

Figure 5

Open in a new tab

Following co-culture as described in Figure 4, B6 DCs were purified from BALB/c DCs by fluorescence activated cell sorting, using the gating strategy as illustrated in flow cytometry plot (A) to isolate a population of ‘H-2Dd high’ and a population of ‘H-2Dd low’ B6 DCs, and 5 × 106 sorted DCs were injected into WT B6 mice. Cytotoxic CD8 T cell alloimmune responses were assayed 10 days later (B); *P = 0.2, Mann Whitney U test. (C) Splenocytes from BALB/c-DTR mice were co-cultured with WT B6 splenocytes for 20 hours, with diphtheria toxin added to culture. The B6 DCs, purified by dead cell removal then positive bead selection, were then injected into WT B6 mice and cytotoxic CD8T cell alloimmunity assayed 10 days later (BALB/c-DTR + B6). As positive and negative controls, BALB/c and BALB/c-DTR splenocytes, respectively, were cultured for 20 hours with DT added, and DCs then purified and injected into naïve B6 mice. Data represent the values for each individual mouse with mean values indicated. * P<0.05; two-tailed Mann Whitney U test.

CD4 T cell- independent CD8 T cell alloimmune responses have been described (29). Nevertheless, the humoral and cellular responses provoked by injection with co-cultured B6 DCs are CD4 T cell dependent, as they are abrogated by administration of depleting anti-CD4 mAb (Figures 4B and 4C). Theoretically, the CD4 T cell help could be delivered in three ways: through direct allorecognition of acquired MHC class II alloantigen (I-Ed or I-Ad) on the surface of the co-cultured DCs; through processing of the acquired MHC class I alloantigen by the co-cultured DCs for indirect presentation; and by acquisition and processing of the intact alloantigen from the co-cultured DCs by host DCs after transfer. As perhaps expected (given the essential requirement for indirect-pathway help in generating alloantibody (30, 31)), injected co-cultured DCs provoked CD4 T cell responses against allopeptide (Figure 6A), as demonstrated by division of adoptively-transferred TCR-transgenic TEa T cells (which recognize self (I-Ab) restricted, dominant I-Eα allopeptide via the indirect-pathway (22)). To distinguish whether indirect-pathway help is due to recognition of alloantigen on the co-cultured DCs or to transfer of intact alloantigen, after injection, to APC in the host, the above experiments were repeated using BMDC cultured from genetically modified B6 mice that lacked MHC class II expression (MHCII−/−). Upon co-culture, MHCII−/− DCs acquired similar amounts of intact MHC class I alloantigen as WT B6 DCs (Figure 6B), but their adoptive transfer resulted in barely detectable humoral and cytotoxic alloimmune responses (Figures 4B and 4C). Notably, despite the presence of intact alloantigen on their surface, co-cultured MHCII−/− DCs provoked minimal or no division of TEa T cells (Figure 6A).

Figure 6. MHC class II deficient DCs acquire intact alloantigen, but do not stimulate indirect-pathway CD4 T cell responses.

Figure 6

Open in a new tab

Wild-type and MHC II−/− B6 DCs were cultured for 20 hours with BALB/c DCs, as described in legend to Figure 4, sorted by flow cytometry and injected into naïve B6 mice. (A) Indirect-pathway CD4 T cell responses were assessed by simultaneous injection of CFSE-labeled TEa CD4 T cells (that recognize self (I-Ab) restricted, I-Ed allopeptide) and division analyzed by flow cytometry 10 days later. Although TEa division was not observed, MHC II−/− DCs did acquire and re-present intact MHC class I alloantigen upon culture with BALB/c DCs, as depicted in (B) by overlay flow histogram, comparing H-2Dd expression on B6 MHC II−/− DCs following standard co-culture (solid line) to control culture stained immediately after mixing (dashed line). Representative flow cytometry plots are also shown (top panel - standard co-culture: bottom panel – control). Number shown is the percentage of MHC II−/− B6 BMDCs staining positive for BALB/c MHC (mean ± SD from four experiments).

The experiments using MHCII−/− DCs exclude direct-pathway CD4 T cell allorecognition of captured allogeneic MHC class II as the source of help for humoral and cellular alloimmunity and also demonstrate that presentation of processed alloantigen by the injected DCs is essential. They are also a particularly convincing control to eliminate carryover of BALB/c DCs as the source of Kd and Dd antigen for alloimmune recognition, because the same potential for contamination with allogeneic DCs is present with the co-cultured MHCII−/− DCs.

DC acquisition of intact alloantigen following heart transplantation

To analyze whether acquisition of alloantigen by host DC could contribute to alloimmunity following solid organ transplantation, splenic CD11c+ve DCs were purified from B6 recipients, on days one to five after transplantation with BALB/c hearts. No BALB/c DCs were recovered, and it was difficult to determine whether MHC class I Kd and Dd alloantigen was expressed on the surface of purified B6 DCs, because if present, levels of staining were barely above background (Figure 7A). Similarly, processed I-E alloantigen could not be detected (Figure 7A). Nevertheless, DCs purified day 5 after transplantation provoked, when transferred to naive B6 mice, indirect-pathway CD4 T cell activation (Figure 7B) and CD4 T cell-dependent CD8 T cell alloimmunity (Figure 7C), indicating effective capture of alloantigen by recipient splenic DCs. Interestingly, alloantibody was not generated (not shown). The necessity for co-presentation of intact and processed alloantigen by the transferred DCs in priming CD8 T cell allo-responses was examined in similar fashion to the in vitro co-culture experiments, by adoptive transfer of CD11c+ve DCs, purified from MHCII−/− recipients the fifth day after challenge with BALB/c hearts. No CD8 T cell responses developed (Figure 7C) and only minimal division of indirect-pathway CD4 T cells was observed (Figure 7B). Thus the findings from the in vivo transplant model replicate the in vitro co-culture experiments in demonstrating that the CD8 T cell alloimmune response is not due to contaminating BALB/c DCs; instead simultaneous presentation of intact and processed alloantigen by the transferred H-2b DC is essential for, respectively, cytotoxic CD8 T cell and indirect-pathway helper CD4 T cell recognition.

Figure 7. Presentation of intact MHC alloantigen by host DCs after heart transplantation primes CD8 T cell alloimmunity.

Figure 7

Open in a new tab

(A) Splenocyte DCs were purified by positive bead selection from B6 recipients on days 1 to 5 after BALB/c heart transplantation and analyzed by flow cytometry for expression of intact donor H-2Dd and I-Ab-bound, I-Eα allopeptide (positive YAe antibody staining). Depicted are overlay histograms, gating on the live DC population. (B) Splenocyte DCs were purified from WT or MHC II−/− B6 recipients five days after BALB/c heart transplantation, and transferred to naïve B6 recipients. Indirect-pathway CD4 T cell responses were assessed by analyzing division of simultaneously-transferred TEa CD4 T cells, as described in Figure 6 legend. As control, CFSE-labeled TEa CD4 T cells were transferred into naïve B6 mice. (C) Corresponding CD8 T cell alloimmune responses, 10 days after transfer of splenocyte DCs, purified as in (B), from WT or MHC II−/− B6 recipients of BALB/c heart grafts. Also depicted is the response observed in CD4 T cell-depleted recipients of DCs purified from WT recipients of a BALB/c heart graft (CD4 depletion). Data presented represent the values for each individual mouse with mean values indicated. * P = 0.01 and † P = 0.05; two-tailed Mann Whitney U test.

Discussion

Although the transfer of membrane proteins between different cell types has been appreciated for several decades (reviewed in (32)), the functional relevance has remained largely unclear, and it is only now becoming apparent that membrane sharing, or ‘trogocytosis’ (33), may confer important immunomodulation. For example, co-stimulatory ligands on APC may continue to trigger pro-survival kinase signaling after antigen-specific acquisition by the responding T cell (34); alternatively, capture of inhibitory ligands such as HLA-G may endow regulatory potential (35). Trogocytosis may play a particularly important role in organ transplantation, as it may solve the long-standing conundrum of how indirect-pathway CD4 T cells that recognize processed alloantigen on recipient APC can provide help to alloreactive cytotoxic CD8 T cells that recognize intact MHC class I alloantigen on donor APC (10), a process that apparently breaks a cardinal requirement for epitope linkage. The key attribute imparted by membrane exchange is the potential for co-presentation of MHC class I alloantigen as both intact and processed alloantigen on the surface of the same recipient APC, thus allowing recognition of self-restricted allopeptide by indirect-pathway CD4 T cells to ‘license’ the APC (6-8) for effective activation of cytotoxic CD8 T cells responding to the intact class I alloantigen on the surface. However, despite evidence of intact alloantigen transfer onto recipient cells following transplantation (18), with simultaneous allopeptide presentation (19), the so-called ‘semi-direct’ pathway (3) remains largely speculative. Our demonstration that in vivo or in vitro acquisition of alloantigen enables recipient DCs to provoke CD8 T cell alloimmunity - a response that, critically, is dependent upon self MHC class II expression by the DC - provides the first functional evidence of a role for membrane exchange in driving destructive cytotoxic responses.

The mechanisms by which DCs retain antigen in unprocessed form have still to be clarified, and were not addressed in our experiments. DCs contain lysosomes with limited protease activity, in which antigen can persist in an un-degraded form (36) and it has been further suggested that internalization through the inhibitory Fc receptor FcγRIIb may target antigen to a non-degradative compartment that permits subsequent recycling of intact antigen to the cell surface (37). This latter pathway is probably not relevant to our experimental systems, because Ig-switched alloantibody was not present in the in vitro co-culture experiments and was unlikely to have developed at the early time-point that DCs were purified from recipients of heart transplants. Whatever the mechanism for redelivery of intact alloantigen to the cell surface, key, conceptually, to the semi-direct pathway is simultaneous presentation of allopeptide for indirect-pathway helper CD4 T cell recognition, and in which case, it is important that some of the acquired alloantigen is diverted into conventional degradative compartments for routine processing. Ideally, our flow-cytometry staining experiments should have stained for processed MHC-class I allopeptide, using an antibody that recognizes the same conformational MHC class II I-Ab/peptide complex as is recognized, for example, by TCR75 TCR transgenic CD4 T cells (38), but this is not available and instead YAe antibody, specific for the I-Ab/I-Eα allopeptide, was used as surrogate, in the expectation that MHC class II and class I alloantigens are acquired and processed similarly.

One of the main concerns of our experimental system was the potential for carry-over of donor cells when purifying recipient DCs either from recipients of heart allografts or following in vitro culture, because even small numbers of donor DCs would be expected to provoke strong cytotoxic CD8 T cell responses. Two main approaches were used to control for confounding contamination. The incorporation of CD11c.DTR mice enabled selective and effective ablation of donor DCs, such that the normally robust cytotoxic response elicited by transfer of donor DCs was abrogated by initial in vitro culture with DT. Hence the restoration of cytotoxicity by co-culture with recipient DCs strongly suggests that this population is solely responsible in this experiment for presenting intact MHC class I alloantigen to the alloreactive CD8 T cell subset. It is perhaps surprising that the recipient DCs did not also acquire the DT receptor from the surface of donor DCs, which would be expected to limit cytotoxicity by invoking vulnerability to DT; certainly there is no a priori reason why MHC class I alloantigen should be acquired any more effectively than the DT receptor. Most likely, DC trogocytosis is limited to only a small number of membrane proteins restricted to the vicinity of membrane fusion with the donor cell, and consequently the recipient DCs acquired generally either the DT receptor or the MHC class I alloantigen, but not both. Alternatively, perhaps only small amounts of class I alloantigen need be acquired for effective presentation to CD8 T lymphocytes, whereas much greater quantities of DT have to be accumulated to raise susceptibility to DT.

The second approach to control for contamination by donor DC incorporated MHC class II deficient mice, and the abrogation of cytotoxic alloimmunity upon transfer of MHC class II-deficient B6 DCs either from in vitro co-culture or from heart-grafted recipients further supports that CD8 T cell allorecognition is centered on recipient and not donor DCs. The experiments utilizing MHC class II deficient mice are, however, much more integral to the interpretation of our results, because they provide the key evidence for the necessity of co-presentation of intact and processed alloantigen in the delivery of effective help by indirect-pathway CD4 T cells to alloreactive CD8 T cells. Without these experiments, the cytotoxic CD8 T cell responses provoked by challenge with WT B6 DC that had acquired intact alloantigen could potentially be explained by transfer of the alloantigen in the second host onto fresh DCs, which then function as the main subset for presenting processed alloantigen to CD4 T cells with indirect allospecificity. In this regard, it is notable, and surprising, that MHC class II-deficient DCs that had acquired intact MHC class I alloantigen upon co-culture did not provoke proliferation of indirect-pathway CD4 T cells upon transfer into WT hosts (Figure 6); presumably the amount of intact alloantigen captured and expressed by the DCs is not sufficient to enable transfer, processing and effective allopeptide presentation in the second host. This could be further addressed by transfer of co-cultured WT DCs into MHC Class II deficient mice, in the expectation that this would provoke proliferation of indirect-pathway CD4 T cells, but was not performed as it is not clear whether transferred T cells would respond normally in a T cell deficient, and essentially MHC class II defective, environment (39).

The antibody-depletion studies confirmed that the humoral and CD8 T cellular alloimmune responses to challenge with co-cultured DCs were dependent upon help from the recipient CD4 T cell population, but despite the ability of DCs to also acquire intact MHC class II alloantigen, the experiments incorporating MHC class II-deficient DCs further highlight that this acquisition is not sufficient to drive help from direct-pathway CD4 T cells; that instead recognition of processed, self-restricted alloantigen by indirect-pathway helper T cells is essential. Although not a surprising finding for humoral alloimmunity (30, 31), direct-pathway recognition of MHC class II alloantigen on the surface of donor DCs is the principal pathway after transplantation for generating CD8 T cell cytotoxicity (9), and it is therefore not clear why co-presentation of intact MHC class I and II alloantigen does not facilitate the recipient DC to act essentially as a surrogate donor DC (Figure 1B). Estimation of the degree of co-presentation of MHC class I and II alloantigens on the surface of recipient DCs was not straightforward, because of difficulty in setting a positive threshold for what, particularly for MHC class II alloantigen, represents a continuum of varying degrees of expression, but the most likely explanation is, as discussed above, that acquisition of alloantigen was limited generally to only a single alloantigen. In any event, the value of the semi-direct pathway does not lie in dual presentation of MHC class I and II alloantigen, but as a means of explaining the ability of indirect-pathway CD4 T cells to provide help for CD8 T cell alloimmunity. This mode of help is probably not relevant until late time points after transplantation, when dominant direct-pathway helper responses can no longer be triggered due to loss of the donor DC population. Consequently, it is important to note that our DC transfer experiments from recipients of heart allografts only indicate the ability of recipient DCs to present intact alloantigen early after transplantation, at a time when donor DCs are presumably present in abundance; they do not provide definitive evidence that the semi-direct pathway is later capable of acting alone in triggering cytotoxic CD8 T cell responses.

In this regard, it is questionable that an approach involving adoptive transfer of DCs from recipients of heart grafts would be applicable for studying late semi-direct pathway responses. The most striking inconsistency in our experimental findings was the inability to detect either intact or processed alloantigen on the surface of recipient DCs that were purified from recipient mice in the first few days after transplantation, and yet their transfer into naïve secondary hosts elicited modest CD8 T cell cytotoxicity. The observation that the CD8 T cell response was abrogated by transfer of DCs purified from MHC class II deficient recipients strongly suggests that this inconsistency simply reflects levels of expression of acquired alloantigen that were below threshold for flow cytometric detection, but nevertheless, alloantigen availability is likely to be too limited at late time points after transplantation for transfer of DCs to elicit a cytotoxic response.

Definitive demonstration of the functional relevance of the semi-direct pathway perhaps lies in resolving the paradox created by the seminal observations that CD8 T cells can effect allograft rejection when target MHC class I alloantigen is expressed by only the parenchymal components of the graft (40), but that conventional lymphoid organs are nevertheless essential for rejection of vascularised allografts (41). Why lymphoid tissue should be required for CD8 T cell allorecognition of graft parenchymal cells remains unexplained, but one hypothesis that offers an elegant and compelling solution is that alloreactive CD8 T cell activation occurs in the regional lymphoid tissue due to recognition of shed, intact MHC class I alloantigen that is presented by recipient APC.

In summary, although transfer of alloantigen from donor cells following transplantation has been demonstrated previously, whether this is an effective mechanism for triggering CD8 T cell alloimmunity has remained unproven. Our demonstration that cellular cytotoxicity is triggered by transfer of recipient DCs that have acquired MHC class I alloantigen identifies a functional relevance to trogocytotic exchange. The requirement for host MHC class II expression on the DCs, moreover, suggests a mechanism whereby indirect-pathway CD4 T cells provide help through simultaneous recognition of processed allopeptide on the surface of the same cell and provides strong support for linked epitope recognition via the semi-direct-pathway.

Acknowledgments

Grant Support

This work was supported by a British Heart Foundation project grant and by the National Institute for Health Research Cambridge Biomedical Research Centre. S.S. was supported by the Commonwealth Scholarship Commission. S.J.H. and K.S.-P. were supported by Wellcome Trust/Academy of Medical Sciences starter grants. R.M. was supported by a clinical research training fellowship from The Wellcome Trust and a Raymond and Beverly Sackler Scholarship.

Non-Standard Abbreviations

alloMHC

allogeneic MHC

BM

bone marrow

BMDCs

bone-marrow dendritic cells

B6

C57BL/6

DCs

dendritic cells

DT

diphtheria toxin

WT

wild-type

References

  • 1.Gould DS, Auchincloss H., Jr. Direct and indirect recognition: the role of MHC antigens in graft rejection. Immunol Today. 1999;20:77–82. doi: 10.1016/s0167-5699(98)01394-2. [DOI] [PubMed] [Google Scholar]
  • 2.Benichou G. Direct and indirect antigen recognition: the pathways to allograft immune rejection. Front Biosci. 1999;4:D476–480. doi: 10.2741/benichou. [DOI] [PubMed] [Google Scholar]
  • 3.Jiang S, Herrera O, Lechler RI. New spectrum of allorecognition pathways: implications for graft rejection and transplantation tolerance. Curr Opin Immunol. 2004;16:550–557. doi: 10.1016/j.coi.2004.07.011. [DOI] [PubMed] [Google Scholar]
  • 4.Kapessidou Y, Habran C, Buonocore S, Flamand V, Barvais L, Goldman M, Braun MY. The replacement of graft endothelium by recipient-type cells conditions allograft rejection mediated by indirect pathway CD4+ T cells. Transplantation. 2006;82:582–591. [PubMed] [Google Scholar]
  • 5.Valujskikh A, Zhang Q, Heeger PS. CD8 T cells specific for a donor-derived, self-restricted transplant antigen are nonpathogenic bystanders after vascularized heart transplantation in mice. J Immunol. 2006;176:2190–2196. doi: 10.4049/jimmunol.176.4.2190. [DOI] [PubMed] [Google Scholar]
  • 6.Ridge JP, Di Rosa F, Matzinger P. A conditioned dendritic cell can be a temporal bridge between a CD4+ T-helper and a T-killer cell. Nature. 1998;393:474–478. doi: 10.1038/30989. [see comments] [DOI] [PubMed] [Google Scholar]
  • 7.Bennett SR, Carbone FR, Karamalis F, Flavell RA, Miller JF, Heath WR. Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature. 1998;393:478–480. doi: 10.1038/30996. [see comments] [DOI] [PubMed] [Google Scholar]
  • 8.Schoenberger SP, Toes RE, van der Voort EI, Offringa R, Melief CJ. T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions. Nature. 1998;393:480–483. doi: 10.1038/31002. [see comments] [DOI] [PubMed] [Google Scholar]
  • 9.Taylor AL, Negus SL, Negus M, Bolton EM, Bradley JA, Pettigrew GJ. Pathways of helper CD4 T cell allorecognition in generating alloantibody and CD8 T cell alloimmunity. Transplantation. 2007;83:931–937. doi: 10.1097/01.tp.0000257960.07783.e3. [DOI] [PubMed] [Google Scholar]
  • 10.Lee RS, Grusby MJ, Glimcher LH, Winn HJ, Auchincloss H., Jr. Indirect recognition by helper cells can induce donor-specific cytotoxic T lymphocytes in vivo. J Exp Med. 1994;179:865–872. doi: 10.1084/jem.179.3.865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Batista FD, Iber D, Neuberger MS. B cells acquire antigen from target cells after synapse formation. Nature. 2001;411:489–494. doi: 10.1038/35078099. [DOI] [PubMed] [Google Scholar]
  • 12.Russo V, Zhou D, Sartirana C, Rovere P, Villa A, Rossini S, Traversari C, Bordignon C. Acquisition of intact allogeneic human leukocyte antigen molecules by human dendritic cells. Blood. 2000;95:3473–3477. [PubMed] [Google Scholar]
  • 13.Huang JF, Yang Y, Sepulveda H, Shi W, Hwang I, Peterson PA, Jackson MR, Sprent J, Cai Z. TCR-Mediated internalization of peptide-MHC complexes acquired by T cells. Science. 1999;286:952–954. doi: 10.1126/science.286.5441.952. [DOI] [PubMed] [Google Scholar]
  • 14.Colino J, Shen Y, Snapper CM. Dendritic cells pulsed with intact Streptococcus pneumoniae elicit both protein- and polysaccharide-specific immunoglobulin isotype responses in vivo through distinct mechanisms. J Exp Med. 2002;195:1–13. doi: 10.1084/jem.20011432. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Herrera OB, Golshayan D, Tibbott R, Salcido Ochoa F, James MJ, Marelli-Berg FM, Lechler RI. A novel pathway of alloantigen presentation by dendritic cells. J Immunol. 2004;173:4828–4837. doi: 10.4049/jimmunol.173.8.4828. [DOI] [PubMed] [Google Scholar]
  • 16.Curry AJ, Pettigrew GJ, Negus MC, Easterfield AJ, Young JL, Bolton EM, Bradley JA. Dendritic cells internalise and re-present conformationally intact soluble MHC class I alloantigen for generation of alloantibody. Eur J Immunol. 2007;37:696–705. doi: 10.1002/eji.200636543. [DOI] [PubMed] [Google Scholar]
  • 17.Xu H, Dhanireddy KK, Kirk AD. Human monocytes as intermediaries between allogeneic endothelial cells and allospecific T cells: a role for direct scavenger receptor-mediated endothelial membrane uptake in the initiation of alloimmunity. J Immunol. 2006;176:750–761. doi: 10.4049/jimmunol.176.2.750. [DOI] [PubMed] [Google Scholar]
  • 18.Brown K, Sacks SH, Wong W. Extensive and bidirectional transfer of major histocompatibility complex class II molecules between donor and recipient cells in vivo following solid organ transplantation. Faseb J. 2008;22:3776–3784. doi: 10.1096/fj.08-107441. [DOI] [PubMed] [Google Scholar]
  • 19.Brown K, Sacks SH, Wong W. Coexpression of donor peptide/recipient MHC complex and intact donor MHC: evidence for a link between the direct and indirect pathways. Am J Transplant. 2011;11:826–831. doi: 10.1111/j.1600-6143.2011.03437.x. [DOI] [PubMed] [Google Scholar]
  • 20.Madsen L, Labrecque N, Engberg J, Dierich A, Svejgaard A, Benoist C, Mathis D, Fugger L. Mice lacking all conventional MHC class II genes. Proc Natl Acad Sci U S A. 1999;96:10338–10343. doi: 10.1073/pnas.96.18.10338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Jung S, Unutmaz D, Wong P, Sano G, De los Santos K, Sparwasser T, Wu S, Vuthoori S, Ko K, Zavala F, Pamer EG, Littman DR, Lang RA. In vivo depletion of CD11c(+) dendritic cells abrogates priming of CD8(+) T cells by exogenous cell-associated antigens. Immunity. 2002;17:211–220. doi: 10.1016/s1074-7613(02)00365-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Grubin CE, Kovats S, deRoos P, Rudensky AY. Deficient positive selection of CD4 T cells in mice displaying altered repertoires of MHC class II-bound self-peptides. Immunity. 1997;7:197–208. doi: 10.1016/s1074-7613(00)80523-3. [DOI] [PubMed] [Google Scholar]
  • 23.Murphy DB, Lo D, Rath S, Brinster RL, Flavell RA, Slanetz A, Janeway CA., Jr. A novel MHC class II epitope expressed in thymic medulla but not cortex. Nature. 1989;338:765–768. doi: 10.1038/338765a0. [DOI] [PubMed] [Google Scholar]
  • 24.Ono K, Lindsey K. Improved technique of heart transplantation. J. Thorac. Cardiovasc. Surg. 1969;57:225–228. [PubMed] [Google Scholar]
  • 25.Motallebzadeh R, Rehakova S, Conlon TM, Win TS, Callaghan CJ, Goddard M, Bolton EM, Ruddle NH, Bradley JA, Pettigrew GJ. Blocking lymphotoxin signaling abrogates the development of ectopic lymphoid tissue within cardiac allografts and inhibits effector antibody responses. FASEB J. 2012;26:51–62. doi: 10.1096/fj.11-186973. [DOI] [PubMed] [Google Scholar]
  • 26.Callaghan CJ, Win TS, Motallebzadeh R, Conlon TM, Chhabra M, Harper I, Sivaganesh S, Bolton EM, Bradley JA, Brownlie RJ, Smith KG, Pettigrew GJ. Regulation of Allograft Survival by Inhibitory FcgammaRIIb Signaling. J Immunol. 2012;189:5694–5702. doi: 10.4049/jimmunol.1202084. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Montecalvo A, Shufesky WJ, Stolz DB, Sullivan MG, Wang Z, Divito SJ, Papworth GD, Watkins SC, Robbins PD, Larregina AT, Morelli AE. Exosomes as a short-range mechanism to spread alloantigen between dendritic cells during T cell allorecognition. J Immunol. 2008;180:3081–3090. doi: 10.4049/jimmunol.180.5.3081. [DOI] [PubMed] [Google Scholar]
  • 28.Viret C, Janeway CA., Jr. Functional and phenotypic evidence for presentation of E alpha 52-68 structurally related self-peptide(s) in I-E alpha-deficient mice. J Immunol. 2000;164:4627–4634. doi: 10.4049/jimmunol.164.9.4627. [DOI] [PubMed] [Google Scholar]
  • 29.Rosenberg AS, Mizuochi T, Singer A. Evidence for involvement of dual-function T cells in rejection of MHC class I disparate skin grafts. Assessment of MHC class I alloantigens as in vivo helper determinants. J Exp Med. 1988;168:33–45. doi: 10.1084/jem.168.1.33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Steele DJ, Laufer TM, Smiley ST, Ando Y, Grusby MJ, Glimcher LH, Auchincloss H., Jr. Two levels of help for B cell alloantibody production. J Exp Med. 1996;183:699–703. doi: 10.1084/jem.183.2.699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Conlon TM, Saeb-Parsy K, Cole JL, Motallebzadeh R, Qureshi MS, Rehakova S, Negus MC, Callaghan CJ, Bolton EM, Bradley JA, Pettigrew GJ. Germinal center alloantibody responses are mediated exclusively by indirect-pathway CD4 T follicular helper cells. J Immunol. 2012;188:2643–2652. doi: 10.4049/jimmunol.1102830. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Rechavi O, Goldstein I, Kloog Y. Intercellular exchange of proteins: the immune cell habit of sharing. FEBS Lett. 2009;583:1792–1799. doi: 10.1016/j.febslet.2009.03.014. [DOI] [PubMed] [Google Scholar]
  • 33.Joly E, Hudrisier D. What is trogocytosis and what is its purpose? Nat Immunol. 2003;4:815. doi: 10.1038/ni0903-815. [DOI] [PubMed] [Google Scholar]
  • 34.Osborne DG, Wetzel SA. Trogocytosis results in sustained intracellular signaling in CD4(+) T cells. J Immunol. 2012;189:4728–4739. doi: 10.4049/jimmunol.1201507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Brown R, Kabani K, Favaloro J, Yang S, Ho PJ, Gibson J, Fromm P, Suen H, Woodland N, Nassif N, Hart D, Joshua D. CD86+ or HLA-G+ can be transferred via trogocytosis from myeloma cells to T cells and are associated with poor prognosis. Blood. 2012;120:2055–2063. doi: 10.1182/blood-2012-03-416792. [DOI] [PubMed] [Google Scholar]
  • 36.Delamarre L, Pack M, Chang H, Mellman I, Trombetta ES. Differential lysosomal proteolysis in antigen-presenting cells determines antigen fate. Science. 2005;307:1630–1634. doi: 10.1126/science.1108003. [DOI] [PubMed] [Google Scholar]
  • 37.Bergtold A, Desai DD, Gavhane A, Clynes R. Cell surface recycling of internalized antigen permits dendritic cell priming of B cells. Immunity. 2005;23:503–514. doi: 10.1016/j.immuni.2005.09.013. [DOI] [PubMed] [Google Scholar]
  • 38.Conlon TM, Cole JL, Motallebzadeh R, Harper I, Callaghan CJ, Bolton EM, Bradley JA, Saeb-Parsy K, Pettigrew GJ. Unlinked memory helper responses promote long-lasting humoral alloimmunity. J Immunol. 2012;189:5703–5712. doi: 10.4049/jimmunol.1202257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Grandjean I, Duban L, Bonney EA, Corcuff E, Di Santo JP, Matzinger P, Lantz O. Are major histocompatibility complex molecules involved in the survival of naive CD4+ T cells? J Exp Med. 2003;198:1089–1102. doi: 10.1084/jem.20030963. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Kreisel D, Krupnick AS, Gelman AE, Engels FH, Popma SH, Krasinskas AM, Balsara KR, Szeto WY, Turka LA, Rosengard BR. Non-hematopoietic allograft cells directly activate CD8+ T cells and trigger acute rejection: an alternative mechanism of allorecognition. Nat Med. 2002;8:233–239. doi: 10.1038/nm0302-233. [DOI] [PubMed] [Google Scholar]
  • 41.Lakkis FG, Arakelov A, Konieczny BT, Inoue Y. Immunologic ‘ignorance’ of vascularized organ transplants in the absence of secondary lymphoid tissue. Nat Med. 2000;6:686–688. doi: 10.1038/76267. [DOI] [PubMed] [Google Scholar]

RESOURCES